Yu, Qingzhao; Medeiros, Kaelen L; Wu, Xiaocheng; Jensen, Roxanne E
2018-04-02
Mediation analysis allows the examination of effects of a third variable (mediator/confounder) in the causal pathway between an exposure and an outcome. The general multiple mediation analysis method (MMA), proposed by Yu et al., improves traditional methods (e.g., estimation of natural and controlled direct effects) to enable consideration of multiple mediators/confounders simultaneously and the use of linear and nonlinear predictive models for estimating mediation/confounding effects. Previous studies find that compared with non-Hispanic cancer survivors, Hispanic survivors are more likely to endure anxiety and depression after cancer diagnoses. In this paper, we applied MMA on MY-Health study to identify mediators/confounders and quantify the indirect effect of each identified mediator/confounder in explaining ethnic disparities in anxiety and depression among cancer survivors who enrolled in the study. We considered a number of socio-demographic variables, tumor characteristics, and treatment factors as potential mediators/confounders and found that most of the ethnic differences in anxiety or depression between Hispanic and non-Hispanic white cancer survivors were explained by younger diagnosis age, lower education level, lower proportions of employment, less likely of being born in the USA, less insurance, and less social support among Hispanic patients.
Methods to control for unmeasured confounding in pharmacoepidemiology: an overview.
Uddin, Md Jamal; Groenwold, Rolf H H; Ali, Mohammed Sanni; de Boer, Anthonius; Roes, Kit C B; Chowdhury, Muhammad A B; Klungel, Olaf H
2016-06-01
Background Unmeasured confounding is one of the principal problems in pharmacoepidemiologic studies. Several methods have been proposed to detect or control for unmeasured confounding either at the study design phase or the data analysis phase. Aim of the Review To provide an overview of commonly used methods to detect or control for unmeasured confounding and to provide recommendations for proper application in pharmacoepidemiology. Methods/Results Methods to control for unmeasured confounding in the design phase of a study are case only designs (e.g., case-crossover, case-time control, self-controlled case series) and the prior event rate ratio adjustment method. Methods that can be applied in the data analysis phase include, negative control method, perturbation variable method, instrumental variable methods, sensitivity analysis, and ecological analysis. A separate group of methods are those in which additional information on confounders is collected from a substudy. The latter group includes external adjustment, propensity score calibration, two-stage sampling, and multiple imputation. Conclusion As the performance and application of the methods to handle unmeasured confounding may differ across studies and across databases, we stress the importance of using both statistical evidence and substantial clinical knowledge for interpretation of the study results.
Instrumental variables as bias amplifiers with general outcome and confounding.
Ding, P; VanderWeele, T J; Robins, J M
2017-06-01
Drawing causal inference with observational studies is the central pillar of many disciplines. One sufficient condition for identifying the causal effect is that the treatment-outcome relationship is unconfounded conditional on the observed covariates. It is often believed that the more covariates we condition on, the more plausible this unconfoundedness assumption is. This belief has had a huge impact on practical causal inference, suggesting that we should adjust for all pretreatment covariates. However, when there is unmeasured confounding between the treatment and outcome, estimators adjusting for some pretreatment covariate might have greater bias than estimators without adjusting for this covariate. This kind of covariate is called a bias amplifier, and includes instrumental variables that are independent of the confounder, and affect the outcome only through the treatment. Previously, theoretical results for this phenomenon have been established only for linear models. We fill in this gap in the literature by providing a general theory, showing that this phenomenon happens under a wide class of models satisfying certain monotonicity assumptions. We further show that when the treatment follows an additive or multiplicative model conditional on the instrumental variable and the confounder, these monotonicity assumptions can be interpreted as the signs of the arrows of the causal diagrams.
ERIC Educational Resources Information Center
Games, Paul A.
1975-01-01
A brief introduction is presented on how multiple regression and linear model techniques can handle data analysis situations that most educators and psychologists think of as appropriate for analysis of variance. (Author/BJG)
Gong, Jing-Bo; Wang, Ya; Lui, Simon S Y; Cheung, Eric F C; Chan, Raymond C K
2017-11-01
Childhood trauma has been shown to be a robust risk factor for mental disorders, and may exacerbate schizotypal traits or contribute to autistic trait severity. However, little is known whether childhood trauma confounds the overlap between schizotypal traits and autistic traits. This study examined whether childhood trauma acts as a confounding variable in the overlap between autistic and schizotypal traits in a large non-clinical adult sample. A total of 2469 participants completed the Autism Spectrum Quotient (AQ), the Schizotypal Personality Questionnaire (SPQ), and the Childhood Trauma Questionnaire-Short Form. Correlation analysis showed that the majority of associations between AQ variables and SPQ variables were significant (p < 0.05). In the multiple regression models predicting scores on the AQ total, scores on the three SPQ subscales were significant predictors(Ps < 0.05). Scores on the Positive schizotypy and Negative schizotypy subscales were significant predictors in the multiple regression model predicting scores on the AQ Social Skill, AQ Attention Switching, AQ Attention to Detail, AQ Communication, and AQ Imagination subscales. The association between autistic and schizotypal traits could not be explained by shared variance in terms of exposure to childhood trauma. The findings point to important overlaps in the conceptualization of ASD and SSD, independent of childhood trauma. Copyright © 2017 Elsevier B.V. All rights reserved.
How Confounder Strength Can Affect Allocation of Resources in Electronic Health Records.
Lynch, Kristine E; Whitcomb, Brian W; DuVall, Scott L
2018-01-01
When electronic health record (EHR) data are used, multiple approaches may be available for measuring the same variable, introducing potentially confounding factors. While additional information may be gleaned and residual confounding reduced through resource-intensive assessment methods such as natural language processing (NLP), whether the added benefits offset the added cost of the additional resources is not straightforward. We evaluated the implications of misclassification of a confounder when using EHRs. Using a combination of simulations and real data surrounding hospital readmission, we considered smoking as a potential confounder. We compared ICD-9 diagnostic code assignment, which is an easily available measure but has the possibility of substantial misclassification of smoking status, with NLP, a method of determining smoking status that more expensive and time-consuming than ICD-9 code assignment but has less potential for misclassification. Classification of smoking status with NLP consistently produced less residual confounding than the use of ICD-9 codes; however, when minimal confounding was present, differences between the approaches were small. When considerable confounding is present, investing in a superior measurement tool becomes advantageous.
Toh, Sengwee; García Rodríguez, Luis A; Hernán, Miguel A
2012-05-01
Electronic healthcare databases are commonly used in comparative effectiveness and safety research of therapeutics. Many databases now include additional confounder information in a subset of the study population through data linkage or data collection. We described and compared existing methods for analyzing such datasets. Using data from The Health Improvement Network and the relation between non-steroidal anti-inflammatory drugs and upper gastrointestinal bleeding as an example, we employed several methods to handle partially missing confounder information. The crude odds ratio (OR) of upper gastrointestinal bleeding was 1.50 (95% confidence interval: 0.98, 2.28) among selective cyclo-oxygenase-2 inhibitor initiators (n = 43 569) compared with traditional non-steroidal anti-inflammatory drug initiators (n = 411 616). The OR dropped to 0.81 (0.52, 1.27) upon adjustment for confounders recorded for all patients. When further considering three additional variables missing in 22% of the study population (smoking, alcohol consumption, body mass index), the OR was between 0.80 and 0.83 for the missing-category approach, the missing-indicator approach, single imputation by the most common category, multiple imputation by chained equations, and propensity score calibration. The OR was 0.65 (0.39, 1.09) and 0.67 (0.38, 1.16) for the unweighted and the inverse probability weighted complete-case analysis, respectively. Existing methods for handling partially missing confounder data require different assumptions and may produce different results. The unweighted complete-case analysis, the missing-category/indicator approach, and single imputation require often unrealistic assumptions and should be avoided. In this study, differences across methods were not substantial, likely due to relatively low proportion of missingness and weak confounding effect by the three additional variables upon adjustment for other variables. Copyright © 2012 John Wiley & Sons, Ltd.
Confounder Detection in High-Dimensional Linear Models Using First Moments of Spectral Measures.
Liu, Furui; Chan, Laiwan
2018-06-12
In this letter, we study the confounder detection problem in the linear model, where the target variable [Formula: see text] is predicted using its [Formula: see text] potential causes [Formula: see text]. Based on an assumption of a rotation-invariant generating process of the model, recent study shows that the spectral measure induced by the regression coefficient vector with respect to the covariance matrix of [Formula: see text] is close to a uniform measure in purely causal cases, but it differs from a uniform measure characteristically in the presence of a scalar confounder. Analyzing spectral measure patterns could help to detect confounding. In this letter, we propose to use the first moment of the spectral measure for confounder detection. We calculate the first moment of the regression vector-induced spectral measure and compare it with the first moment of a uniform spectral measure, both defined with respect to the covariance matrix of [Formula: see text]. The two moments coincide in nonconfounding cases and differ from each other in the presence of confounding. This statistical causal-confounding asymmetry can be used for confounder detection. Without the need to analyze the spectral measure pattern, our method avoids the difficulty of metric choice and multiple parameter optimization. Experiments on synthetic and real data show the performance of this method.
Confounder summary scores when comparing the effects of multiple drug exposures.
Cadarette, Suzanne M; Gagne, Joshua J; Solomon, Daniel H; Katz, Jeffrey N; Stürmer, Til
2010-01-01
Little information is available comparing methods to adjust for confounding when considering multiple drug exposures. We compared three analytic strategies to control for confounding based on measured variables: conventional multivariable, exposure propensity score (EPS), and disease risk score (DRS). Each method was applied to a dataset (2000-2006) recently used to examine the comparative effectiveness of four drugs. The relative effectiveness of risedronate, nasal calcitonin, and raloxifene in preventing non-vertebral fracture, were each compared to alendronate. EPSs were derived both by using multinomial logistic regression (single model EPS) and by three separate logistic regression models (separate model EPS). DRSs were derived and event rates compared using Cox proportional hazard models. DRSs derived among the entire cohort (full cohort DRS) was compared to DRSs derived only among the referent alendronate (unexposed cohort DRS). Less than 8% deviation from the base estimate (conventional multivariable) was observed applying single model EPS, separate model EPS or full cohort DRS. Applying the unexposed cohort DRS when background risk for fracture differed between comparison drug exposure cohorts resulted in -7 to + 13% deviation from our base estimate. With sufficient numbers of exposed and outcomes, either conventional multivariable, EPS or full cohort DRS may be used to adjust for confounding to compare the effects of multiple drug exposures. However, our data also suggest that unexposed cohort DRS may be problematic when background risks differ between referent and exposed groups. Further empirical and simulation studies will help to clarify the generalizability of our findings.
de Vocht, Frank; Lee, Brian
2014-08-01
Studies have suggested that residential exposure to extremely low frequency (50 Hz) electromagnetic fields (ELF-EMF) from high voltage cables, overhead power lines, electricity substations or towers are associated with reduced birth weight and may be associated with adverse birth outcomes or even miscarriages. We previously conducted a study of 140,356 singleton live births between 2004 and 2008 in Northwest England, which suggested that close residential proximity (≤ 50 m) to ELF-EMF sources was associated with reduced average birth weight of 212 g (95%CI: -395 to -29 g) but not with statistically significant increased risks for other adverse perinatal outcomes. However, the cohort was limited by missing data for most potentially confounding variables including maternal smoking during pregnancy, which was only available for a small subgroup, while also residual confounding could not be excluded. This study, using the same cohort, was conducted to minimize the effects of these problems using multiple imputation to address missing data and propensity score matching to minimize residual confounding. Missing data were imputed using multiple imputation using chained equations to generate five datasets. For each dataset 115 exposed women (residing ≤ 50 m from a residential ELF-EMF source) were propensity score matched to 1150 unexposed women. After doubly robust confounder adjustment, close proximity to a residential ELF-EMF source remained associated with a reduction in birth weight of -116 g (95% confidence interval: -224:-7 g). No effect was found for proximity ≤ 100 m compared to women living further away. These results indicate that although the effect size was about half of the effect previously reported, close maternal residential proximity to sources of ELF-EMF remained associated with suboptimal fetal growth. Copyright © 2014 Elsevier Ltd. All rights reserved.
Understanding logistic regression analysis.
Sperandei, Sandro
2014-01-01
Logistic regression is used to obtain odds ratio in the presence of more than one explanatory variable. The procedure is quite similar to multiple linear regression, with the exception that the response variable is binomial. The result is the impact of each variable on the odds ratio of the observed event of interest. The main advantage is to avoid confounding effects by analyzing the association of all variables together. In this article, we explain the logistic regression procedure using examples to make it as simple as possible. After definition of the technique, the basic interpretation of the results is highlighted and then some special issues are discussed.
Hallager, Dennis Winge; Hansen, Lars Valentin; Dragsted, Casper Rokkjær; Peytz, Nina; Gehrchen, Martin; Dahl, Benny
2016-05-01
Cross-sectional analyses on a consecutive, prospective cohort. To evaluate the ability of the Scoliosis Research Society (SRS)-Schwab Adult Spinal Deformity Classification to group patients by widely used health-related quality-of-life (HRQOL) scores and examine possible confounding variables. The SRS-Schwab Adult Spinal Deformity Classification includes sagittal modifiers considered important for HRQOL and the clinical impact of the classification has been validated in patients from the International Spine Study Group database; however, equivocal results were reported for the Pelvic Tilt modifier and potential confounding variables were not evaluated. Between March 2013 and May 2014, all adult spinal deformity patients from our outpatient clinic with sufficient radiographs were prospectively enrolled. Analyses of HRQOL variance and post hoc analyses were performed for each SRS-Schwab modifier. Age, history of spine surgery, and aetiology of spinal deformity were considered potential confounders and their influence on the association between SRS-Schwab modifiers and aggregated Oswestry Disability Index (ODI) scores was evaluated with multivariate proportional odds regressions. P values were adjusted for multiple testing. Two hundred ninety-two of 460 eligible patients were included for analyses. The SRS-Schwab Classification significantly discriminated HRQOL scores between normal and abnormal sagittal modifier classifications. Individual grade comparisons showed equivocal results; however, Pelvic Tilt grade + versus + + did not discriminate patients according to any HRQOL score. All modifiers showed significant proportional odds for worse aggregated ODI scores with increasing grade levels and the effects were robust to confounding. However, age group and aetiology had individual significant effects. The SRS-Schwab sagittal modifiers reliably grouped patients graded 0 versus + / + + according to the most widely used HRQOL scores and the effects of increasing grade level on odds for worse ODI scores remained significant after adjusting for potential confounders. However, effects of age group and aetiology should not be neglected. 3.
ERIC Educational Resources Information Center
An, Brian P.; Loes, Chad N.; Trolian, Teniell L.
2017-01-01
Using longitudinal data from multiple institutions, we focused on the relation between binge drinking and academic performance. Binge drinking exerts a negative influence on grade point average, even after accounting for a host of precollege confounding variables. Furthermore, the number of times a student binge drinks in college is less…
Pregnant women with the sickle cell trait are not at increased risk for developing preeclampsia.
Stamilio, David M; Sehdev, Harish M; Macones, George A
2003-01-01
The primary objective of this study was to determine whether having the sickle cell trait is independently associated with preeclampsia. We performed a retrospective cohort study of 1998 pregnant patients who either did or did not have the sickle cell trait. All patients were screened for the sickle trait using the "Sickledex" test. Data on neonatal and maternal outcome, including preeclampsia, and potential confounding variables were abstracted from medical records. Unadjusted, stratified, and multiple logistic regression analyses were used to identify interactions, and confounding between multiple variables and the association between sickle cell trait and preeclampsia. With an anticipated 6.5% rate of preeclampsia, and alpha = 0.05, this cohort study has 80% power to detect a relative risk (RR) of 2.3 for preeclampsia. Univariate analysis revealed that the two cohorts were similar with regard to primiparity, maternal age, chronic diseases, birth weight, and gestational age at delivery, but the sickle cell trait cohort was more likely to have gestational diabetes and had a higher mean body mass index (BMI). In the univariate analysis, the sickle cell trait cohort was not at increased risk for preeclampsia [unadjusted RR = 0.5, 95% CI (0.2-1.6)]. After controlling for potential confounding variables with logistic regression analysis, sickle trait was not independently associated with preeclampsia [adjusted RR = 0.5, 95% CI (0.2- 1.6)]. In contrast to prior work, these data suggest that the sickle cell trait is not an independent risk factor for preeclampsia or postpartum complications. In fact, the data are more consistent with the sickle trait being protective for developing preeclampsia.
Mismeasurement and the resonance of strong confounders: uncorrelated errors.
Marshall, J R; Hastrup, J L
1996-05-15
Greenland first documented (Am J Epidemiol 1980; 112:564-9) that error in the measurement of a confounder could resonate--that it could bias estimates of other study variables, and that the bias could persist even with statistical adjustment for the confounder as measured. An important question is raised by this finding: can such bias be more than trivial within the bounds of realistic data configurations? The authors examine several situations involving dichotomous and continuous data in which a confounder and a null variable are measured with error, and they assess the extent of resultant bias in estimates of the effect of the null variable. They show that, with continuous variables, measurement error amounting to 40% of observed variance in the confounder could cause the observed impact of the null study variable to appear to alter risk by as much as 30%. Similarly, they show, with dichotomous independent variables, that 15% measurement error in the form of misclassification could lead the null study variable to appear to alter risk by as much as 50%. Such bias would result only from strong confounding. Measurement error would obscure the evidence that strong confounding is a likely problem. These results support the need for every epidemiologic inquiry to include evaluations of measurement error in each variable considered.
Association between prolonged breast-feeding and early childhood caries: a hierarchical approach.
Nunes, Ana Margarida Melo; Alves, Claudia Maria Coelho; Borba de Araújo, Fernando; Ortiz, Tânia Mara Lopes; Ribeiro, Marizélia Rodrigues Costa; Silva, Antônio Augusto Moura da; Ribeiro, Cecília Claudia Costa
2012-12-01
This study was conducted to investigate the association between prolonged breastfeeding and early childhood caries(ECC) with adjustment for important confounders, using hieraschical approach. This retrospective cohort study involved 260 low-income children (18-42 months). The number of decayed teeth was used as a measure of caries. Following a theoretical framework, the hierarchical model was built in a forward fashion, by adding the following levels in succession: level 1: age; level 2: social variables; level 3: health variables; level 4: behavioral variables; level 5: oral hygiene-related variables; level 6: oral hygiene quality measured by visible plaque; and level 7: contamination by mutans streptococci. Sequential forward multiple Poisson regression analysis was employed. Breast-feeding was not a risk factor for ECC after adjustment for some confounders (incidence density ratio, 1.15; 95% confidence interval, 0.84-1.59, P = 0.363). Prolonged breast-feeding was not a risk factor for ECC while age, high sucrose comption between main meals and the quality of oral higiene were associated with disease in children. © 2012 John Wiley & Sons A/S.
Smith, Eric G.
2015-01-01
Background: Nonrandomized studies typically cannot account for confounding from unmeasured factors. Method: A method is presented that exploits the recently-identified phenomenon of “confounding amplification” to produce, in principle, a quantitative estimate of total residual confounding resulting from both measured and unmeasured factors. Two nested propensity score models are constructed that differ only in the deliberate introduction of an additional variable(s) that substantially predicts treatment exposure. Residual confounding is then estimated by dividing the change in treatment effect estimate between models by the degree of confounding amplification estimated to occur, adjusting for any association between the additional variable(s) and outcome. Results: Several hypothetical examples are provided to illustrate how the method produces a quantitative estimate of residual confounding if the method’s requirements and assumptions are met. Previously published data is used to illustrate that, whether or not the method routinely provides precise quantitative estimates of residual confounding, the method appears to produce a valuable qualitative estimate of the likely direction and general size of residual confounding. Limitations: Uncertainties exist, including identifying the best approaches for: 1) predicting the amount of confounding amplification, 2) minimizing changes between the nested models unrelated to confounding amplification, 3) adjusting for the association of the introduced variable(s) with outcome, and 4) deriving confidence intervals for the method’s estimates (although bootstrapping is one plausible approach). Conclusions: To this author’s knowledge, it has not been previously suggested that the phenomenon of confounding amplification, if such amplification is as predictable as suggested by a recent simulation, provides a logical basis for estimating total residual confounding. The method's basic approach is straightforward. The method's routine usefulness, however, has not yet been established, nor has the method been fully validated. Rapid further investigation of this novel method is clearly indicated, given the potential value of its quantitative or qualitative output. PMID:25580226
Fischer, Susanne; Duncko, Roman; Hatch, Stephani L; Papadopoulos, Andrew; Goodwin, Laura; Frissa, Souci; Hotopf, Matthew; Cleare, Anthony J
2017-02-01
Hypothalamic-pituitary-adrenal (HPA) measures are crucial for research into stress and stress-related disorders. Most HPA measures fluctuate depending on diurnal rhythms and state confounders. Hair cortisol concentrations (HCC) are less susceptible to such fluctuations, but less is known about trait-like confounders. Using a community sample, we tested the relationship between HCC and a range of variables including demographic variables, hair treatment, and medication, as well as psychosocial variables, namely childhood trauma, critical life events, and depressive symptoms. Hair samples were collected from 144 individuals from the South East London Community Health (SELCoH) study. Childhood trauma, life events, and depressive symptoms were measured, together with age, sex, ethnicity, relationship status, educational attainment, employment status, occupational social class, hair washing frequency, hair treatments, season reflected in the hair sample, hazardous drinking, smoking, medication intake, and body mass index. Hair samples reflecting the past 3 months were collected and analysed using immunoassays. First, correlations (continuous variables) and simple linear regressions (dichotomous variables) were conducted to identify sociodemographic, hair-related, and lifestyle determinants of HCC. Next, multiple linear regressions were conducted to test the relationship between psychosocial variables and HCC when controlling for the identified confounders. Age (r=-0.17, p=0.050), White British ethnicity (β=-0.19, p=0.023), heat-based treatments (β=-0.22, p=0.010), and winter season (β=-0.18, p=0.024) were associated with lower HCC, whereas summer season (β=0.24, p=0.024), painkillers (β=0.25, p=0.003), anxiolytics/antidepressants (β=0.21, p=0.014), and hormonal contraceptives (β=0.27, p=0.006) were associated with higher HCC. Controlling for these confounders, physical neglect during childhood (β=-0.17, p=0.057), war-related experiences (β=0.20, p=0.027), separation (β=0.18, p=0.054), and being the victim of a serious crime (β=-0.17, p=0.062) were linked with altered HCC. Our findings suggest that variation in HCC occurs according to sociodemographic, hair-related, and lifestyle variables, and that certain associations between stress and altered HCC can only be revealed when accounting for these confounders. Copyright © 2016 Elsevier Ltd. All rights reserved.
Grover, Sandeep; Del Greco M, Fabiola; Stein, Catherine M; Ziegler, Andreas
2017-01-01
Confounding and reverse causality have prevented us from drawing meaningful clinical interpretation even in well-powered observational studies. Confounding may be attributed to our inability to randomize the exposure variable in observational studies. Mendelian randomization (MR) is one approach to overcome confounding. It utilizes one or more genetic polymorphisms as a proxy for the exposure variable of interest. Polymorphisms are randomly distributed in a population, they are static throughout an individual's lifetime, and may thus help in inferring directionality in exposure-outcome associations. Genome-wide association studies (GWAS) or meta-analyses of GWAS are characterized by large sample sizes and the availability of many single nucleotide polymorphisms (SNPs), making GWAS-based MR an attractive approach. GWAS-based MR comes with specific challenges, including multiple causality. Despite shortcomings, it still remains one of the most powerful techniques for inferring causality.With MR still an evolving concept with complex statistical challenges, the literature is relatively scarce in terms of providing working examples incorporating real datasets. In this chapter, we provide a step-by-step guide for causal inference based on the principles of MR with a real dataset using both individual and summary data from unrelated individuals. We suggest best possible practices and give recommendations based on the current literature.
Prenatal Maternal Stress Associated with ADHD and Autistic Traits in early Childhood
Ronald, Angelica; Pennell, Craig E.; Whitehouse, Andrew J. O.
2010-01-01
Research suggests that offspring of mothers who experience high levels of stress during pregnancy are more likely to have problems in neurobehavioral development. There is preliminary evidence that prenatal maternal stress (PNMS) is a risk factor for both autism and attention deficit hyperactivity disorder (ADHD), however most studies do not control for confounding factors and no study has investigated PNMS as a risk factor for behaviors characteristic of these disorders in early childhood. A population cohort of 2900 pregnant women were recruited before their 18th week of pregnancy and investigated prospectively. Maternal experience of stressful life events was assessed during pregnancy. When offspring were age 2 years, mothers completed the child behavior checklist. Multiple regression showed that maternal stressful events during pregnancy significantly predicted ADHD behaviors in offspring, after controlling for autistic traits and other confounding variables, in both males (p = 0.03) and females (p = 0.01). Similarly, stressful events during pregnancy significantly predicted autistic traits in the offspring after controlling for ADHD behaviors and confounding variables, in males only (p = 0.04). In conclusion, this study suggests that PNMS, in the form of typical stressful life events such as divorce or a residential move, show a small but significant association with both autistic traits and ADHD behaviors independently, in offspring at age 2 years, after controlling for multiple antenatal, obstetric, postnatal, and sociodemographic covariates. This finding supports future research using epigenetic, cross-fostering, and gene–environment interaction designs to identify the causal processes underlying this association. PMID:21833278
Lasserson, Daniel S; Scherpbier de Haan, Nynke; de Grauw, Wim; van der Wel, Mark; Wetzels, Jack F; O'Callaghan, Christopher A
2016-06-10
To determine the relationship between renal function and visit-to-visit blood pressure (BP) variability in a cohort of primary care patients. Retrospective cohort study from routinely collected healthcare data. Primary care in Nijmegen, the Netherlands, from 2007 to 2012. 19 175 patients who had a measure of renal function, and 7 separate visits with BP readings in the primary care record. Visit-to-visit variability in systolic BP, calculated from the first 7 office measurements, including SD, successive variation, absolute real variation and metrics of variability shown to be independent of mean. Multiple linear regression was used to analyse the influence of estimated glomerular filtration rate (eGFR) on BP variability measures with adjustment for age, sex, diabetes, mean BP, proteinuria, cardiovascular disease, time interval between measures and antihypertensive use. In the patient cohort, 57% were women, mean (SD) age was 65.5 (12.3) years, mean (SD) eGFR was 75.6 (18.0) mL/min/1.73m(2) and SD systolic BP 148.3 (21.4) mm Hg. All BP variability measures were negatively correlated with eGFR and positively correlated with age. However, multiple linear regressions demonstrated consistent, small magnitude negative relationships between eGFR and all measures of BP variability adjusting for confounding variables. Worsening renal function is associated with small increases in measures of visit-to-visit BP variability after adjustment for confounding factors. This is seen across the spectrum of renal function in the population, and provides a mechanism whereby chronic kidney disease may raise the risk of cardiovascular events. Published by the BMJ Publishing Group Limited. For permission to use (where not already granted under a licence) please go to http://www.bmj.com/company/products-services/rights-and-licensing/
Rudolph, Kara E.; Sánchez, Brisa N.; Stuart, Elizabeth A.; Greenberg, Benjamin; Fujishiro, Kaori; Wand, Gary S.; Shrager, Sandi; Seeman, Teresa; Diez Roux, Ana V.; Golden, Sherita H.
2016-01-01
Evidence of the link between job strain and cortisol levels has been inconsistent. This could be due to failure to account for cortisol variability leading to underestimated standard errors. Our objective was to model the relationship between job strain and the whole cortisol curve, accounting for sources of cortisol variability. Our functional mixed-model approach incorporated all available data—18 samples over 3 days—and uncertainty in estimated relationships. We used employed participants from the Multi-Ethnic Study of Atherosclerosis Stress I Study and data collected between 2002 and 2006. We used propensity score matching on an extensive set of variables to control for sources of confounding. We found that job strain was associated with lower salivary cortisol levels and lower total area under the curve. We found no relationship between job strain and the cortisol awakening response. Our findings differed from those of several previous studies. It is plausible that our results were unique to middle- to older-aged racially, ethnically, and occupationally diverse adults and were therefore not inconsistent with previous research among younger, mostly white samples. However, it is also plausible that previous findings were influenced by residual confounding and failure to propagate uncertainty (i.e., account for the multiple sources of variability) in estimating cortisol features. PMID:26905339
Dai, James Y.; Chan, Kwun Chuen Gary; Hsu, Li
2014-01-01
Instrumental variable regression is one way to overcome unmeasured confounding and estimate causal effect in observational studies. Built on structural mean models, there has been considerale work recently developed for consistent estimation of causal relative risk and causal odds ratio. Such models can sometimes suffer from identification issues for weak instruments. This hampered the applicability of Mendelian randomization analysis in genetic epidemiology. When there are multiple genetic variants available as instrumental variables, and causal effect is defined in a generalized linear model in the presence of unmeasured confounders, we propose to test concordance between instrumental variable effects on the intermediate exposure and instrumental variable effects on the disease outcome, as a means to test the causal effect. We show that a class of generalized least squares estimators provide valid and consistent tests of causality. For causal effect of a continuous exposure on a dichotomous outcome in logistic models, the proposed estimators are shown to be asymptotically conservative. When the disease outcome is rare, such estimators are consistent due to the log-linear approximation of the logistic function. Optimality of such estimators relative to the well-known two-stage least squares estimator and the double-logistic structural mean model is further discussed. PMID:24863158
Bias, Confounding, and Interaction: Lions and Tigers, and Bears, Oh My!
Vetter, Thomas R; Mascha, Edward J
2017-09-01
Epidemiologists seek to make a valid inference about the causal effect between an exposure and a disease in a specific population, using representative sample data from a specific population. Clinical researchers likewise seek to make a valid inference about the association between an intervention and outcome(s) in a specific population, based upon their randomly collected, representative sample data. Both do so by using the available data about the sample variable to make a valid estimate about its corresponding or underlying, but unknown population parameter. Random error in an experiment can be due to the natural, periodic fluctuation or variation in the accuracy or precision of virtually any data sampling technique or health measurement tool or scale. In a clinical research study, random error can be due to not only innate human variability but also purely chance. Systematic error in an experiment arises from an innate flaw in the data sampling technique or measurement instrument. In the clinical research setting, systematic error is more commonly referred to as systematic bias. The most commonly encountered types of bias in anesthesia, perioperative, critical care, and pain medicine research include recall bias, observational bias (Hawthorne effect), attrition bias, misclassification or informational bias, and selection bias. A confounding variable is a factor associated with both the exposure of interest and the outcome of interest. A confounding variable (confounding factor or confounder) is a variable that correlates (positively or negatively) with both the exposure and outcome. Confounding is typically not an issue in a randomized trial because the randomized groups are sufficiently balanced on all potential confounding variables, both observed and nonobserved. However, confounding can be a major problem with any observational (nonrandomized) study. Ignoring confounding in an observational study will often result in a "distorted" or incorrect estimate of the association or treatment effect. Interaction among variables, also known as effect modification, exists when the effect of 1 explanatory variable on the outcome depends on the particular level or value of another explanatory variable. Bias and confounding are common potential explanations for statistically significant associations between exposure and outcome when the true relationship is noncausal. Understanding interactions is vital to proper interpretation of treatment effects. These complex concepts should be consistently and appropriately considered whenever one is not only designing but also analyzing and interpreting data from a randomized trial or observational study.
Schroeder, Krista; Jia, Haomiao; Smaldone, Arlene
Propensity score (PS) methods are increasingly being employed by researchers to reduce bias arising from confounder imbalance when using observational data to examine intervention effects. The purpose of this study was to examine PS theory and methodology and compare application of three PS methods (matching, stratification, weighting) to determine which best improves confounder balance. Baseline characteristics of a sample of 20,518 school-aged children with severe obesity (of whom 1,054 received an obesity intervention) were assessed prior to PS application. Three PS methods were then applied to the data to determine which showed the greatest improvement in confounder balance between the intervention and control group. The effect of each PS method on the outcome variable-body mass index percentile change at one year-was also examined. SAS 9.4 and Comprehensive Meta-analysis statistical software were used for analyses. Prior to PS adjustment, the intervention and control groups differed significantly on seven of 11 potential confounders. PS matching removed all differences. PS stratification and weighting both removed one difference but created two new differences. Sensitivity analyses did not change these results. Body mass index percentile at 1 year decreased in both groups. The size of the decrease was smaller in the intervention group, and the estimate of the decrease varied by PS method. Selection of a PS method should be guided by insight from statistical theory and simulation experiments, in addition to observed improvement in confounder balance. For this data set, PS matching worked best to correct confounder imbalance. Because each method varied in correcting confounder imbalance, we recommend that multiple PS methods be compared for ability to improve confounder balance before implementation in evaluating treatment effects in observational data.
A review of covariate selection for non-experimental comparative effectiveness research.
Sauer, Brian C; Brookhart, M Alan; Roy, Jason; VanderWeele, Tyler
2013-11-01
This paper addresses strategies for selecting variables for adjustment in non-experimental comparative effectiveness research and uses causal graphs to illustrate the causal network that relates treatment to outcome. Variables in the causal network take on multiple structural forms. Adjustment for a common cause pathway between treatment and outcome can remove confounding, whereas adjustment for other structural types may increase bias. For this reason, variable selection would ideally be based on an understanding of the causal network; however, the true causal network is rarely known. Therefore, we describe more practical variable selection approaches based on background knowledge when the causal structure is only partially known. These approaches include adjustment for all observed pretreatment variables thought to have some connection to the outcome, all known risk factors for the outcome, and all direct causes of the treatment or the outcome. Empirical approaches, such as forward and backward selection and automatic high-dimensional proxy adjustment, are also discussed. As there is a continuum between knowing and not knowing the causal, structural relations of variables, we recommend addressing variable selection in a practical way that involves a combination of background knowledge and empirical selection and that uses high-dimensional approaches. This empirical approach can be used to select from a set of a priori variables based on the researcher's knowledge to be included in the final analysis or to identify additional variables for consideration. This more limited use of empirically derived variables may reduce confounding while simultaneously reducing the risk of including variables that may increase bias. Copyright © 2013 John Wiley & Sons, Ltd.
A Review of Covariate Selection for Nonexperimental Comparative Effectiveness Research
Sauer, Brian C.; Brookhart, Alan; Roy, Jason; Vanderweele, Tyler
2014-01-01
This paper addresses strategies for selecting variables for adjustment in non-experimental comparative effectiveness research (CER), and uses causal graphs to illustrate the causal network that relates treatment to outcome. Variables in the causal network take on multiple structural forms. Adjustment for on a common cause pathway between treatment and outcome can remove confounding, while adjustment for other structural types may increase bias. For this reason variable selection would ideally be based on an understanding of the causal network; however, the true causal network is rarely know. Therefore, we describe more practical variable selection approaches based on background knowledge when the causal structure is only partially known. These approaches include adjustment for all observed pretreatment variables thought to have some connection to the outcome, all known risk factors for the outcome, and all direct causes of the treatment or the outcome. Empirical approaches, such as forward and backward selection and automatic high-dimensional proxy adjustment, are also discussed. As there is a continuum between knowing and not knowing the causal, structural relations of variables, we recommend addressing variable selection in a practical way that involves a combination of background knowledge and empirical selection and that uses the high-dimensional approaches. This empirical approach can be used to select from a set of a priori variables based on the researcher’s knowledge to be included in the final analysis or to identify additional variables for consideration. This more limited use of empirically-derived variables may reduce confounding while simultaneously reducing the risk of including variables that may increase bias. PMID:24006330
La Gamba, Fabiola; Corrao, Giovanni; Romio, Silvana; Sturkenboom, Miriam; Trifirò, Gianluca; Schink, Tania; de Ridder, Maria
2017-10-01
Clustering of patients in databases is usually ignored in one-stage meta-analysis of multi-database studies using matched case-control data. The aim of this study was to compare bias and efficiency of such a one-stage meta-analysis with a two-stage meta-analysis. First, we compared the approaches by generating matched case-control data under 5 simulated scenarios, built by varying: (1) the exposure-outcome association; (2) its variability among databases; (3) the confounding strength of one covariate on this association; (4) its variability; and (5) the (heterogeneous) confounding strength of two covariates. Second, we made the same comparison using empirical data from the ARITMO project, a multiple database study investigating the risk of ventricular arrhythmia following the use of medications with arrhythmogenic potential. In our study, we specifically investigated the effect of current use of promethazine. Bias increased for one-stage meta-analysis with increasing (1) between-database variance of exposure effect and (2) heterogeneous confounding generated by two covariates. The efficiency of one-stage meta-analysis was slightly lower than that of two-stage meta-analysis for the majority of investigated scenarios. Based on ARITMO data, there were no evident differences between one-stage (OR = 1.50, CI = [1.08; 2.08]) and two-stage (OR = 1.55, CI = [1.12; 2.16]) approaches. When the effect of interest is heterogeneous, a one-stage meta-analysis ignoring clustering gives biased estimates. Two-stage meta-analysis generates estimates at least as accurate and precise as one-stage meta-analysis. However, in a study using small databases and rare exposures and/or outcomes, a correct one-stage meta-analysis becomes essential. Copyright © 2017 John Wiley & Sons, Ltd.
Tang, Zheng-Zheng; Chen, Guanhua; Alekseyenko, Alexander V
2016-09-01
Recent advances in sequencing technology have made it possible to obtain high-throughput data on the composition of microbial communities and to study the effects of dysbiosis on the human host. Analysis of pairwise intersample distances quantifies the association between the microbiome diversity and covariates of interest (e.g. environmental factors, clinical outcomes, treatment groups). In the design of these analyses, multiple choices for distance metrics are available. Most distance-based methods, however, use a single distance and are underpowered if the distance is poorly chosen. In addition, distance-based tests cannot flexibly handle confounding variables, which can result in excessive false-positive findings. We derive presence-weighted UniFrac to complement the existing UniFrac distances for more powerful detection of the variation in species richness. We develop PERMANOVA-S, a new distance-based method that tests the association of microbiome composition with any covariates of interest. PERMANOVA-S improves the commonly-used Permutation Multivariate Analysis of Variance (PERMANOVA) test by allowing flexible confounder adjustments and ensembling multiple distances. We conducted extensive simulation studies to evaluate the performance of different distances under various patterns of association. Our simulation studies demonstrate that the power of the test relies on how well the selected distance captures the nature of the association. The PERMANOVA-S unified test combines multiple distances and achieves good power regardless of the patterns of the underlying association. We demonstrate the usefulness of our approach by reanalyzing several real microbiome datasets. miProfile software is freely available at https://medschool.vanderbilt.edu/tang-lab/software/miProfile z.tang@vanderbilt.edu or g.chen@vanderbilt.edu Supplementary data are available at Bioinformatics online. © The Author 2016. Published by Oxford University Press.
Tang, Zheng-Zheng; Chen, Guanhua; Alekseyenko, Alexander V.
2016-01-01
Motivation: Recent advances in sequencing technology have made it possible to obtain high-throughput data on the composition of microbial communities and to study the effects of dysbiosis on the human host. Analysis of pairwise intersample distances quantifies the association between the microbiome diversity and covariates of interest (e.g. environmental factors, clinical outcomes, treatment groups). In the design of these analyses, multiple choices for distance metrics are available. Most distance-based methods, however, use a single distance and are underpowered if the distance is poorly chosen. In addition, distance-based tests cannot flexibly handle confounding variables, which can result in excessive false-positive findings. Results: We derive presence-weighted UniFrac to complement the existing UniFrac distances for more powerful detection of the variation in species richness. We develop PERMANOVA-S, a new distance-based method that tests the association of microbiome composition with any covariates of interest. PERMANOVA-S improves the commonly-used Permutation Multivariate Analysis of Variance (PERMANOVA) test by allowing flexible confounder adjustments and ensembling multiple distances. We conducted extensive simulation studies to evaluate the performance of different distances under various patterns of association. Our simulation studies demonstrate that the power of the test relies on how well the selected distance captures the nature of the association. The PERMANOVA-S unified test combines multiple distances and achieves good power regardless of the patterns of the underlying association. We demonstrate the usefulness of our approach by reanalyzing several real microbiome datasets. Availability and Implementation: miProfile software is freely available at https://medschool.vanderbilt.edu/tang-lab/software/miProfile. Contact: z.tang@vanderbilt.edu or g.chen@vanderbilt.edu Supplementary information: Supplementary data are available at Bioinformatics online. PMID:27197815
DOE Office of Scientific and Technical Information (OSTI.GOV)
Mundt, D.J.; van Wijngaarden, E.; Mundt, K.A.
2007-07-01
We evaluated the likelihood and extent to which the observed increased risk of lung cancer may be due to confounding (a mixing of effects of multiple exposures) by co-exposure to other potential carcinogens present in roofing or to lifestyle variables. We conducted a review of the epidemiological and industrial hygiene literature of asphalt-exposed workers. Peer-reviewed epidemiological studies of asphalt fumes, related occupational exposures, and confounding factors were identified from MEDLINE (1966 early 2004). Industrial hygiene studies of asphalt workers were identified through MEDLINE, publicly available government documents, and asphalt industry documents. Using well established statistical methods, we quantified the extentmore » to which lung cancer relative risk estimates among roofers reflect confounding from other exposures, using different prevalence and risk scenarios. The relative risk of lung cancer varied from 1.2 to 5.0 in 13 epidemiological studies of roofers; most studies reported a relative risk between 1.2 and 1.4. Smoking, asbestos and coal tar were the most likely confounders, but the prevalence of these factors varied over time. The results of the study indicate that much of the observed risk reported in epidemiological studies of cancer among roofers is well within the range of what may have resulted from confounding by reasonable and expected levels of smoking, asbestos or coal tar. This may be particularly true for those studies that did not adjust for these confounders and where the exposure was defined as employment in the roofing industry. In addition to poorly defined asphalt exposure, uncontrolled confounding cannot reliably be ruled out in studies of lung cancer among asphalt-exposed roofers. Therefore, it is not possible to conclude whether roofers are at increased risk of lung cancer due to asphalt exposure.« less
Jou, Jerwen
2014-10-01
Subjects performed Sternberg-type memory recognition tasks (Sternberg paradigm) in four experiments. Category-instance names were used as learning and testing materials. Sternberg's original experiments demonstrated a linear relation between reaction time (RT) and memory-set size (MSS). A few later studies found no relation, and other studies found a nonlinear relation (logarithmic) between the two variables. These deviations were used as evidence undermining Sternberg's serial scan theory. This study identified two confounding variables in the fixed-set procedure of the paradigm (where multiple probes are presented at test for a learned memory set) that could generate a MSS RT function that was either flat or logarithmic rather than linearly increasing. These two confounding variables were task-switching cost and repetition priming. The former factor worked against smaller memory sets and in favour of larger sets whereas the latter factor worked in the opposite way. Results demonstrated that a null or a logarithmic RT-to-MSS relation could be the artefact of the combined effects of these two variables. The Sternberg paradigm has been used widely in memory research, and a thorough understanding of the subtle methodological pitfalls is crucial. It is suggested that a varied-set procedure (where only one probe is presented at test for a learned memory set) is a more contamination-free procedure for measuring the MSS effects, and that if a fixed-set procedure is used, it is worthwhile examining the RT function of the very first trials across the MSSs, which are presumably relatively free of contamination by the subsequent trials.
Barnwell-Ménard, Jean-Louis; Li, Qing; Cohen, Alan A
2015-03-15
The loss of signal associated with categorizing a continuous variable is well known, and previous studies have demonstrated that this can lead to an inflation of Type-I error when the categorized variable is a confounder in a regression analysis estimating the effect of an exposure on an outcome. However, it is not known how the Type-I error may vary under different circumstances, including logistic versus linear regression, different distributions of the confounder, and different categorization methods. Here, we analytically quantified the effect of categorization and then performed a series of 9600 Monte Carlo simulations to estimate the Type-I error inflation associated with categorization of a confounder under different regression scenarios. We show that Type-I error is unacceptably high (>10% in most scenarios and often 100%). The only exception was when the variable categorized was a continuous mixture proxy for a genuinely dichotomous latent variable, where both the continuous proxy and the categorized variable are error-ridden proxies for the dichotomous latent variable. As expected, error inflation was also higher with larger sample size, fewer categories, and stronger associations between the confounder and the exposure or outcome. We provide online tools that can help researchers estimate the potential error inflation and understand how serious a problem this is. Copyright © 2014 John Wiley & Sons, Ltd.
Tong, I S; Lu, Y
2001-01-01
To explore the best approach to identify and adjust for confounders in epidemiologic practice. In the Port Pirie cohort study, the selection of covariates was based on both a priori and an empirical consideration. In an assessment of the relationship between exposure to environmental lead and child development, change-in-estimate (CE) and significance testing (ST) criteria were compared in identifying potential confounders. The Pearson correlation coefficients were used to evaluate the potential for collinearity between pairs of major quantitative covariates. In multivariate analyses, the effects of confounding factors were assessed with multiple linear regression models. The nature and number of covariates selected varied with different confounder selection criteria and different cutoffs. Four covariates (i.e., quality of home environment, socioeconomic status (SES), maternal intelligence, and parental smoking behaviour) met the conventional CE criterion (> or =10%), whereas 14 variables met the ST criterion (p < or = 0.25). However, the magnitude of the relationship between blood lead concentration and children's IQ differed slightly after adjustment for confounding, using either the CE (partial regression coefficient: -4.4; 95% confidence interval (CI): -0.5 to -8.3) or ST criterion (-4.3; 95% CI: -0.2 to -8.4). Identification and selection of confounding factors need to be viewed cautiously in epidemiologic studies. Either the CE (e.g., > or = 10%) or ST (e.g., p < or = 0.25) criterion may be implemented in identification of a potential confounder if a study sample is sufficiently large, and both the methods are subject to arbitrariness of selecting a cut-off point. In this study, the CE criterion (i.e., > or = 10%) appears to be more stringent than the ST method (i.e., p < or = 0.25) in the identification of confounders. However, the ST rule cannot be used to determine the trueness of confounding because it cannot reflect the causal relationship between the confounder and outcome. This study shows the complexities one can expect to encounter in the identification of and adjustment for confounders.
Ananth, Cande V; Schisterman, Enrique F
2017-08-01
Prospective and retrospective cohorts and case-control studies are some of the most important study designs in epidemiology because, under certain assumptions, they can mimic a randomized trial when done well. These assumptions include, but are not limited to, properly accounting for 2 important sources of bias: confounding and selection bias. While not adjusting the causal association for an intermediate variable will yield an unbiased estimate of the exposure-outcome's total causal effect, it is often that obstetricians will want to adjust for an intermediate variable to assess if the intermediate is the underlying driver of the association. Such a practice must be weighed in light of the underlying research question and whether such an adjustment is necessary should be carefully considered. Gestational age is, by far, the most commonly encountered variable in obstetrics that is often mislabeled as a confounder when, in fact, it may be an intermediate. If, indeed, gestational age is an intermediate but if mistakenly labeled as a confounding variable and consequently adjusted in an analysis, the conclusions can be unexpected. The implications of this overadjustment of an intermediate as though it were a confounder can render an otherwise persuasive study downright meaningless. This commentary provides an exposition of confounding bias, collider stratification, and selection biases, with applications in obstetrics and perinatal epidemiology. Copyright © 2017 Elsevier Inc. All rights reserved.
A long term study of pulmonary function among US refractory ceramic fibre workers
LeMasters, Grace K; Hilbert, Timothy J; Levin, Linda S; Rice, Carol H; Borton, Eric K; Lockey, James E
2010-01-01
Background Cross-sectional studies have shown declines in lung function among refractory ceramic fibre (RCF) workers with increasing fibre exposure. This study followed current and former workers (n=1396) for up to 17 years and collected 5243 pulmonary function tests. Methods Cumulative fibre exposure and production years were categorised into exposure levels at five manufacturing locations. Conventional longitudinal models did not adequately partition age-related changes from other time-dependent variables. Therefore, a restricted cubic spline model was developed to account for the non-linear decline with age. Results Cumulative fibre >60 fibre-months/cc showed a significant loss in lung function at the first test. When results were examined longitudinally, cumulative exposure was confounded with age as workers with the highest cumulative exposure were generally older. A longitudinal model adjusted by age groups was implemented to control for this confounding. No consistent longitudinal loss in lung function was observed with RCF exposure. Smoking, initial weight and weight increase were significant factors. Conclusion No consistent decline was observed longitudinally with exposure to RCF, although cross-sectional and longitudinal findings were discordant. Confounding and accelerated lung function declines with ageing and the correlation of multiple time-dependent variables should be considered in order to minimise error and maximise precision. An innovative statistical methodology for these types of data is described. PMID:20798015
Burgess, Stephen; Daniel, Rhian M; Butterworth, Adam S; Thompson, Simon G
2015-01-01
Background: Mendelian randomization uses genetic variants, assumed to be instrumental variables for a particular exposure, to estimate the causal effect of that exposure on an outcome. If the instrumental variable criteria are satisfied, the resulting estimator is consistent even in the presence of unmeasured confounding and reverse causation. Methods: We extend the Mendelian randomization paradigm to investigate more complex networks of relationships between variables, in particular where some of the effect of an exposure on the outcome may operate through an intermediate variable (a mediator). If instrumental variables for the exposure and mediator are available, direct and indirect effects of the exposure on the outcome can be estimated, for example using either a regression-based method or structural equation models. The direction of effect between the exposure and a possible mediator can also be assessed. Methods are illustrated in an applied example considering causal relationships between body mass index, C-reactive protein and uric acid. Results: These estimators are consistent in the presence of unmeasured confounding if, in addition to the instrumental variable assumptions, the effects of both the exposure on the mediator and the mediator on the outcome are homogeneous across individuals and linear without interactions. Nevertheless, a simulation study demonstrates that even considerable heterogeneity in these effects does not lead to bias in the estimates. Conclusions: These methods can be used to estimate direct and indirect causal effects in a mediation setting, and have potential for the investigation of more complex networks between multiple interrelated exposures and disease outcomes. PMID:25150977
Lutz, Sharon M; Thwing, Annie; Schmiege, Sarah; Kroehl, Miranda; Baker, Christopher D; Starling, Anne P; Hokanson, John E; Ghosh, Debashis
2017-07-19
In mediation analysis if unmeasured confounding is present, the estimates for the direct and mediated effects may be over or under estimated. Most methods for the sensitivity analysis of unmeasured confounding in mediation have focused on the mediator-outcome relationship. The Umediation R package enables the user to simulate unmeasured confounding of the exposure-mediator, exposure-outcome, and mediator-outcome relationships in order to see how the results of the mediation analysis would change in the presence of unmeasured confounding. We apply the Umediation package to the Genetic Epidemiology of Chronic Obstructive Pulmonary Disease (COPDGene) study to examine the role of unmeasured confounding due to population stratification on the effect of a single nucleotide polymorphism (SNP) in the CHRNA5/3/B4 locus on pulmonary function decline as mediated by cigarette smoking. Umediation is a flexible R package that examines the role of unmeasured confounding in mediation analysis allowing for normally distributed or Bernoulli distributed exposures, outcomes, mediators, measured confounders, and unmeasured confounders. Umediation also accommodates multiple measured confounders, multiple unmeasured confounders, and allows for a mediator-exposure interaction on the outcome. Umediation is available as an R package at https://github.com/SharonLutz/Umediation A tutorial on how to install and use the Umediation package is available in the Additional file 1.
Cacace, Anthony T; McFarland, Dennis J
2013-01-01
Tests of auditory perception, such as those used in the assessment of central auditory processing disorders ([C]APDs), represent a domain in audiological assessment where measurement of this theoretical construct is often confounded by nonauditory abilities due to methodological shortcomings. These confounds include the effects of cognitive variables such as memory and attention and suboptimal testing paradigms, including the use of verbal reproduction as a form of response selection. We argue that these factors need to be controlled more carefully and/or modified so that their impact on tests of auditory and visual perception is only minimal. To advocate for a stronger theoretical framework than currently exists and to suggest better methodological strategies to improve assessment of auditory processing disorders (APDs). Emphasis is placed on adaptive forced-choice psychophysical methods and the use of matched tasks in multiple sensory modalities to achieve these goals. Together, this approach has potential to improve the construct validity of the diagnosis, enhance and develop theory, and evolve into a preferred method of testing. Examination of methods commonly used in studies of APDs. Where possible, currently used methodology is compared to contemporary psychophysical methods that emphasize computer-controlled forced-choice paradigms. In many cases, the procedures used in studies of APD introduce confounding factors that could be minimized if computer-controlled forced-choice psychophysical methods were utilized. Ambiguities of interpretation, indeterminate diagnoses, and unwanted confounds can be avoided by minimizing memory and attentional demands on the input end and precluding the use of response-selection strategies that use complex motor processes on the output end. Advocated are the use of computer-controlled forced-choice psychophysical paradigms in combination with matched tasks in multiple sensory modalities to enhance the prospect of obtaining a valid diagnosis. American Academy of Audiology.
Rudolph, Kara E; Sánchez, Brisa N; Stuart, Elizabeth A; Greenberg, Benjamin; Fujishiro, Kaori; Wand, Gary S; Shrager, Sandi; Seeman, Teresa; Diez Roux, Ana V; Golden, Sherita H
2016-03-01
Evidence of the link between job strain and cortisol levels has been inconsistent. This could be due to failure to account for cortisol variability leading to underestimated standard errors. Our objective was to model the relationship between job strain and the whole cortisol curve, accounting for sources of cortisol variability. Our functional mixed-model approach incorporated all available data-18 samples over 3 days-and uncertainty in estimated relationships. We used employed participants from the Multi-Ethnic Study of Atherosclerosis Stress I Study and data collected between 2002 and 2006. We used propensity score matching on an extensive set of variables to control for sources of confounding. We found that job strain was associated with lower salivary cortisol levels and lower total area under the curve. We found no relationship between job strain and the cortisol awakening response. Our findings differed from those of several previous studies. It is plausible that our results were unique to middle- to older-aged racially, ethnically, and occupationally diverse adults and were therefore not inconsistent with previous research among younger, mostly white samples. However, it is also plausible that previous findings were influenced by residual confounding and failure to propagate uncertainty (i.e., account for the multiple sources of variability) in estimating cortisol features. © The Author 2016. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Dipnall, Joanna F.
2016-01-01
Background Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study. Methods The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009–2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators. Results After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001). Conclusion The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin. PMID:26848571
Dipnall, Joanna F; Pasco, Julie A; Berk, Michael; Williams, Lana J; Dodd, Seetal; Jacka, Felice N; Meyer, Denny
2016-01-01
Atheoretical large-scale data mining techniques using machine learning algorithms have promise in the analysis of large epidemiological datasets. This study illustrates the use of a hybrid methodology for variable selection that took account of missing data and complex survey design to identify key biomarkers associated with depression from a large epidemiological study. The study used a three-step methodology amalgamating multiple imputation, a machine learning boosted regression algorithm and logistic regression, to identify key biomarkers associated with depression in the National Health and Nutrition Examination Study (2009-2010). Depression was measured using the Patient Health Questionnaire-9 and 67 biomarkers were analysed. Covariates in this study included gender, age, race, smoking, food security, Poverty Income Ratio, Body Mass Index, physical activity, alcohol use, medical conditions and medications. The final imputed weighted multiple logistic regression model included possible confounders and moderators. After the creation of 20 imputation data sets from multiple chained regression sequences, machine learning boosted regression initially identified 21 biomarkers associated with depression. Using traditional logistic regression methods, including controlling for possible confounders and moderators, a final set of three biomarkers were selected. The final three biomarkers from the novel hybrid variable selection methodology were red cell distribution width (OR 1.15; 95% CI 1.01, 1.30), serum glucose (OR 1.01; 95% CI 1.00, 1.01) and total bilirubin (OR 0.12; 95% CI 0.05, 0.28). Significant interactions were found between total bilirubin with Mexican American/Hispanic group (p = 0.016), and current smokers (p<0.001). The systematic use of a hybrid methodology for variable selection, fusing data mining techniques using a machine learning algorithm with traditional statistical modelling, accounted for missing data and complex survey sampling methodology and was demonstrated to be a useful tool for detecting three biomarkers associated with depression for future hypothesis generation: red cell distribution width, serum glucose and total bilirubin.
Field, Matt; Gage, Suzanne; Hammerton, Gemma; Heron, Jon; Hickman, Matt; Munafò, Marcus R
2018-01-01
Abstract Aims The study aimed to examine the association between adolescent alcohol use and working memory (WM) using a large population sample. Methods Data from the Avon Longitudinal Study of Parents and Children were used to investigate the association between alcohol use at age 15 years and WM 3 years later, assessed using the N-back task (N ~ 3300). A three-category ordinal variable captured mutually exclusive alcohol groupings ranging in order of severity (i.e. low alcohol users, frequent drinkers and frequent/binge drinkers). Differential dropout was accounted for using multiple imputation and inverse probability weighting. Adjustment was made for potential confounders. Results There was evidence of an association between frequent/binge drinking (compared to the low alcohol group) and poorer performance on the 3-back task after adjusting for sociodemographic confounding variables, WM at age 11 years, and experience of a head injury/unconsciousness before age 11 years (β = −0.23, 95% CI = −0.37 to −0.09, P = 0.001). However, this association was attenuated (β = −0.12, 95% CI = −0.27 to 0.03, P = 0.11) when further adjusted for baseline measures of weekly cigarette tobacco and cannabis use. Weaker associations were found for the less demanding 2-back task. We found no evidence to suggest frequent drinking was associated with performance on either task. Conclusions We found weak evidence of an association between sustained heavy alcohol use in mid-adolescence and impaired WM 3 years later. Although we cannot fully rule out the possibility of reverse causation, several potential confounding variables were included to address the directionality of the relationship between WM and alcohol use problems. PMID:29329371
Mahedy, Liam; Field, Matt; Gage, Suzanne; Hammerton, Gemma; Heron, Jon; Hickman, Matt; Munafò, Marcus R
2018-05-01
The study aimed to examine the association between adolescent alcohol use and working memory (WM) using a large population sample. Data from the Avon Longitudinal Study of Parents and Children were used to investigate the association between alcohol use at age 15 years and WM 3 years later, assessed using the N-back task (N ~ 3300). A three-category ordinal variable captured mutually exclusive alcohol groupings ranging in order of severity (i.e. low alcohol users, frequent drinkers and frequent/binge drinkers). Differential dropout was accounted for using multiple imputation and inverse probability weighting. Adjustment was made for potential confounders. There was evidence of an association between frequent/binge drinking (compared to the low alcohol group) and poorer performance on the 3-back task after adjusting for sociodemographic confounding variables, WM at age 11 years, and experience of a head injury/unconsciousness before age 11 years (β = -0.23, 95% CI = -0.37 to -0.09, P = 0.001). However, this association was attenuated (β = -0.12, 95% CI = -0.27 to 0.03, P = 0.11) when further adjusted for baseline measures of weekly cigarette tobacco and cannabis use. Weaker associations were found for the less demanding 2-back task. We found no evidence to suggest frequent drinking was associated with performance on either task. We found weak evidence of an association between sustained heavy alcohol use in mid-adolescence and impaired WM 3 years later. Although we cannot fully rule out the possibility of reverse causation, several potential confounding variables were included to address the directionality of the relationship between WM and alcohol use problems.
An introduction to instrumental variables analysis: part 1.
Bennett, Derrick A
2010-01-01
There are several examples in the medical literature where the associations of treatment effects predicted by observational studies have been refuted by evidence from subsequent large-scale randomised trials. This is because of the fact that non-experimental studies are subject to confounding - and confounding cannot be entirely eliminated even if all known confounders have been measured in the study as there may be unknown confounders. The aim of this 2-part methodological primer is to introduce an emerging methodology for estimating treatment effects using observational data in the absence of good randomised evidence known as the method of instrumental variables. Copyright © 2010 S. Karger AG, Basel.
Motl, Robert W; McAuley, Edward; Sandroff, Brian M
2013-08-01
Physical activity is beneficial for people with multiple sclerosis (MS), but this population is largely inactive. There is minimal information on change in physical activity and its correlates for informing the development of behavioral interventions. This study examined change in physical activity and its symptomatic, social-cognitive, and ambulatory or disability correlates over a 2.5-year period of time in people with relapsing-remitting multiple sclerosis. On 6 occasions, each separated by 6 months, people (N=269) with relapsing-remitting multiple sclerosis completed assessments of symptoms, self-efficacy, walking impairment, disability, and physical activity. The participants wore an accelerometer for 7 days. The change in study variables over 6 time points was examined with unconditional latent growth curve modeling. The association among changes in study variables over time was examined using conditional latent growth curve modeling, and the associations were expressed as standardized path coefficients (β). There were significant linear changes in self-reported and objectively measured physical activity, self-efficacy, walking impairment, and disability over the 2.5-year period; there were no changes in fatigue, depression, and pain. The changes in self-reported and objective physical activity were associated with change in self-efficacy (β=.49 and β=.61, respectively), after controlling for other variables and confounders. The primary limitations of the study were the generalizability of results among those with progressive multiple sclerosis and inclusion of a single variable from social-cognitive theory. Researchers should consider designing interventions that target self-efficacy for the promotion and maintenance of physical activity in this population.
Baldock, Deanna; Miller, Justin B; Leger, Gabriel C; Banks, Sarah Jane
2016-01-01
Patients with frontotemporal dementia (FTD) typically have initial deficits in language or changes in personality, while the defining characteristic of Alzheimer's disease (AD) is memory impairment. Neuropsychological findings in the two diseases tend to differ, but can be confounded by verbal impairment in FTD impacting performance on memory tests in these patients. Twenty-seven patients with FTD and 102 patients with AD underwent a neuropsychological assessment before diagnosis. By utilizing analogous versions of a verbal and nonverbal memory test, we demonstrated differences in these two modalities between AD and FTD. Better differentiation between AD and FTD is found in a nonverbal memory test, possibly because it eliminates the confounding variable of language deficits found in patients with FTD. These results highlight the importance of nonverbal learning tests with multiple learning trials in diagnostic testing.
Agogo, George O; van der Voet, Hilko; van 't Veer, Pieter; Ferrari, Pietro; Muller, David C; Sánchez-Cantalejo, Emilio; Bamia, Christina; Braaten, Tonje; Knüppel, Sven; Johansson, Ingegerd; van Eeuwijk, Fred A; Boshuizen, Hendriek C
2016-10-13
Measurement error in self-reported dietary intakes is known to bias the association between dietary intake and a health outcome of interest such as risk of a disease. The association can be distorted further by mismeasured confounders, leading to invalid results and conclusions. It is, however, difficult to adjust for the bias in the association when there is no internal validation data. We proposed a method to adjust for the bias in the diet-disease association (hereafter, association), due to measurement error in dietary intake and a mismeasured confounder, when there is no internal validation data. The method combines prior information on the validity of the self-report instrument with the observed data to adjust for the bias in the association. We compared the proposed method with the method that ignores the confounder effect, and with the method that ignores measurement errors completely. We assessed the sensitivity of the estimates to various magnitudes of measurement error, error correlations and uncertainty in the literature-reported validation data. We applied the methods to fruits and vegetables (FV) intakes, cigarette smoking (confounder) and all-cause mortality data from the European Prospective Investigation into Cancer and Nutrition study. Using the proposed method resulted in about four times increase in the strength of association between FV intake and mortality. For weakly correlated errors, measurement error in the confounder minimally affected the hazard ratio estimate for FV intake. The effect was more pronounced for strong error correlations. The proposed method permits sensitivity analysis on measurement error structures and accounts for uncertainties in the reported validity coefficients. The method is useful in assessing the direction and quantifying the magnitude of bias in the association due to measurement errors in the confounders.
Integrated presentation of ecological risk from multiple stressors
NASA Astrophysics Data System (ADS)
Goussen, Benoit; Price, Oliver R.; Rendal, Cecilie; Ashauer, Roman
2016-10-01
Current environmental risk assessments (ERA) do not account explicitly for ecological factors (e.g. species composition, temperature or food availability) and multiple stressors. Assessing mixtures of chemical and ecological stressors is needed as well as accounting for variability in environmental conditions and uncertainty of data and models. Here we propose a novel probabilistic ERA framework to overcome these limitations, which focusses on visualising assessment outcomes by construct-ing and interpreting prevalence plots as a quantitative prediction of risk. Key components include environmental scenarios that integrate exposure and ecology, and ecological modelling of relevant endpoints to assess the effect of a combination of stressors. Our illustrative results demonstrate the importance of regional differences in environmental conditions and the confounding interactions of stressors. Using this framework and prevalence plots provides a risk-based approach that combines risk assessment and risk management in a meaningful way and presents a truly mechanistic alternative to the threshold approach. Even whilst research continues to improve the underlying models and data, regulators and decision makers can already use the framework and prevalence plots. The integration of multiple stressors, environmental conditions and variability makes ERA more relevant and realistic.
Integrated presentation of ecological risk from multiple stressors.
Goussen, Benoit; Price, Oliver R; Rendal, Cecilie; Ashauer, Roman
2016-10-26
Current environmental risk assessments (ERA) do not account explicitly for ecological factors (e.g. species composition, temperature or food availability) and multiple stressors. Assessing mixtures of chemical and ecological stressors is needed as well as accounting for variability in environmental conditions and uncertainty of data and models. Here we propose a novel probabilistic ERA framework to overcome these limitations, which focusses on visualising assessment outcomes by construct-ing and interpreting prevalence plots as a quantitative prediction of risk. Key components include environmental scenarios that integrate exposure and ecology, and ecological modelling of relevant endpoints to assess the effect of a combination of stressors. Our illustrative results demonstrate the importance of regional differences in environmental conditions and the confounding interactions of stressors. Using this framework and prevalence plots provides a risk-based approach that combines risk assessment and risk management in a meaningful way and presents a truly mechanistic alternative to the threshold approach. Even whilst research continues to improve the underlying models and data, regulators and decision makers can already use the framework and prevalence plots. The integration of multiple stressors, environmental conditions and variability makes ERA more relevant and realistic.
ERIC Educational Resources Information Center
Lee, Paul H.
2017-01-01
Purpose: Some confounders are nonlinearly associated with dependent variables, but they are often adjusted using a linear term. The purpose of this study was to examine the error of mis-specifying the nonlinear confounding effect. Methods: We carried out a simulation study to investigate the effect of adjusting for a nonlinear confounder in the…
Lester L. Yuan,; Amina I. Pollard,; Carlisle, Daren M.
2009-01-01
Analyses of observational data can provide insights into relationships between environmental conditions and biological responses across a broader range of natural conditions than experimental studies, potentially complementing insights gained from experiments. However, observational data must be analyzed carefully to minimize the likelihood that confounding variables bias observed relationships. Propensity scores provide a robust approach for controlling for the effects of measured confounding variables when analyzing observational data. Here, we use propensity scores to estimate changes in mean invertebrate taxon richness in streams that have experienced insecticide concentrations that exceed aquatic life use benchmark concentrations. A simple comparison of richness in sites exposed to elevated insecticides with those that were not exposed suggests that exposed sites had on average 6.8 fewer taxa compared to unexposed sites. The presence of potential confounding variables makes it difficult to assert a causal relationship from this simple comparison. After controlling for confounding factors using propensity scores, the difference in richness between exposed and unexposed sites was reduced to 4.1 taxa, a difference that was still statistically significant. Because the propensity score analysis controlled for the effects of a wide variety of possible confounding variables, we infer that the change in richness observed in the propensity score analysis was likely caused by insecticide exposure.
Yuan, L.L.; Pollard, A.I.; Carlisle, D.M.
2009-01-01
Analyses of observational data can provide insights into relationships between environmental conditions and biological responses across a broader range of natural conditions than experimental studies, potentially complementing insights gained from experiments. However, observational data must be analyzed carefully to minimize the likelihood that confounding variables bias observed relationships. Propensity scores provide a robust approach for controlling for the effects of measured confounding variables when analyzing observational data. Here, we use propensity scores to estimate changes in mean invertebrate taxon richness in streams that have experienced insecticide concentrations that exceed aquatic life use benchmark concentrations. A simple comparison of richness in sites exposed to elevated insecticides with those that were not exposed suggests that exposed sites had on average 6.8 fewer taxa compared to unexposed sites. The presence of potential confounding variables makes it difficult to assert a causal relationship from this simple comparison. After controlling for confounding factors using propensity scores, the difference in richness between exposed and unexposed sites was reduced to 4.1 taxa, a difference that was still statistically significant. Because the propensity score analysis controlled for the effects of a wide variety of possible confounding variables, we infer that the change in richness observed in the propensity score analysis was likely caused by insecticide exposure. ?? 2009 SETAC.
Kaptein, Maurits; van Emden, Robin; Iannuzzi, Davide
2017-01-01
Due to the ubiquitous presence of treatment heterogeneity, measurement error, and contextual confounders, numerous social phenomena are hard to study. Precise control of treatment variables and possible confounders is often key to the success of studies in the social sciences, yet often proves out of the realm of control of the experimenter. To amend this situation we propose a novel approach coined "lock-in feedback" which is based on a method that is routinely used in high-precision physics experiments to extract small signals out of a noisy environment. Here, we adapt the method to noisy social signals in multiple dimensions and evaluate it by studying an inherently noisy topic: the perception of (subjective) beauty. We show that the lock-in feedback approach allows one to select optimal treatment levels despite the presence of considerable noise. Furthermore, through the introduction of an external contextual shock we demonstrate that we can find relationships between noisy variables that were hitherto unknown. We therefore argue that lock-in methods may provide a valuable addition to the social scientist's experimental toolbox and we explicitly discuss a number of future applications.
2017-01-01
Due to the ubiquitous presence of treatment heterogeneity, measurement error, and contextual confounders, numerous social phenomena are hard to study. Precise control of treatment variables and possible confounders is often key to the success of studies in the social sciences, yet often proves out of the realm of control of the experimenter. To amend this situation we propose a novel approach coined “lock-in feedback” which is based on a method that is routinely used in high-precision physics experiments to extract small signals out of a noisy environment. Here, we adapt the method to noisy social signals in multiple dimensions and evaluate it by studying an inherently noisy topic: the perception of (subjective) beauty. We show that the lock-in feedback approach allows one to select optimal treatment levels despite the presence of considerable noise. Furthermore, through the introduction of an external contextual shock we demonstrate that we can find relationships between noisy variables that were hitherto unknown. We therefore argue that lock-in methods may provide a valuable addition to the social scientist’s experimental toolbox and we explicitly discuss a number of future applications. PMID:28306728
Gibbs, Jenna C; Giangregorio, Lora M; Wong, Andy K O; Josse, Robert G; Cheung, Angela M
2017-10-01
The purpose of this cross-sectional study was to determine how appendicular lean mass index (ALMI), and whole body lean (LMI) and fat mass indices (FMI) associate with estimated bone strength outcomes at the distal radius and tibia in adults aged 40 years and older. Dual energy X-ray absorptiometry (DXA) scans were performed to determine body composition, including whole body lean and fat mass, and appendicular lean mass. ALMI (appendicular lean mass/height 2 ), LMI (lean tissue mass/height 2 ) and FMI (fat mass/height 2 ) were calculated. High-resolution peripheral quantitative computed tomography (HRpQCT) scans were performed to assess bone structural properties at the distal radius and tibia. Using finite element analysis, failure load (N), stiffness (N/mm), ultimate stress (MPa), and cortical-to-trabecular load ratio were estimated from HRpQCT scans. The associations between body composition (ALMI, LMI, FMI) and estimated bone strength were examined using bivariate and multivariable linear regression analyses adjusting for age, sex, and other confounding variables. In 197 participants (127 women; mean±SD, age: 69.5±10.3y, body mass index: 27.95±4.95kg/m 2 , ALMI: 7.31±1.31kg/m 2 ), ALMI and LMI were significantly associated with failure load at the distal radius and tibia (explained 39%-48% of the variance) and remained significant after adjusting for confounding variables and multiple testing (R 2 =0.586-0.645, p<0.001). ALMI, LMI, and FMI did not have significant associations with ultimate stress in our multivariable models. FMI was significantly associated with cortical-to-trabecular load ratio at the distal radius and tibia (explained 6%-12% of the variance) and remained significant after adjusting for confounders and multiple testing (R 2 =0.208-0.243, p<0.001). FMI was no longer significantly associated with failure load after adjusting for confounders. These findings suggest that ALMI and LMI are important determinants of estimated bone strength, particularly failure load, at the distal radius and tibia, and may contribute to preservation of bone strength in middle-to-late adulthood. Copyright © 2017 Elsevier Inc. All rights reserved.
Correction of confounding bias in non-randomized studies by appropriate weighting.
Schmoor, Claudia; Gall, Christine; Stampf, Susanne; Graf, Erika
2011-03-01
In non-randomized studies, the assessment of a causal effect of treatment or exposure on outcome is hampered by possible confounding. Applying multiple regression models including the effects of treatment and covariates on outcome is the well-known classical approach to adjust for confounding. In recent years other approaches have been promoted. One of them is based on the propensity score and considers the effect of possible confounders on treatment as a relevant criterion for adjustment. Another proposal is based on using an instrumental variable. Here inference relies on a factor, the instrument, which affects treatment but is thought to be otherwise unrelated to outcome, so that it mimics randomization. Each of these approaches can basically be interpreted as a simple reweighting scheme, designed to address confounding. The procedures will be compared with respect to their fundamental properties, namely, which bias they aim to eliminate, which effect they aim to estimate, and which parameter is modelled. We will expand our overview of methods for analysis of non-randomized studies to methods for analysis of randomized controlled trials and show that analyses of both study types may target different effects and different parameters. The considerations will be illustrated using a breast cancer study with a so-called Comprehensive Cohort Study design, including a randomized controlled trial and a non-randomized study in the same patient population as sub-cohorts. This design offers ideal opportunities to discuss and illustrate the properties of the different approaches. Copyright © 2011 WILEY-VCH Verlag GmbH & Co. KGaA, Weinheim.
Tutorial in Biostatistics: Instrumental Variable Methods for Causal Inference*
Baiocchi, Michael; Cheng, Jing; Small, Dylan S.
2014-01-01
A goal of many health studies is to determine the causal effect of a treatment or intervention on health outcomes. Often, it is not ethically or practically possible to conduct a perfectly randomized experiment and instead an observational study must be used. A major challenge to the validity of observational studies is the possibility of unmeasured confounding (i.e., unmeasured ways in which the treatment and control groups differ before treatment administration which also affect the outcome). Instrumental variables analysis is a method for controlling for unmeasured confounding. This type of analysis requires the measurement of a valid instrumental variable, which is a variable that (i) is independent of the unmeasured confounding; (ii) affects the treatment; and (iii) affects the outcome only indirectly through its effect on the treatment. This tutorial discusses the types of causal effects that can be estimated by instrumental variables analysis; the assumptions needed for instrumental variables analysis to provide valid estimates of causal effects and sensitivity analysis for those assumptions; methods of estimation of causal effects using instrumental variables; and sources of instrumental variables in health studies. PMID:24599889
Mink, Pamela J; Goodman, Michael; Barraj, Leila M; Imrey, Harriet; Kelsh, Michael A; Yager, Janice
2004-07-01
Neurobehavioral tests are commonly used in studies of children exposed to low-level environmental concentrations of compounds known to be neurotoxic at higher levels. However, uncontrolled or incomplete control for confounding makes interpretation of results problematic because effects of confounders are often stronger than the effects of primary interest. We examined a priori the potential impact of confounding in a hypothetical study evaluating the association of a potentially neurotoxic environmental exposure with neurobehavioral function in children. We used 2 outcome measures: the Bayley Scales of Infant Development Mental Development Index and the Stanford-Binet Intelligence Scale Composite Score. We selected 3 potential confounders: maternal intelligence, home environment, and socioeconomic status as measured by years of parental education. We conducted 3 sets of analyses measuring the effect of each of the 3 confounding factors alone, 2 confounders acting simultaneously, and all 3 confounders acting simultaneously. Relatively small differences (0.5 standard deviations) in confounding variables between "exposed" and "unexposed" groups, if unmeasured and unaccounted for in the analysis, could produce spurious differences in cognitive test scores. The magnitude of this difference (3-10 points) has been suggested to have a meaningful impact in populations. The method of measuring confounders (eg, maternal intelligence) could also substantially affect the results. It is important to carefully consider the impact of potential confounders during the planning stages of an observational study. Study-to-study differences in neurobehavioral outcomes with similar environmental exposures could be partially explained by differences in the adjustment for confounding variables. Copyright 2004 Lippincott Williams and Wilkins
Coping strategies and quality of life in caregivers of dependent elderly relatives.
Rodríguez-Pérez, Margarita; Abreu-Sánchez, Ana; Rojas-Ocaña, María Jesús; Del-Pino-Casado, Rafael
2017-04-14
Despite the importance of coping in caregiving, there are few studies on the relationship between coping and quality of life in caregivers of the frail dependent elderly. Thus, this study aims to analyze the relationship between coping strategies and quality of life dimensions in primary caregivers of dependent elderly relatives. A cross-sectional study was conducted from 86 caregivers. Predictive variables were coping strategies (problem-focused, emotion-focused, socially-supported, and dysfunctional); dependent variables were quality of life dimensions (psychological, physical, relational, and environmental); and potential confounding variables were age, gender, perceived health and burden of caregiver, and functional capacity of care receiver. Correlation coefficients were calculated and multiple linear regression analysis was performed. After controlling for potential confounders, dysfunctional coping was related to worse quality of life in the psychological dimension, while emotion-focused and socially-supported coping were related to superior psychological and environmental dimensions of quality of life. The physical and relational dimensions of quality of life were not related to coping strategies. 1) it is important to consider coping strategies in the assessment of primary caregivers of dependent elderly relatives; 2) the quality of life of caregivers is related to their coping strategies, 3) their quality of life can be worsened by avoidance-type coping, and 4) their quality of life can be improved by active emotion-focused coping and socially-supported coping.
Bollen, Kenneth A
2007-06-01
R. D. Howell, E. Breivik, and J. B. Wilcox (2007) have argued that causal (formative) indicators are inherently subject to interpretational confounding. That is, they have argued that using causal (formative) indicators leads the empirical meaning of a latent variable to be other than that assigned to it by a researcher. Their critique of causal (formative) indicators rests on several claims: (a) A latent variable exists apart from the model when there are effect (reflective) indicators but not when there are causal (formative) indicators, (b) causal (formative) indicators need not have the same consequences, (c) causal (formative) indicators are inherently subject to interpretational confounding, and (d) a researcher cannot detect interpretational confounding when using causal (formative) indicators. This article shows that each claim is false. Rather, interpretational confounding is more a problem of structural misspecification of a model combined with an underidentified model that leaves these misspecifications undetected. Interpretational confounding does not occur if the model is correctly specified whether a researcher has causal (formative) or effect (reflective) indicators. It is the validity of a model not the type of indicator that determines the potential for interpretational confounding. Copyright 2007 APA, all rights reserved.
American black duck summer range versus winter range: a dichotomy of riches
Longcore, J.R.; Perry, Matthew C.
2002-01-01
The status of the American black duck (Anas rubripes) population has more often been attributed to a single event than to multiple events over time and throughout space. The difference in the quality of the habitat, however defined, within breeding areas in the North and in the southerly wintering areas, especially Chesapeake Bay, also has been proposed as affecting black duck status. The obvious question is 'What variable cuts across all habitats, time, and space to affect black ducks?' This paper attempts to answer that question by examining the connectivity of seemingly unrelated variables and events associated with the black duck's summer range and its winter range relative to population change. Insights from examples of relations among these variables reveal how results may be confounded and even misleading. A perspective that may be required to ensure future black duck populations is discussed.
Introduction to the use of regression models in epidemiology.
Bender, Ralf
2009-01-01
Regression modeling is one of the most important statistical techniques used in analytical epidemiology. By means of regression models the effect of one or several explanatory variables (e.g., exposures, subject characteristics, risk factors) on a response variable such as mortality or cancer can be investigated. From multiple regression models, adjusted effect estimates can be obtained that take the effect of potential confounders into account. Regression methods can be applied in all epidemiologic study designs so that they represent a universal tool for data analysis in epidemiology. Different kinds of regression models have been developed in dependence on the measurement scale of the response variable and the study design. The most important methods are linear regression for continuous outcomes, logistic regression for binary outcomes, Cox regression for time-to-event data, and Poisson regression for frequencies and rates. This chapter provides a nontechnical introduction to these regression models with illustrating examples from cancer research.
Integrated presentation of ecological risk from multiple stressors
Goussen, Benoit; Price, Oliver R.; Rendal, Cecilie; Ashauer, Roman
2016-01-01
Current environmental risk assessments (ERA) do not account explicitly for ecological factors (e.g. species composition, temperature or food availability) and multiple stressors. Assessing mixtures of chemical and ecological stressors is needed as well as accounting for variability in environmental conditions and uncertainty of data and models. Here we propose a novel probabilistic ERA framework to overcome these limitations, which focusses on visualising assessment outcomes by construct-ing and interpreting prevalence plots as a quantitative prediction of risk. Key components include environmental scenarios that integrate exposure and ecology, and ecological modelling of relevant endpoints to assess the effect of a combination of stressors. Our illustrative results demonstrate the importance of regional differences in environmental conditions and the confounding interactions of stressors. Using this framework and prevalence plots provides a risk-based approach that combines risk assessment and risk management in a meaningful way and presents a truly mechanistic alternative to the threshold approach. Even whilst research continues to improve the underlying models and data, regulators and decision makers can already use the framework and prevalence plots. The integration of multiple stressors, environmental conditions and variability makes ERA more relevant and realistic. PMID:27782171
The Problem of Confounding in Studies of the Effect of Maternal Drug Use on Pregnancy Outcome
Källén, Bengt
2012-01-01
In most epidemilogical studies, the problem of confounding adds to the uncertainty in conclusions drawn. This is also true for studies on the effect of maternal drug use on birth defect risks. This paper describes various types of such confounders and discusses methods to identify and adjust for them. Such confounders can be found in maternal characteristics like age, parity, smoking, use of alcohol, and body mass index, subfertility, and previous pregnancies including previous birth of a malformed child, socioeconomy, race/ethnicity, or country of birth. Confounding by concomitant maternal drug use may occur. A geographical or seasonal confounding can exist. In rare instances, infant sex and multiple birth can appear as confounders. The most difficult problem to solve is often confounding by indication. The problem of confounding is less important for congenital malformations than for many other pregnancy outcomes. PMID:22190949
Association of Lifestyle-Related Comorbidities With Periodontitis
Lee, Jae-Hong; Lee, Jung-Seok; Park, Jin-Young; Choi, Jung-Kyu; Kim, Dong-Wook; Kim, Young-Taek; Choi, Seong-Ho
2015-01-01
Abstract The aim of this study was to determine the association of periodontitis with lifestyle-related comorbidities (LCs) using data in the Korean National Health Insurance Cohort Database from 2002 to 2013. This was a retrospective study involving a large national cohort with patient samples (representing 2% of the total Korean population) stratified on the basis of sociodemographic information. Using this precisely extracted database, the correlations between LCs (cerebral infarction, angina pectoris, myocardial infarction, hypertension, diabetes mellitus, rheumatoid arthritis, erectile dysfunction, osteoporosis, and obesity) and periodontitis were investigated while adjusting for confounding bias. Univariate and multiple logistic regression analyses were used to evaluate differences in variable factors. Among a total of 1,025,340 samples, 321,103 (31.3%) cases were diagnosed with periodontitis. Statistically significant associations were found between all LCs except myocardial infarction and periodontitis (P < 0.005). Periodontitis is significantly and positively correlated with LCs (except for myocardial infarction) after adjusting for confounding bias. In particular, lifestyle-related diseases, erectile dysfunction, and osteoporosis seem to be intimately related to periodontitis. PMID:26376407
Periodontitis is associated with diabetic retinopathy in non-obese adults.
Song, Su Jeong; Lee, Seong-Su; Han, Kyungdo; Park, Jun-Beom
2017-04-01
Patients with diabetes retinopathy appear to show increased susceptibility to periodontal disease. This study was performed to assess the relationship between periodontitis and the prevalence of diabetic retinopathy in a large probability sample of the Korean population. A subgroup analysis was performed using body mass index <25 kg/m 2 as the criterion to evaluate the effect of obesity on this relationship. This study is based on data from the Korean National Health and Nutrition Examination Survey of the Korean population, conducted between 2008 and 2010. The presence of diabetic retinopathy in relation to demographic variables and anthropometric characteristics of the participants is presented as means with their standard errors. The presence of periodontitis and presence of retinopathy categorized by body mass index (<25 and ≥25 kg/m 2 ) were evaluated. Multiple logistic regression analyses were used to assess the associations between periodontitis and diabetic retinopathy after adjustment with variables, including age, sex, smoking, drinking, exercise, hypertension, metabolic syndrome, HbA1c, and duration of diabetes mellitus. There was a statistically significant increase in the prevalence of periodontitis in individuals who had proliferative diabetic retinopathy. The odds ratios [95% confidence intervals] of prevalence of diabetic retinopathy were 1.193 [0.757-1.881] for the whole population after adjustments with confounding factors. Subgroup analysis after adjustments with confounding factors showed that the odds ratios [95% confidence intervals] of prevalence were 2.206 [1.114-4.366] and 0.588 [0.326-1.061] among participants with body mass index <25 kg/m 2 and body mass index 37 ≥25 kg/m 2 , respectively. The diabetic retinopathy was positively associated with the presence of periodontitis in non-obese diabetic Korean adults after adjustment with confounding variables. Our findings suggest that when a periodontist finds the presence of periodontitis in non-obese diabetic patients, timely evaluation of the patient's ophthalmic evaluation should be 44 recommended.
Ertefaie, Ashkan; Flory, James H; Hennessy, Sean; Small, Dylan S
2017-06-15
Instrumental variable (IV) methods provide unbiased treatment effect estimation in the presence of unmeasured confounders under certain assumptions. To provide valid estimates of treatment effect, treatment effect confounders that are associated with the IV (IV-confounders) must be included in the analysis, and not including observations with missing values may lead to bias. Missing covariate data are particularly problematic when the probability that a value is missing is related to the value itself, which is known as nonignorable missingness. In such cases, imputation-based methods are biased. Using health-care provider preference as an IV method, we propose a 2-step procedure with which to estimate a valid treatment effect in the presence of baseline variables with nonignorable missing values. First, the provider preference IV value is estimated by performing a complete-case analysis using a random-effects model that includes IV-confounders. Second, the treatment effect is estimated using a 2-stage least squares IV approach that excludes IV-confounders with missing values. Simulation results are presented, and the method is applied to an analysis comparing the effects of sulfonylureas versus metformin on body mass index, where the variables baseline body mass index and glycosylated hemoglobin have missing values. Our result supports the association of sulfonylureas with weight gain. © The Author 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Simulation of Medical Imaging Systems: Emission and Transmission Tomography
NASA Astrophysics Data System (ADS)
Harrison, Robert L.
Simulation is an important tool in medical imaging research. In patient scans the true underlying anatomy and physiology is unknown. We have no way of knowing in a given scan how various factors are confounding the data: statistical noise; biological variability; patient motion; scattered radiation, dead time, and other data contaminants. Simulation allows us to isolate a single factor of interest, for instance when researchers perform multiple simulations of the same imaging situation to determine the effect of statistical noise or biological variability. Simulations are also increasingly used as a design optimization tool for tomographic scanners. This article gives an overview of the mechanics of emission and transmission tomography simulation, reviews some of the publicly available simulation tools, and discusses trade-offs between the accuracy and efficiency of simulations.
Napping during the night shift and recovery after work among hospital nurses1
Palermo, Thaís Aparecida de Castro; Rotenberg, Lúcia; Zeitoune, Regina Célia Gollner; Silva-Costa, Aline; Souto, Ester Paiva; Griep, Rosane Härter
2015-01-01
OBJECTIVE: To analyze the association between the length of napping during the night shift and the recovery after work among nurses. METHOD: Cross-sectional epidemiological study involving 1940 nurses from 18 public hospitals in the City of Rio de Janeiro. A multidimensional and self-applied questionnaire was used with information about health, sociodemographic and occupational characteristics, health-related behaviors and housework. Multiple logistic regression was applied to identify the association, adjusted for confounding variables. RESULTS: The gross analyses showed 44%, 127% and 66% higher chances of a high level of recovery after work for nurses who sleep up to two hours, between 2.1 and 3 hours and 3.1 hours or more, respectively, when compared to the nurses who do not sleep. After adjusting for confounding variables, the association only continues significant for the group that sleeps 2.1 to 3 hours during the night shift (OR=1.79; 95%CI=1.33-2.41). CONCLUSION: The association between the length of napping and the high level of recovery after work, confirmed in the present results, can be included in the studies that aim to support more appropriate policies aimed at improving the workers' work, life and health conditions, not only in nursing, but night-shift workers in general. PMID:25806639
Adult Lifespan Cognitive Variability in the Cross-Sectional Cam-CAN Cohort
Green, Emma; Shafto, Meredith A.; Matthews, Fiona E.; White, Simon R.
2015-01-01
This study examines variability across the age span in cognitive performance in a cross-sectional, population-based, adult lifespan cohort from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) study (n = 2680). A key question we highlight is whether using measures that are designed to detect age-related cognitive pathology may not be sensitive to, or reflective of, individual variability among younger adults. We present three issues that contribute to the debate for and against age-related increases in variability. Firstly, the need to formally define measures of central tendency and measures of variability. Secondly, in addition to the commonly addressed location-confounding (adjusting for covariates) there may exist changes in measures of variability due to confounder sub-groups. Finally, that increases in spread may be a result of floor or ceiling effects; where the measure is not sensitive enough at all ages. From the Cam-CAN study, a large population-based dataset, we demonstrate the existence of variability-confounding for the immediate episodic memory task; and show that increasing variance with age in our general cognitive measures is driven by a ceiling effect in younger age groups. PMID:26690191
Adult Lifespan Cognitive Variability in the Cross-Sectional Cam-CAN Cohort.
Green, Emma; Shafto, Meredith A; Matthews, Fiona E; White, Simon R
2015-12-07
This study examines variability across the age span in cognitive performance in a cross-sectional, population-based, adult lifespan cohort from the Cambridge Centre for Ageing and Neuroscience (Cam-CAN) study (n = 2680). A key question we highlight is whether using measures that are designed to detect age-related cognitive pathology may not be sensitive to, or reflective of, individual variability among younger adults. We present three issues that contribute to the debate for and against age-related increases in variability. Firstly, the need to formally define measures of central tendency and measures of variability. Secondly, in addition to the commonly addressed location-confounding (adjusting for covariates) there may exist changes in measures of variability due to confounder sub-groups. Finally, that increases in spread may be a result of floor or ceiling effects; where the measure is not sensitive enough at all ages. From the Cam-CAN study, a large population-based dataset, we demonstrate the existence of variability-confounding for the immediate episodic memory task; and show that increasing variance with age in our general cognitive measures is driven by a ceiling effect in younger age groups.
Tosteson, Tor D.; Morden, Nancy E.; Stukel, Therese A.; O'Malley, A. James
2014-01-01
The estimation of treatment effects is one of the primary goals of statistics in medicine. Estimation based on observational studies is subject to confounding. Statistical methods for controlling bias due to confounding include regression adjustment, propensity scores and inverse probability weighted estimators. These methods require that all confounders are recorded in the data. The method of instrumental variables (IVs) can eliminate bias in observational studies even in the absence of information on confounders. We propose a method for integrating IVs within the framework of Cox's proportional hazards model and demonstrate the conditions under which it recovers the causal effect of treatment. The methodology is based on the approximate orthogonality of an instrument with unobserved confounders among those at risk. We derive an estimator as the solution to an estimating equation that resembles the score equation of the partial likelihood in much the same way as the traditional IV estimator resembles the normal equations. To justify this IV estimator for a Cox model we perform simulations to evaluate its operating characteristics. Finally, we apply the estimator to an observational study of the effect of coronary catheterization on survival. PMID:25506259
MacKenzie, Todd A; Tosteson, Tor D; Morden, Nancy E; Stukel, Therese A; O'Malley, A James
2014-06-01
The estimation of treatment effects is one of the primary goals of statistics in medicine. Estimation based on observational studies is subject to confounding. Statistical methods for controlling bias due to confounding include regression adjustment, propensity scores and inverse probability weighted estimators. These methods require that all confounders are recorded in the data. The method of instrumental variables (IVs) can eliminate bias in observational studies even in the absence of information on confounders. We propose a method for integrating IVs within the framework of Cox's proportional hazards model and demonstrate the conditions under which it recovers the causal effect of treatment. The methodology is based on the approximate orthogonality of an instrument with unobserved confounders among those at risk. We derive an estimator as the solution to an estimating equation that resembles the score equation of the partial likelihood in much the same way as the traditional IV estimator resembles the normal equations. To justify this IV estimator for a Cox model we perform simulations to evaluate its operating characteristics. Finally, we apply the estimator to an observational study of the effect of coronary catheterization on survival.
Empowerment of women and its association with the health of the community.
Varkey, Prathibha; Mbbs; Kureshi, Sarah; Lesnick, Timothy
2010-01-01
Empowerment and opportunities to experience power and control in one's life contribute to health and wellness. Although studies have assessed specific factors related to women's empowerment and their influence on health outcomes, there is a dearth of published literature assessing the relationship of the empowerment of women with the overall health of a community. By means of this article, we aim to assess the relationship of women's empowerment with health in 75 countries. We used the gender empowerment measure (GEM), a composite index measuring gender inequality in economic participation and decision making, political participation and decision making, and power over economic resources. All 75 countries with GEM values in the 2006 Human Development Report (HDR) were included in the study. Association between the GEM values and seven health indicators was evaluated using descriptive statistics, scatter plots, and simple and multiple linear regression models. We also controlled for gross domestic product (GDP) as a possible confounding factor and included this variable in the multiple regression models. When GDP was not considered, GEM had a statistically significant association with all health indicator variables except for proportion of 1-year-olds immunized against measles (correlation coefficient 0.063, p = 0.597). After adjusting for GDP, GEM was significantly associated with low birth weight, fertility rate, infant mortality, and age
Bayesian effect estimation accounting for adjustment uncertainty.
Wang, Chi; Parmigiani, Giovanni; Dominici, Francesca
2012-09-01
Model-based estimation of the effect of an exposure on an outcome is generally sensitive to the choice of which confounding factors are included in the model. We propose a new approach, which we call Bayesian adjustment for confounding (BAC), to estimate the effect of an exposure of interest on the outcome, while accounting for the uncertainty in the choice of confounders. Our approach is based on specifying two models: (1) the outcome as a function of the exposure and the potential confounders (the outcome model); and (2) the exposure as a function of the potential confounders (the exposure model). We consider Bayesian variable selection on both models and link the two by introducing a dependence parameter, ω, denoting the prior odds of including a predictor in the outcome model, given that the same predictor is in the exposure model. In the absence of dependence (ω= 1), BAC reduces to traditional Bayesian model averaging (BMA). In simulation studies, we show that BAC, with ω > 1, estimates the exposure effect with smaller bias than traditional BMA, and improved coverage. We, then, compare BAC, a recent approach of Crainiceanu, Dominici, and Parmigiani (2008, Biometrika 95, 635-651), and traditional BMA in a time series data set of hospital admissions, air pollution levels, and weather variables in Nassau, NY for the period 1999-2005. Using each approach, we estimate the short-term effects of on emergency admissions for cardiovascular diseases, accounting for confounding. This application illustrates the potentially significant pitfalls of misusing variable selection methods in the context of adjustment uncertainty. © 2012, The International Biometric Society.
Collins, Linda M.; Dziak, John J.; Li, Runze
2009-01-01
An investigator who plans to conduct experiments with multiple independent variables must decide whether to use a complete or reduced factorial design. This article advocates a resource management perspective on making this decision, in which the investigator seeks a strategic balance between service to scientific objectives and economy. Considerations in making design decisions include whether research questions are framed as main effects or simple effects; whether and which effects are aliased (confounded) in a particular design; the number of experimental conditions that must be implemented in a particular design and the number of experimental subjects the design requires to maintain the desired level of statistical power; and the costs associated with implementing experimental conditions and obtaining experimental subjects. In this article four design options are compared: complete factorial, individual experiments, single factor, and fractional factorial designs. Complete and fractional factorial designs and single factor designs are generally more economical than conducting individual experiments on each factor. Although relatively unfamiliar to behavioral scientists, fractional factorial designs merit serious consideration because of their economy and versatility. PMID:19719358
DeMaris, Alfred
2014-01-01
Unmeasured confounding is the principal threat to unbiased estimation of treatment “effects” (i.e., regression parameters for binary regressors) in nonexperimental research. It refers to unmeasured characteristics of individuals that lead them both to be in a particular “treatment” category and to register higher or lower values than others on a response variable. In this article, I introduce readers to 2 econometric techniques designed to control the problem, with a particular emphasis on the Heckman selection model (HSM). Both techniques can be used with only cross-sectional data. Using a Monte Carlo experiment, I compare the performance of instrumental-variable regression (IVR) and HSM to that of ordinary least squares (OLS) under conditions with treatment and unmeasured confounding both present and absent. I find HSM generally to outperform IVR with respect to mean-square-error of treatment estimates, as well as power for detecting either a treatment effect or unobserved confounding. However, both HSM and IVR require a large sample to be fully effective. The use of HSM and IVR in tandem with OLS to untangle unobserved confounding bias in cross-sectional data is further demonstrated with an empirical application. Using data from the 2006–2010 General Social Survey (National Opinion Research Center, 2014), I examine the association between being married and subjective well-being. PMID:25110904
Nelson, Suchitra; Albert, Jeffrey M.
2013-01-01
Mediators are intermediate variables in the causal pathway between an exposure and an outcome. Mediation analysis investigates the extent to which exposure effects occur through these variables, thus revealing causal mechanisms. In this paper, we consider the estimation of the mediation effect when the outcome is binary and multiple mediators of different types exist. We give a precise definition of the total mediation effect as well as decomposed mediation effects through individual or sets of mediators using the potential outcomes framework. We formulate a model of joint distribution (probit-normal) using continuous latent variables for any binary mediators to account for correlations among multiple mediators. A mediation formula approach is proposed to estimate the total mediation effect and decomposed mediation effects based on this parametric model. Estimation of mediation effects through individual or subsets of mediators requires an assumption involving the joint distribution of multiple counterfactuals. We conduct a simulation study that demonstrates low bias of mediation effect estimators for two-mediator models with various combinations of mediator types. The results also show that the power to detect a non-zero total mediation effect increases as the correlation coefficient between two mediators increases, while power for individual mediation effects reaches a maximum when the mediators are uncorrelated. We illustrate our approach by applying it to a retrospective cohort study of dental caries in adolescents with low and high socioeconomic status. Sensitivity analysis is performed to assess the robustness of conclusions regarding mediation effects when the assumption of no unmeasured mediator-outcome confounders is violated. PMID:23650048
Wang, Wei; Nelson, Suchitra; Albert, Jeffrey M
2013-10-30
Mediators are intermediate variables in the causal pathway between an exposure and an outcome. Mediation analysis investigates the extent to which exposure effects occur through these variables, thus revealing causal mechanisms. In this paper, we consider the estimation of the mediation effect when the outcome is binary and multiple mediators of different types exist. We give a precise definition of the total mediation effect as well as decomposed mediation effects through individual or sets of mediators using the potential outcomes framework. We formulate a model of joint distribution (probit-normal) using continuous latent variables for any binary mediators to account for correlations among multiple mediators. A mediation formula approach is proposed to estimate the total mediation effect and decomposed mediation effects based on this parametric model. Estimation of mediation effects through individual or subsets of mediators requires an assumption involving the joint distribution of multiple counterfactuals. We conduct a simulation study that demonstrates low bias of mediation effect estimators for two-mediator models with various combinations of mediator types. The results also show that the power to detect a nonzero total mediation effect increases as the correlation coefficient between two mediators increases, whereas power for individual mediation effects reaches a maximum when the mediators are uncorrelated. We illustrate our approach by applying it to a retrospective cohort study of dental caries in adolescents with low and high socioeconomic status. Sensitivity analysis is performed to assess the robustness of conclusions regarding mediation effects when the assumption of no unmeasured mediator-outcome confounders is violated. Copyright © 2013 John Wiley & Sons, Ltd.
Poppers, Kaposi's sarcoma, and HIV infection: empirical example of a strong confounding effect?
Morabia, A
1995-01-01
Are there empirical examples of strong confounding effects? Textbooks usually show examples of weak confounding or use hypothetical examples of strong confounding to illustrate the paradoxical consequences of not separating out the effect of the studied exposure from that of second factor acting as a confounder. HIV infection is a candidate strong confounder of the spuriously high association reported between consumption of poppers, a sexual stimulant, and risk of Kaposi's sarcoma in the early phase of the AIDS epidemic. To examine this hypothesis, assumptions must be made on the prevalence of HIV infection among cases of Kaposi's sarcoma and on the prevalence of heavy popper consumption according to HIV infection in cases and controls. Results show that HIV infection may have confounded the poppers-Kaposi's sarcoma association. However, it cannot be ruled out that HIV did not qualify as a confounder because it was either an intermediate variable or an effect modifier of the association between popper inhalation and Kaposi's sarcoma. This example provides a basis to discuss the mechanism by which confounding occurs as well as the practical importance of confounding in epidemiologic research.
Castro, V M; Kong, S W; Clements, C C; Brady, R; Kaimal, A J; Doyle, A E; Robinson, E B; Churchill, S E; Kohane, I S; Perlis, R H
2016-01-05
Multiple studies have examined the risk of prenatal antidepressant exposure and risk for autism spectrum disorder (ASD) or attention-deficit hyperactivity disorder (ADHD), with inconsistent results. Precisely estimating such risk, if any, is of great importance in light of the need to balance such risk with the benefit of depression and anxiety treatment. We developed a method to integrate data from multiple New England health systems, matching offspring and maternal health data in electronic health records to characterize diagnoses and medication exposure. Children with ASD or ADHD were matched 1:3 with children without neurodevelopmental disorders. Association between maternal antidepressant exposure and ASD or ADHD liability was examined using logistic regression, adjusting for potential sociodemographic and psychiatric confounding variables. In new cohorts of 1245 ASD cases and 1701 ADHD cases, along with age-, sex- and socioeconomic status matched controls, neither disorder was significantly associated with prenatal antidepressant exposure in crude or adjusted models (adjusted odds ratio 0.90, 95% confidence interval 0.50-1.54 for ASD; 0.97, 95% confidence interval 0.53-1.69 for ADHD). Pre-pregnancy antidepressant exposure significantly increased risk for both disorders. These results suggest that prior reports of association between prenatal antidepressant exposure and neurodevelopmental disease are likely to represent a false-positive finding, which may arise in part through confounding by indication. They further demonstrate the potential to integrate data across electronic health records studies spanning multiple health systems to enable efficient pharmacovigilance investigation.
Zhang, Ling; Xue, Chengbing; Wang, Youjie; Zhang, Liuyi; Liang, Yuan
2016-01-01
Despite the benefits of maternal health services, these services are often underutilized, especially in the developing countries. The aim of the present study is to provide insight regarding factors affecting maternal health services use from the family perspective. We use data from the fourth National Health Services Survey in Jiangsu province of Eastern China to investigate the effect of family characteristics on the use of maternal health services. Family characteristics included whether or not living with parents, age of husband, husband's education, and husband's work status as well as family economic status. Demographic variables, social and environmental factors, and previous reproductive history were taken as potential confounders. Multiple logistic regression models were used to examine the independent effects of the family characteristic variables on maternal health service utilization. The data indicate that the percentages of prenatal care, postnatal visits and hospital delivery were 85.44, 65.12 and 99.59 % respectively. Living with parents was associated with less use of prenatal care and husband's age, education and employment status had no effect on the use of prenatal care after adjusting for potential confounding variables. Our findings suggest that maternal health education (especially the role of prenatal care) needs to be extended beyond the expectant mothers themselves to their parents and husbands. The difference of health care delivery as a result of traditional family culture may highlight the differences in factors influencing the use of postnatal visits and those influencing the use of prenatal care; which may be worthy of further study.
Gallagher, Patience J; Castro, Victor; Fava, Maurizio; Weilburg, Jeffrey B; Murphy, Shawn N; Gainer, Vivian S; Churchill, Susanne E; Kohane, Isaac S; Iosifescu, Dan V; Smoller, Jordan W; Perlis, Roy H
2012-10-01
OBJECTIVE It has been suggested that there is a mechanism by which nonsteroidal anti-inflammatory drugs (NSAIDs) may interfere with antidepressant response, and poorer outcomes among NSAID-treated patients were reported in the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study. To attempt to confirm this association in an independent population-based treatment cohort and explore potential confounding variables, the authors examined use of NSAIDs and related medications among 1,528 outpatients in a New England health care system. METHOD Treatment outcomes were classified using a validated machine learning tool applied to electronic medical records. Logistic regression was used to examine the association between medication exposure and treatment outcomes, adjusted for potential confounding variables. To further elucidate confounding and treatment specificity of the observed effects, data from the STAR*D study were reanalyzed. RESULTS NSAID exposure was associated with a greater likelihood of depression classified as treatment resistant compared with depression classified as responsive to selective serotonin reuptake inhibitors (odds ratio=1.55, 95% CI=1.21-2.00). This association was apparent in the NSAIDs-only group but not in those using other agents with NSAID-like mechanisms (cyclooxygenase-2 inhibitors and salicylates). Inclusion of age, sex, ethnicity, and measures of comorbidity and health care utilization in regression models indicated confounding; association with outcome was no longer significant in fully adjusted models. Reanalysis of STAR*D results likewise identified an association in NSAIDs but not NSAID-like drugs, with more modest effects persisting after adjustment for potential confounding variables. CONCLUSIONS These results support an association between NSAID use and poorer antidepressant outcomes in major depressive disorder but indicate that some of the observed effect may be a result of confounding.
Yang, Wei; Yi, Dan-Hui; Xie, Yan-Ming; Yang, Wei; Dai, Yi; Zhi, Ying-Jie; Zhuang, Yan; Yang, Hu
2013-09-01
To estimate treatment effects of Shuxuetong injection on abnormal changes on ALT index, that is, to explore whether the Shuxuetong injection harms liver function in clinical settings and to provide clinical guidance for its safe application. Clinical information of traditional Chinese medicine (TCM) injections is gathered from hospital information system (HIS) of eighteen general hospitals. This is a retrospective cohort study, using abnormal changes in ALT index as an outcome. A large number of confounding biases are taken into account through the generalized boosted models (GBM) and multiple logistic regression model (MLRM) to estimate the treatment effects of Shuxuetong injections on abnormal changes in ALT index and to explore possible influencing factors. The advantages and process of application of GBM has been demonstrated with examples which eliminate the biases from most confounding variables between groups. This serves to modify the estimation of treatment effects of Shuxuetong injection on ALT index making the results more reliable. Based on large scale clinical observational data from HIS database, significant effects of Shuxuetong injection on abnormal changes in ALT have not been found.
Tehran Survey of Potential Risk Factors for Multiple Births.
Omani Samani, Reza; Almasi-Hashiani, Amir; Vesali, Samira; Shokri, Fatemeh; Cheraghi, Rezvaneh; Torkestani, Farahnaz; Sepidarkish, Mahdi
2017-10-01
The multiple pregnancy incidence is increasing worldwide. This increased incidence is concerning to the health care system. This study aims to determine the frequency of multiple pregnancy and identify factors that affect this frequency in Tehran, Iran. This cross-sectional study included 5170 mothers in labor between July 6-21, 2015 from 103 hospitals with Obstetrics and Gynecology Wards. The questionnaire used in this study consisted of five parts: demographic characteristics; information related to pregnancy; information related to the infant; information regarding the multiple pregnancy; and information associated with infertility. We recruited 103 trained midwives to collect data related to the questionnaire from eligible participants through an interview and medical records review. Frequencies and odds ratios (OR) for the association between multiple pregnancy and the selected characteristics (maternal age, economic status, history of multiple pregnancy in first-degree relatives, and reproductive history) were computed by multiple logistic regression. Stata software, version 13 (Stata Corp, College Station, TX, USA) was used for all statistical analyses. Multiple pregnancy had a prevalence of 1.48% [95% confidence interval (CI): 1.19-1.85]. After controlling for confounding variables, we observed a significant association between frequency of multiple pregnancy and mother's age (OR=1.04, 95% CI: 1.001-1.09, P=0.044), assisted reproductive technique (ART, OR=6.11, 95% CI: 1.7- 21.97, P=0.006), and history of multiple pregnancy in the mother's family (OR=5.49, 95% CI: 3.55-9.93, P=0.001). The frequency of multiple pregnancy approximated results reported in previous studies in Iran. Based on the results, we observed significantly greater frequency of multiple pregnancy in older women, those with a history of ART, and a history of multiple pregnancy in the mother's family compared to the other variables. Copyright© by Royan Institute. All rights reserved.
Chiba, Yasutaka
2014-01-01
Questions of mediation are often of interest in reasoning about mechanisms, and methods have been developed to address these questions. However, these methods make strong assumptions about the absence of confounding. Even if exposure is randomized, there may be mediator-outcome confounding variables. Inference about direct and indirect effects is particularly challenging if these mediator-outcome confounders are affected by the exposure because in this case these effects are not identified irrespective of whether data is available on these exposure-induced mediator-outcome confounders. In this paper, we provide a sensitivity analysis technique for natural direct and indirect effects that is applicable even if there are mediator-outcome confounders affected by the exposure. We give techniques for both the difference and risk ratio scales and compare the technique to other possible approaches. PMID:25580387
[Adoption of new technologies by health services: the challenge of analyzing relevant factors].
Trindade, Evelinda
2008-05-01
The exponential increase in the incorporation of health technologies has been considered a key factor in increased expenditures by the health sector. Such decisions involve multiple levels and stakeholders. Decentralization has multiplied the decision-making levels, with numerous difficult choices and limited resources. The interrelationship between stakeholders is complex, in creative systems with multiple determinants and confounders. The current review discusses the interaction between the factors influencing the decisions to incorporate technologies by health services, and proposes a structure for their analysis. The application and intensity of these factors in decision-making and the incorporation of products and programs by health services shapes the installed capacity of local and regional networks and modifies the health system. Empirical observation of decision-making and technology incorporation in Brazilian health services poses an important challenge. The structured recognition and measurement of these variables can assist proactive planning of health services.
Controlling for confounding variables in MS-omics protocol: why modularity matters.
Smith, Rob; Ventura, Dan; Prince, John T
2014-09-01
As the field of bioinformatics research continues to grow, more and more novel techniques are proposed to meet new challenges and improvements upon solutions to long-standing problems. These include data processing techniques and wet lab protocol techniques. Although the literature is consistently thorough in experimental detail and variable-controlling rigor for wet lab protocol techniques, bioinformatics techniques tend to be less described and less controlled. As the validation or rejection of hypotheses rests on the experiment's ability to isolate and measure a variable of interest, we urge the importance of reducing confounding variables in bioinformatics techniques during mass spectrometry experimentation. © The Author 2013. Published by Oxford University Press. For Permissions, please email: journals.permissions@oup.com.
Assessing mediation using marginal structural models in the presence of confounding and moderation
Coffman, Donna L.; Zhong, Wei
2012-01-01
This paper presents marginal structural models (MSMs) with inverse propensity weighting (IPW) for assessing mediation. Generally, individuals are not randomly assigned to levels of the mediator. Therefore, confounders of the mediator and outcome may exist that limit causal inferences, a goal of mediation analysis. Either regression adjustment or IPW can be used to take confounding into account, but IPW has several advantages. Regression adjustment of even one confounder of the mediator and outcome that has been influenced by treatment results in biased estimates of the direct effect (i.e., the effect of treatment on the outcome that does not go through the mediator). One advantage of IPW is that it can properly adjust for this type of confounding, assuming there are no unmeasured confounders. Further, we illustrate that IPW estimation provides unbiased estimates of all effects when there is a baseline moderator variable that interacts with the treatment, when there is a baseline moderator variable that interacts with the mediator, and when the treatment interacts with the mediator. IPW estimation also provides unbiased estimates of all effects in the presence of non-randomized treatments. In addition, for testing mediation we propose a test of the null hypothesis of no mediation. Finally, we illustrate this approach with an empirical data set in which the mediator is continuous, as is often the case in psychological research. PMID:22905648
Assessing mediation using marginal structural models in the presence of confounding and moderation.
Coffman, Donna L; Zhong, Wei
2012-12-01
This article presents marginal structural models with inverse propensity weighting (IPW) for assessing mediation. Generally, individuals are not randomly assigned to levels of the mediator. Therefore, confounders of the mediator and outcome may exist that limit causal inferences, a goal of mediation analysis. Either regression adjustment or IPW can be used to take confounding into account, but IPW has several advantages. Regression adjustment of even one confounder of the mediator and outcome that has been influenced by treatment results in biased estimates of the direct effect (i.e., the effect of treatment on the outcome that does not go through the mediator). One advantage of IPW is that it can properly adjust for this type of confounding, assuming there are no unmeasured confounders. Further, we illustrate that IPW estimation provides unbiased estimates of all effects when there is a baseline moderator variable that interacts with the treatment, when there is a baseline moderator variable that interacts with the mediator, and when the treatment interacts with the mediator. IPW estimation also provides unbiased estimates of all effects in the presence of nonrandomized treatments. In addition, for testing mediation we propose a test of the null hypothesis of no mediation. Finally, we illustrate this approach with an empirical data set in which the mediator is continuous, as is often the case in psychological research. PsycINFO Database Record (c) 2013 APA, all rights reserved.
Directed acyclic graphs (DAGs): an aid to assess confounding in dental research.
Merchant, Anwar T; Pitiphat, Waranuch
2002-12-01
Confounding, a special type of bias, occurs when an extraneous factor is associated with the exposure and independently affects the outcome. In order to get an unbiased estimate of the exposure-outcome relationship, we need to identify potential confounders, collect information on them, design appropriate studies, and adjust for confounding in data analysis. However, it is not always clear which variables to collect information on and adjust for in the analyses. Inappropriate adjustment for confounding can even introduce bias where none existed. Directed acyclic graphs (DAGs) provide a method to select potential confounders and minimize bias in the design and analysis of epidemiological studies. DAGs have been used extensively in expert systems and robotics. Robins (1987) introduced the application of DAGs in epidemiology to overcome shortcomings of traditional methods to control for confounding, especially as they related to unmeasured confounding. DAGs provide a quick and visual way to assess confounding without making parametric assumptions. We introduce DAGs, starting with definitions and rules for basic manipulation, stressing more on applications than theory. We then demonstrate their application in the control of confounding through examples of observational and cross-sectional epidemiological studies.
Peres, Maria Fernanda Tourinho; Cardia, Nancy; de Mesquita Neto, Paulo; Dos Santos, Patrícia Carla; Adorno, Sérgio
2008-04-01
To analyze the association between police violence and homicide mortality rates taking into consideration the effect of contextual variables. This was an environmental, cross-sectional study that included the 96 census districts in the City of São Paulo. The association between the variables was analyzed using Spearman's rank correlation and simple and multiple regression analysis. Univariate analysis revealed a strong and significant association between homicide mortality coefficients and all the indicators of socioeconomic development and police violence. After controlling for potential confounding factors, the association between police violence and homicide mortality coefficients remained strong and significant. This significance was lost only after control for the size of the resident population. The results indicate that police action that violates basic human rights is not the right answer to urban violence. The combination of homicides from interpersonal violence and deaths from police violence results in negative socialization and promotes further violence.
Farbu, Jorunn; Haugen, Margaretha; Meltzer, Helle Margrete; Brantsæter, Anne Lise
2014-12-05
Little attention has been given to the impact of singlehood during pregnancy. The aim of this study was to examine the impact of marital status on diet during pregnancy and pregnancy outcome. The study population comprised 62,773 women participating in the Norwegian Mother and Child Cohort Study. Marital status was categorised into singles living alone, singles living with parents and married/cohabiting (reference group). Participants answered a general health questionnaire in gestational week 15-17 and a food frequency questionnaire in gestational week 22. We used nonparametric tests to compare dietary intakes by marital status, and multiple logistic regression to estimate odds ratios (OR) and 95% confidence intervals (CI) for infants being small for gestational age (SGA), large for gestational age (LGA), and preterm delivery (defined as delivery before gestational week 37). Single women living with parents had lower intakes of fruits and vegetables, higher intake of total energy, higher proportion of energy from added sugar, and lower intake of fibre than the reference group. Singles living alone also had a higher intake of added sugar. In both of the single groups, daily smoking was more prevalent than in women living with a partner. In analyses adjusted for maternal age, pre-pregnancy BMI, energy intake, energy contributed by protein, education, income, parity and nausea, single women living alone had increased risk of SGA with OR = 1.27 (95% CI: 1.05, 1.55). When smoking was included among the confounding variables, the association was no longer significant. Likewise, singles living alone had increased risk of preterm delivery, with OR = 1.32 (95% CI: 1.01, 1.72) in a partly adjusted model, but the association did not remain significant in a model fully adjusted for confounding variables. Single mothers had lower dietary quality and included more smokers than women who lived with a partner. Single mothers living alone had higher prevalence of SGA and preterm delivery, but the associations with adverse pregnancy outcomes were confounded by other variables. This study shows that single mothers should be given special attention during antenatal care and counselling.
Jones, Jason D.; Calkins, Monica E.; Scott, J. Cobb; Bach, Emily C.; Gur, Raquel E.
2017-01-01
Purpose To examine how cannabis use and polysubstance use among cannabis users relate to psychosis spectrum (PS) symptoms in a large community-based sample of US youth. Methods 4,171 youth (age 14–21; mean=16.90, SD=1.85; 55% female) from the Philadelphia Neurodevelopmental Cohort (PNC) completed assessments of substance use, PS symptoms, and confounding variables (e.g., demographics, comorbid psychopathology, trauma exposure). Results After adjusting for confounds, cannabis use by itself was not associated with increased odds of being classified as “psychosis spectrum.” However, cannabis use in combination with tobacco or other substance use was associated with increased odds of PS classification (adjusted odds ratios [ORs] = 1.37 to 1.76). Follow-up symptom-level analyses revealed that cannabis use in combination with other substances was associated with subclinical positive symptoms (ORs=1.95 and 2.24) and frequent cannabis use was associated with subclinical negative/disorganized symptoms (OR=2.14). However, these symptom-level findings were reduced to trends after correction for multiple comparisons. Neither cannabis use nor polysubstance use was associated with threshold delusions or hallucinations. Conclusions After adjusting for important confounds, there was minimal evidence for associations between cannabis use by itself and PS symptoms. More compelling evidence emerged for associations between polysubstance use among cannabis users and PS symptoms. This study highlights the importance of considering polysubstance use and confounds when examining associations between cannabis use and PS symptoms. Further longitudinal research is necessary to determine whether these findings represent causal associations or shared genetic and/or environmental vulnerability for substance use and PS symptoms. PMID:28318911
Luvizutto, Gustavo José; Moliga, Augusta Fabiana; Rizzatti, Gabriela Rizzo Soares; Fogaroli, Marcelo Ortolani; Moura Neto, Eduardo de; Nunes, Hélio Rubens de Carvalho; Resende, Luiz Antônio de Lima; Bazan, Rodrigo
2018-05-21
The aim of this study was to assess the relationship between the degree of unilateral spatial neglect during the acute phase of stroke and long-term functional independence. This was a prospective study of right ischemic stroke patients in which the independent variable was the degree of spatial neglect and the outcome that was measured was functional independence. The potential confounding factors included sex, age, stroke severity, topography of the lesion, risk factors, glycemia and the treatment received. Unilateral spatial neglect was measured using the line cancellation test, the star cancellation test and the line bisection test within 48 hours of the onset of symptoms. Functional independence was measured using the modified Rankin and Barthel scales at 90 days after discharge. The relationship between unilateral spatial neglect and functional independence was analyzed using multiple logistic regression that was corrected for confounding factors. We studied 60 patients with a median age of 68 (34-89) years, 52% of whom were male and 74% of whom were Caucasian. The risk for moderate to severe disability increased with increasing star cancellation test scores (OR=1.14 [1.03-1.26], p=0.01) corrected for the stroke severity, which was a confounding factor that had a statistically positive association with disability (OR=1.63 [1.13-2.65], p=0.01). The best chance of functional independence decreased with increasing star cancellation test scores (OR=0.86 [0.78-0.96], p=0.006) corrected for the stroke severity, which was a confounding factor that had a statistically negative association with independence (OR=0.66 [0.48-0.92], p=0.017). The severity of unilateral spatial neglect in acute stroke worsens the degree of long-term disability and functional independence.
Jones, Jason D; Calkins, Monica E; Scott, J Cobb; Bach, Emily C; Gur, Raquel E
2017-06-01
To examine how cannabis use and polysubstance use among cannabis users relate to psychosis spectrum (PS) symptoms in a large community-based sample of U.S. youth. Four thousand one hundred seventy-one youths (aged 14-21 years; mean = 16.90 years, SD = 1.85; 55% female) from the Philadelphia Neurodevelopmental Cohort completed assessments of substance use, PS symptoms, and confounding variables (e.g., demographics, comorbid psychopathology, and trauma exposure). After adjusting for confounds, cannabis use by itself was not associated with increased odds of being classified as "psychosis spectrum." However, cannabis use in combination with tobacco or other substance use was associated with increased odds of PS classification (adjusted odds ratios [ORs] = 1.37-1.76). Follow-up symptom-level analyses revealed that cannabis use in combination with other substances was associated with subclinical positive symptoms (ORs = 1.95 and 2.24) and frequent cannabis use was associated with subclinical negative/disorganized symptoms (OR = 2.14). However, these symptom-level findings were reduced to trends after correction for multiple comparisons. Neither cannabis use nor polysubstance use was associated with threshold delusions or hallucinations. After adjusting for important confounds, there was minimal evidence for associations between cannabis use by itself and PS symptoms. More compelling evidence emerged for associations between polysubstance use among cannabis users and PS symptoms. This study highlights the importance of considering polysubstance use and confounds when examining associations between cannabis use and PS symptoms. Further longitudinal research is necessary to determine whether these findings represent causal associations or shared genetic and/or environmental vulnerability for substance use and PS symptoms. Copyright © 2017 Society for Adolescent Health and Medicine. Published by Elsevier Inc. All rights reserved.
Overcoming ecologic bias using the two-phase study design.
Wakefield, Jon; Haneuse, Sebastien J-P A
2008-04-15
Ecologic (aggregate) data are widely available and widely utilized in epidemiologic studies. However, ecologic bias, which arises because aggregate data cannot characterize within-group variability in exposure and confounder variables, can only be removed by supplementing ecologic data with individual-level data. Here the authors describe the two-phase study design as a framework for achieving this objective. In phase 1, outcomes are stratified by any combination of area, confounders, and error-prone (or discretized) versions of exposures of interest. Phase 2 data, sampled within each phase 1 stratum, provide accurate measures of exposure and possibly of additional confounders. The phase 1 aggregate-level data provide a high level of statistical power and a cross-classification by which individuals may be efficiently sampled in phase 2. The phase 2 individual-level data then provide a control for ecologic bias by characterizing the within-area variability in exposures and confounders. In this paper, the authors illustrate the two-phase study design by estimating the association between infant mortality and birth weight in several regions of North Carolina for 2000-2004, controlling for gender and race. This example shows that the two-phase design removes ecologic bias and produces gains in efficiency over the use of case-control data alone. The authors discuss the advantages and disadvantages of the approach.
Keller, Matthew C
2014-01-01
Candidate gene × environment (G × E) interaction research tests the hypothesis that the effects of some environmental variable (e.g., childhood maltreatment) on some outcome measure (e.g., depression) depend on a particular genetic polymorphism. Because this research is inherently nonexperimental, investigators have been rightly concerned that detected interactions could be driven by confounders (e.g., ethnicity, gender, age, socioeconomic status) rather than by the specified genetic or environmental variables per se. In an attempt to eliminate such alternative explanations for detected G × E interactions, investigators routinely enter the potential confounders as covariates in general linear models. However, this practice does not control for the effects these variables might have on the G × E interaction. Rather, to properly control for confounders, researchers need to enter the covariate × environment and the covariate × gene interaction terms in the same model that tests the G × E term. In this manuscript, I demonstrate this point analytically and show that the practice of improperly controlling for covariates is the norm in the G × E interaction literature to date. Thus, many alternative explanations for G × E findings that investigators had thought were eliminated have not been. © 2013 Society of Biological Psychiatry Published by Society of Biological Psychiatry All rights reserved.
Solar radiation increases suicide rate after adjusting for other climate factors in South Korea.
Jee, Hee-Jung; Cho, Chul-Hyun; Lee, Yu Jin; Choi, Nari; An, Hyonggin; Lee, Heon-Jeong
2017-03-01
Previous studies have indicated that suicide rates have significant seasonal variations. There is seasonal discordance between temperature and solar radiation due to the monsoon season in South Korea. We investigated the seasonality of suicide and assessed its association with climate variables in South Korea. Suicide rates were obtained from the National Statistical Office of South Korea, and climatic data were obtained from the Korea Meteorological Administration for the period of 1992-2010. We conducted analyses using a generalized additive model (GAM). First, we explored the seasonality of suicide and climate variables such as mean temperature, daily temperature range, solar radiation, and relative humidity. Next, we identified confounding climate variables associated with suicide rate. To estimate the adjusted effect of solar radiation on the suicide rate, we investigated the confounding variables using a multivariable GAM. Suicide rate showed seasonality with a pattern similar to that of solar radiation. We found that the suicide rate increased 1.008 times when solar radiation increased by 1 MJ/m 2 after adjusting for other confounding climate factors (P < 0.001). Solar radiation has a significant linear relationship with suicide after adjusting for region, other climate variables, and time trends. © 2016 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Fisk, William J.; Eliseeva, Ekaterina A.; Mendell, Mark J.
Dampness and mold have been shown in qualitative reviews to be associated with a variety of adverse respiratory health effects, including respiratory tract infections. Several published meta-analyses have provided quantitative summaries for some of these associations, but not for respiratory infections. Demonstrating a causal relationship between dampness-related agents, which are preventable exposures, and respiratory tract infections would suggest important new public health strategies. We report the results of quantitative meta-analyses of published studies that examined the association of dampness or mold in homes with respiratory infections and bronchitis. For primary studies meeting eligibility criteria, we transformed reported odds ratios (ORs)more » and confidence intervals (CIs) to the log scale. Both fixed and random effects models were applied to the log ORs and their variances. Most studies contained multiple estimated ORs. Models accounted for the correlation between multiple results within the studies analyzed. One set of analyses was performed with all eligible studies, and another set restricted to studies that controlled for age, gender, smoking, and socioeconomic status. Subgroups of studies were assessed to explore heterogeneity. Funnel plots were used to assess publication bias. The resulting summary estimates of ORs from random effects models based on all studies ranged from 1.38 to 1.50, with 95% CIs excluding the null in all cases. Use of different analysis models and restricting analyses based on control of multiple confounding variables changed findings only slightly. ORs (95% CIs) from random effects models using studies adjusting for major confounding variables were, for bronchitis, 1.45 (1.32-1.59); for respiratory infections, 1.44 (1.31-1.59); for respiratory infections excluding nonspecific upper respiratory infections, 1.50 (1.32-1.70), and for respiratory infections in children or infants, 1.48 (1.33-1.65). Little effect of publication bias was evident. Estimated attributable risk proportions ranged from 8% to 20%. Residential dampness and mold are associated with substantial and statistically significant increases in both respiratory infections and bronchitis. If these associations were confirmed as causal, effective control of dampness and mold in buildings would prevent a substantial proportion of respiratory infections.« less
Parker, Nicole R; Hudson, Amanda L; Khong, Peter; Parkinson, Jonathon F; Dwight, Trisha; Ikin, Rowan J; Zhu, Ying; Cheng, Zhangkai Jason; Vafaee, Fatemeh; Chen, Jason; Wheeler, Helen R; Howell, Viive M
2016-03-04
Heterogeneity is a hallmark of glioblastoma with intratumoral heterogeneity contributing to variability in responses and resistance to standard treatments. Promoter methylation status of the DNA repair enzyme O(6)-methylguanine DNA methyltransferase (MGMT) is the most important clinical biomarker in glioblastoma, predicting for therapeutic response. However, it does not always correlate with response. This may be due to intratumoral heterogeneity, with a single biopsy unlikely to represent the entire lesion. Aberrations in other DNA repair mechanisms may also contribute. This study investigated intratumoral heterogeneity in multiple glioblastoma tumors with a particular focus on the DNA repair pathways. Transcriptional intratumoral heterogeneity was identified in 40% of cases with variability in MGMT methylation status found in 14% of cases. As well as identifying intratumoral heterogeneity at the transcriptional and epigenetic levels, targeted next generation sequencing identified between 1 and 37 unique sequence variants per specimen. In-silico tools were then able to identify deleterious variants in both the base excision repair and the mismatch repair pathways that may contribute to therapeutic response. As these pathways have roles in temozolomide response, these findings may confound patient management and highlight the importance of assessing multiple tumor biopsies.
Obesity and the community food environment: a systematic review.
Holsten, Joanna E
2009-03-01
To examine the relationship between obesity and the community and/or consumer food environment. A comprehensive literature search of multiple databases was conducted and seven studies were identified for review. Studies were selected if they measured BMI and environmental variables related to food outlets. Environmental variables included the geographic arrangement of food stores or restaurants in communities and consumer conditions such as food price and availability within each outlet. The study designs, methods, limitations and results related to obesity and the food environment were reviewed, and implications for future research were synthesized. The reviewed studies used cross-sectional designs to examine the community food environment defined as the number per capita, proximity or density of food outlets. Most studies indirectly identified food outlets through large databases. The studies varied substantially in sample populations, outcome variables, units of measurement and data analysis. Two studies did not find any significant association between obesity rates and community food environment variables. Five studies found significant results. Many of the studies were subject to limitations that may have mitigated the validity of the results. Research examining obesity and the community or consumer food environment is at an early stage. The most pertinent gaps include primary data at the individual level, direct measures of the environment, studies examining the consumer environment and study designs involving a time sequence. Future research should directly measure multiple levels of the food environment and key confounders at the individual level.
Fischer, A; Delagarde, R; Faverdin, P
2018-05-01
Residual feed intake, which is usually used to estimate individual variation of feed efficiency, requires frequent and accurate measurements of individual feed intake to be carried out. Developing a breeding scheme based on residual feed intake in dairy cows is therefore complicated, especially because feed intake is not measurable for a large population. Another solution could be to focus on biological determinants of feed efficiency, which could potentially be directly and broadband measurable on farm. Several phenotypes have been identified in literature as being associated with differences in feed efficiency. The present study therefore aims to identify which biological mechanisms are associated with residual energy intake (REI) differences among dairy cows. Several candidate phenotypes were recorded frequently and simultaneously throughout the first 238 d in milk for 60 Holstein cows fed on a constant diet based on maize silage. A multiple linear regression of the 238 d in milk average of net energy intake was fitted on the 238 d in milk averages for milk energy output, metabolic body weight, the sum over the 238 d in milk of both, body condition score loss and gain, and the residuals were defined as REI. A partial least square regression was fitted over all biological traits to explain REI variability. Linear multiple regression explained 93.6% of net energy intake phenotypic variation, with 65.5% associated with lactation requirement, 23.2% with maintenance, and 4.9% with body reserves change; the 6.4% residuals represented REI. Overall, measured biological traits contributed to 58.9% of REI phenotypic variability, which were mainly explained by activity (26.5%) and feeding behavior (21.3%). However, apparent confounding was observed between behavior, activity, digestibility, and rumen-temperature variables. Drawing a conclusion on biological traits that explain feed efficiency differences among dairy cows was not possible due to this apparent confounding between the measured variables. Further investigation is needed to validate these results and to characterize the causal relationship of feed efficiency with feeding behavior, digestibility, body reserves change, activity, and rumen temperature. The Authors. Published by FASS Inc. and Elsevier Inc. on behalf of the American Dairy Science Association®. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/3.0/).
Mismeasurement and the resonance of strong confounders: correlated errors.
Marshall, J R; Hastrup, J L; Ross, J S
1999-07-01
Confounding in epidemiology, and the limits of standard methods of control for an imperfectly measured confounder, have been understood for some time. However, most treatments of this problem are based on the assumption that errors of measurement in confounding and confounded variables are independent. This paper considers the situation in which a strong risk factor (confounder) and an inconsequential but suspected risk factor (confounded) are each measured with errors that are correlated; the situation appears especially likely to occur in the field of nutritional epidemiology. Error correlation appears to add little to measurement error as a source of bias in estimating the impact of a strong risk factor: it can add to, diminish, or reverse the bias induced by measurement error in estimating the impact of the inconsequential risk factor. Correlation of measurement errors can add to the difficulty involved in evaluating structures in which confounding and measurement error are present. In its presence, observed correlations among risk factors can be greater than, less than, or even opposite to the true correlations. Interpretation of multivariate epidemiologic structures in which confounding is likely requires evaluation of measurement error structures, including correlations among measurement errors.
Mascha, Edward J; Dalton, Jarrod E; Kurz, Andrea; Saager, Leif
2013-10-01
In comparative clinical studies, a common goal is to assess whether an exposure, or intervention, affects the outcome of interest. However, just as important is to understand the mechanism(s) for how the intervention affects outcome. For example, if preoperative anemia was shown to increase the risk of postoperative complications by 15%, it would be important to quantify how much of that effect was due to patients receiving intraoperative transfusions. Mediation analysis attempts to quantify how much, if any, of the effect of an intervention on outcome goes though prespecified mediator, or "mechanism" variable(s), that is, variables sitting on the causal pathway between exposure and outcome. Effects of an exposure on outcome can thus be divided into direct and indirect, or mediated, effects. Mediation is claimed when 2 conditions are true: the exposure affects the mediator and the mediator (adjusting for the exposure) affects the outcome. Understanding how an intervention affects outcome can validate or invalidate one's original hypothesis and also facilitate further research to modify the responsible factors, and thus improve patient outcome. We discuss the proper design and analysis of studies investigating mediation, including the importance of distinguishing mediator variables from confounding variables, the challenge of identifying potential mediators when the exposure is chronic versus acute, and the requirements for claiming mediation. Simple designs are considered, as well as those containing multiple mediators, multiple outcomes, and mixed data types. Methods are illustrated with data collected by the National Surgical Quality Improvement Project (NSQIP) and utilized in a companion paper which assessed the effects of preoperative anemic status on postoperative outcomes.
Geeleher, Paul; Cox, Nancy J; Huang, R Stephanie
2016-09-21
We show that variability in general levels of drug sensitivity in pre-clinical cancer models confounds biomarker discovery. However, using a very large panel of cell lines, each treated with many drugs, we could estimate a general level of sensitivity to all drugs in each cell line. By conditioning on this variable, biomarkers were identified that were more likely to be effective in clinical trials than those identified using a conventional uncorrected approach. We find that differences in general levels of drug sensitivity are driven by biologically relevant processes. We developed a gene expression based method that can be used to correct for this confounder in future studies.
Johnston, K M; Gustafson, P; Levy, A R; Grootendorst, P
2008-04-30
A major, often unstated, concern of researchers carrying out epidemiological studies of medical therapy is the potential impact on validity if estimates of treatment are biased due to unmeasured confounders. One technique for obtaining consistent estimates of treatment effects in the presence of unmeasured confounders is instrumental variables analysis (IVA). This technique has been well developed in the econometrics literature and is being increasingly used in epidemiological studies. However, the approach to IVA that is most commonly used in such studies is based on linear models, while many epidemiological applications make use of non-linear models, specifically generalized linear models (GLMs) such as logistic or Poisson regression. Here we present a simple method for applying IVA within the class of GLMs using the generalized method of moments approach. We explore some of the theoretical properties of the method and illustrate its use within both a simulation example and an epidemiological study where unmeasured confounding is suspected to be present. We estimate the effects of beta-blocker therapy on one-year all-cause mortality after an incident hospitalization for heart failure, in the absence of data describing disease severity, which is believed to be a confounder. 2008 John Wiley & Sons, Ltd
Color discrimination errors associate with axial motor impairments in Parkinson's disease.
Bohnen, Nicolaas I; Haugen, Jacob; Ridder, Andrew; Kotagal, Vikas; Albin, Roger L; Frey, Kirk A; Müller, Martijn L T M
2017-01-01
Visual function deficits are more common in imbalance-predominant compared to tremor-predominant PD suggesting a pathophysiological role of impaired visual functions in axial motor impairments. To investigate the relationship between changes in color discrimination and motor impairments in PD while accounting for cognitive or other confounder factors. PD subjects (n=49, age 66.7±8.3 years; Hoehn & Yahr stage 2.6±0.6) completed color discrimination assessment using the Farnsworth-Munsell 100 Hue Color Vision Test, neuropsychological, motor assessments and [ 11 C]dihydrotetrabenazine vesicular monoamine transporter type 2 PET imaging. MDS-UPDRS sub-scores for cardinal motor features were computed. Timed up and go mobility and walking tests were assessed in 48 subjects. Bivariate correlation coefficients between color discrimination and motor variables were significant only for the Timed up and go (R S =0.44, P=0.0018) and the MDS-UPDRS axial motor scores (R S =0.38, P=0.0068). Multiple regression confounder analysis using the Timed up and go as outcome parameter showed a significant total model (F (5,43) = 7.3, P<0.0001) with significant regressor effects for color discrimination (standardized β=0.32, t=2.6, P=0.012), global cognitive Z-score (β=-0.33, t=-2.5, P=0.018), duration of disease (β=0.26, t=1.8, P=0.038), but not for age or striatal dopaminergic binding. The color discrimination test was also a significant independent regressor in the MDS-UPDRS axial motor model (standardized β=0.29, t=2.4, P=0.022; total model t (5,43) = 6.4, P=0.0002). Color discrimination errors associate with axial motor features in PD independent of cognitive deficits, nigrostriatal dopaminergic denervation, and other confounder variables. These findings may reflect shared pathophysiology between color discrimination visual impairments and axial motor burden in PD.
Kangaroo-mother care method and neurobehavior of preterm infants.
Silva, Margareth Gurgel de Castro; Barros, Marina Carvalho de Moraes; Pessoa, Úrsula Maria Lima; Guinsburg, Ruth
2016-04-01
To evaluate the effect of kangaroo-mother care (KMC) in preterm (PT) neurobehavior between 36 and 41 weeks post-conceptual age (PCA). A prospective cohort of 61 preterm infants with gestational age (GA) of 28-32 w evaluated by the Neonatal Intensive Care Unit Network Neurobehavioral Scale (NNNS), with 36-41 w PCA. Infants with clinical instability were excluded. They were analyzed in 2 groups: - Kangaroo (KAN): KMC for 7 or more days; Conventional (CON): did not receive KMC. Scores of the 13 NNNS variables were compared between groups and the effect of KMC in the scores of the variables of NNNS were evaluated by multiple linear regression, controlling for confounders. The KAN groups (n=24) and CON (n=37) were similar regarding main demographic and clinical maternal and neonatal characteristics. Mean GA was 30.3 w; and birth weight was 1170 g for both groups. PT of KAN group were admitted in KMC with PCA of 35.8 w (38.5 days of life) and remained with this care for 14.3 days. The NNNS was applied 13 days after the start of KMC. PT submitted to KMC showed higher quality of movements (KAN: 4.98 ± 0.53 vs CON: 4.53 ± 0.47; p=0.001) and lower scores on Signs of stress and abstinence (KAN: 0.03 ± 0.03 vs CON: 0.05 ± 0.03; p=0.001). Controlling for confounders, the KMC was associated with higher scores on the variables Attention, Quality of movements, and lower scores on Asymmetry and Signs of stress and abstinence. PT submitted to the KMC, compared to those non-submitted, have better neurobehavior performance between 36 and 41 weeks of post-conceptual age. Copyright © 2016. Published by Elsevier Ireland Ltd.
The association between subgingival periodontal pathogens and systemic inflammation.
Winning, Lewis; Patterson, Christopher C; Cullen, Kathy M; Stevenson, Kathryn A; Lundy, Fionnuala T; Kee, Frank; Linden, Gerard J
2015-09-01
To investigate associations between periodontal disease pathogens and levels of systemic inflammation measured by C-reactive protein (CRP). A representative sample of dentate 60-70-year-old men in Northern Ireland had a comprehensive periodontal examination. Men taking statins were excluded. Subgingival plaque samples were analysed by quantitative real time PCR to identify the presence of Aggregatibacter actinomycetemcomitans, Porphyromonas gingivalis, Treponema denticola and Tannerella forsythia. High-sensitivity CRP (mg/l) was measured from fasting blood samples. Multiple linear regression analysis was performed using log-transformed CRP concentration as the dependent variable, with the presence of each periodontal pathogen as predictor variables, with adjustment for various potential confounders. A total of 518 men (mean age 63.6 SD 3.0 years) were included in the analysis. Multiple regression analysis showed that body mass index (p < 0.001), current smoking (p < 0.01), the detectable presence of P. gingivalis (p < 0.01) and hypertension (p = 0.01), were independently associated with an increased CRP. The detectable presence of P. gingivalis was associated with a 20% (95% confidence interval 4-35%) increase in CRP (mg/l) after adjustment for all other predictor variables. In these 60-70-year-old dentate men, the presence of P. gingivalis in subgingival plaque was significantly associated with a raised level of C-reactive protein. © 2015 John Wiley & Sons A/S. Published by John Wiley & Sons Ltd.
Instrumental variables and Mendelian randomization with invalid instruments
NASA Astrophysics Data System (ADS)
Kang, Hyunseung
Instrumental variables (IV) methods have been widely used to determine the causal effect of a treatment, exposure, policy, or an intervention on an outcome of interest. The IV method relies on having a valid instrument, a variable that is (A1) associated with the exposure, (A2) has no direct effect on the outcome, and (A3) is unrelated to the unmeasured confounders associated with the exposure and the outcome. However, in practice, finding a valid instrument, especially those that satisfy (A2) and (A3), can be challenging. For example, in Mendelian randomization studies where genetic markers are used as instruments, complete knowledge about instruments' validity is equivalent to complete knowledge about the involved genes' functions. The dissertation explores the theory, methods, and application of IV methods when invalid instruments are present. First, when we have multiple candidate instruments, we establish a theoretical bound whereby causal effects are only identified as long as less than 50% of instruments are invalid, without knowing which of the instruments are invalid. We also propose a fast penalized method, called sisVIVE, to estimate the causal effect. We find that sisVIVE outperforms traditional IV methods when invalid instruments are present both in simulation studies as well as in real data analysis. Second, we propose a robust confidence interval under the multiple invalid IV setting. This work is an extension of our work on sisVIVE. However, unlike sisVIVE which is robust to violations of (A2) and (A3), our confidence interval procedure provides honest coverage even if all three assumptions, (A1)-(A3), are violated. Third, we study the single IV setting where the one IV we have may actually be invalid. We propose a nonparametric IV estimation method based on full matching, a technique popular in causal inference for observational data, that leverages observed covariates to make the instrument more valid. We propose an estimator along with inferential results that are robust to mis-specifications of the covariate-outcome model. We also provide a sensitivity analysis should the instrument turn out to be invalid, specifically violate (A3). Fourth, in application work, we study the causal effect of malaria on stunting among children in Ghana. Previous studies on the effect of malaria and stunting were observational and contained various unobserved confounders, most notably nutritional deficiencies. To infer causality, we use the sickle cell genotype, a trait that confers some protection against malaria and was randomly assigned at birth, as an IV and apply our nonparametric IV method. We find that the risk of stunting increases by 0.22 (95% CI: 0.044,1) for every malaria episode and is sensitive to unmeasured confounders.
Siegmund, Gunter P; Sanderson, David J; Myers, Barry S; Inglis, J Timothy
2003-04-01
To examine whether habituation confounds the study of whiplash injury using human subjects, we quantified changes in the magnitude and temporal development of the neck muscle electromyogram and peak linear and angular head/torso kinematics of subjects exposed to sequential whiplash-like perturbations. Forty-four seated subjects (23F, 21M) underwent 11 consecutive forward horizontal perturbations (peak sled acceleration=1.5 g). Electromyographic (EMG) activity was recorded over the sternocleidomastoid (SCM) and cervical paraspinal (PARA) muscles with surface electrodes, and head and torso kinematics were measured using linear and angular accelerometers and a 3D motion analysis system. EMG onset occurred at reflex latencies (67-75 ms in SCM) and did not vary with repeated perturbations. EMG amplitude was significantly attenuated by the second perturbation in PARA muscles and by the third perturbation in SCM muscles. The mean decrement in EMG amplitude between the first trial and the mean of the last five trials was between 41% and 64%. Related kinematic changes ranged from a 21% increase in head extension angle to a 29% decrease in forward acceleration at the forehead, and were also significantly different by the second exposure in some variables. Although a wider range of perturbation intensities and inter-perturbation intervals need to be studied, the significant changes observed in both muscle and kinematic variables by the second perturbation indicated that habituation was a potential confounder of whiplash injury studies using repeated perturbations of human subjects.
Simple and multiple linear regression: sample size considerations.
Hanley, James A
2016-11-01
The suggested "two subjects per variable" (2SPV) rule of thumb in the Austin and Steyerberg article is a chance to bring out some long-established and quite intuitive sample size considerations for both simple and multiple linear regression. This article distinguishes two of the major uses of regression models that imply very different sample size considerations, neither served well by the 2SPV rule. The first is etiological research, which contrasts mean Y levels at differing "exposure" (X) values and thus tends to focus on a single regression coefficient, possibly adjusted for confounders. The second research genre guides clinical practice. It addresses Y levels for individuals with different covariate patterns or "profiles." It focuses on the profile-specific (mean) Y levels themselves, estimating them via linear compounds of regression coefficients and covariates. By drawing on long-established closed-form variance formulae that lie beneath the standard errors in multiple regression, and by rearranging them for heuristic purposes, one arrives at quite intuitive sample size considerations for both research genres. Copyright © 2016 Elsevier Inc. All rights reserved.
Effects of multiple predator species on green treefrog (Hyla cinerea) tadpoles
Gunzburger, M.S.; Travis, J.
2005-01-01
Prey species that occur across a range of habitats may be exposed to variable communities of multiple predator species across habitats. Predicting the combined effects of multiple predators can be complex. Many experiments evaluating the effects of multiple predators on prey confound either variation in predator density with predator identity or variation in relative predator frequency with overall predation rates. We develop a new experimental design of factorial predator combinations that maintains a constant expected predation rate, under the null hypothesis of additive predator effects. We implement this design to evaluate the combined effects of three predator species (bass, aeshnid and libellulid odonate naiads) on mortality rate of a prey species, Hyla cinerea (Schneider, 1799) tadpoles, that occurs across a range of aquatic habitats. Two predator treatments (libellulid and aeshnid + libellulid) resulted in lower tadpole mortality than any of the other predator treatments. Variation in tadpole mortality across treatments was not related to coarse variation in microhabitat use, but was likely due to intraguild predation, which occurred in all predator treatments. Hyla cinerea tadpoles have constant, low survival values when exposed to many different combinations of predator species, and predation rate probably increases linearly with predator density.
Mollenhauer, Robert; Mouser, Joshua B.; Brewer, Shannon K.
2018-01-01
Temporal and spatial variability in streams result in heterogeneous gear capture probability (i.e., the proportion of available individuals identified) that confounds interpretation of data used to monitor fish abundance. We modeled tow-barge electrofishing capture probability at multiple spatial scales for nine Ozark Highland stream fishes. In addition to fish size, we identified seven reach-scale environmental characteristics associated with variable capture probability: stream discharge, water depth, conductivity, water clarity, emergent vegetation, wetted width–depth ratio, and proportion of riffle habitat. The magnitude of the relationship between capture probability and both discharge and depth varied among stream fishes. We also identified lithological characteristics among stream segments as a coarse-scale source of variable capture probability. The resulting capture probability model can be used to adjust catch data and derive reach-scale absolute abundance estimates across a wide range of sampling conditions with similar effort as used in more traditional fisheries surveys (i.e., catch per unit effort). Adjusting catch data based on variable capture probability improves the comparability of data sets, thus promoting both well-informed conservation and management decisions and advances in stream-fish ecology.
Echo characteristics of two salmon species
NASA Astrophysics Data System (ADS)
Nealson, Patrick A.; Horne, John K.; Burwen, Debby L.
2005-04-01
The Alaska Department of Fish and Game relies on split-beam hydroacoustic techniques to estimate Chinook salmon (Oncorhynchus tshawytscha) returns to the Kenai River. Chinook counts are periodically confounded by large numbers of smaller sockeye salmon (O. nerka). Echo target-strength has been used to distinguish fish length classes, but was too variable to separate Kenai River chinook and sockeye distributions. To evaluate the efficacy of alternate echo metrics, controlled acoustic measurements of tethered chinook and sockeye salmon were collected at 200 kHz. Echo returns were digitally sampled at 48 kHz. A suite of descriptive metrics were collected from a series of 1,000 echoes per fish. Measurements of echo width were least variable at the -3 dB power point. Initial results show echo elongation and ping-to-ping variability in echo envelope width were significantly greater for chinook than for sockeye salmon. Chinook were also observed to return multiple discrete peaks from a single broadcast echo. These characteristics were attributed to the physical width of chinook exceeding half of the broadcast echo pulse width at certain orientations. Echo phase variability, correlation coefficient and fractal dimension distributions did not demonstrate significant discriminatory power between the two species. [Work supported by ADF&G, ONR.
Brain natriuretic peptide in patients with congestive heart failure and central sleep apnea.
Carmona-Bernal, Carmen; Quintana-Gallego, Esther; Villa-Gil, Manuel; Sánchez-Armengol, Angeles; Martínez-Martínez, Angel; Capote, Francisco
2005-05-01
To assess the possible relationship between Cheyne-Stokes respiration (CSR) associated with central sleep apnea (CSA) syndrome and brain natriuretic peptide (BNP) in an outpatient population presenting with stable congestive heart failure (CHF). Ninety patients with CHF due to systolic dysfunction (left ventricular ejection fraction
DeBruine, Lisa M; Jones, Benedict C; Smith, Finlay G; Little, Anthony C
2010-06-01
Women's preferences for male masculinity are highly variable. Although many researchers explain this variability as reflecting systematic individual differences in how women resolve the tradeoff between the costs and benefits of choosing a masculine partner, others suggest that methodological differences between studies are responsible. A recent study found general femininity preferences for judgments of faces that were manipulated in sexual dimorphism of shape but general masculinity preferences for judgments of faces that were based on perceived masculinity. Using the original stimuli, we replicated these previous results but found equivalent general femininity preferences for both types of faces when nonface confounds in the stimuli (e.g. hairstyle) were eliminated through masking. We conclude that care must be taken to control potential confounds in stimuli and that the influence of nonface cues on preferences for facial masculinity deserves further study.
Eliminating Survivor Bias in Two-stage Instrumental Variable Estimators.
Vansteelandt, Stijn; Walter, Stefan; Tchetgen Tchetgen, Eric
2018-07-01
Mendelian randomization studies commonly focus on elderly populations. This makes the instrumental variables analysis of such studies sensitive to survivor bias, a type of selection bias. A particular concern is that the instrumental variable conditions, even when valid for the source population, may be violated for the selective population of individuals who survive the onset of the study. This is potentially very damaging because Mendelian randomization studies are known to be sensitive to bias due to even minor violations of the instrumental variable conditions. Interestingly, the instrumental variable conditions continue to hold within certain risk sets of individuals who are still alive at a given age when the instrument and unmeasured confounders exert additive effects on the exposure, and moreover, the exposure and unmeasured confounders exert additive effects on the hazard of death. In this article, we will exploit this property to derive a two-stage instrumental variable estimator for the effect of exposure on mortality, which is insulated against the above described selection bias under these additivity assumptions.
Vega-Franco, L; Mejía, A M; Robles, B; Moreno, L; Pérez, Y
1991-11-01
This study gave us the opportunity to know the roles iron deficiency and the presence of lead in blood play, as confounding variables, in relation to the state of malnutrition and the intellect of those children. A sample of 169 school children were classified according to their state of nutrition, their condition in reference to serum iron and lead concentrations. In addition, their intelligence was evaluated. The results confirmed that those children with lower weights and heights registered lesser points of intelligence; in fact, iron deficiency cancels out the difference in favor of those taller and weighing more. Lead did not contribute as a confounding variable, but more than half of the children showed possible toxic levels of this metal.
NASA Technical Reports Server (NTRS)
Morton, Douglas C.; Nagol, Jyoteshwar; Carabajal, Claudia C.; Rosette, Jacqueline; Palace, Michael; Cook, Bruce D.; Vermote, Eric F.; Harding, David J.; North, Peter R. J.
2016-01-01
Multiple mechanisms could lead to up-regulation of dry-season photosynthesis in Amazon forests, including canopy phenology and illumination geometry. We specifically tested two mechanisms for phenology-driven changes in Amazon forests during dry-season months, and the combined evidence from passive optical and lidar satellite data was incompatible with large net changes in canopy leaf area or leaf reflectance suggested by previous studies. We therefore hypothesized that seasonal changes in the fraction of sunlit and shaded canopies, one aspect of bidirectional reflectance effects in Moderate Resolution Imaging Spectroradiometer (MODIS) data, could alter light availability for dry-season photosynthesis and the photosynthetic capacity of Amazon forests without large net changes in canopy composition. Subsequent work supports the hypothesis that seasonal changes in illumination geometry and diffuse light regulate light saturation in Amazon forests. These studies clarify the physical mechanisms that govern light availability in Amazon forests from seasonal variability in direct and diffuse illumination. Previously, in the debate over light limitation of Amazon forest productivity, seasonal changes in the distribution of light within complex Amazon forest canopies were confounded with dry-season increases in total incoming photosynthetically active radiation. In the accompanying Comment, Saleska et al. do not fully account for this confounding effect of forest structure on photosynthetic capacity.
Imamura, Fumiaki; Lichtenstein, Alice H; Dallal, Gerard E; Meigs, James B; Jacques, Paul F
2009-07-01
The ability to interpret epidemiologic observations is limited because of potential residual confounding by correlated dietary components. Dietary pattern analyses by factor analysis or partial least squares may overcome the limitation. To examine confounding by dietary pattern as well as standard risk factors and selected nutrients, the authors modeled the longitudinal association between alcohol consumption and 7-year risk of type 2 diabetes mellitus in 2,879 healthy adults enrolled in the Framingham Offspring Study (1991-2001) by Cox proportional hazard models. After adjustment for standard risk factors, consumers of > or =9.0 drinks/week had a significantly lower risk of type 2 diabetes mellitus compared with abstainers (hazard ratio = 0.47, 95% confidence interval (CI): 0.27, 0.81). Adjustment for selected nutrients had little effect on the hazard ratio, whereas adjustment for dietary pattern variables by factor analysis significantly shifted the hazard ratio away from null (hazard ratio = 0.33, 95% CI: 0.17, 0.64) by 40.0% (95% CI: 16.8, 57.0; P = 0.002). Dietary pattern variables by partial least squares showed similar results. Therefore, the observed inverse association, consistent with past studies, was confounded by dietary patterns, and this confounding was not captured by individual nutrient adjustment. The data suggest that alcohol intake, not dietary patterns associated with alcohol intake, is responsible for the observed inverse association with type 2 diabetes mellitus risk.
Xu, Ronghui; Hou, Jue; Chambers, Christina D
2018-06-01
Our work was motivated by small cohort studies on the risk of birth defects in infants born to pregnant women exposed to medications. We controlled for confounding using propensity scores (PS). The extremely rare events setting renders the matching or stratification infeasible. In addition, the PS itself may be formed via different approaches to select confounders from a relatively long list of potential confounders. We carried out simulation experiments to compare different combinations of approaches: IPW or regression adjustment, with 1) including all potential confounders without selection, 2) selection based on univariate association between the candidate variable and the outcome, 3) selection based on change in effects (CIE). The simulation showed that IPW without selection leads to extremely large variances in the estimated odds ratio, which help to explain the empirical data analysis results that we had observed. Copyright © 2018 Elsevier Inc. All rights reserved.
Tsutsumi, Chiyo; Kakuma, Tatsuyuki
2015-01-01
Associations have been reported between periodontal disease and increased cardiovascular disease risk, as well as between healthy self-reported tooth brushing behavior and reduced cardiovascular disease risk. We examined the association between self-reported tooth brushing behavior and the risk of metabolic syndrome (MetS) using a large medical check-up database. A total of 12,548 medical checkup records from a medical screening center in a city in southwest Japan were assessed. Subjects were aged 30 to 59 years. As lifestyle is associated with both tooth brushing and MetS, many lifestyle variables were considered as potential confounders. Logistic regression model were employed with a list of 127 lifestyle variables. Twenty variables associated with both tooth brushing and MetS were selected for both males and females. Furthermore, final confounding variables were selected by principal component analysis to avoid collinearity problems. The association of tooth brushing and MetS was evaluated for males and females separately by calculating adjusted odds ratios (ORs) with selected confounders plus age. The association between daily tooth brushing frequency and MetS risk for both genders was significant after adjusting for confounders, with the risk of MetS decreasing with increased frequency of daily tooth brushing (adjusted OR = 0.57 (95%Cl:0.48-0.81), 0.50 (95%Cl:0.35-0.71), 0.42 (95%Cl: 0.29-0.61) for males and adjusted OR = 0.65 (95%Cl:0.48-0.87), 0.44 (95%Cl:0.32-0.62) for females). Therefore, oral hygiene, including the frequency of tooth brushing as a significant component, may be an important factor in preventing MetS.
The Combined Effects of Measurement Error and Omitting Confounders in the Single-Mediator Model
Fritz, Matthew S.; Kenny, David A.; MacKinnon, David P.
2016-01-01
Mediation analysis requires a number of strong assumptions be met in order to make valid causal inferences. Failing to account for violations of these assumptions, such as not modeling measurement error or omitting a common cause of the effects in the model, can bias the parameter estimates of the mediated effect. When the independent variable is perfectly reliable, for example when participants are randomly assigned to levels of treatment, measurement error in the mediator tends to underestimate the mediated effect, while the omission of a confounding variable of the mediator to outcome relation tends to overestimate the mediated effect. Violations of these two assumptions often co-occur, however, in which case the mediated effect could be overestimated, underestimated, or even, in very rare circumstances, unbiased. In order to explore the combined effect of measurement error and omitted confounders in the same model, the impact of each violation on the single-mediator model is first examined individually. Then the combined effect of having measurement error and omitted confounders in the same model is discussed. Throughout, an empirical example is provided to illustrate the effect of violating these assumptions on the mediated effect. PMID:27739903
NASA Astrophysics Data System (ADS)
Oroza, C.; Bales, R. C.; Zheng, Z.; Glaser, S. D.
2017-12-01
Predicting the spatial distribution of soil moisture in mountain environments is confounded by multiple factors, including complex topography, spatial variably of soil texture, sub-surface flow paths, and snow-soil interactions. While remote-sensing tools such as passive-microwave monitoring can measure spatial variability of soil moisture, they only capture near-surface soil layers. Large-scale sensor networks are increasingly providing soil-moisture measurements at high temporal resolution across a broader range of depths than are accessible from remote sensing. It may be possible to combine these in-situ measurements with high-resolution LIDAR topography and canopy cover to estimate the spatial distribution of soil moisture at high spatial resolution at multiple depths. We study the feasibility of this approach using six years (2009-2014) of daily volumetric water content measurements at 10-, 30-, and 60-cm depths from the Southern Sierra Critical Zone Observatory. A non-parametric, multivariate regression algorithm, Random Forest, was used to predict the spatial distribution of depth-integrated soil-water storage, based on the in-situ measurements and a combination of node attributes (topographic wetness, northness, elevation, soil texture, and location with respect to canopy cover). We observe predictable patterns of predictor accuracy and independent variable ranking during the six-year study period. Predictor accuracy is highest during the snow-cover and early recession periods but declines during the dry period. Soil texture has consistently high feature importance. Other landscape attributes exhibit seasonal trends: northness peaks during the wet-up period, and elevation and topographic-wetness index peak during the recession and dry period, respectively.
Third and Fourth Degree Perineal Injury After Vaginal Delivery: Does Race Make a Difference?
de Silva, Kanoe-Lehua; Tsai, Pai-Jong Stacy; Kon, Leanne M; Kessel, Bruce; Seto, Todd; Kaneshiro, Bliss
2014-01-01
Severe perineal injury (third and fourth degree laceration) at the time of vaginal delivery increases the risk of fecal incontinence, chronic perineal pain, and dyspareunia.1–5 Studies suggest the prevalence of severe perineal injury may vary by racial group.6 The purpose of the current study was to examine rates of severe perineal injury in different Asian and Pacific Islander subgroups. A retrospective cohort study was performed among all patients who had a vaginal delivery at Queens Medical Center in Honolulu, Hawai‘i between January 1, 2002 and December 31, 2003. Demographic and health related variables were obtained for each participant. Maternal race/ethnicity (Japanese, Filipino, Chinese, other Asian, Part-Hawaiian/Hawaiian, Micronesian, other Pacific Islander, Caucasian, multiracial [non-Hawaiian], and other) was self-reported by the patient at the time admission. The significance of associations between racial/ethnic groups and demographic and health related variables was determined using chi-square tests for categorical variables and analysis of variance for continuous factors. Multiple logistic regression was performed to adjust for potential confounders when examining severe laceration rates. A total of 1842 subjects met inclusion criteria. The proportion of severe perineal lacerations did not differ significantly between racial groups. In the multiple logistic regression analysis, operative vaginal delivery was related to both race and severe perineal laceration. However, despite adjusting for this variable, race was not associated with an increased risk of having a severe laceration (P = .70). The results of this study indicate the risk of severe perineal laceration does not differ based on maternal race/ethnicity. PMID:24660124
Biasogram: Visualization of Confounding Technical Bias in Gene Expression Data
Krzystanek, Marcin; Szallasi, Zoltan; Eklund, Aron C.
2013-01-01
Gene expression profiles of clinical cohorts can be used to identify genes that are correlated with a clinical variable of interest such as patient outcome or response to a particular drug. However, expression measurements are susceptible to technical bias caused by variation in extraneous factors such as RNA quality and array hybridization conditions. If such technical bias is correlated with the clinical variable of interest, the likelihood of identifying false positive genes is increased. Here we describe a method to visualize an expression matrix as a projection of all genes onto a plane defined by a clinical variable and a technical nuisance variable. The resulting plot indicates the extent to which each gene is correlated with the clinical variable or the technical variable. We demonstrate this method by applying it to three clinical trial microarray data sets, one of which identified genes that may have been driven by a confounding technical variable. This approach can be used as a quality control step to identify data sets that are likely to yield false positive results. PMID:23613961
Nguyen, Trang Quynh; Webb-Vargas, Yenny; Koning, Ina M; Stuart, Elizabeth A
We investigate a method to estimate the combined effect of multiple continuous/ordinal mediators on a binary outcome: 1) fit a structural equation model with probit link for the outcome and identity/probit link for continuous/ordinal mediators, 2) predict potential outcome probabilities, and 3) compute natural direct and indirect effects. Step 2 involves rescaling the latent continuous variable underlying the outcome to address residual mediator variance/covariance. We evaluate the estimation of risk-difference- and risk-ratio-based effects (RDs, RRs) using the ML, WLSMV and Bayes estimators in Mplus. Across most variations in path-coefficient and mediator-residual-correlation signs and strengths, and confounding situations investigated, the method performs well with all estimators, but favors ML/WLSMV for RDs with continuous mediators, and Bayes for RRs with ordinal mediators. Bayes outperforms WLSMV/ML regardless of mediator type when estimating RRs with small potential outcome probabilities and in two other special cases. An adolescent alcohol prevention study is used for illustration.
Leveraging Hierarchical Population Structure in Discrete Association Studies
Carlson, Jonathan; Kadie, Carl; Mallal, Simon; Heckerman, David
2007-01-01
Population structure can confound the identification of correlations in biological data. Such confounding has been recognized in multiple biological disciplines, resulting in a disparate collection of proposed solutions. We examine several methods that correct for confounding on discrete data with hierarchical population structure and identify two distinct confounding processes, which we call coevolution and conditional influence. We describe these processes in terms of generative models and show that these generative models can be used to correct for the confounding effects. Finally, we apply the models to three applications: identification of escape mutations in HIV-1 in response to specific HLA-mediated immune pressure, prediction of coevolving residues in an HIV-1 peptide, and a search for genotypes that are associated with bacterial resistance traits in Arabidopsis thaliana. We show that coevolution is a better description of confounding in some applications and conditional influence is better in others. That is, we show that no single method is best for addressing all forms of confounding. Analysis tools based on these models are available on the internet as both web based applications and downloadable source code at http://atom.research.microsoft.com/bio/phylod.aspx. PMID:17611623
On the definition of a confounder
VanderWeele, Tyler J.; Shpitser, Ilya
2014-01-01
Summary The causal inference literature has provided a clear formal definition of confounding expressed in terms of counterfactual independence. The causal inference literature has not, however, produced a clear formal definition of a confounder, as it has given priority to the concept of confounding over that of a confounder. We consider a number of candidate definitions arising from various more informal statements made in the literature. We consider the properties satisfied by each candidate definition, principally focusing on (i) whether under the candidate definition control for all “confounders” suffices to control for “confounding” and (ii) whether each confounder in some context helps eliminate or reduce confounding bias. Several of the candidate definitions do not have these two properties. Only one candidate definition of those considered satisfies both properties. We propose that a “confounder” be defined as a pre-exposure covariate C for which there exists a set of other covariates X such that effect of the exposure on the outcome is unconfounded conditional on (X, C) but such that for no proper subset of (X, C) is the effect of the exposure on the outcome unconfounded given the subset. A variable that helps reduce bias but not eliminate bias we propose referring to as a “surrogate confounder.” PMID:25544784
Assessing Omitted Confounder Bias in Multilevel Mediation Models.
Tofighi, Davood; Kelley, Ken
2016-01-01
To draw valid inference about an indirect effect in a mediation model, there must be no omitted confounders. No omitted confounders means that there are no common causes of hypothesized causal relationships. When the no-omitted-confounder assumption is violated, inference about indirect effects can be severely biased and the results potentially misleading. Despite the increasing attention to address confounder bias in single-level mediation, this topic has received little attention in the growing area of multilevel mediation analysis. A formidable challenge is that the no-omitted-confounder assumption is untestable. To address this challenge, we first analytically examined the biasing effects of potential violations of this critical assumption in a two-level mediation model with random intercepts and slopes, in which all the variables are measured at Level 1. Our analytic results show that omitting a Level 1 confounder can yield misleading results about key quantities of interest, such as Level 1 and Level 2 indirect effects. Second, we proposed a sensitivity analysis technique to assess the extent to which potential violation of the no-omitted-confounder assumption might invalidate or alter the conclusions about the indirect effects observed. We illustrated the methods using an empirical study and provided computer code so that researchers can implement the methods discussed.
Lee, Paul H
2017-06-01
Some confounders are nonlinearly associated with dependent variables, but they are often adjusted using a linear term. The purpose of this study was to examine the error of mis-specifying the nonlinear confounding effect. We carried out a simulation study to investigate the effect of adjusting for a nonlinear confounder in the estimation of a causal relationship between the exposure and outcome in 3 ways: using a linear term, binning into 5 equal-size categories, or using a restricted cubic spline of the confounder. Continuous, binary, and survival outcomes were simulated. We examined the confounder across varying measurement error. In addition, we performed a real data analysis examining the 3 strategies to handle the nonlinear effects of accelerometer-measured physical activity in the National Health and Nutrition Examination Survey 2003-2006 data. The mis-specification of a nonlinear confounder had little impact on causal effect estimation for continuous outcomes. For binary and survival outcomes, this mis-specification introduced bias, which could be eliminated using spline adjustment only when there is small measurement error of the confounder. Real data analysis showed that the associations between high blood pressure, high cholesterol, and diabetes and mortality adjusted for physical activity with restricted cubic spline were about 3% to 11% larger than their counterparts adjusted with a linear term. For continuous outcomes, confounders with nonlinear effects can be adjusting with a linear term. Spline adjustment should be used for binary and survival outcomes on confounders with small measurement error.
Lifetime Alcohol Use & Cognitive Performance in Older Adults
Kalapatapu, Raj K.; Ventura, Maria I.; Barnes, Deborah E.
2016-01-01
Background Substance use is an important clinical issue in the older adult population. As older adults are susceptible to cognitive disorders, the intersection of the fields of substance use and cognitive neuroscience is an active area of research. Prior studies of alcohol use and cognitive performance are mixed, and inconsistencies may be due to under- or over-adjustment for confounders. Aim This manuscript adds to this literature by conducting a secondary analysis of self-reported lifetime history of alcohol use and cognitive performance in older adults (n = 133). We hypothesized that current alcohol users would have poorer cognitive performance compared to never/minimal and former alcohol users. Methods Older adult participants were classified into never/minimal alcohol users, former alcohol users, and current alcohol users. A neurocognitive battery included a global cognitive measure and individual measures of attention, memory, fluency, and executive function. A directed acyclic graph (DAG)-based approach was used to select variables to be included in the multiple linear regression models. Results Though unadjusted analyses showed some significant associations between alcohol use and cognitive performance, all associations between alcohol use and cognitive performance were eliminated after adjusting for age, education, sex, race and smoking pack years. Alcohol drink years were not significantly associated with cognitive performance among current and former alcohol users. Discussion These results suggest that lifetime alcohol use is not significantly associated with cognitive performance in older adults after adjustment for key confounders. Inconsistencies in prior studies may be due to uncontrolled confounding and/or unnecessary adjustment of mediators and/or colliders. PMID:27719514
Association of Fatigue With Sarcopenia and its Elements: A Secondary Analysis of SABE-Bogotá
Patino-Hernandez, Daniela; David-Pardo, David Gabriel; Borda, Miguel Germán; Pérez-Zepeda, Mario Ulises; Cano-Gutiérrez, Carlos
2017-01-01
Objective: Sarcopenia, fatigue, and depression are associated with higher mortality rates and adverse outcomes in the aging population. Understanding the association among clinical variables, mainly symptoms, is important for screening and appropriately managing these conditions. The aim of this article is to evaluate the association among sarcopenia and its elements with depression and fatigue. Method: We used cross-sectional data from 2012 SABE (Salud, Bienestar y Envejecimiento)-Bogotá study, which included 2,000 participants of ages ≥60 years. Sarcopenia and its elements were taken as the dependent variable, while fatigue and depression were the main independent variables. We tested the association among these through multiple logistic regression models, which were fitted for each dependent variable and adjusted for confounding variables. Results: Our findings showed that gait speed was associated with fatigue (adjusted odds ratio [OR] = 1.41, 95% confidence interval [CI] = [1.05, 1.90], p = .02) as well as abnormal handgrip strength (adjusted OR = 1.40, 95% CI = [1.02, 1.93], p = .04). No other associations were significant. Conclusion: While sarcopenia and fatigue are not associated, two of the sarcopenia-defining variables are associated with fatigue; this suggests that lack of sarcopenia does not exclude undesirable outcomes related to fatigue in aging adults. Also, the lack of association between sarcopenia-defining elements and depression demonstrates that depression and fatigue are different concepts. PMID:28474000
Drosophila melanogaster as a model system for the evaluation of anti-aging compounds.
Jafari, Mahtab
2010-01-01
Understanding the causes of aging is a complex problem due to the multiple factors that influence aging, which include genetics, environment, metabolism and reproduction, among others. These multiple factors create logistical difficulties in the evaluation of anti-aging agents. There is a need for good model systems to evaluate potential anti-aging compounds. The model systems used should represent the complexities of aging in humans, so that the findings may be extrapolated to human studies, but they should also present an opportunity to minimize the variables so that the experimental results can be accurately interpreted. In addition to positively affecting lifespan, the impact of the compound on the physiologic confounders of aging, including fecundity and the health span--the period of life where an organism is generally healthy and free from serious or chronic illness--of the model organism needs to be evaluated. Fecundity is considered a major confounder of aging in fruit flies. It is well established that female flies that are exposed to toxic substances typically reduce their dietary intake and their reproductive output and display an artifactual lifespan extension. As a result, drugs that achieve longevity benefits by reducing fecundity as a result of diminished food intake are probably not useful candidates for eventual treatment of aging in humans and should be eliminated during the screening process. Drosophila melanogaster provides a suitable model system for the screening of anti-aging compounds as D. melanogaster and humans have many conserved physiological and biological pathways. In this paper, I propose an algorithm to screen anti-aging compounds using Drosophila melanogaster as a model system.
Benseñor, Isabela M; Nunes, Maria Angélica; Sander Diniz, Maria de Fátima; Santos, Itamar S; Brunoni, André R; Lotufo, Paulo A
2016-02-01
To evaluate the association between subclinical thyroid dysfunction and psychiatric disorders using baseline data from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil). Cross-sectional study. The study included 12 437 participants from the ELSA-Brasil with normal thyroid function (92·8%), 193 (1·4%) with subclinical hyperthyroidism and 784 (5·8%) with subclinical hypothyroidism, totalling 13 414 participants (50·6% of women). The mental health diagnoses of participants were assessed by trained raters using the Clinical Interview Schedule - Revised (CIS-R) and grouped according to the International Classification of Diseases 10 (ICD-10). Thyroid dysfunction was assessed using TSH and FT4 as well as routine use of thyroid hormones or antithyroid medications. Logistic models were presented using psychiatric disorders as the dependent variable and subclinical thyroid disorders as the independent variable. All logistic models were corrected for multiple comparisons using Bonferroni correction. After multivariate adjustment for possible confounders, we found a direct association between subclinical hyperthyroidism and panic disorder odds ratio [OR], 2·55; 95% confidence Interval (95% CI), 1·09-5·94; and an inverse association between subclinical hypothyroidism and generalized anxiety disorder (OR, 0·75; 95% CI, 0·59-0·96). However, both lost significance after correction for multiple comparisons. Subclinical hyperthyroidism was positively associated with panic disorder and negatively associated with anxiety disorder, although not significant after adjustment for multiple comparisons. © 2015 John Wiley & Sons Ltd.
Reynard, Linda M; Burt, Nicole; Koon, Hannah E C; Tuross, Noreen
2016-01-01
We examined multiple natural abundance isotope ratios of human hair to assess biological variability within and between geographic locations and, further, to determine how well these isotope values predict location of origin. Sampling locations feature differing seasonality and mobile populations as a robust test of the method. Serially-sampled hair from Cambridge, MA, USA, shows lower δ(2)H and δ(18)O variability over a one-year time course than model-predicted precipitation isotope ratios, but exhibits considerable differences between individuals. Along a ∼13° north-south transect in the eastern USA (Brookline, MA, 42.3 ° N, College Park, MD, 39.0 ° N, and Gainesville, FL, 29.7 ° N) δ(18)O in human hair shows relatively greater differences and tracks changes in drinking water isotope ratios more sensitively than δ(2)H. Determining the domicile of humans using isotope ratios of hair can be confounded by differing variability in hair δ(18)O and δ(2)H between locations, differential incorporation of H and O into this protein and, in some cases, by tap water δ(18)O and δ(2)H that differ significantly from predicted precipitation values. With these caveats, randomly chosen people in Florida are separated from those in the two more northerly sites on the basis of the natural abundance isotopes of carbon, nitrogen, hydrogen, and oxygen.
ERIC Educational Resources Information Center
Bollen, Kenneth A.
2007-01-01
R. D. Howell, E. Breivik, and J. B. Wilcox (2007) have argued that causal (formative) indicators are inherently subject to interpretational confounding. That is, they have argued that using causal (formative) indicators leads the empirical meaning of a latent variable to be other than that assigned to it by a researcher. Their critique of causal…
Causal diagrams and multivariate analysis III: confound it!
Jupiter, Daniel C
2015-01-01
This commentary concludes my series concerning inclusion of variables in multivariate analyses. We take up the issues of confounding and effect modification and summarize the work we have thus far done. Finally, we provide a rough algorithm to help guide us through the maze of possibilities that we have outlined. Copyright © 2015 American College of Foot and Ankle Surgeons. Published by Elsevier Inc. All rights reserved.
Heo, Moonseong; Kim, Ryung S.; Wylie-Rosett, Judith; Allison, David B.; Heymsfield, Steve B.; Faith, Myles S.
2011-01-01
Objective To estimate fruit and vegetable (FV) intake levels of US adult population and evaluate the association between FV intake and BMI status after controlling for confounding demographic, socioeconomic and lifestyle factors. We also sought to identify moderating factors. Methods We used 2007 Behavior Risk Factors Surveillance System (N > 400,000) data. FV intake was dichotomized as ≥5 servings (FV5+) versus <5 servings/day. BMI status was categorized as normal, overweight, and obese. Identification of moderators was performed by testing interactions between BMI status and other variables using bivariate analyses followed by multiple logistic regression analysis incorporating complex survey sampling design features. Results Only 24.6% of US adults consumed ≥5 servings per day and less than 4% consumed 9 or more servings. Overweight (% FV5+ = 23.9%) and obese (21.9%) groups consumed significantly less FV than the normal-weight (27.4%) group (p < 0.0001). This inverse association remained significant even after controlling for potential confounding factors. Multivariate analysis identified five significant moderators (p < 0.0001) after controlling for all evaluated variables: race, sex, smoking status, health coverage, and physical activity. Notably, physically inactive obese males tended to consume the least FV (% FV5+ = 14.7%). Conclusion Current US population FV intake level is below recommended levels. The inverse association between FV intake and obesity was significant and was moderated by demographic, socioeconomic status, and lifestyle factors. These factors should be considered when developing policies and interventions to increase FV intake. PMID:22248995
Impact of maternal work outside the home during pregnancy on neonatal outcome.
Zuckerman, B S; Frank, D A; Hingson, R; Morelock, S; Kayne, H L
1986-04-01
The impact of mothers' work during pregnancy upon neonatal outcome has not been fully elucidated. A study of maternal health habits and neonatal outcome conducted from 1977 to 1979 at Boston City Hospital assessed the effects of paid work outside the home during pregnancy, particularly third trimester work in a standing position, upon duration of gestation and intrauterine growth, while controlling for potentially confounding variables. Postpartum, 1,690 low-income women were interviewed to provide detailed information on their sociodemographic and health characteristics. Of the 1,507 women whose work status during pregnancy could be classified, 55% did not perform paid work outside the home or attend school, 7% worked into the third trimester in a standing position, and 38% had other work histories or attended school. Infants were examined according to a standard protocol by pediatricians blind to mothers' work history. There was no significant relationship between the mothers' work history during pregnancy and the infants' length of gestation, weight, or head circumference at birth, when confounding variables were controlled statistically by multiple regression analyses. However, women who worked in a standing position into the third trimester delivered infants who were longer than infants born either to mothers with other work histories or to mothers who did not work or attend school. Healthy women with uncomplicated low-risk pregnancies who feel well enough to attend school or work during pregnancy, even in a job requiring some standing, may do so without fear that their infant will suffer shortened gestation or impaired intrauterine growth.
Peerenboom, L; Collard, R M; Naarding, P; Comijs, H C
2015-08-15
We investigated the association between old age depression and emotional and social loneliness. A cross-sectional study was performed using data from the Netherlands Study of Depression in Older Persons (NESDO). A total of 341 participants diagnosed with a depressive disorder, and 125 non-depressed participants were included. Depression diagnosis was confirmed with the Composite International Diagnostic Interview. Emotional and social loneliness were assessed using the De Jong Gierveld Loneliness Scale. Socio-demographic variables, social support variables, depression characteristics (Inventory of Depressive Symptoms), cognitive functioning (Mini Mental State Examination) and personality factors (the NEO- Five Factor Inventory and the Pearlin Mastery Scale) were considered as possible explanatory factors or confounders. (Multiple) logistic regression analyses were performed. Depression was strongly associated with emotional loneliness, but not with social loneliness. A higher sense of neuroticism and lower sense of mastery were the most important explanatory factors. Also, we found several other explanatory and confounding factors in the association of depression and emotional loneliness; a lower sense of extraversion and higher severity of depression. We performed a cross-sectional observational study. Therefore we cannot add evidence in regard to causation; whether depression leads to loneliness or vice versa. Depression in older persons is strongly associated with emotional loneliness but not with social loneliness. Several personality traits and the severity of depression are important in regard to the association of depression and emotional loneliness. It is important to develop interventions in which both can be treated. Copyright © 2015 Elsevier B.V. All rights reserved.
Polychlorinated biphenyl exposure and effects in transformer repair workers.
Emmett, E A
1985-01-01
Fifty-five present and past transformer repair workers exposed to polychlorinated biphenyls (PCBs) and 56 unexposed comparison workers were evaluated in a clinical-epidemiologic study. The groups were similar in most demographic variables. Adipose tissue lipid and serum PCBs concentrations were higher in current exposed workers (geometric means adipose 2.1 ppm, serum 12.2 ppb). Concentrations in comparison (0.6 ppm and 4.6 ppb) and previously exposed (0.83 ppm and 5.9 ppb) workers were lower. Statistically significant differences in serum albumin and lactic dehydrogenase, but not in other liver function tests, were seen between the exposed and comparison groups; however, after adjustment for confounding variables, no correlations were observed between liver function tests and either adipose or serum PCBs concentrations. Statistically significant correlation both before and after adjustment for confounding variables were seen with adipose PCBs and 24-hr urinary 17-hydroxycorticosteroid excretion and with serum PCBs and serum gamma-glutamyl transpeptidase. Both associations could reflect microsomal enzyme induction among other possibilities. No differences were seen in fasting serum triglycerides, total cholesterol, LDL, HDL or VLDL cholesterol between the two exposure groups. A statistically significant correlation between serum PCBs and serum triglycerides, total cholesterol, and VLDL cholesterol was removed by adjusting for confounding variables. No correlation was seen between adipose PCBs concentrations and any serum lipid component. Partition phenomena could account for these findings.(ABSTRACT TRUNCATED AT 250 WORDS) PMID:2863134
On causal interpretation of race in regressions adjusting for confounding and mediating variables
VanderWeele, Tyler J.; Robinson, Whitney R.
2014-01-01
We consider several possible interpretations of the “effect of race” when regressions are run with race as an exposure variable, controlling also for various confounding and mediating variables. When adjustment is made for socioeconomic status early in a person’s life, we discuss under what contexts the regression coefficients for race can be interpreted as corresponding to the extent to which a racial inequality would remain if various socioeconomic distributions early in life across racial groups could be equalized. When adjustment is also made for adult socioeconomic status, we note how the overall racial inequality can be decomposed into the portion that would be eliminated by equalizing adult socioeconomic status across racial groups and the portion of the inequality that would remain even if adult socioeconomic status across racial groups were equalized. We also discuss a stronger interpretation of the “effect of race” (stronger in terms of assumptions) involving the joint effects of race-associated physical phenotype (e.g. skin color), parental physical phenotype, genetic background and cultural context when such variables are thought to be hypothetically manipulable and if adequate control for confounding were possible. We discuss some of the challenges with such an interpretation. Further discussion is given as to how the use of selected populations in examining racial disparities can additionally complicate the interpretation of the effects. PMID:24887159
How multiple causes combine: independence constraints on causal inference.
Liljeholm, Mimi
2015-01-01
According to the causal power view, two core constraints-that causes occur independently (i.e., no confounding) and influence their effects independently-serve as boundary conditions for causal induction. This study investigated how violations of these constraints modulate uncertainty about the existence and strength of a causal relationship. Participants were presented with pairs of candidate causes that were either confounded or not, and that either interacted or exerted their influences independently. Consistent with the causal power view, uncertainty about the existence and strength of causal relationships was greater when causes were confounded or interacted than when unconfounded and acting independently. An elemental Bayesian causal model captured differences in uncertainty due to confounding but not those due to an interaction. Implications of distinct sources of uncertainty for the selection of contingency information and causal generalization are discussed.
Flanders, W Dana; Strickland, Matthew J; Klein, Mitchel
2017-05-15
Methods exist to detect residual confounding in epidemiologic studies. One requires a negative control exposure with 2 key properties: 1) conditional independence of the negative control and the outcome (given modeled variables) absent confounding and other model misspecification, and 2) associations of the negative control with uncontrolled confounders and the outcome. We present a new method to partially correct for residual confounding: When confounding is present and our assumptions hold, we argue that estimators from models that include a negative control exposure with these 2 properties tend to be less biased than those from models without it. Using regression theory, we provide theoretical arguments that support our claims. In simulations, we empirically evaluated the approach using a time-series study of ozone effects on asthma emergency department visits. In simulations, effect estimators from models that included the negative control exposure (ozone concentrations 1 day after the emergency department visit) had slightly or modestly less residual confounding than those from models without it. Theory and simulations show that including the negative control can reduce residual confounding, if our assumptions hold. Our method differs from available methods because it uses a regression approach involving an exposure-based indicator rather than a negative control outcome to partially correct for confounding. © The Author 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Reps, Jenna M; Aickelin, Uwe; Hubbard, Richard B
2016-02-01
To develop a framework for identifying and incorporating candidate confounding interaction terms into a regularised cox regression analysis to refine adverse drug reaction signals obtained via longitudinal observational data. We considered six drug families that are commonly associated with myocardial infarction in observational healthcare data, but where the causal relationship ground truth is known (adverse drug reaction or not). We applied emergent pattern mining to find itemsets of drugs and medical events that are associated with the development of myocardial infarction. These are the candidate confounding interaction terms. We then implemented a cohort study design using regularised cox regression that incorporated and accounted for the candidate confounding interaction terms. The methodology was able to account for signals generated due to confounding and a cox regression with elastic net regularisation correctly ranking the drug families known to be true adverse drug reactions above those that are not. This was not the case without the inclusion of the candidate confounding interaction terms, where confounding leads to a non-adverse drug reaction being ranked highest. The methodology is efficient, can identify high-order confounding interactions and does not require expert input to specify outcome specific confounders, so it can be applied for any outcome of interest to quickly refine its signals. The proposed method shows excellent potential to overcome some forms of confounding and therefore reduce the false positive rate for signal analysis using longitudinal data. Copyright © 2015 Elsevier Ltd. All rights reserved.
Nam, Kijoeng; Henderson, Nicholas C; Rohan, Patricia; Woo, Emily Jane; Russek-Cohen, Estelle
2017-01-01
The Vaccine Adverse Event Reporting System (VAERS) and other product surveillance systems compile reports of product-associated adverse events (AEs), and these reports may include a wide range of information including age, gender, and concomitant vaccines. Controlling for possible confounding variables such as these is an important task when utilizing surveillance systems to monitor post-market product safety. A common method for handling possible confounders is to compare observed product-AE combinations with adjusted baseline frequencies where the adjustments are made by stratifying on observable characteristics. Though approaches such as these have proven to be useful, in this article we propose a more flexible logistic regression approach which allows for covariates of all types rather than relying solely on stratification. Indeed, a main advantage of our approach is that the general regression framework provides flexibility to incorporate additional information such as demographic factors and concomitant vaccines. As part of our covariate-adjusted method, we outline a procedure for signal detection that accounts for multiple comparisons and controls the overall Type 1 error rate. To demonstrate the effectiveness of our approach, we illustrate our method with an example involving febrile convulsion, and we further evaluate its performance in a series of simulation studies.
Vegetarian diets and depressive symptoms among men.
Hibbeln, Joseph R; Northstone, Kate; Evans, Jonathan; Golding, Jean
2018-01-01
Vegetarian diets are associate with cardiovascular and other health benefits, but little is known about mental health benefits or risks. To determine whether self-identification of vegetarian dietary habits is associated with significant depressive symptoms in men. Self-report data from 9668 adult male partners of pregnant women in the Avon Longitudinal Study of Parents and Children (ALSPAC) included identification as vegetarian or vegan, dietary frequency data and the Edinburgh Post Natal Depression Scale (EPDS). Continuous and binary outcomes were assessed using multiple linear and logistic regression taking account of potential confounding variables including: age, marital status, employment status, housing tenure, number of children in the household, religion, family history of depression previous childhood psychiatric contact, cigarette and alcohol consumption. Vegetarians [n = 350 (3.6% of sample)], had higher depression scores on average than non-vegetarians (mean difference 0.96 points [95%CI + 0.53, + 1.40]) and a greater risk for EPDS scores above 10 (adjusted OR = 1.67 [95% CI: 1.14,2.44]) than non-vegetarians after adjustment for potential confounding factors. Vegetarian men have more depressive symptoms after adjustment for socio-demographic factors. Nutritional deficiencies (e.g. in cobalamin or iron) are a possible explanation for these findings, however reverse causation cannot be ruled out. Copyright © 2017. Published by Elsevier B.V.
Environmental confounding in gene-environment interaction studies.
Vanderweele, Tyler J; Ko, Yi-An; Mukherjee, Bhramar
2013-07-01
We show that, in the presence of uncontrolled environmental confounding, joint tests for the presence of a main genetic effect and gene-environment interaction will be biased if the genetic and environmental factors are correlated, even if there is no effect of either the genetic factor or the environmental factor on the disease. When environmental confounding is ignored, such tests will in fact reject the joint null of no genetic effect with a probability that tends to 1 as the sample size increases. This problem with the joint test vanishes under gene-environment independence, but it still persists if estimating the gene-environment interaction parameter itself is of interest. Uncontrolled environmental confounding will bias estimates of gene-environment interaction parameters even under gene-environment independence, but it will not do so if the unmeasured confounding variable itself does not interact with the genetic factor. Under gene-environment independence, if the interaction parameter without controlling for the environmental confounder is nonzero, then there is gene-environment interaction either between the genetic factor and the environmental factor of interest or between the genetic factor and the unmeasured environmental confounder. We evaluate several recently proposed joint tests in a simulation study and discuss the implications of these results for the conduct of gene-environment interaction studies.
Ambs, Stefan; Tang, Wei; Zhu, Shuotun; Dorsey, Tiffany H.; Goudie, Megan; Masters, Jonathan G.; Ferguson, Lynnette R.
2018-01-01
Introduction The prostate-specific antigen (PSA) based prostate cancer (PC) screening is currently being debated. The current assessment is to understand the variability of detecting high-risk PC in a NZ cohort in comparison to a US cohort with better PSA screening facilities. Aldo-keto reductase 1C3 (AKR1C3) is known for multiple functions with a potential to regulate subsequent PSA levels. Therefore, we wish to understand the influence of tobacco smoking and the AKR1C3 rs12529 gene polymorphism in this variability. Method NZ cohort (n = 376) consisted of 94% Caucasians while the US cohort consisted of African Americans (AA), n = 202, and European Americans (EA), n = 232. PSA level, PC grade and stage at diagnosis were collected from hospital databases for assigning high-risk PC status. Tobacco smoking status and the AKR1C3 rs12529 SNP genotype were considered as confounding variables. Variation of the cumulative % high-risk PC (outcome variable) with increasing PSA intervals (exposure factor) was compared between the cohorts using the Kolmogorov-Smirnov test. Comparisons were carried out with and without stratifications made using confounding variables. Results NZ cohort has been diagnosed at a significantly higher mean age (66.67± (8.08) y) compared to both AA (62.65±8.17y) and EA (64.83+8.56y); median PSA (NZ 8.90ng/ml compared to AA 6.86ng/ml and EA 5.80ng/ml); and Gleason sum (NZ (7) compared EA (6)) (p<0.05). The cumulative % high-risk PC detection shows NZ cohort with a significantly lower diagnosis rates at PSA levels between >6 - <10ng/ml compared to both US groups (p<0.05). These were further compounded significantly by smoking status and genetics. Conclusions High-risk PCs recorded at higher PSA levels in NZ could be due to factors including lower levels of PSA screening and subsequent specialist referrals for biopsies. These consequences could be pronounced among NZ ever smokers carrying the AKR1C3 rs12529 G alleles making them a group that requires increased PSA screening attention. PMID:29920533
Hart, Carole L; Hole, David J; Lawlor, Debbie A; Smith, George Davey; Lever, Tony F
2007-01-01
Objectives To investigate how loss of a spouse affects mortality risk in the bereaved partner. Design and setting Prospective cohort study in Renfrew and Paisley in Scotland. Participants 4395 married couples aged 45–64 years when the study was carried out between 1972 and 1976. Methods The date of bereavement for the bereaved spouse was the date of death of his or her spouse. Bereavement could occur at any time during the follow‐up period, so it was considered as a time‐dependent exposure variable and the Cox proportional hazards model for time‐dependent variables was used. The relative rate (RR) of mortality was calculated for bereaved versus non‐bereaved spouses and adjusted for confounding variables. Main outcome measures Causes of death to 31 March 2004. Results Bereaved participants were at higher risk than non‐bereaved participants of dying from any cause (RR 1.27; 95% CI 1.2 to 1.35). These risks remained but were attenuated after adjustment for confounding variables. There were raised RRs for bereaved participants dying of cardiovascular disease, coronary heart disease, stroke, all cancer, lung cancer, smoking‐related cancer, and accidents or violence. After adjustment for confounding variables, RRs remained higher for bereaved participants for all these causes except for mortality from lung cancer. There was no strong statistical evidence that the increased risks of death associated with bereavement changed with time after bereavement. Conclusions Conjugal bereavement, in addition to existing risk factors, is related to mortality risk for major causes of death. PMID:17435215
Sanderson, Eleanor; Macdonald-Wallis, Corrie; Davey Smith, George
2018-01-01
Abstract Background Negative control exposure studies are increasingly being used in epidemiological studies to strengthen causal inference regarding an exposure-outcome association when unobserved confounding is thought to be present. Negative control exposure studies contrast the magnitude of association of the negative control, which has no causal effect on the outcome but is associated with the unmeasured confounders in the same way as the exposure, with the magnitude of the association of the exposure with the outcome. A markedly larger effect of the exposure on the outcome than the negative control on the outcome strengthens inference that the exposure has a causal effect on the outcome. Methods We investigate the effect of measurement error in the exposure and negative control variables on the results obtained from a negative control exposure study. We do this in models with continuous and binary exposure and negative control variables using analysis of the bias of the estimated coefficients and Monte Carlo simulations. Results Our results show that measurement error in either the exposure or negative control variables can bias the estimated results from the negative control exposure study. Conclusions Measurement error is common in the variables used in epidemiological studies; these results show that negative control exposure studies cannot be used to precisely determine the size of the effect of the exposure variable, or adequately adjust for unobserved confounding; however, they can be used as part of a body of evidence to aid inference as to whether a causal effect of the exposure on the outcome is present. PMID:29088358
Sanderson, Eleanor; Macdonald-Wallis, Corrie; Davey Smith, George
2018-04-01
Negative control exposure studies are increasingly being used in epidemiological studies to strengthen causal inference regarding an exposure-outcome association when unobserved confounding is thought to be present. Negative control exposure studies contrast the magnitude of association of the negative control, which has no causal effect on the outcome but is associated with the unmeasured confounders in the same way as the exposure, with the magnitude of the association of the exposure with the outcome. A markedly larger effect of the exposure on the outcome than the negative control on the outcome strengthens inference that the exposure has a causal effect on the outcome. We investigate the effect of measurement error in the exposure and negative control variables on the results obtained from a negative control exposure study. We do this in models with continuous and binary exposure and negative control variables using analysis of the bias of the estimated coefficients and Monte Carlo simulations. Our results show that measurement error in either the exposure or negative control variables can bias the estimated results from the negative control exposure study. Measurement error is common in the variables used in epidemiological studies; these results show that negative control exposure studies cannot be used to precisely determine the size of the effect of the exposure variable, or adequately adjust for unobserved confounding; however, they can be used as part of a body of evidence to aid inference as to whether a causal effect of the exposure on the outcome is present.
NASA Astrophysics Data System (ADS)
Sandrini-Neto, L.; Lana, P. C.
2012-06-01
Heterogeneity in the distribution of organisms occurs at a range of spatial scales, which may vary from few centimeters to hundreds of kilometers. The exclusion of small-scale variability from routine sampling designs may confound comparisons at larger scales and lead to inconsistent interpretation of data. Despite its ecological and social-economic importance, little is known about the spatial structure of the mangrove crab Ucides cordatus in the southwest Atlantic. Previous studies have commonly compared densities at relatively broad scales, relying on alleged distribution patterns (e.g., mangroves of distinct composition and structure). We have assessed variability patterns of U. cordatus in mangroves of Paranaguá Bay at four levels of spatial hierarchy (10 s km, km, 10 s m and m) using a nested ANOVA and variance components measures. The potential role of sediment parameters, pneumatophore density, and organic matter content in regulating observed patterns was assessed by multiple regression models. Densities of total and non-commercial size crabs varied mostly at 10 s m to km scales. Densities of commercial size crabs differed at the scales of 10 s m and 10 s km. Variance components indicated that small-scale variation was the most important, contributing up to 70% of the crab density variability. Multiple regression models could not explain the observed variations. Processes driving differences in crab abundance were not related to the measured variables. Small-scale patchy distribution has direct implications to current management practices of U. cordatus. Future studies should consider processes operating at smaller scales, which are responsible for a complex mosaic of patches within previously described patterns.
Gender effects in gaming research: a case for regression residuals?
Pfister, Roland
2011-10-01
Numerous recent studies have examined the impact of video gaming on various dependent variables, including the players' affective reactions, positive as well as detrimental cognitive effects, and real-world aggression. These target variables are typically analyzed as a function of game characteristics and player attributes-especially gender. However, findings on the uneven distribution of gaming experience between males and females, on the one hand, and the effect of gaming experience on several target variables, on the other hand, point at a possible confound when gaming experiments are analyzed with a standard analysis of variance. This study uses simulated data to exemplify analysis of regression residuals as a potentially beneficial data analysis strategy for such datasets. As the actual impact of gaming experience on each of the various dependent variables differs, the ultimate benefits of analysis of regression residuals entirely depend on the research question, but it offers a powerful statistical approach to video game research whenever gaming experience is a confounding factor.
Kischkel, Sabine; Miekisch, Wolfram; Sawacki, Annika; Straker, Eva M; Trefz, Phillip; Amann, Anton; Schubert, Jochen K
2010-11-11
Up to now, none of the breath biomarkers or marker sets proposed for cancer recognition has reached clinical relevance. Possible reasons are the lack of standardized methods of sampling, analysis and data processing and effects of environmental contaminants. Concentration profiles of endogenous and exogenous breath markers were determined in exhaled breath of 31 lung cancer patients, 31 smokers and 31 healthy controls by means of SPME-GC-MS. Different correcting and normalization algorithms and a principal component analysis were applied to the data. Differences of exhalation profiles in cancer and non-cancer patients did not persist if physiology and confounding variables were taken into account. Smoking history, inspired substance concentrations, age and gender were recognized as the most important confounding variables. Normalization onto PCO2 or BSA or correction for inspired concentrations only partially solved the problem. In contrast, previous smoking behaviour could be recognized unequivocally. Exhaled substance concentrations may depend on a variety of parameters other than the disease under investigation. Normalization and correcting parameters have to be chosen with care as compensating effects may be different from one substance to the other. Only well-founded biomarker identification, normalization and data processing will provide clinically relevant information from breath analysis. 2010 Elsevier B.V. All rights reserved.
Howe, Piers D. L.
2017-01-01
To understand how the visual system represents multiple moving objects and how those representations contribute to tracking, it is essential that we understand how the processes of attention and working memory interact. In the work described here we present an investigation of that interaction via a series of tracking and working memory dual-task experiments. Previously, it has been argued that tracking is resistant to disruption by a concurrent working memory task and that any apparent disruption is in fact due to observers making a response to the working memory task, rather than due to competition for shared resources. Contrary to this, in our experiments we find that when task order and response order confounds are avoided, all participants show a similar decrease in both tracking and working memory performance. However, if task and response order confounds are not adequately controlled for we find substantial individual differences, which could explain the previous conflicting reports on this topic. Our results provide clear evidence that tracking and working memory tasks share processing resources. PMID:28410383
Lapierre, Mark D; Cropper, Simon J; Howe, Piers D L
2017-01-01
To understand how the visual system represents multiple moving objects and how those representations contribute to tracking, it is essential that we understand how the processes of attention and working memory interact. In the work described here we present an investigation of that interaction via a series of tracking and working memory dual-task experiments. Previously, it has been argued that tracking is resistant to disruption by a concurrent working memory task and that any apparent disruption is in fact due to observers making a response to the working memory task, rather than due to competition for shared resources. Contrary to this, in our experiments we find that when task order and response order confounds are avoided, all participants show a similar decrease in both tracking and working memory performance. However, if task and response order confounds are not adequately controlled for we find substantial individual differences, which could explain the previous conflicting reports on this topic. Our results provide clear evidence that tracking and working memory tasks share processing resources.
Enhancing the estimation of blood pressure using pulse arrival time and two confounding factors.
Baek, Hyun Jae; Kim, Ko Keun; Kim, Jung Soo; Lee, Boreom; Park, Kwang Suk
2010-02-01
A new method of blood pressure (BP) estimation using multiple regression with pulse arrival time (PAT) and two confounding factors was evaluated in clinical and unconstrained monitoring situations. For the first analysis with clinical data, electrocardiogram (ECG), photoplethysmogram (PPG) and invasive BP signals were obtained by a conventional patient monitoring device during surgery. In the second analysis, ECG, PPG and non-invasive BP were measured using systems developed to obtain data under conditions in which the subject was not constrained. To enhance the performance of BP estimation methods, heart rate (HR) and arterial stiffness were considered as confounding factors in regression analysis. The PAT and HR were easily extracted from ECG and PPG signals. For arterial stiffness, the duration from the maximum derivative point to the maximum of the dicrotic notch in the PPG signal, a parameter called TDB, was employed. In two experiments that normally cause BP variation, the correlation between measured BP and the estimated BP was investigated. Multiple-regression analysis with the two confounding factors improved correlation coefficients for diastolic blood pressure and systolic blood pressure to acceptable confidence levels, compared to existing methods that consider PAT only. In addition, reproducibility for the proposed method was determined using constructed test sets. Our results demonstrate that non-invasive, non-intrusive BP estimation can be obtained using methods that can be applied in both clinical and daily healthcare situations.
Fatigue and crashes: the case of freight transport in Colombia.
Torregroza-Vargas, Nathaly M; Bocarejo, Juan Pablo; Ramos-Bonilla, Juan P
2014-11-01
Truck drivers have been involved in a significant number of road fatalities in Colombia. To identify variables that could be associated with crashes in which truck drivers are involved, a logistic regression model was constructed. The model had as the response variable a dichotomous variable that included the presence or absence of a crash during a specific trip. As independent variables the model included information regarding a driver's work shift, with variables that could be associated with driver's fatigue. The model also included potential confounders related with road conditions. With the model, it was possible to determine the odds ratio of a crash in relation to several variables, adjusting for confounding. To collect the information about the trips included in the model, a survey among truck drivers was conducted. The results suggest strong associations between crashes (i.e., some of them statistically significant) with the number of stops made during the trip, and the average time of each stop. Survey analysis allowed us to identify the practices that contribute to generating fatigue and unhealthy conditions on the road among professional drivers. A review of national regulations confirmed the lack of legislation on this topic. Copyright © 2014 Elsevier Ltd. All rights reserved.
Traskin, Mikhail; Wang, Wei; Ten Have, Thomas R; Small, Dylan S
2013-01-01
The PAF for an exposure is the fraction of disease cases in a population that can be attributed to that exposure. One method of estimating the PAF involves estimating the probability of having the disease given the exposure and confounding variables. In many settings, the exposure will interact with the confounders and the confounders will interact with each other. Also, in many settings, the probability of having the disease is thought, based on subject matter knowledge, to be a monotone increasing function of the exposure and possibly of some of the confounders. We develop an efficient approach for estimating logistic regression models with interactions and monotonicity constraints, and apply this approach to estimating the population attributable fraction (PAF). Our approach produces substantially more accurate estimates of the PAF in some settings than the usual approach which uses logistic regression without monotonicity constraints.
Ju, Jin Hyun; Crystal, Ronald G.
2017-01-01
Genome-wide expression Quantitative Trait Loci (eQTL) studies in humans have provided numerous insights into the genetics of both gene expression and complex diseases. While the majority of eQTL identified in genome-wide analyses impact a single gene, eQTL that impact many genes are particularly valuable for network modeling and disease analysis. To enable the identification of such broad impact eQTL, we introduce CONFETI: Confounding Factor Estimation Through Independent component analysis. CONFETI is designed to address two conflicting issues when searching for broad impact eQTL: the need to account for non-genetic confounding factors that can lower the power of the analysis or produce broad impact eQTL false positives, and the tendency of methods that account for confounding factors to model broad impact eQTL as non-genetic variation. The key advance of the CONFETI framework is the use of Independent Component Analysis (ICA) to identify variation likely caused by broad impact eQTL when constructing the sample covariance matrix used for the random effect in a mixed model. We show that CONFETI has better performance than other mixed model confounding factor methods when considering broad impact eQTL recovery from synthetic data. We also used the CONFETI framework and these same confounding factor methods to identify eQTL that replicate between matched twin pair datasets in the Multiple Tissue Human Expression Resource (MuTHER), the Depression Genes Networks study (DGN), the Netherlands Study of Depression and Anxiety (NESDA), and multiple tissue types in the Genotype-Tissue Expression (GTEx) consortium. These analyses identified both cis-eQTL and trans-eQTL impacting individual genes, and CONFETI had better or comparable performance to other mixed model confounding factor analysis methods when identifying such eQTL. In these analyses, we were able to identify and replicate a few broad impact eQTL although the overall number was small even when applying CONFETI. In light of these results, we discuss the broad impact eQTL that have been previously reported from the analysis of human data and suggest that considerable caution should be exercised when making biological inferences based on these reported eQTL. PMID:28505156
Ju, Jin Hyun; Shenoy, Sushila A; Crystal, Ronald G; Mezey, Jason G
2017-05-01
Genome-wide expression Quantitative Trait Loci (eQTL) studies in humans have provided numerous insights into the genetics of both gene expression and complex diseases. While the majority of eQTL identified in genome-wide analyses impact a single gene, eQTL that impact many genes are particularly valuable for network modeling and disease analysis. To enable the identification of such broad impact eQTL, we introduce CONFETI: Confounding Factor Estimation Through Independent component analysis. CONFETI is designed to address two conflicting issues when searching for broad impact eQTL: the need to account for non-genetic confounding factors that can lower the power of the analysis or produce broad impact eQTL false positives, and the tendency of methods that account for confounding factors to model broad impact eQTL as non-genetic variation. The key advance of the CONFETI framework is the use of Independent Component Analysis (ICA) to identify variation likely caused by broad impact eQTL when constructing the sample covariance matrix used for the random effect in a mixed model. We show that CONFETI has better performance than other mixed model confounding factor methods when considering broad impact eQTL recovery from synthetic data. We also used the CONFETI framework and these same confounding factor methods to identify eQTL that replicate between matched twin pair datasets in the Multiple Tissue Human Expression Resource (MuTHER), the Depression Genes Networks study (DGN), the Netherlands Study of Depression and Anxiety (NESDA), and multiple tissue types in the Genotype-Tissue Expression (GTEx) consortium. These analyses identified both cis-eQTL and trans-eQTL impacting individual genes, and CONFETI had better or comparable performance to other mixed model confounding factor analysis methods when identifying such eQTL. In these analyses, we were able to identify and replicate a few broad impact eQTL although the overall number was small even when applying CONFETI. In light of these results, we discuss the broad impact eQTL that have been previously reported from the analysis of human data and suggest that considerable caution should be exercised when making biological inferences based on these reported eQTL.
Zhang, Rui; Beyer, Frauke; Lampe, Leonie; Luck, Tobias; Riedel-Heller, Steffi G; Loeffler, Markus; Schroeter, Matthias L; Stumvoll, Michael; Villringer, Arno; Witte, A Veronica
2018-05-15
Obesity has been linked with structural and functional brain changes. However, the impact of obesity on brain and cognition in aging remains debatable, especially for white matter. We therefore aimed to determine the effects of obesity on white matter microstructure and potential implications for cognition in a well-characterized large cohort of healthy adults. In total, 1255 participants (50% females, 19-80 years, BMI 16.8-50.2 kg/m 2 ) with diffusion-weighted magnetic resonance imaging at 3T were analysed. Tract-based spatial statistics (TBSS) probed whether body mass index (BMI) and waist-to-hip ratio (WHR) were related to fractional anisotropy (FA). We conducted partial correlations and mediation analyses to explore whether obesity or regional FA were related to cognitive performance. Analyses were adjusted for demographic, genetic, and obesity-associated confounders. Results showed that higher BMI and higher WHR were associated with lower FA in multiple white matter tracts (p < 0.05, FWE-corrected). Mediation analyses provided evidence for indirect negative effects of higher BMI and higher WHR on executive functions and processing speed through lower FA in fiber tracts connecting (pre)frontal, visual, and associative areas (indirect paths, |ß| ≥ 0.01; 99% |CI| > 0). This large cross-sectional study showed that obesity is correlated with lower FA in multiple white matter tracts in otherwise healthy adults, independent of confounders. Moreover, although effect sizes were small, mediation results indicated that visceral obesity was linked to poorer executive functions and lower processing speed through lower FA in callosal and associative fiber tracts. Longitudinal studies are needed to support this hypothesis. Copyright © 2018 Elsevier Inc. All rights reserved.
Koch, E; Bogado, M; Araya, F; Romero, T; Díaz, C; Manriquez, L; Paredes, M; Román, C; Taylor, A; Kirschbaum, A
2008-05-01
To find out whether there is an association between parity and obesity, evaluated through body mass index (BMI), waist circumference (WC), waist-to-hip ratio (WHR) and waist-to-height ratio (WHtR) in Chilean women after controlling for sociodemographic characteristics, health risk and gynaeco-obstetric factors. Cross-sectional study, using baseline data of the San Francisco Project. San Francisco de Mostazal, located in the central region of Chile, 6512 Chilean-Hispanic women (Spanish heritage with a variable indigenous component). A weighted random sample of 508 women who had their first pregnancy inside the primary child-bearing ages. Data were collected between 1997 and 1999. Statistical associations between parity and different anthropometric measurements of adiposity in multiple linear (MLnR) and logistic regression models (MLtR) were evaluated. In MLnR a modest parity-related increment in BMI and practically null increment in WC, WHR and WHtR was observed. Covariates that showed a statistically significant association with anthropometric measures of adiposity were age, low education, marital status, employment, smoking, smoking cessation, hypertension, diabetes, dyslipidaemia, parent's obesity, menarche and fetal macrosomia. Crude odds ratio (OR) showed a strong association between parity and anthropometric markers of obesity. Nevertheless, after adjustments in MLtR models, the association remained only for BMI. All the measures of abdominal obesity related to parous women showed OR smaller than 1 (95% confidence intervals 0.57 to 0.96). Parity modestly influences BMI, but does not seem to be related to WC, WHR and WHtR after controlling by confounders. Parity can increase adiposity but not necessarily following an abdominal pattern.
THE EXPERIMENTAL DESIGN FOR BeO IRAADIATION EXPERIMENTS ORNL 41-8 AND ORNL 41-9
DOE Office of Scientific and Technical Information (OSTI.GOV)
Gardiner, D.A.
1962-07-18
The experimental plan for irradiating BeO pellets in Experiments ORNL 41- 8 and ORNL 41-9 was chosen in accordance with the principles of experimental design. The design is known by statisticians as a 2/sup 5/ factorial experiment confound'' in six replications. Five variables---size, density, grain size, temperature and time--are controlled at two levels to form the basic 2i factorial experiment. The sixth variable, neutron flux, is introduced by confounding on higher-order interactions. An explanation is presented in nontechnical language the means by which the aims of the experimenters and the physical conditions affecting the experiment were utilized in constructing themore » experimental design. (auth)« less
ERIC Educational Resources Information Center
Hevel, Michael S.; Weeden, Dustin D.; Pasquesi, Kira; Pascarella, Ernest T.
2015-01-01
We use a longitudinal dataset to explore the effect of fraternity/sorority membership on political orientation and social/political activism. After controlling for a variety of potentially confounding variables, including pretests on the dependent variables, fraternity and sorority members were significantly less liberal that their unaffiliated…
Beyond Free Lunch: Which Family Background Measures Matter?
ERIC Educational Resources Information Center
Lubienski, Sarah Theule; Crane, Corinna Crawford
2010-01-01
Most studies of school achievement use free-lunch eligibility or other basic indicators to adjust for differences in students' socioeconomic backgrounds. This study determines whether these variables are enough to separate the confounding effects of students' backgrounds from the main variables of interest in education studies. The Early Childhood…
The Causal Mediation Formula - A Guide to the Assessment of Pathways and Mechanisms
2011-10-01
for estimating CDE(z) in observational studies (in the presence of unobserved confounders) can be derived using do- calculus (Pearl, 2009, pp. 85–88...2011; Albert and Nelson , 2011). 7In the presence of measured and unmeasured confounders, the general conditions under which NDE is estimable from...achieve this independence (see footnote 7), that Z may represent a vector of variables, and that integrals should replace summations when dealing with
Nguyen, Trang Quynh; Webb-Vargas, Yenny; Koning, Ina M.; Stuart, Elizabeth A.
2016-01-01
We investigate a method to estimate the combined effect of multiple continuous/ordinal mediators on a binary outcome: 1) fit a structural equation model with probit link for the outcome and identity/probit link for continuous/ordinal mediators, 2) predict potential outcome probabilities, and 3) compute natural direct and indirect effects. Step 2 involves rescaling the latent continuous variable underlying the outcome to address residual mediator variance/covariance. We evaluate the estimation of risk-difference- and risk-ratio-based effects (RDs, RRs) using the ML, WLSMV and Bayes estimators in Mplus. Across most variations in path-coefficient and mediator-residual-correlation signs and strengths, and confounding situations investigated, the method performs well with all estimators, but favors ML/WLSMV for RDs with continuous mediators, and Bayes for RRs with ordinal mediators. Bayes outperforms WLSMV/ML regardless of mediator type when estimating RRs with small potential outcome probabilities and in two other special cases. An adolescent alcohol prevention study is used for illustration. PMID:27158217
Measuring the surgical 'learning curve': methods, variables and competency.
Khan, Nuzhath; Abboudi, Hamid; Khan, Mohammed Shamim; Dasgupta, Prokar; Ahmed, Kamran
2014-03-01
To describe how learning curves are measured and what procedural variables are used to establish a 'learning curve' (LC). To assess whether LCs are a valuable measure of competency. A review of the surgical literature pertaining to LCs was conducted using the Medline and OVID databases. Variables should be fully defined and when possible, patient-specific variables should be used. Trainee's prior experience and level of supervision should be quantified; the case mix and complexity should ideally be constant. Logistic regression may be used to control for confounding variables. Ideally, a learning plateau should reach a predefined/expert-derived competency level, which should be fully defined. When the group splitting method is used, smaller cohorts should be used in order to narrow the range of the LC. Simulation technology and competence-based objective assessments may be used in training and assessment in LC studies. Measuring the surgical LC has potential benefits for patient safety and surgical education. However, standardisation in the methods and variables used to measure LCs is required. Confounding variables, such as participant's prior experience, case mix, difficulty of procedures and level of supervision, should be controlled. Competency and expert performance should be fully defined. © 2013 The Authors. BJU International © 2013 BJU International.
McLaughlin, C.J.; Smith, C.A.; Buddemeier, R.W.; Bartley, J.D.; Maxwell, B.A.
2003-01-01
The role of terrigenous sediment in controlling the occurrence of coral reef ecosystems is qualitatively understood and has been studied at local scales, but has not been systematically evaluated on a global-to-regional scale. Current concerns about degradation of reef environments and alteration of the hydrologic and sediment cycles place the issue at a focal point of multiple environmental concerns. We use a geospatial clustering of a coastal zone database of river and local runoff identified with 0.5?? grid cells to identify areas of high potential runoff effects, and combine this with a database of reported coral reef locations. Coastal cells with high runoff values are much less likely to contain reefs than low runoff cells and GIS buffer analysis demonstrates that this inhibition extends to offshore ocean cells as well. This analysis does not uniquely define the effects of sediment, since salinity, nutrients, and contaminants are potentially confounding variables also associated with runoff. However, sediment effects are likely to be a major factor and a basis is provided for extending the study to higher resolution with more specific variables. ?? 2003 Elsevier B.V. All rights reserved.
Amagasa, Takashi; Nakayama, Takeo
2012-07-01
To test the hypothesis that relationship reported between long working hours and depression was inconsistent in previous studies because job demand was treated as a confounder. Structural equation modeling was used to construct five models, using work-related factors and depressive mood scale obtained from 218 clerical workers, to test for goodness of fit and was externally validated with data obtained from 1160 sales workers. Multiple logistic regression analysis was also performed. The model that showed that long working hours increased depression risk when job demand was regarded as an intermediate variable was the best fitted model (goodness-of-fit index/root-mean-square error of approximation: 0.981 to 0.996/0.042 to 0.044). The odds ratio for depression risk with work that was high demand and 60 hours or more per week was estimated at 2 to 4 versus work that was low demand and less than 60 hours per week. Long working hours increased depression risk, with job demand being an intermediate variable.
Milner, Allison; Aitken, Zoe; Kavanagh, Anne; LaMontagne, Anthony D; Pega, Frank; Petrie, Dennis
2017-06-23
Previous studies suggest that poor psychosocial job quality is a risk factor for mental health problems, but they use conventional regression analytic methods that cannot rule out reverse causation, unmeasured time-invariant confounding and reporting bias. This study combines two quasi-experimental approaches to improve causal inference by better accounting for these biases: (i) linear fixed effects regression analysis and (ii) linear instrumental variable analysis. We extract 13 annual waves of national cohort data including 13 260 working-age (18-64 years) employees. The exposure variable is self-reported level of psychosocial job quality. The instruments used are two common workplace entitlements. The outcome variable is the Mental Health Inventory (MHI-5). We adjust for measured time-varying confounders. In the fixed effects regression analysis adjusted for time-varying confounders, a 1-point increase in psychosocial job quality is associated with a 1.28-point improvement in mental health on the MHI-5 scale (95% CI: 1.17, 1.40; P < 0.001). When the fixed effects was combined with the instrumental variable analysis, a 1-point increase psychosocial job quality is related to 1.62-point improvement on the MHI-5 scale (95% CI: -0.24, 3.48; P = 0.088). Our quasi-experimental results provide evidence to confirm job stressors as risk factors for mental ill health using methods that improve causal inference. © The Author 2017. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
Cloern, J.E.; Canuel, E.A.; Harris, D.
2002-01-01
We report measurements of seasonal variability in the C-N stable isotope ratios of plants collected across the habitat mosaic of San Francisco Bay, its marshes, and its tributary river system. Analyses of 868 plant samples were binned into 10 groups (e.g., terrestrial riparian, freshwater phytoplankton, salt marsh) to determine whether C-N isotopes can be used as biomarkers for tracing the origins of organic matter in this river-marsh-estuary complex. Variability of ??13C and ??15N was high (???5-10???) within each plant group, and we identified three modes of variability: (1) between species and their microhabitats, (2) over annual cycles of plant growth and senescence, and (3) between living and decomposing biomass. These modes of within-group variability obscure any source-specific isotopic signatures, confounding the application of C-N isotopes for identifying the origins of organic matter. A second confounding factor was large dissimilarity between the ??13C-??15N of primary producers and the organic-matter pools in the seston and sediments. Both confounding factors impede the application of C-N isotopes to reveal the food supply to primary consumers in ecosystems supporting diverse autotrophs and where the isotopic composition of organic matter has been transformed and become distinct from that of its parent plant sources. Our results support the advice of others: variability of C-N stable isotopes within all organic-matter pools is high and must be considered in applications of these isotopes to trace trophic linkages from primary producers to primary consumers. Isotope-based approaches are perhaps most powerful when used to complement other tools, such as molecular biomarkers, bioassays, direct measures of production, and compilations of organic-matter budgets.
Goodnight, Jackson A.; D’Onofrio, Brian M.; Cherlin, Andrew J.; Emery, Robert E.; Van Hulle, Carol A.; Lahey, Benjamin B.
2012-01-01
Previous studies of the association between multiple parental relationship transitions (i.e., when a parent begins or terminates an intimate relationship involving cohabitation) and offspring antisocial behavior have varied in their efforts to rule out confounding influences, such as parental antisocial behavior and low income. They also have been limited in the representativeness of their samples. Thus, it remains unclear to what degree parents’ multiple relationship transitions have independent effects on children’s antisocial behavior. Analyses were conducted using data on 8,652 6–9-year-old, 6,911 10–13-year-old, and 6,495 14-17-year-old offspring of a nationally representative sample of U.S. women. Cousin-comparisons were used in combination with statistical covariates to evaluate the associations between maternal relationship transitions and offspring antisocial behavior in childhood and adolescence. Cousin-comparisons suggested that associations between maternal relationship transitions and antisocial behavior in childhood and early adolescence are largely explained by confounding factors. In contrast, the associations between maternal relationship transitions and offspring delinquency in late adolescence were robust to measured and unmeasured confounds. The present findings suggest that interventions aimed at reducing exposure to parental relationship transitions or addressing the psychosocial consequences of exposure to parental relationship transitions could reduce risk for offspring delinquency in late adolescence. PMID:22829173
Saez, Marc; Barceló, Maria A; Farrerons, Mònica; López-Casasnovas, Guillem
2018-06-08
A number of factors contribute to attention deficit hyperactivity disorder (ADHD) and although they are not fully known, the occurrence of ADHD seems to be a consequence of an interaction between multiple genetic and environmental factors. However, apart from pesticides, the evidence is inadequate and inconsistent as it differs not only in the population and time period analysed, but also in the type of study, the control of the confounding variables and the statistical methods used. In the latter case, the studies also differ in the adjustment of spatial and temporal variability. Our objective here, is to provide evidence on an association between environmental factors and ADHD. In our study, we used a population-based retrospective cohort in which we matched cases and controls (children free of the disease) by sex and year of birth (n = 5193, 78.9% boys). The cases were children born between 1998 and 2012 and diagnosed with ADHD (n = 116). To evaluate whether there was a geographical pattern in the incidence of ADHD, we first represented the smoothed standardized incidence rates on a map of the region being studied. We then estimated the probability of being a case by using a generalized liner mixed model with a binomial link. As explanatory variables of interest, we included the following environmental variables: distance to agricultural areas, distance to roads (stratified into three categories according to traffic density and intensity), distance to petrol stations, distance to industrial estates, and land use. We control for both observed (individual and family specific variables and deprivation index) and unobserved confounders (in particular, individual and familial heterogeneity). In addition, we adjusted for spatial extra variability. We found a north-south pattern containing two clusters (one in the centre of the study region and another in the south) in relation to the risk of developing ADHD. The results from the multivariate model suggest that these clusters could be related to some of the environmental variables. Specifically, living within 100 m from an agricultural area or a residential street and/or living fewer than 300 m from a motorway, dual carriageway or one of the industrial estates analysed was associated (statistically significant) with an increased risk of ADHD. Our results indicate that some environmental factors could be associated with ADHD occurring, particularly those associated with exposure to pesticides, organochlorine compounds and air pollutants because of traffic. Copyright © 2018 Elsevier Inc. All rights reserved.
Karim, Mohammad Ehsanul; Petkau, John; Gustafson, Paul; Platt, Robert W.; Tremlett, Helen
2017-01-01
In longitudinal studies, if the time-dependent covariates are affected by the past treatment, time-dependent confounding may be present. For a time-to-event response, marginal structural Cox models (MSCMs) are frequently used to deal with such confounding. To avoid some of the problems of fitting MSCM, the sequential Cox approach has been suggested as an alternative. Although the estimation mechanisms are different, both approaches claim to estimate the causal effect of treatment by appropriately adjusting for time-dependent confounding. We carry out simulation studies to assess the suitability of the sequential Cox approach for analyzing time-to-event data in the presence of a time-dependent covariate that may or may not be a time-dependent confounder. Results from these simulations revealed that the sequential Cox approach is not as effective as MSCM in addressing the time-dependent confounding. The sequential Cox approach was also found to be inadequate in the presence of a time-dependent covariate. We propose a modified version of the sequential Cox approach that correctly estimates the treatment effect in both of the above scenarios. All approaches are applied to investigate the impact of beta-interferon treatment in delaying disability progression in the British Columbia Multiple Sclerosis cohort (1995 – 2008). PMID:27659168
Karim, Mohammad Ehsanul; Petkau, John; Gustafson, Paul; Platt, Robert W; Tremlett, Helen
2018-06-01
In longitudinal studies, if the time-dependent covariates are affected by the past treatment, time-dependent confounding may be present. For a time-to-event response, marginal structural Cox models are frequently used to deal with such confounding. To avoid some of the problems of fitting marginal structural Cox model, the sequential Cox approach has been suggested as an alternative. Although the estimation mechanisms are different, both approaches claim to estimate the causal effect of treatment by appropriately adjusting for time-dependent confounding. We carry out simulation studies to assess the suitability of the sequential Cox approach for analyzing time-to-event data in the presence of a time-dependent covariate that may or may not be a time-dependent confounder. Results from these simulations revealed that the sequential Cox approach is not as effective as marginal structural Cox model in addressing the time-dependent confounding. The sequential Cox approach was also found to be inadequate in the presence of a time-dependent covariate. We propose a modified version of the sequential Cox approach that correctly estimates the treatment effect in both of the above scenarios. All approaches are applied to investigate the impact of beta-interferon treatment in delaying disability progression in the British Columbia Multiple Sclerosis cohort (1995-2008).
Identifying cis-mediators for trans-eQTLs across many human tissues using genomic mediation analysis
Yang, Fan; Wang, Jiebiao; Pierce, Brandon L.; Chen, Lin S.
2017-01-01
The impact of inherited genetic variation on gene expression in humans is well-established. The majority of known expression quantitative trait loci (eQTLs) impact expression of local genes (cis-eQTLs). More research is needed to identify effects of genetic variation on distant genes (trans-eQTLs) and understand their biological mechanisms. One common trans-eQTLs mechanism is “mediation” by a local (cis) transcript. Thus, mediation analysis can be applied to genome-wide SNP and expression data in order to identify transcripts that are “cis-mediators” of trans-eQTLs, including those “cis-hubs” involved in regulation of many trans-genes. Identifying such mediators helps us understand regulatory networks and suggests biological mechanisms underlying trans-eQTLs, both of which are relevant for understanding susceptibility to complex diseases. The multitissue expression data from the Genotype-Tissue Expression (GTEx) program provides a unique opportunity to study cis-mediation across human tissue types. However, the presence of complex hidden confounding effects in biological systems can make mediation analyses challenging and prone to confounding bias, particularly when conducted among diverse samples. To address this problem, we propose a new method: Genomic Mediation analysis with Adaptive Confounding adjustment (GMAC). It enables the search of a very large pool of variables, and adaptively selects potential confounding variables for each mediation test. Analyses of simulated data and GTEx data demonstrate that the adaptive selection of confounders by GMAC improves the power and precision of mediation analysis. Application of GMAC to GTEx data provides new insights into the observed patterns of cis-hubs and trans-eQTL regulation across tissue types. PMID:29021290
Yang, Fan; Wang, Jiebiao; Pierce, Brandon L; Chen, Lin S
2017-11-01
The impact of inherited genetic variation on gene expression in humans is well-established. The majority of known expression quantitative trait loci (eQTLs) impact expression of local genes ( cis -eQTLs). More research is needed to identify effects of genetic variation on distant genes ( trans -eQTLs) and understand their biological mechanisms. One common trans -eQTLs mechanism is "mediation" by a local ( cis ) transcript. Thus, mediation analysis can be applied to genome-wide SNP and expression data in order to identify transcripts that are " cis -mediators" of trans -eQTLs, including those " cis -hubs" involved in regulation of many trans -genes. Identifying such mediators helps us understand regulatory networks and suggests biological mechanisms underlying trans -eQTLs, both of which are relevant for understanding susceptibility to complex diseases. The multitissue expression data from the Genotype-Tissue Expression (GTEx) program provides a unique opportunity to study cis -mediation across human tissue types. However, the presence of complex hidden confounding effects in biological systems can make mediation analyses challenging and prone to confounding bias, particularly when conducted among diverse samples. To address this problem, we propose a new method: Genomic Mediation analysis with Adaptive Confounding adjustment (GMAC). It enables the search of a very large pool of variables, and adaptively selects potential confounding variables for each mediation test. Analyses of simulated data and GTEx data demonstrate that the adaptive selection of confounders by GMAC improves the power and precision of mediation analysis. Application of GMAC to GTEx data provides new insights into the observed patterns of cis -hubs and trans -eQTL regulation across tissue types. © 2017 Yang et al.; Published by Cold Spring Harbor Laboratory Press.
“Skill of Generalized Additive Model to Detect PM2.5 Health ...
Summary. Measures of health outcomes are collinear with meteorology and air quality, making analysis of connections between human health and air quality difficult. The purpose of this analysis was to determine time scales and periods shared by the variables of interest (and by implication scales and periods that are not shared). Hospital admissions, meteorology (temperature and relative humidity), and air quality (PM2.5 and daily maximum ozone) for New York City during the period 2000-2006 were decomposed into temporal scales ranging from 2 days to greater than two years using a complex wavelet transform. Health effects were modeled as functions of the wavelet components of meteorology and air quality using the generalized additive model (GAM) framework. This simulation study showed that GAM is extremely successful at extracting and estimating a health effect embedded in a dataset. It also shows that, if the objective in mind is to estimate the health signal but not to fully explain this signal, a simple GAM model with a single confounder (calendar time) whose smooth representation includes a sufficient number of constraints is as good as a more complex model.Introduction. In the context of wavelet regression, confounding occurs when two or more independent variables interact with the dependent variable at the same frequency. Confounding also acts on a variety of time scales, changing the PM2.5 coefficient (magnitude and sign) and its significance ac
Cannabis use and risk of lung cancer: a case-control study.
Aldington, S; Harwood, M; Cox, B; Weatherall, M; Beckert, L; Hansell, A; Pritchard, A; Robinson, G; Beasley, R
2008-02-01
The aim of the present study was to determine the risk of lung cancer associated with cannabis smoking. A case-control study of lung cancer in adults
Effect of Placenta Previa on Preeclampsia
Ying, Hao; Lu, Yi; Dong, Yi-Nuo; Wang, De-Fen
2016-01-01
Background The correlation between gestational hypertension-preeclampsia (GH-PE) and placenta previa (PP) is controversial. Specifically, it is unknown whether placenta previa has any effect on the various types of preeclampsia (PE), and the role PP with concurrent placenta accreta (PA) play in the occurrence of GH-PE are not well understood. Objective The aim of this study was to identify the effects of PP on GH, mild and severe preeclampsia (MPE and SPE), and early- and late-onset preeclampsia (EPE and LPE). Another aim of the study was to determine if concurrent PA impacts the relationship between PP and GH-PE. Methods A retrospective single-center study of 1,058 patients having singleton pregnancies with PP was performed, and 2,116 pregnant women were randomly included as controls. These cases were collected from a tertiary hospital and met the inclusion criteria for the study. Clinical information, including PP and the gestational age at the onset of GH-PE were collected. Binary and multiple logistic regression analyses were conducted after the confounding variables were controlled to assess the effects of PP on different types of GH-PE. Results There were 155 patients with GH-PE in the two groups. The incidences of GH-PE in the PP group and the control group were 2.5% (26/1058) and 6.1% (129/2116), respectively (P = 0.000). Binary and multiple regression analyses were conducted after controlling for confounding variables. Compared to the control group, in the PP group, the risk of GH-PE was reduced significantly by 78% (AOR: 0.216; 95% CI: 0.135–0.345); the risks of GH and PE were reduced by 55% (AOR: 0.451; 95% CI: 0.233–0.873) and 86% (AOR: 0.141; 95% CI: 0.073–0.271), respectively; the risks of MPE and SPE were reduced by 73% (AOR: 0.269; 95% CI: 0.087–0828) and 88% (AOR: 0.123; 95% CI: 0.055–0.279), respectively; and the risks of EPE and LPE were reduced by 95% (AOR: 0.047; 95% CI: 0.012–0.190) and 67% (AOR: 0.330; 95% CI: 0.153–0.715), respectively. The incidence of concurrent PA in women with PP was 5.86%; PP with PA did not significantly further reduce the incidence of GH-PE compared with PP without PA (1.64% vs. 2.51%, P>0.05). Binary logistic regression analyses were conducted after controlling for confounding variables, compared with the non-PP + GH-PE group, and the AOR of FGR in the non-PP + non-GH-PE group was 0.206 (0.124–0.342). Compared with the PP + GH-PE group, the AOR of FGR in the PP + non-GH-PE group was 0.430 (0.123–1.500). Conclusion PP is not only associated with a significant reduction in the incidence of GH-PE, but also is associated with a reduction in incidence of various types of PE. Concurrent PA and PP do not show association with a reduction in incidence of GH-PE. PMID:26731265
"Fundamental Causes" of Social Inequalities in Mortality: A Test of the Theory
ERIC Educational Resources Information Center
Phelan, Jo C.; Link, Bruce G.; Diez-Roux, Ana; Kawachi, Ichiro; Levin, Bruce
2004-01-01
Medicine and epidemiology currently dominate the study of the strong association between socioeconomic status and mortality. Socioeconomic status typically is viewed as a causally irrelevant "confounding variable" or as a less critical variable marking only the beginning of a causal chain in which intervening risk factors are given prominence. Yet…
ERIC Educational Resources Information Center
Street, Nathan Lee
2017-01-01
Teacher value-added measures (VAM) are designed to provide information regarding teachers' causal impact on the academic growth of students while controlling for exogenous variables. While some researchers contend VAMs successfully and authentically measure teacher causality on learning, others suggest VAMs cannot adequately control for exogenous…
ERIC Educational Resources Information Center
Debruine, Lisa M.; Jones, Benedict C.; Smith, Finlay G.; Little, Anthony C.
2010-01-01
Women's preferences for male masculinity are highly variable. Although many researchers explain this variability as reflecting systematic individual differences in how women resolve the tradeoff between the costs and benefits of choosing a masculine partner, others suggest that methodological differences between studies are responsible. A recent…
Adolescent Substance Use, Sleep, and Academic Achievement: Evidence of Harm Due to Caffeine
ERIC Educational Resources Information Center
James, Jack E.; Kristjansson, Alfgeir Logi; Sigfusdottir, Inga Dora
2011-01-01
Using academic achievement as the key outcome variable, 7377 Icelandic adolescents were surveyed for cigarette smoking, alcohol use, daytime sleepiness, caffeine use, and potential confounders. Structural equation modeling (SEM) was used to examine direct and indirect effects of measured and latent variables in two models: the first with caffeine…
Goldstein, R.M.; Carlisle, D.M.; Meador, M.R.; Short, T.M.
2007-01-01
The environmental setting (e.g., climate, topography, geology) and land use affect stream physical characteristics singly and cumulatively. At broad geographic scales, we determined the importance of environmental setting and land use in explaining variation in stream physical characteristics. We hypothesized that as the spatial scale decreased from national to regional, land use would explain more of the variation in stream physical characteristics because environmental settings become more homogeneous. At a national scale, stepwise linear regression indicated that environmental setting was more important in explaining variability in stream physical characteristics. Although statistically discernible, the amount of variation explained by land use was not remarkable due to low partial correlations. At level II ecoregion spatial scales (southeastern USA plains, central USA plains, and a combination of the western Cordillera and the western interior basins and ranges), environmental setting variables were again more important predictors of stream physical characteristics, however, as the spatial scale decreased from national to regional, the portion of variability in stream physical characteristics explained by basin land use increased. Development of stream habitat indicators of land use will depend upon an understanding of relations between stream physical characteristics and environmental factors at multiple spatial scales. Smaller spatial scales will be necessary to reduce the confounding effects of variable environmental settings before the effects of land use can be reliably assessed. ?? Springer Science+Business Media B.V. 2006.
Yiin, James H; Anderson, Jeri L; Bertke, Stephen J; Tollerud, David J
2018-05-09
To examine dose-response relationships between internal uranium exposures and select outcomes among a cohort of uranium enrichment workers. Cox regression was conducted to examine associations between selected health outcomes and cumulative internal uranium with consideration for external ionizing radiation, work-related medical X-rays and contaminant radionuclides technetium ( 99 Tc) and plutonium ( 239 Pu) as potential confounders. Elevated and monotonically increasing mortality risks were observed for kidney cancer, chronic renal diseases, and multiple myeloma, and the association with internal uranium absorbed organ dose was statistically significant for multiple myeloma. Adjustment for potential confounders had minimal impact on the risk estimates. Kidney cancer, chronic renal disease, and multiple myeloma mortality risks were elevated with increasing internal uranium absorbed organ dose. The findings add to evidence of an association between internal exposure to uranium and cancer. Future investigation includes a study of cancer incidence in this cohort. © 2018 Wiley Periodicals, Inc.
NASA Astrophysics Data System (ADS)
Chaibub Neto, Elias
2016-11-01
Clinical trials traditionally employ blinding as a design mechanism to reduce the influence of placebo effects. In practice, however, it can be difficult or impossible to blind study participants and unblinded trials are common in medical research. Here we show how instrumental variables can be used to quantify and disentangle treatment and placebo effects in randomized clinical trials comparing control and active treatments in the presence of confounders. The key idea is to use randomization to separately manipulate treatment assignment and psychological encouragement conversations/interactions that increase the participants’ desire for improved symptoms. The proposed approach is able to improve the estimation of treatment effects in blinded studies and, most importantly, opens the doors to account for placebo effects in unblinded trials.
Brody, Gene H.; Lei, Man-Kit; Chae, David H.; Yu, Tianyi; Kogan, Steven M.; Beach, Steven R. H.
2013-01-01
This study was designed to examine the prospective relations of perceived racial discrimination with allostatic load (AL), along with a possible buffer of the association. A sample of 331 African Americans in the rural South provided assessments of perceived discrimination from ages 16 to 18 years. When youths were 18, caregivers reported parental emotional support, and youths assessed peer emotional support. AL and potential confounder variables were assessed when youths were 20. Latent Growth Mixture Modeling identified two perceived discrimination classes: high and stable and low and increasing. Adolescents in the high and stable class evinced heightened AL even with confounder variables controlled. The racial discrimination to AL link was not significant for young adults who received high emotional support. PMID:24673162
Okello, James; Nakimuli-Mpungu, Etheldreda; Musisi, Seggane; Broekaert, Eric; Derluyn, Ilse
2013-11-01
The relationship between war-related trauma exposure, depressive symptoms and multiple risk behaviors among adolescents is less clear in sub-Saharan Africa. We analyzed data collected from a sample of school-going adolescents four years postwar. Participants completed interviews assessing various risk behaviors defined by the Youth Self Report (YSR) and a sexual risk behavior survey, and were screened for post-traumatic stress, anxiety and depression symptoms based on the Impact of Events Scale Revised (IESR) and Hopkins Symptom Checklist for Adolescents (HSCL-37A) respectively. Multivariate logistic regression was used to assess factors independently associated with multiple risk behaviors. The logistic regression model of Baron and Kenny (1986) was used to evaluate the mediating role of depression in the relationship between stressful war events and multiple risk behaviors. Of 551 participants, 139 (25%) reported multiple (three or more) risk behaviors in the past year. In the multivariate analyses, depression symptoms remained uniquely associated with multiple risk behavior after adjusting for potential confounders including socio-demographic characteristics, war-related trauma exposure variables, anxiety and post-traumatic stress symptoms. In mediation analysis, depression symptoms mediated the associations between stressful war events and multiple risk behaviors. The psychometric properties of the questionnaires used in this study are not well established in war affected African samples thus ethno cultural variation may decrease the validity of our measures. Adolescents with depression may be at a greater risk of increased engagement in multiple risk behaviors. Culturally sensitive and integrated interventions to treat and prevent depression among adolescents in post-conflict settings are urgently needed. © 2013 Elsevier B.V. All rights reserved.
The role of sex and body weight on the metabolic effects of high-fat diet in C57BL/6N mice.
Ingvorsen, C; Karp, N A; Lelliott, C J
2017-04-10
Metabolic disorders are commonly investigated using knockout and transgenic mouse models on the C57BL/6N genetic background due to its genetic susceptibility to the deleterious metabolic effects of high-fat diet (HFD). There is growing awareness of the need to consider sex in disease progression, but limited attention has been paid to sexual dimorphism in mouse models and its impact in metabolic phenotypes. We assessed the effect of HFD and the impact of sex on metabolic variables in this strain. We generated a reference data set encompassing glucose tolerance, body composition and plasma chemistry data from 586 C57BL/6N mice fed a standard chow and 733 fed a HFD collected as part of a high-throughput phenotyping pipeline. Linear mixed model regression analysis was used in a dual analysis to assess the effect of HFD as an absolute change in phenotype, but also as a relative change accounting for the potential confounding effect of body weight. HFD had a significant impact on all variables tested with an average absolute effect size of 29%. For the majority of variables (78%), the treatment effect was modified by sex and this was dominated by male-specific or a male stronger effect. On average, there was a 13.2% difference in the effect size between the male and female mice for sexually dimorphic variables. HFD led to a significant body weight phenotype (24% increase), which acts as a confounding effect on the other analysed variables. For 79% of the variables, body weight was found to be a significant source of variation, but even after accounting for this confounding effect, similar HFD-induced phenotypic changes were found to when not accounting for weight. HFD and sex are powerful modifiers of metabolic parameters in C57BL/6N mice. We also demonstrate the value of considering body size as a covariate to obtain a richer understanding of metabolic phenotypes.
Liu, Tianyi; Nie, Xiaolu; Wu, Zehao; Zhang, Ying; Feng, Guoshuang; Cai, Siyu; Lv, Yaqi; Peng, Xiaoxia
2017-12-29
Different confounder adjustment strategies were used to estimate odds ratios (ORs) in case-control study, i.e. how many confounders original studies adjusted and what the variables are. This secondary data analysis is aimed to detect whether there are potential biases caused by difference of confounding factor adjustment strategies in case-control study, and whether such bias would impact the summary effect size of meta-analysis. We included all meta-analyses that focused on the association between breast cancer and passive smoking among non-smoking women, as well as each original case-control studies included in these meta-analyses. The relative deviations (RDs) of each original study were calculated to detect how magnitude the adjustment would impact the estimation of ORs, compared with crude ORs. At the same time, a scatter diagram was sketched to describe the distribution of adjusted ORs with different number of adjusted confounders. Substantial inconsistency existed in meta-analysis of case-control studies, which would influence the precision of the summary effect size. First, mixed unadjusted and adjusted ORs were used to combine individual OR in majority of meta-analysis. Second, original studies with different adjustment strategies of confounders were combined, i.e. the number of adjusted confounders and different factors being adjusted in each original study. Third, adjustment did not make the effect size of original studies trend to constringency, which suggested that model fitting might have failed to correct the systematic error caused by confounding. The heterogeneity of confounder adjustment strategies in case-control studies may lead to further bias for summary effect size in meta-analyses, especially for weak or medium associations so that the direction of causal inference would be even reversed. Therefore, further methodological researches are needed, referring to the assessment of confounder adjustment strategies, as well as how to take this kind of bias into consideration when drawing conclusion based on summary estimation of meta-analyses.
Incidence of workers compensation indemnity claims across socio-demographic and job characteristics.
Du, Juan; Leigh, J Paul
2011-10-01
We hypothesized that low socioeconomic status, employer-provided health insurance, low wages, and overtime were predictors of reporting workers compensation indemnity claims. We also tested for gender and race disparities. Responses from 17,190 (person-years) Americans participating in the Panel Study of Income Dynamics, 1997-2005, were analyzed with logistic regressions. The dependent variable indicated whether the subject collected benefits from a claim. Odds ratios for men and African-Americans were relatively large and strongly significant predictors of claims; significance for Hispanics was moderate and confounded by education. Odds ratios for variables measuring education were the largest for all statistically significant covariates. Neither low wages nor employer-provided health insurance was a consistent predictor. Due to confounding from the "not salaried" variable, overtime was not a consistently significant predictor. Few studies use nationally representative longitudinal data to consider which demographic and job characteristics predict reporting workers compensation indemnity cases. This study did and tested some common hypotheses about predictors. Copyright © 2011 Wiley-Liss, Inc.
Doubly robust matching estimators for high dimensional confounding adjustment.
Antonelli, Joseph; Cefalu, Matthew; Palmer, Nathan; Agniel, Denis
2018-05-11
Valid estimation of treatment effects from observational data requires proper control of confounding. If the number of covariates is large relative to the number of observations, then controlling for all available covariates is infeasible. In cases where a sparsity condition holds, variable selection or penalization can reduce the dimension of the covariate space in a manner that allows for valid estimation of treatment effects. In this article, we propose matching on both the estimated propensity score and the estimated prognostic scores when the number of covariates is large relative to the number of observations. We derive asymptotic results for the matching estimator and show that it is doubly robust in the sense that only one of the two score models need be correct to obtain a consistent estimator. We show via simulation its effectiveness in controlling for confounding and highlight its potential to address nonlinear confounding. Finally, we apply the proposed procedure to analyze the effect of gender on prescription opioid use using insurance claims data. © 2018, The International Biometric Society.
Effect decomposition in the presence of an exposure-induced mediator-outcome confounder
VanderWeele, Tyler J.; Vansteelandt, Stijn; Robins, James M.
2014-01-01
Methods from causal mediation analysis have generalized the traditional approach to direct and indirect effects in the epidemiologic and social science literature by allowing for interaction and non-linearities. However, the methods from the causal inference literature have themselves been subject to a major limitation in that the so-called natural direct and indirect effects that are employed are not identified from data whenever there is a variable that is affected by the exposure, which also confounds the relationship between the mediator and the outcome. In this paper we describe three alternative approaches to effect decomposition that give quantities that can be interpreted as direct and indirect effects, and that can be identified from data even in the presence of an exposure-induced mediator-outcome confounder. We describe a simple weighting-based estimation method for each of these three approaches, illustrated with data from perinatal epidemiology. The methods described here can shed insight into pathways and questions of mediation even when an exposure-induced mediator-outcome confounder is present. PMID:24487213
CADDIS Volume 4. Data Analysis: Basic Principles & Issues
Use of inferential statistics in causal analysis, introduction to data independence and autocorrelation, methods to identifying and control for confounding variables, references for the Basic Principles section of Data Analysis.
Demographic factors, fatigue, and driving accidents: An examination of the published literature.
Di Milia, Lee; Smolensky, Michael H; Costa, Giovanni; Howarth, Heidi D; Ohayon, Maurice M; Philip, Pierre
2011-03-01
This article reviews the literature pertaining to the association between demographic variables (e.g., age, sex, race, socio-economic status) with fatigue, and when feasible, accident risk. It also explores their potential influence and interaction with some working arrangements, commute time, personality characteristics, and circadian chronotype. Fatigue has been implicated in a range of impairments that can have detrimental effects on individuals, and it is differentially associated with conventional demographic variables. However, several major methodological limitations prevent clear conclusions. First, there is absence of a shared definition both within and across disciplines. Second, although fatigue has been investigated using a variety of diverse designs, they have either been too weak to substantiate causality or lacked ecological validity. Third, while both subjective and objective measures have been used as dependent variables, fatigue has been more often found to be more strongly linked with the former. Fourth, with the exception of age and sex, the influence of other demographic variables is unknown, since they have not yet been concomitantly assessed. In instances when they have been assessed and included in statistical analyses, they are considered as covariates or confounders; thus, their contribution to the outcome variable is controlled for, rather than being a planned aspect of investigation. Because the interaction of demographic factors with fatigue is largely a neglected area of study, we recommend greater interdisciplinary collaborations, incorporation of multiple demographic variables as independent factors, and use of within-participant analyses. These recommendations would provide meaningful results that may be used to inform public policy and preventive strategies. Copyright © 2009 Elsevier Ltd. All rights reserved.
Clustering Words to Match Conditions: An Algorithm for Stimuli Selection in Factorial Designs
ERIC Educational Resources Information Center
Guasch, Marc; Haro, Juan; Boada, Roger
2017-01-01
With the increasing refinement of language processing models and the new discoveries about which variables can modulate these processes, stimuli selection for experiments with a factorial design is becoming a tough task. Selecting sets of words that differ in one variable, while matching these same words into dozens of other confounding variables…
Pendleton, Richard M.; Hoeinghaus, David J.; Gomes, Luiz C.; Agostinho, Angelo A.
2014-01-01
Experiments with realistic scenarios of species loss from multitrophic ecosystems may improve insight into how biodiversity affects ecosystem functioning. Using 1000 L mesocoms, we examined effects of nonrandom species loss on community structure and ecosystem functioning of experimental food webs based on multitrophic tropical floodplain lagoon ecosystems. Realistic biodiversity scenarios were developed based on long-term field surveys, and experimental assemblages replicated sequential loss of rare species which occurred across all trophic levels of these complex food webs. Response variables represented multiple components of ecosystem functioning, including nutrient cycling, primary and secondary production, organic matter accumulation and whole ecosystem metabolism. Species richness significantly affected ecosystem function, even after statistically controlling for potentially confounding factors such as total biomass and direct trophic interactions. Overall, loss of rare species was generally associated with lower nutrient concentrations, phytoplankton and zooplankton densities, and whole ecosystem metabolism when compared with more diverse assemblages. This pattern was also observed for overall ecosystem multifunctionality, a combined metric representing the ability of an ecosystem to simultaneously maintain multiple functions. One key exception was attributed to time-dependent effects of intraguild predation, which initially increased values for most ecosystem response variables, but resulted in decreases over time likely due to reduced nutrient remineralization by surviving predators. At the same time, loss of species did not result in strong trophic cascades, possibly a result of compensation and complexity of these multitrophic ecosystems along with a dominance of bottom-up effects. Our results indicate that although rare species may comprise minor components of communities, their loss can have profound ecosystem consequences across multiple trophic levels due to a combination of direct and indirect effects in diverse multitrophic ecosystems. PMID:24416246
Uddin, Md Jamal; Groenwold, Rolf H H; de Boer, Anthonius; Gardarsdottir, Helga; Martin, Elisa; Candore, Gianmario; Belitser, Svetlana V; Hoes, Arno W; Roes, Kit C B; Klungel, Olaf H
2016-03-01
Instrumental variable (IV) analysis can control for unmeasured confounding, yet it has not been widely used in pharmacoepidemiology. We aimed to assess the performance of IV analysis using different IVs in multiple databases in a study of antidepressant use and hip fracture. Information on adults with at least one prescription of a selective serotonin reuptake inhibitor (SSRI) or tricyclic antidepressant (TCA) during 2001-2009 was extracted from the THIN (UK), BIFAP (Spain), and Mondriaan (Netherlands) databases. IVs were created using the proportion of SSRI prescriptions per practice or using the one, five, or ten previous prescriptions by a physician. Data were analysed using conventional Cox regression and two-stage IV models. In the conventional analysis, SSRI (vs. TCA) was associated with an increased risk of hip fracture, which was consistently found across databases: the adjusted hazard ratio (HR) was approximately 1.35 for time-fixed and 1.50 to 2.49 for time-varying SSRI use, while the IV analysis based on the IVs that appeared to satisfy the IV assumptions showed conflicting results, e.g. the adjusted HRs ranged from 0.55 to 2.75 for time-fixed exposure. IVs for time-varying exposure violated at least one IV assumption and were therefore invalid. This multiple database study shows that the performance of IV analysis varied across the databases for time-fixed and time-varying exposures and strongly depends on the definition of IVs. It remains challenging to obtain valid IVs in pharmacoepidemiological studies, particularly for time-varying exposure, and IV analysis should therefore be interpreted cautiously. Copyright © 2016 John Wiley & Sons, Ltd.
Social roles and alcohol consumption: a study of 10 industrialised countries*
Kuntsche, Sandra; Knibbe, Ronald A.; Gmel, Gerhard
2014-01-01
The empirical evidence as regards the precise association between alcohol use and social roles, and these associations across genders and cultures is heterogeneous. The literature tends to focus on two central but conflicting theories. The first - classic role theory - assumes that a higher number of social roles is associated with a more structured life and thus fewer opportunities to drink heavily. The second - the multiple burden hypothesis - posits that the increasing complexity of multiple social roles leads to higher stress levels, and thus to increased alcohol use. Survey data on 25- to 54-year olds in ten western industrialised countries which participate in the GenACIS project were used to test whether holding the three main social roles - partnership, parenthood, and paid labour - had a more protective or a more detrimental association with problematic alcohol use than holding fewer roles. Age and education were included as possible confounders, while the outcome variables were risky single occasion drinking (RSOD) and heavy-volume drinking. For both genders and in almost all countries, the study found that those who had all three roles were least likely to drink heavily or engage in RSOD, thus supporting the assumptions of classic role theory. It also found that the protective effect of multiple roles was more consistent for RSOD. There were a few countries where a two-role model gave a better fit. Results for Germany (RSOD), Switzerland, and the US (heavy volume) indicate that the role of paid labour appears to be particularly relevant for risky alcohol use among women. Despite some variability in the association between paid labour and heavy drinking or RSOD among women, in almost all countries the greater the number of roles a person held, the lower their risk of this type of alcohol use was. PMID:19232807
Twentieth century warming of the tropical Atlantic captured by Sr-U paleothermometry
NASA Astrophysics Data System (ADS)
Alpert, Alice E.; Cohen, Anne L.; Oppo, Delia W.; DeCarlo, Thomas M.; Gaetani, Glenn A.; Hernandez-Delgado, Edwin A.; Winter, Amos; Gonneea, Meagan E.
2017-02-01
Coral skeletons are valuable archives of past ocean conditions. However, interpretation of coral paleotemperature records is confounded by uncertainties associated with single-element ratio thermometers, including Sr/Ca. A new approach, Sr-U, uses U/Ca to constrain the influence of Rayleigh fractionation on Sr/Ca. Here we build on the initial Pacific Porites Sr-U calibration to include multiple Atlantic and Pacific coral genera from multiple coral reef locations spanning a temperature range of 23.15-30.12°C. Accounting for the wintertime growth cessation of one Bermuda coral, we show that Sr-U is strongly correlated with the average water temperature at each location (r2 = 0.91, P < 0.001, n = 19). We applied the multispecies spatial calibration between Sr-U and temperature to reconstruct a 96 year long temperature record at Mona Island, Puerto Rico, using a coral not included in the calibration. Average Sr-U derived temperature for the period 1900-1996 is within 0.12°C of the average instrumental temperature at this site and captures the twentieth century warming trend of 0.06°C per decade. Sr-U also captures the timing of multiyear variability but with higher amplitude than implied by the instrumental data. Mean Sr-U temperatures and patterns of multiyear variability were replicated in a second coral in the same grid box. Conversely, Sr/Ca records from the same two corals were inconsistent with each other and failed to capture absolute sea temperatures, timing of multiyear variability, or the twentieth century warming trend. Our results suggest that coral Sr-U paleothermometry is a promising new tool for reconstruction of past ocean temperatures.
Costa, Andréa A; Serra-Negra, Júnia M; Bendo, Cristiane B; Pordeus, Isabela A; Paiva, Saul M
2016-01-01
To investigate the impact of wearing a fixed orthodontic appliance on oral health-related quality of life (OHRQoL) among adolescents. A case-control study (1 ∶ 2) was carried out with a population-based randomized sample of 327 adolescents aged 11 to 14 years enrolled at public and private schools in the City of Brumadinho, southeast of Brazil. The case group (n = 109) was made up of adolescents with a high negative impact on OHRQoL, and the control group (n = 218) was made up of adolescents with a low negative impact. The outcome variable was the impact on OHRQoL measured by the Brazilian version of the Child Perceptions Questionnaire (CPQ 11-14) - Impact Short Form (ISF:16). The main independent variable was wearing fixed orthodontic appliances. Malocclusion and the type of school were identified as possible confounding variables. Bivariate and multiple conditional logistic regressions were employed in the statistical analysis. A multiple conditional logistic regression model demonstrated that adolescents wearing fixed orthodontic appliances had a 4.88-fold greater chance of presenting high negative impact on OHRQoL (95% CI: 2.93-8.13; P < .001) than those who did not wear fixed orthodontic appliances. A bivariate conditional logistic regression demonstrated that malocclusion was significantly associated with OHRQoL (P = .017), whereas no statistically significant association was found between the type of school and OHRQoL (P = .108). Adolescents who wore fixed orthodontic appliances had a greater chance of reporting a negative impact on OHRQoL than those who did not wear such appliances.
Twentieth century warming of the tropical Atlantic captured by Sr-U paleothermometry
Alpert, Alice E.; Cohen, Anne L.; Oppo, Delia W.; DeCarlo, Thomas M.; Gaetani, Glenn A.; Hernandez-Delgado, Edwin A.; Winter, Amos; Gonneea, Meagan
2017-01-01
Coral skeletons are valuable archives of past ocean conditions. However, interpretation of coral paleotemperature records is confounded by uncertainties associated with single-element ratio thermometers, including Sr/Ca. A new approach, Sr-U, uses U/Ca to constrain the influence of Rayleigh fractionation on Sr/Ca. Here we build on the initial Pacific Porites Sr-U calibration to include multiple Atlantic and Pacific coral genera from multiple coral reef locations spanning a temperature range of 23.15–30.12°C. Accounting for the wintertime growth cessation of one Bermuda coral, we show that Sr-U is strongly correlated with the average water temperature at each location (r2 = 0.91, P < 0.001, n = 19). We applied the multispecies spatial calibration between Sr-U and temperature to reconstruct a 96 year long temperature record at Mona Island, Puerto Rico, using a coral not included in the calibration. Average Sr-U derived temperature for the period 1900–1996 is within 0.12°C of the average instrumental temperature at this site and captures the twentieth century warming trend of 0.06°C per decade. Sr-U also captures the timing of multiyear variability but with higher amplitude than implied by the instrumental data. Mean Sr-U temperatures and patterns of multiyear variability were replicated in a second coral in the same grid box. Conversely, Sr/Ca records from the same two corals were inconsistent with each other and failed to capture absolute sea temperatures, timing of multiyear variability, or the twentieth century warming trend. Our results suggest that coral Sr-U paleothermometry is a promising new tool for reconstruction of past ocean temperatures.
Effects of diazoxide in multiple sclerosis
Rovira, Alex; Montalban, Xavier; Arroyo, Rafael; Paul, Friedemann; Meca-Lallana, Virginia; Ramo, Cristina; Fernandez, Oscar; Saiz, Albert; Garcia-Merino, Antonio; Ramió-Torrentà, Lluís; Casanova, Bonaventura; Oreja-Guevara, Celia; Muñoz, Delicias; Martinez-Rodriguez, Jose Enrique; Lensch, Eckart; Prieto, Jose Maria; Meuth, Sven G.; Nuñez, Xavier; Campás, Clara; Pugliese, Marco
2015-01-01
Objective: The aim of this study was to test the safety of diazoxide and to search for signs of efficacy in patients with relapsing-remitting multiple sclerosis (RRMS). Methods: In this multicenter, randomized, placebo-controlled, double-blind trial (treatment allocation was concealed), 102 patients with RRMS were randomized to receive a daily oral dose of diazoxide (0.3 and 4 mg/d) or placebo for 24 weeks (NCT01428726). The primary endpoint was the cumulative number of new T1 gadolinium-enhancing lesions per patient, recorded every 4 weeks from week 4 to week 24. Secondary endpoints included brain MRI variables such as the number of new/enlarging T2 lesions and the percentage brain volume change (PBVC); clinical variables such as the percentage of relapse-free patients, relapse rate, and change in the Expanded Disability Status Scale score; and safety and tolerability. Results: Diazoxide was well-tolerated and it produced no serious adverse events other than 1 case of Hashimoto disease. At the 2 doses tested, diazoxide did not improve the primary endpoint or the MRI and clinical variables related to the presence of new lesions or relapses. Patients treated with diazoxide showed reduced PBVC compared with the placebo group, although such changes could be confounded by the higher disease activity of the treated group and the vascular effects of diazoxide. Conclusion: At the doses tested, oral diazoxide did not decrease the appearance of new lesions evident by MRI. The effects in slowing the progression of brain atrophy require further validation. Classification of evidence: This study provides Class I evidence that for patients with RRMS, diazoxide (0.3 and 4 mg/d) does not significantly change the number of new MRI T1 gadolinium-enhancing lesions. PMID:26405686
Ploubidis, G B; Edwards, P; Kendrick, D
2015-12-15
This paper reports the development and testing of a construct measuring parental fire safety behaviours for planning escape from a house fire. Latent variable modelling of data on parental-reported fire safety behaviours and plans for escaping from a house fire and multivariable logistic regression to quantify the association between groups defined by the latent variable modelling and parental-report of having a plan for escaping from a house fire. Data comes from 1112 participants in a cluster randomised controlled trial set in children's centres in 4 study centres in the UK. A two class model provided the best fit to the data, combining responses to five fire safety planning behaviours. The first group ('more behaviours for escaping from a house fire') comprised 86% of participants who were most likely to have a torch, be aware of how their smoke alarm sounds, to have external door and window keys accessible, and exits clear. The second group ('fewer behaviours for escaping from a house fire') comprised 14% of participants who were less likely to report these five behaviours. After adjusting for potential confounders, participants allocated to the 'more behaviours for escaping from a house fire group were 2.5 times more likely to report having an escape plan (OR 2.48; 95% CI 1.59-3.86) than those in the "fewer behaviours for escaping from a house fire" group. Multiple fire safety behaviour questions can be combined into a single binary summary measure of fire safety behaviours for escaping from a house fire. Our findings will be useful to future studies wishing to use a single measure of fire safety planning behaviour as measures of outcome or exposure. NCT 01452191. Date of registration 13/10/2011.
Tan, Min-Min; Chan, Carina K Y; Reidpath, Daniel D
2016-08-01
Religion has been shown to be salutary on health, and a possible link between religion and positive health outcomes is diet. Research has shown that religiosity is associated with better diet but most studies were conducted in a multi-denominational context, which might be confounded with theological differences. This study examined the relationship between religiosity and diet within a homogenous group of believers. Data from survey of 574 Seventh-Day Adventists residing in West Malaysia, aged 18-80, were analyzed using multiple regressions. While none of the religious variables were significantly associated with fruit and vegetable intake, a higher level of religiosity was associated with a better dietary habit and vegetarian status. The mixed relationship between religiosity and diet suggest that further research is needed to explore how religion might influence the diet of adherents.
Health-Related Quality of Life Among US Workers: Variability Across Occupation Groups.
Shockey, Taylor M; Zack, Matthew; Sussell, Aaron
2017-08-01
To examine the health-related quality of life among workers in 22 standard occupation groups using data from the 2013-2014 US Behavioral Risk Factor Surveillance System. We examined the health-related quality of life measures of self-rated health, frequent physical distress, frequent mental distress, frequent activity limitation, and frequent overall unhealthy days by occupation group for 155 839 currently employed adults among 17 states. We performed multiple logistic regression analyses that accounted for the Behavioral Risk Factor Surveillance System's complex survey design to obtain prevalence estimates adjusted for potential confounders. Among all occupation groups, the arts, design, entertainment, sports, and media occupation group reported the highest adjusted prevalence of frequent physical distress, frequent mental distress, frequent activity limitation, and frequent overall unhealthy days. The personal care and service occupation group had the highest adjusted prevalence for fair or poor self-rated health. Workers' jobs affect their health-related quality of life.
[Psychomotor development in offspring of mothers with post partum depression].
Podestá L, Loreto; Alarcón, Ana María; Muñoz, Sergio; Legüe C, Marcela; Bustos, Luis; Barría P, Mauricio
2013-04-01
Postpartum depression (PPD) has adverse effects on psychomotor development of the offspring. To evaluate the relationship between PPD and psychomotor development in children aged 18 months, consulting in primary care. Cross-sectional study with 360 infants and their mothers. Children had their psychomotor evaluation at l8 months and mothers completed the Edinburgh Postnatal Depression Scale at 4 and 12 weeks postpartum. The prevalence of both PPD and psychomotor alteration was estimated. The association between PPD and psychomotor alteration, including confounding variables, was estimated through logistic multiple regression analysis. The prevalence of PPD and psychomotor alteration was 29 and 16%, respectively Mothers with PPD had twice the probability of having an offspring with psychomotor alteration (Odds ratio = 2.0, confidence intervals = 1.07-3.68). This probability was significantly higher among single mothers or those with an unstable partner. PPD has a detrimental impact on psychomotor development of children.
2011-01-01
Background Being an immigrant in a high-income country is a risk factor for severe mental ill health. Studies on mental ill health among immigrants have found significant differences in mental health outcome between immigrants from high income countries and low-income countries. Being an asylum seeker or a refugee is also associated with mental ill health. This study aimed to assess if there is a difference in mental ill health problems between male and female refugee and non-refugee immigrants from six low-income countries in Sweden. Methods A cross-sectional, population-based study design was used comparing refugees with non-refugees. The study size was determined by the number of persons in Sweden fulfilling the inclusion criteria at the time of the study during 2006. Outcome: Mental ill health, as measured with the proxy variable psychotropic drugs purchased. Refugee/Non-refugee: Sweden grants asylum to refugees according to the Geneva Convention and those with a well-grounded fear of death penalty, torture or who need protection due to an internal or external armed conflict or an environmental disaster. The non-refugees were all family members of those granted asylum in Sweden. Covariates: Gender and origin. Potential confounders: Age, marital status, education and duration of stay in Sweden. Background variables were analysed using chi square tests. The association between outcome, exposure and possible confounders was analysed using logistic regression analyses. Multiple logistic regression analysis was used to adjust for potential confounders. Results The study population comprised 43,168 refugees and non-refugees, of whom 20,940 (48.5%) were women and 24,403 (56.5%) were refugees. Gender, age, origin, marital status and education were all associated with the outcome. For female, but not male, refugees there was a significantly higher likelihood of purchasing psychotropic drugs than non-refugees (OR = 1.27, 95% CI = 1.15 - 1.40). Conclusions Female refugees from low-income countries seem to be a risk group among immigrant women from low-income countries, whereas male refugees had the same risk patterns as non-refugee immigrants from low-income countries. This underlines the need for training of clinicians in order to focus on pre-migration stress and the asylum process, among female newcomers. PMID:21435212
Hollander, Anna-Clara; Bruce, Daniel; Burström, Bo; Ekblad, Solvig
2011-03-24
Being an immigrant in a high-income country is a risk factor for severe mental ill health. Studies on mental ill health among immigrants have found significant differences in mental health outcome between immigrants from high income countries and low-income countries. Being an asylum seeker or a refugee is also associated with mental ill health. This study aimed to assess if there is a difference in mental ill health problems between male and female refugee and non-refugee immigrants from six low-income countries in Sweden. A cross-sectional, population-based study design was used comparing refugees with non-refugees. The study size was determined by the number of persons in Sweden fulfilling the inclusion criteria at the time of the study during 2006. Mental ill health, as measured with the proxy variable psychotropic drugs purchased. Refugee/Non-refugee: Sweden grants asylum to refugees according to the Geneva Convention and those with a well-grounded fear of death penalty, torture or who need protection due to an internal or external armed conflict or an environmental disaster. The non-refugees were all family members of those granted asylum in Sweden. Covariates: Gender and origin. Potential confounders: Age, marital status, education and duration of stay in Sweden. Background variables were analysed using chi square tests. The association between outcome, exposure and possible confounders was analysed using logistic regression analyses. Multiple logistic regression analysis was used to adjust for potential confounders. The study population comprised 43,168 refugees and non-refugees, of whom 20,940 (48.5%) were women and 24,403 (56.5%) were refugees. Gender, age, origin, marital status and education were all associated with the outcome. For female, but not male, refugees there was a significantly higher likelihood of purchasing psychotropic drugs than non-refugees (OR = 1.27, 95% CI = 1.15-1.40). Female refugees from low-income countries seem to be a risk group among immigrant women from low-income countries, whereas male refugees had the same risk patterns as non-refugee immigrants from low-income countries. This underlines the need for training of clinicians in order to focus on pre-migration stress and the asylum process, among female newcomers.
Huesch, Marco D
2013-06-01
Assessing the real-world comparative effectiveness of common interventions is challenged by unmeasured confounding. To determine whether the mortality benefit shown for drug-eluting stents (DES) over bare metal stents (BMS) in observational studies persists after controls for/tests for confounding. Retrospective observational study involving 38,019 patients, 65 years or older admitted for an index percutaneous coronary intervention receiving DES or BMS in Pennsylvania in 2004-2005 followed up for death through 3 years. Analysis was at the patient level. Mortality was analyzed with Cox proportional hazards models allowing for stratification by disease severity or DES use propensity, accounting for clustering of patients. Instrumental variables analysis used lagged physician stent usage to proxy for the focal stent type decision. A method originating in work by Cornfield and others in 1954 and popularized by Greenland in 1996 was used to assess robustness to confounding. DES was associated with a significantly lower adjusted risk of death at 3 years in Cox and in instrumented analyses. An implausibly strong hypothetical unobserved confounder would be required to fully explain these results. Confounding by indication can bias observational studies. No strong evidence of such selection biases was found in the reduced risk of death among elderly patients receiving DES instead of BMS in a Pennsylvanian state-wide population. © Health Research and Educational Trust.
Dale, Caroline; Nüesch, Eveline; Prieto-Merino, David; Choi, Minkyoung; Amuzu, Antoinette; Ebrahim, Shah; Casas, Juan P.; Davey-Smith, George
2015-01-01
Low adiposity has been linked to elevated mortality from several causes including respiratory disease. However, this could arise from confounding or reverse causality. We explore the association between two measures of adiposity (BMI and WHR) with COPD in the British Women’s Heart and Health Study including a detailed assessment of the potential for confounding and reverse causality for each adiposity measure. Low BMI was found to be associated with increased COPD risk while low WHR was not (OR = 2.2; 95% CI 1.3 – 3.1 versus OR = 1.2; 95% CI 0.7 – 1.6). Potential confounding variables (e.g. smoking) and markers of ill-health (e.g. unintentional weight loss) were found to be higher in low BMI but not in low WHR. Women with low BMI have a detrimental profile across a broad range of health markers compared to women with low WHR, and women with low WHR do not appear to have an elevated COPD risk, lending support to the hypothesis that WHR is a less confounded measure of adiposity than BMI. Low adiposity does not in itself appear to increase the risk of respiratory disease, and the apparent adverse consequences of low BMI may be due to reverse causation and confounding. PMID:25884834
Interventional effects for mediation analysis with multiple mediators
Vansteelandt, Stijn; Daniel, Rhian M.
2016-01-01
The mediation formula for the identification of natural (in)direct effects has facilitated mediation analyses that better respect the nature of the data, with greater consideration of the need for confounding control. The default assumptions on which it relies are strong, however. In particular, they are known to be violated when confounders of the mediator–outcome association are affected by the exposure. This complicates extensions of counterfactual-based mediation analysis to settings that involve repeatedly measured mediators, or multiple correlated mediators. VanderWeele, Vansteelandt, and Robins21 introduced so-called interventional (in)direct effects. These can be identified under much weaker conditions than natural (in)direct effects, but have the drawback of not adding up to the total effect. In this article, we adapt their proposal in order to achieve an exact decomposition of the total effect, and extend it to the multiple mediator setting. Interestingly, the proposed effects capture the path-specific effects of an exposure on an outcome that are mediated by distinct mediators, even when – as often – the structural dependence between the multiple mediators is unknown; for instance, when the direction of the causal effects between the mediators is unknown, or there may be unmeasured common causes of the mediators. PMID:27922534
Interventional Effects for Mediation Analysis with Multiple Mediators.
Vansteelandt, Stijn; Daniel, Rhian M
2017-03-01
The mediation formula for the identification of natural (in)direct effects has facilitated mediation analyses that better respect the nature of the data, with greater consideration of the need for confounding control. The default assumptions on which it relies are strong, however. In particular, they are known to be violated when confounders of the mediator-outcome association are affected by the exposure. This complicates extensions of counterfactual-based mediation analysis to settings that involve repeatedly measured mediators, or multiple correlated mediators. VanderWeele, Vansteelandt, and Robins introduced so-called interventional (in)direct effects. These can be identified under much weaker conditions than natural (in)direct effects, but have the drawback of not adding up to the total effect. In this article, we adapt their proposal to achieve an exact decomposition of the total effect, and extend it to the multiple mediator setting. Interestingly, the proposed effects capture the path-specific effects of an exposure on an outcome that are mediated by distinct mediators, even when-as often-the structural dependence between the multiple mediators is unknown, for instance, when the direction of the causal effects between the mediators is unknown, or there may be unmeasured common causes of the mediators.
Zandvliet, Maarten L.; Koolen, Stijn L. W.; Mathijssen, Ron H. J.; van der Rijt, Carin C. D.
2016-01-01
Fentanyl is a strong opioid that is available for various administration routes, and which is widely used to treat cancer‐related pain. Many factors influence the fentanyl pharmacokinetics leading to a wide inter‐ and intrapatient variability. This systematic review summarizes multiple studied factors that potentially influence fentanyl pharmacokinetics with a focus on implications for cancer patients. The use of CYP3A4 inhibitors and inducers, impaired liver function, and heating of the patch potentially influence fentanyl pharmacokinetics in a clinically relevant way. In elderly patients, current data suggest that we should carefully dose fentanyl due to alterations in absorption and metabolism. The influence of BMI and gender on fentanyl pharmacokinetics is questionable, most probably due to a large heterogeneity in the published studies. Pharmacogenetics, e.g. the CYP3A5*3 gene polymorphism, may influence fentanyl pharmacokinetics as well, although further study is warranted. Several other factors have been studied but did not show significant and clinically relevant effects on fentanyl pharmacokinetics. Unfortunately, most of the published papers that studied factors influencing fentanyl pharmacokinetics describe healthy volunteers instead of cancer patients. Results from the studies in volunteers may not be simply extrapolated to cancer patients because of multiple confounding factors. To handle fentanyl treatment in a population of cancer patients, it is essential that physicians recognize factors that influence fentanyl pharmacokinetics, thereby preventing potential side‐effects and increasing its efficacy. PMID:27619152
O'Malley, A James; Cotterill, Philip; Schermerhorn, Marc L; Landon, Bruce E
2011-12-01
When 2 treatment approaches are available, there are likely to be unmeasured confounders that influence choice of procedure, which complicates estimation of the causal effect of treatment on outcomes using observational data. To estimate the effect of endovascular (endo) versus open surgical (open) repair, including possible modification by institutional volume, on survival after treatment for abdominal aortic aneurysm, accounting for observed and unobserved confounding variables. Observational study of data from the Medicare program using a joint model of treatment selection and survival given treatment to estimate the effects of type of surgery and institutional volume on survival. We studied 61,414 eligible repairs of intact abdominal aortic aneurysms during 2001 to 2004. The outcome, perioperative death, is defined as in-hospital death or death within 30 days of operation. The key predictors are use of endo, transformed endo and open volume, and endo-volume interactions. There is strong evidence of nonrandom selection of treatment with potential confounding variables including institutional volume and procedure date, variables not typically adjusted for in clinical trials. The best fitting model included heterogeneous transformations of endo volume for endo cases and open volume for open cases as predictors. Consistent with our hypothesis, accounting for unmeasured selection reduced the mortality benefit of endo. The effect of endo versus open surgery varies nonlinearly with endo and open volume. Accounting for institutional experience and unmeasured selection enables better decision-making by physicians making treatment referrals, investigators evaluating treatments, and policy makers.
Ngerncham, Monawat; Barnhart, Douglas C; Haricharan, Ramanath N; Roseman, Jeffrey M; Georgeson, Keith E; Harmon, Carroll M
2007-09-01
Recurrent gastroesophageal reflux disease (rGERD) is a common problem after fundoplication. Previous studies attempting to identify risk factors for rGERD have failed to control for confounding variables. The purpose of this study was to identify significant risk factors for rGERD after controlling for potential confounding variables. A retrospective, matched case-control study was conducted at a tertiary children's hospital. Cases (n = 116) met 1 of these criteria: reoperation for rGERD, symptomatic rGERD (confirmed by upper gastrointestinal series, esophagogastroduodenoscopy, or pH monitoring), or postoperative reinstitution of antireflux medication for more than 8 weeks. Controls (n = 209) were matched for surgeon, approach (laparoscopic/open), technique (partial/complete), and approximate operative date. Univariate and multivariable associations were analyzed by conditional logistic regression. Significant risk factors for rGERD were age of less than 6 years (odds ratio [OR], 3.6; 95% confidence interval [CI], 1.7-7.5), preoperative hiatal hernia (OR, 3.2; 95% CI, 1.4-7.3), postoperative retching (OR, 5.1; 95% CI, 2.6-10.0), and postoperative esophageal dilatation (OR, 10.8; 95% CI, 1.8-65.4). Interestingly, significant association was not found between neurologic impairment and rGERD after controlling for potential confounding variables. Age of less than 6 years, preoperative hiatal hernia, postoperative retching, and postoperative esophageal dilatation are independently associated with increased risk of rGERD. Neurologic impairment alone does not increase the risk of developing rGERD.
Wright, Wilson J.; Irvine, Kathryn M.
2017-01-01
We examined data on white pine blister rust (blister rust) collected during the monitoring of whitebark pine trees in the Greater Yellowstone Ecosystem (from 2004-2015). Summaries of repeat observations performed by multiple independent observers are reviewed and discussed. These summaries show variability among observers and the potential for errors being made in blister rust status. Based on this assessment, we utilized occupancy models to analyze blister rust prevalence while explicitly accounting for imperfect detection. Available covariates were used to model both the probability of a tree being infected with blister rust and the probability of an observer detecting the infection. The fitted model provided strong evidence that the probability of blister rust infection increases as tree diameter increases and decreases as site elevation increases. Most importantly, we found evidence of heterogeneity in detection probabilities related to tree size and average slope of a transect. These results suggested that detecting the presence of blister rust was more difficult in larger trees. Also, there was evidence that blister rust was easier to detect on transects located on steeper slopes. Our model accounted for potential impacts of observer experience on blister rust detection probabilities and also showed moderate variability among the different observers in their ability to detect blister rust. Based on these model results, we suggest that multiple observer sampling continue in future field seasons in order to allow blister rust prevalence estimates to be corrected for imperfect detection. We suggest that the multiple observer effort be spread out across many transects (instead of concentrated at a few each field season) while retaining the overall proportion of trees with multiple observers around 5-20%. Estimates of prevalence are confounded with detection unless it is explicitly accounted for in an analysis and we demonstrate how an occupancy model can be used to do account for this source of observation error.
The individualistic fallacy, ecological studies and instrumental variables: a causal interpretation.
Loney, Tom; Nagelkerke, Nico J
2014-01-01
The validity of ecological studies in epidemiology for inferring causal relationships has been widely challenged as observed associations could be biased by the Ecological Fallacy. We reconsider the important design components of ecological studies, and discuss the conditions that may lead to spurious associations. Ecological associations are useful and valid when the ecological exposures can be interpreted as Instrumental Variables. A suitable example may be a time series analysis of environmental pollution (e.g. particulate matter with an aerodynamic diameter of <10 micrometres; PM10) and health outcomes (e.g. hospital admissions for acute myocardial infarction) as environmental pollution levels are a cause of individual exposure levels and not just an aggregate measurement. Ecological exposures may also be employed in situations (perhaps rare) where individual exposures are known but their associations with health outcomes are confounded by unknown or unquantifiable factors. Ecological associations have a notorious reputation in epidemiology and individualistic associations are considered superior to ecological associations because of the "ecological fallacy". We have argued that this is incorrect in situations in which ecological or aggregate exposures can serve as an instrumental variable and associations between individual exposure and outcome are likely to be confounded by unmeasured variables.
2010-06-01
without influences that may confound the re- sults (e.g., pain, anxiety , transport conditions, caregiver in- terventions). Second, rather than being...32. Ryan KL, Rickards CA, Muniz GW, Moralez G, Convertino VA: Interindi- vidual variability in heart rate variability ( HRV ) and complexity measure...Raimondi G, Legramante JM, Macerata A: Revisiting the potential of time-domain indexes in short-term HRV analysis. Biomed Tech (Berl) 51:190Y193, 2006
NASA Astrophysics Data System (ADS)
Nelson, N.; Munoz-Carpena, R.; Kaplan, D. A.; Phlips, E. J.
2016-12-01
In many aquatic systems, cyanobacteria form harmful blooms capable of producing toxins, prompting hypoxia, and/or introducing internal nitrogen loads via N2-fixation, among other impacts. Traditionally, system-specific cyanobacteria drivers are determined by performing controlled experiments and bioassays, but these approaches may neglect the influences of confounding factors and over assign importance to only those variables considered within experimental designs. For example, a bioassay may conclude that the cyanobacteria in a particular system are limited by phosphorus, but will not explicitly take into account the role of flow as a control on phosphorus delivery. This study aims to address this analytical gap by identifying environmental controls on cyanobacteria while removing the effects of potentially confounding variables. In the present work, we evaluate a unique long-term (17 year) dataset composed of monthly observations of phytoplankton and zooplankton species abundances, water quality constituents, and hydrologic variables from Lake George, a flow-through lake of the St. Johns River (FL) impacted by cyanobacterial blooms. Using conditional Granger causality analysis, a time series approach that infers causality while removing the effects of confounding variables, data were evaluated to identify biological and physicochemical drivers of cyanobacteria. The analysis was performed for three response variable sets: total cyanobacteria, N2-fixers and non-fixers, and cyanobacteria genera. Results depicted increasing levels of ecological complexity as subdivisions of cyanobacteria became more detailed; whereas causal networks produced from analyses of cyanobacteria genera provided novel insights relevant for management (i.e. nutrients, flow), the total cyanobacteria network only included water temperature as a significant driver. Additionally, the more detailed cyanobacteria subdivisions uncovered that N2-fixation was only evident with the earliest season bloomer, thus indicating that early season bloomers may be facilitating later-season growth. These findings highlight the importance of detailed monitoring efforts, and indicate that greater emphasis needs to be placed on the role of early bloomers in phytoplankton dynamics, as it relates to management efforts.
Paz-Zulueta, María; Llorca, Javier; Sarabia-Lavín, Raquel; Bolumar, Francisco; Rioja, Luis; Delgado, Abraham; Santibáñez, Miguel
2015-01-01
Literature evaluating association between neonatal morbidity and immigrant status presents contradictory results. Poorer compliance with prenatal care and greater social risk factors among immigrants could play roles as major confounding variables, thus explaining contradictions. We examined whether prenatal care and social risk factors are confounding variables in the relationship between immigrant status and neonatal morbidity. Retrospective cohort study: 231 pregnant African immigrant women were recruited from 2007-2010 in northern Spain. A Spanish population sample was obtained by simple random sampling at 1:3 ratio. Immigrant status (Spanish, Sub-Saharan and Northern African), prenatal care (Kessner Index adequate, intermediate or inadequate), and social risk factors were treated as independent variables. Low birth weight (LBW < 2500 grams) and preterm birth (< 37 weeks) were collected as neonatal morbidity variables. Crude and adjusted odds ratios (OR) were estimated by unconditional logistic regression with 95% confidence intervals (95% CI). Positive associations between immigrant women and higher risk of neonatal morbidity were obtained. Crude OR for preterm births in Northern Africans with respect to nonimmigrants was 2.28 (95% CI: 1.04-5.00), and crude OR for LBW was 1.77 (95% CI: 0.74-4.22). However, after adjusting for prenatal care and social risk factors, associations became protective: adjusted OR for preterm birth = 0.42 (95% CI: 0.14-1.32); LBW = 0.48 (95% CI: 0.15-1.52). Poor compliance with prenatal care was the main independent risk factor associated with both preterm birth (adjusted OR inadequate care = 17.05; 95% CI: 3.92-74.24) and LBW (adjusted OR inadequate care = 6.25; 95% CI: 1.28-30.46). Social risk was an important independent risk factor associated with LBW (adjusted OR = 5.42; 95% CI: 1.58-18.62). Prenatal care and social risk factors were major confounding variables in the relationship between immigrant status and neonatal morbidity.
Practical applications of the bioinformatics toolbox for narrowing quantitative trait loci.
Burgess-Herbert, Sarah L; Cox, Allison; Tsaih, Shirng-Wern; Paigen, Beverly
2008-12-01
Dissecting the genes involved in complex traits can be confounded by multiple factors, including extensive epistatic interactions among genes, the involvement of epigenetic regulators, and the variable expressivity of traits. Although quantitative trait locus (QTL) analysis has been a powerful tool for localizing the chromosomal regions underlying complex traits, systematically identifying the causal genes remains challenging. Here, through its application to plasma levels of high-density lipoprotein cholesterol (HDL) in mice, we demonstrate a strategy for narrowing QTL that utilizes comparative genomics and bioinformatics techniques. We show how QTL detected in multiple crosses are subjected to both combined cross analysis and haplotype block analysis; how QTL from one species are mapped to the concordant regions in another species; and how genomewide scans associating haplotype groups with their phenotypes can be used to prioritize the narrowed regions. Then we illustrate how these individual methods for narrowing QTL can be systematically integrated for mouse chromosomes 12 and 15, resulting in a significantly reduced number of candidate genes, often from hundreds to <10. Finally, we give an example of how additional bioinformatics resources can be combined with experiments to determine the most likely quantitative trait genes.
Analyzing the association between fish consumption and osteoporosis in a sample of Chinese men.
Li, Xia; Lei, Tao; Tang, Zihui; Dong, Jingcheng
2017-04-19
The main purpose of this study was to estimate the associations between frequency of fish food consumption and osteoporosis (OP) in general Chinese men. We conducted a large-scale, community-based, cross-sectional study to investigate the associations by using self-report questionnaire to access frequency of fish food intake. A total of 1092 men were available for data analysis in this study. Multiple regression models controlling for confounding factors to include frequency of fish food consumption variable were performed to investigate the relationships for OP. Positive correlations between frequency of fish food consumption and T score were reported (β = 0.084, P value = 0.025). Multiple regression analysis indicated that the frequency of fish food consumption was significantly associated with OP (P < 0.05 for model 1 and model 2). The men with high frequency of fish food consumption had a lower prevalence of OP. The findings indicated that frequency of fish food consumption was independently and significantly associated with OP. The prevalence of OP was less frequent in Chinese men preferring fish food habits. ClinicalTrials.gov Identifier: NCT02451397 retrospectively registered 28 May 2015.
Sheikh, Mashhood Ahmed
2017-01-01
The life course perspective, the risky families model, and stress-and-coping models provide the rationale for assessing the role of smoking as a mediator in the association between childhood adversity and anxious and depressive symptomatology (ADS) in adulthood. However, no previous study has assessed the independent mediating role of smoking in the association between childhood adversity and ADS in adulthood. Moreover, the importance of mediator-response confounding variables has rarely been demonstrated empirically in social and psychiatric epidemiology. The aim of this paper was to (i) assess the mediating role of smoking in adulthood in the association between childhood adversity and ADS in adulthood, and (ii) assess the change in estimates due to different mediator-response confounding factors (education, alcohol intake, and social support). The present analysis used data collected from 1994 to 2008 within the framework of the Tromsø Study ( N = 4,530), a representative prospective cohort study of men and women. Seven childhood adversities (low mother's education, low father's education, low financial conditions, exposure to passive smoke, psychological abuse, physical abuse, and substance abuse distress) were used to create a childhood adversity score. Smoking status was measured at a mean age of 54.7 years (Tromsø IV), and ADS in adulthood was measured at a mean age of 61.7 years (Tromsø V). Mediation analysis was used to assess the indirect effect and the proportion of mediated effect (%) of childhood adversity on ADS in adulthood via smoking in adulthood. The test-retest reliability of smoking was good (Kappa: 0.67, 95% CI: 0.63; 0.71) in this sample. Childhood adversity was associated with a 10% increased risk of smoking in adulthood (Relative risk: 1.10, 95% CI: 1.03; 1.18), and both childhood adversity and smoking in adulthood were associated with greater levels of ADS in adulthood ( p < 0.001). Smoking in adulthood did not significantly mediate the association between childhood adversity and ADS in adulthood. However, when education was excluded as a mediator-response confounding variable, the indirect effect of childhood adversity on ADS in adulthood was statistically significant ( p < 0.05). This study shows that a careful inclusion of potential confounding variables is important when assessing mediation.
Sheikh, Mashhood Ahmed
2017-01-01
The life course perspective, the risky families model, and stress-and-coping models provide the rationale for assessing the role of smoking as a mediator in the association between childhood adversity and anxious and depressive symptomatology (ADS) in adulthood. However, no previous study has assessed the independent mediating role of smoking in the association between childhood adversity and ADS in adulthood. Moreover, the importance of mediator-response confounding variables has rarely been demonstrated empirically in social and psychiatric epidemiology. The aim of this paper was to (i) assess the mediating role of smoking in adulthood in the association between childhood adversity and ADS in adulthood, and (ii) assess the change in estimates due to different mediator-response confounding factors (education, alcohol intake, and social support). The present analysis used data collected from 1994 to 2008 within the framework of the Tromsø Study (N = 4,530), a representative prospective cohort study of men and women. Seven childhood adversities (low mother's education, low father's education, low financial conditions, exposure to passive smoke, psychological abuse, physical abuse, and substance abuse distress) were used to create a childhood adversity score. Smoking status was measured at a mean age of 54.7 years (Tromsø IV), and ADS in adulthood was measured at a mean age of 61.7 years (Tromsø V). Mediation analysis was used to assess the indirect effect and the proportion of mediated effect (%) of childhood adversity on ADS in adulthood via smoking in adulthood. The test-retest reliability of smoking was good (Kappa: 0.67, 95% CI: 0.63; 0.71) in this sample. Childhood adversity was associated with a 10% increased risk of smoking in adulthood (Relative risk: 1.10, 95% CI: 1.03; 1.18), and both childhood adversity and smoking in adulthood were associated with greater levels of ADS in adulthood (p < 0.001). Smoking in adulthood did not significantly mediate the association between childhood adversity and ADS in adulthood. However, when education was excluded as a mediator-response confounding variable, the indirect effect of childhood adversity on ADS in adulthood was statistically significant (p < 0.05). This study shows that a careful inclusion of potential confounding variables is important when assessing mediation. PMID:28824498
Weiss, Noel S; Dhillon, Preet K; Etzioni, Ruth
2004-07-01
A case-control study of the efficacy of cancer screening, like any other case-control study, must deal with potential confounding. There are 2 categories of confounding variables that pose special problems for studies of screening: (1) age and calendar time resulting from different temporal distributions of screening between cases and controls irrespective of whether the screening test leads to a reduction in mortality; and (2) the administration of other screening tests for the cancer in question when it is not clear whether the result of the other test had a bearing on the decision to order the test under study. We describe circumstances in which confounding from these sources can be dealt with satisfactorily by means of restriction or adjustment, and other circumstances in which it cannot. Copyright 2004 Lippincott Williams and Wilkins
Sinclair, Thomas R; Manandhar, Anju; Shekoofa, Avat; Rosas-Anderson, Pablo; Bagherzadi, Laleh; Schoppach, Remy; Sadok, Walid; Rufty, Thomas W
2017-04-01
Theoretical derivation predicted growth retardation due to pot water limitations, i.e., pot binding. Experimental observations were consistent with these limitations. Combined, these results indicate a need for caution in high-throughput screening and phenotyping. Pot experiments are a mainstay in many plant studies, including the current emphasis on developing high-throughput, phenotyping systems. Pot studies can be vulnerable to decreased physiological activity of the plants particularly when pot volume is small, i.e., "pot binding". It is necessary to understand the conditions under which pot binding may exist to avoid the confounding influence of pot binding in interpreting experimental results. In this paper, a derivation is offered that gives well-defined conditions for the occurrence of pot binding based on restricted water availability. These results showed that not only are pot volume and plant size important variables, but the potting media is critical. Artificial potting mixtures used in many studies, including many high-throughput phenotyping systems, are particularly susceptible to the confounding influences of pot binding. Experimental studies for several crop species are presented that clearly show the existence of thresholds of plant leaf area at which various pot sizes and potting media result in the induction of pot binding even though there may be no immediate, visual plant symptoms. The derivation and experimental results showed that pot binding can readily occur in plant experiments if care is not given to have sufficiently large pots, suitable potting media, and maintenance of pot water status. Clear guidelines are provided for avoiding the confounding effects of water-limited pot binding in studying plant phenotype.
Simpson's paradox: A statistician's case study.
Chu, Kevin H; Brown, Nathan J; Pelecanos, Anita; Brown, Anthony Ft
2018-02-26
Gender equality and workforce diversity has recently been in the forefront of College discussions. Reasons for the difference between various groups may not be as they initially appeared. The results of comparing the outcome between two groups can sometimes be confounded and even reversed by an unrecognised third variable. This concept is known as Simpson's Paradox, and is illustrated here using a renowned case study on potential gender bias for acceptance to Graduate School at the University of California, Berkeley. The investigation showed that males were 1.8 times more likely to be admitted to Graduate School than females in 1973. Initially it appeared that women were discriminated against in the selection process. However, when admissions were re-examined at individual Departments of the School, admission tended to be better for women than men in four of six Departments. This contradiction or paradox tells us that the association between admission and gender was dependent upon on Department. The confounding effect of Department was defined by two characteristics. Firstly, a strong association between Department and admission: some Departments admitted much smaller percentages of applicants than others. Secondly, a strong association between Department and gender: females tended to apply to Departments with lower admission rates. The explanation of differences between groups can be multifactorial. A search for possible confounders will assist in this understanding. This could apply whenever two groups initially appear to differ, but on closer analysis this difference is either unfounded, or even reversed by reference to a third, confounding variable. © 2018 Australasian College for Emergency Medicine and Australasian Society for Emergency Medicine.
Keogh, Ruth; Frost, Chris; Owen, Gail; Daniel, Rhian M; Langbehn, Doug R; Leavitt, Blair; Durr, Alexandra; Roos, Raymund A C; Landwehrmeyer, G Bernhard; Reilmann, Ralf; Borowsky, Beth; Stout, Julie; Craufurd, David; Tabrizi, Sarah J
2016-01-11
Insufficient evidence exists to guide the long-term pharmacological management of Huntington's disease (HD) although most current interventions rely on symptomatic management. The effect of many frontline treatments on potential endpoints for HD clinical trials remains unknown. Our objective was to investigate how therapies widely used to manage HD affect the symptom for which they are prescribed and other endpoints using data from TRACK-HD. We used longitudinal models to estimate effects of medication use on performance on tests of motor, cognitive and neuropsychiatric function using data from 123 TRACK-HD stage 1/2 participants across four study visits. Adjustment for confounding by prior medication use, prior clinical performance, concomitant use of other medications, and baseline variables (sex, disease group, age, CAG, study site, education) enabled a closer-to-causal interpretation of the associations. Adjusting for baseline variables only, medication use was typically associated with worse clinical performance, reflecting greater medication use in more advanced patients. After additional adjustment for longitudinal confounders such "inverse" associations were generally eliminated and in the expected directions: participants taking neuroleptics tended to have better motor performance, improved affect and poorer cognitive performance, and those taking SSRI/SNRIs had less apathy, less affect and better total behaviour scores. However, we uncovered few statistically significant associations. Limitations include sample size and unmeasured confounding. In conclusion, adjustment for confounding by prior measurements largely eliminated associations between medication use and poorer clinical performance from simple analyses. However, there was little convincing evidence of causal effects of medication on clinical performance and larger cohorts or trials are needed.
Drug induced liver injury with analysis of alternative causes as confounding variables.
Teschke, Rolf; Danan, Gaby
2018-04-01
Drug-induced liver injury (DILI) is rare compared to the worldwide frequent acute or chronic liver diseases. Therefore, patients included in series of suspected DILI are at high risk of not having DILI, whereby alternative causes may confound the DILI diagnosis. The aim of this review is to evaluate published case series of DILI for alternative causes. Relevant studies were identified using a computerized search of the Medline database for publications from 1993 through 30 October 2017. We used the following terms: drug hepatotoxicity, drug induced liver injury, hepatotoxic drugs combined with diagnosis, causality assessment and alternative causes. Alternative causes as variables confounding the DILI diagnosis emerged in 22 published DILI case series, ranging from 4 to 47%. Among 13 335 cases of suspected DILI, alternative causes were found to be more likely in 4555 patients (34.2%), suggesting that the suspected DILI was probably not DILI. Biliary diseases such as biliary obstruction, cholangitis, choledocholithiasis, primary biliary cholangitis and primary sclerosing cholangitis were among the most missed diagnoses. Alternative causes included hepatitis B, C and E, cytomegalovirus, Epstein-Barr virus, ischemic hepatitis, cardiac hepatopathy, autoimmune hepatitis, nonalcoholic fatty liver disease, nonalcoholic steatohepatitis, and alcoholic liver disease. In more than one-third of published global DILI case series, alternative causes as published in these reports confounded the DILI diagnosis. In the future, published DILI case series should include only patients with secured DILI diagnosis, preferentially established by prospective use of scored items provided by robust diagnostic algorithms such as the updated Roussel Uclaf causality assessment method. © 2018 The British Pharmacological Society.
Domingues, Érica Prado; Silva, Guilherme Gomes; Oliveira, Andrei Barbassa; Mota, Lorrany Marins; Santos, Vanessa Santana Vieira; de Campos, Edimar Olegário; Pereira, Boscolli Barbosa
2018-03-14
Workers in several occupational environments are exposed to pollutants. Street vendors, for example, typically work in a high-traffic urban environment and are exposed to numerous air pollutants, including genotoxic substances emitted by motor vehicles. This study examined the genotoxic effects of exposure to air pollution. We conducted cytological analyses to assess frequencies of micronucleated (MN) and binucleated (BN) cells in a sample of exfoliated oral mucosa cells. We compared street vendors and control subjects in the city of Uberlândia, Minas Gerais, Brazil, and also collected quantitative information on exposure conditions of all test subjects, including concentrations of particulate matter. We found street vendors to exhibit higher frequencies of MN cells compared to the control group. We evaluated the effects of possible confounding variables on MN frequencies, namely the body mass index (BMI), age, as well as smoking and alcohol habits. Multiple linear regression analysis found no significant effects of any of those variables. Our results suggest that continued exposure to air pollution from traffic represents a major source of genotoxicity and raises concerns regarding disease prevention not only in street vendors but also other groups of people working in urban environments.
Sauchyn, David J.; St-Jacques, Jeannine-Marie; Luckman, Brian H.
2015-01-01
Exploitation of the Alberta oil sands, the world’s third-largest crude oil reserve, requires fresh water from the Athabasca River, an allocation of 4.4% of the mean annual flow. This allocation takes into account seasonal fluctuations but not long-term climatic variability and change. This paper examines the decadal-scale variability in river discharge in the Athabasca River Basin (ARB) with (i) a generalized least-squares (GLS) regression analysis of the trend and variability in gauged flow and (ii) a 900-y tree-ring reconstruction of the water-year flow of the Athabasca River at Athabasca, Alberta. The GLS analysis removes confounding transient trends related to the Pacific Decadal Oscillation (PDO) and Pacific North American mode (PNA). It shows long-term declining flows throughout the ARB. The tree-ring record reveals a larger range of flows and severity of hydrologic deficits than those captured by the instrumental records that are the basis for surface water allocation. It includes periods of sustained low flow of multiple decades in duration, suggesting the influence of the PDO and PNA teleconnections. These results together demonstrate that low-frequency variability must be considered in ARB water allocation, which has not been the case. We show that the current and projected surface water allocations from the Athabasca River for the exploitation of the Alberta oil sands are based on an untenable assumption of the representativeness of the short instrumental record. PMID:26392554
Toh, Sengwee; Gagne, Joshua J; Rassen, Jeremy A; Fireman, Bruce H; Kulldorff, Martin; Brown, Jeffrey S
2013-08-01
A distributed research network (DRN) of electronic health care databases, in which data reside behind the firewall of each data partner, can support a wide range of comparative effectiveness research (CER) activities. An essential component of a fully functional DRN is the capability to perform robust statistical analyses to produce valid, actionable evidence without compromising patient privacy, data security, or proprietary interests. We describe the strengths and limitations of different confounding adjustment approaches that can be considered in observational CER studies conducted within DRNs, and the theoretical and practical issues to consider when selecting among them in various study settings. Several methods can be used to adjust for multiple confounders simultaneously, either as individual covariates or as confounder summary scores (eg, propensity scores and disease risk scores), including: (1) centralized analysis of patient-level data, (2) case-centered logistic regression of risk set data, (3) stratified or matched analysis of aggregated data, (4) distributed regression analysis, and (5) meta-analysis of site-specific effect estimates. These methods require different granularities of information be shared across sites and afford investigators different levels of analytic flexibility. DRNs are growing in use and sharing of highly detailed patient-level information is not always feasible in DRNs. Methods that incorporate confounder summary scores allow investigators to adjust for a large number of confounding factors without the need to transfer potentially identifiable information in DRNs. They have the potential to let investigators perform many analyses traditionally conducted through a centralized dataset with detailed patient-level information.
Gallagher, Lisa G; Webster, Thomas F; Aschengrau, Ann
2017-03-11
Investigating the effects of prenatal and childhood exposures on behavioral health outcomes in adolescence is challenging given the lengthy period between the exposure and outcomes. We conducted a retrospective cohort study in Cape Cod, Massachusetts to evaluate the impact of prenatal and early childhood exposure to tetrachloroethylene (PCE)-contaminated drinking water on the occurrence of risk-taking behaviors as a teenager. An increased occurrence of risk-taking behaviors, particularly illicit drug use, was observed in those highly exposed to PCE. We hypothesized that there may be other sources of prenatal solvent exposure such as maternal consumption of alcoholic beverages during pregnancy which might modify the previously observed associations between PCE and risk-taking behaviors and so we conducted an exploratory analysis using available cohort data. The current report presents the results of these analyses and describes the difficulties in conducting research on long-term behavioral effects of early life exposures. The exploratory analysis compared a referent group of subjects with no early life exposure to PCE or alcohol (n = 242) to subjects with only alcohol exposure (n = 201), subjects with only PCE exposure (n = 361), and subjects with exposure to both PCE and alcohol (n = 302). Surveys completed by the subject's mother included questions on prenatal alcoholic beverage consumption and available confounding variables such as cigarette smoking and marijuana use. Surveys completed by the subjects included questions on risk-taking behaviors such as alcoholic beverage consumption and illicit drug use as a teenager and available confounding variables. PCE exposure was modeled using a leaching and transport algorithm embedded in water distribution system modeling software that estimated the amount of PCE delivered to a subject's residence during gestation and early childhood. Subjects with early life exposure to both PCE and alcohol had an increased risk of using two or more major drugs as a teen (RR = 1.9 (95% CI 1.2, 3.0)) compared to unexposed subjects. Increased risks for only PCE exposure (RR = 1.6 (95% CI 1.0, 2.4) and only alcohol exposure (RR = 1.3 (95% CI 0.7, 2.1)) were also evident but were smaller than the increased risk associated with both exposures. While available confounding variables were controlled, many relevant social risk factors were not obtained due to limitations in the retrospective study design. This exploratory analysis found evidence for an additive effect of early life exposure to PCE and alcohol on the risk of use of multiple illicit drugs as a teenager. Because of numerous limitations in this retrospective study, further research is needed to examine longstanding behavioral effects of early life exposures. To be most informative, this research should involve long-term prospective data collection.
DiPrete, Thomas A.; Burik, Casper A. P.; Koellinger, Philipp D.
2018-01-01
Identifying causal effects in nonexperimental data is an enduring challenge. One proposed solution that recently gained popularity is the idea to use genes as instrumental variables [i.e., Mendelian randomization (MR)]. However, this approach is problematic because many variables of interest are genetically correlated, which implies the possibility that many genes could affect both the exposure and the outcome directly or via unobserved confounding factors. Thus, pleiotropic effects of genes are themselves a source of bias in nonexperimental data that would also undermine the ability of MR to correct for endogeneity bias from nongenetic sources. Here, we propose an alternative approach, genetic instrumental variable (GIV) regression, that provides estimates for the effect of an exposure on an outcome in the presence of pleiotropy. As a valuable byproduct, GIV regression also provides accurate estimates of the chip heritability of the outcome variable. GIV regression uses polygenic scores (PGSs) for the outcome of interest which can be constructed from genome-wide association study (GWAS) results. By splitting the GWAS sample for the outcome into nonoverlapping subsamples, we obtain multiple indicators of the outcome PGSs that can be used as instruments for each other and, in combination with other methods such as sibling fixed effects, can address endogeneity bias from both pleiotropy and the environment. In two empirical applications, we demonstrate that our approach produces reasonable estimates of the chip heritability of educational attainment (EA) and show that standard regression and MR provide upwardly biased estimates of the effect of body height on EA. PMID:29686100
DiPrete, Thomas A; Burik, Casper A P; Koellinger, Philipp D
2018-05-29
Identifying causal effects in nonexperimental data is an enduring challenge. One proposed solution that recently gained popularity is the idea to use genes as instrumental variables [i.e., Mendelian randomization (MR)]. However, this approach is problematic because many variables of interest are genetically correlated, which implies the possibility that many genes could affect both the exposure and the outcome directly or via unobserved confounding factors. Thus, pleiotropic effects of genes are themselves a source of bias in nonexperimental data that would also undermine the ability of MR to correct for endogeneity bias from nongenetic sources. Here, we propose an alternative approach, genetic instrumental variable (GIV) regression, that provides estimates for the effect of an exposure on an outcome in the presence of pleiotropy. As a valuable byproduct, GIV regression also provides accurate estimates of the chip heritability of the outcome variable. GIV regression uses polygenic scores (PGSs) for the outcome of interest which can be constructed from genome-wide association study (GWAS) results. By splitting the GWAS sample for the outcome into nonoverlapping subsamples, we obtain multiple indicators of the outcome PGSs that can be used as instruments for each other and, in combination with other methods such as sibling fixed effects, can address endogeneity bias from both pleiotropy and the environment. In two empirical applications, we demonstrate that our approach produces reasonable estimates of the chip heritability of educational attainment (EA) and show that standard regression and MR provide upwardly biased estimates of the effect of body height on EA. Copyright © 2018 the Author(s). Published by PNAS.
Rosen, Rebecca L; Levy-Carrick, Nomi; Reibman, Joan; Xu, Ning; Shao, Yongzhao; Liu, Mengling; Ferri, Lucia; Kazeros, Angeliki; Caplan-Shaw, Caralee E; Pradhan, Deepak R; Marmor, Michael; Galatzer-Levy, Isaac R
2017-06-01
Systemic inflammation has emerged as a promising marker and potential mechanism underlying post-traumatic stress disorder (PTSD). The relationship between posttraumatic stress pathology and systemic inflammation has not, however, been consistently replicated and is potentially confounded by comorbid illness or injury, common complications of trauma exposure. We analyzed a large naturalistic cohort sharing a discrete physical and mental health trauma from the destruction of the World Trade Center (WTC) towers on September 11, 2001 (n = 641). We evaluated the relationship between multiple physical and mental health related indices collected through routine evaluations at the WTC Environmental Health Center (WTC EHC), a treatment program for community members exposed to the disaster. C-Reactive Protein (CRP), a marker of systemic inflammation, was examined in relation to scores for PTSD, PTSD symptom clusters (re-experiencing, avoidance, negative cognitions/mood, arousal), depression and anxiety, while controlling for WTC exposures, lower respiratory symptoms, age, sex, BMI and smoking as potential risks or confounders. CRP was positively associated with PTSD severity (p < 0.001), trending toward association with depression (p = 0.06), but not with anxiety (p = 0.27). CRP was positively associated with re-experiencing (p < 0.001) and avoidance (p < 0.05) symptom clusters, and trended toward associations with negative cognitions/mood (p = 0.06) and arousal (p = 0.08). In this large study of the relationship between CRP and posttraumatic stress pathology, we demonstrated an association between systemic inflammation and stress pathology (PTSD; trending with depression), which remained after adjusting for potentially confounding variables. These results contribute to research findings suggesting a salient relationship between inflammation and posttraumatic stress pathology. Copyright © 2017 The Authors. Published by Elsevier Ltd.. All rights reserved.
Lynch, Wendy D; Sherman, Bruce W
2014-01-01
When corporate health researchers examine the effects of health on business outcomes or the effect of health interventions on health and business outcomes, results will necessarily be confounded by the corporate environment(s) in which they are studied. In this research setting, most studies control for factors traditionally identified in public health, such as demographics and health status. Nevertheless, often overlooked is the extent to which company policies can also independently impact health care cost, work attendance, and productivity outcomes. With changes in employment and benefits practices resulting from health care reform, including incentives and plan design options, consideration of these largely neglected variables in research design has become increasingly important. This commentary summarizes existing knowledge regarding the implications of policy variations in research outcomes and provides a framework for incorporating them into future employer-based research.
Instrumental variable methods in comparative safety and effectiveness research.
Brookhart, M Alan; Rassen, Jeremy A; Schneeweiss, Sebastian
2010-06-01
Instrumental variable (IV) methods have been proposed as a potential approach to the common problem of uncontrolled confounding in comparative studies of medical interventions, but IV methods are unfamiliar to many researchers. The goal of this article is to provide a non-technical, practical introduction to IV methods for comparative safety and effectiveness research. We outline the principles and basic assumptions necessary for valid IV estimation, discuss how to interpret the results of an IV study, provide a review of instruments that have been used in comparative effectiveness research, and suggest some minimal reporting standards for an IV analysis. Finally, we offer our perspective of the role of IV estimation vis-à-vis more traditional approaches based on statistical modeling of the exposure or outcome. We anticipate that IV methods will be often underpowered for drug safety studies of very rare outcomes, but may be potentially useful in studies of intended effects where uncontrolled confounding may be substantial.
Male fertility: psychiatric considerations.
Hall, Elise; Burt, Vivien K
2012-02-01
To examine: 1) current knowledge on normal biologic variation of seminal parameters; 2) how stress and psychological factors affect sperm quality in fertile and infertile males; and 3) how mental illness and psychopharmacologic agents can affect male fertility. English-language Medline, Embase, and Psycinfo were searched for relevant publications (from 1970 to January 2011) for systematic review. None. None. None. Possible effects of stress, mood, and psychotropic medications on male factor fertility. Male-factor infertility is influenced by myriad factors (obesity, tobacco, etc.). Stress alone may reduce testosterone levels and spermatogenesis. Infertility assessment and treatment can lead to distress and negatively affect sperm samples. Available research has failed to control for potentially confounding variables. Although some trends have been identified, larger-scale studies that adequately control all confounding variables are needed before conclusions can be made about the relationship between stress, psychotropic agents, and male infertility. Copyright © 2012 American Society for Reproductive Medicine. Published by Elsevier Inc. All rights reserved.
Perrakis, Konstantinos; Gryparis, Alexandros; Schwartz, Joel; Le Tertre, Alain; Katsouyanni, Klea; Forastiere, Francesco; Stafoggia, Massimo; Samoli, Evangelia
2014-12-10
An important topic when estimating the effect of air pollutants on human health is choosing the best method to control for seasonal patterns and time varying confounders, such as temperature and humidity. Semi-parametric Poisson time-series models include smooth functions of calendar time and weather effects to control for potential confounders. Case-crossover (CC) approaches are considered efficient alternatives that control seasonal confounding by design and allow inclusion of smooth functions of weather confounders through their equivalent Poisson representations. We evaluate both methodological designs with respect to seasonal control and compare spline-based approaches, using natural splines and penalized splines, and two time-stratified CC approaches. For the spline-based methods, we consider fixed degrees of freedom, minimization of the partial autocorrelation function, and general cross-validation as smoothing criteria. Issues of model misspecification with respect to weather confounding are investigated under simulation scenarios, which allow quantifying omitted, misspecified, and irrelevant-variable bias. The simulations are based on fully parametric mechanisms designed to replicate two datasets with different mortality and atmospheric patterns. Overall, minimum partial autocorrelation function approaches provide more stable results for high mortality counts and strong seasonal trends, whereas natural splines with fixed degrees of freedom perform better for low mortality counts and weak seasonal trends followed by the time-season-stratified CC model, which performs equally well in terms of bias but yields higher standard errors. Copyright © 2014 John Wiley & Sons, Ltd.
Russa, A D; Bouma, A; Vernooij, J C M; Jacobs-Reitsma, W; Stegeman, J A
2005-01-01
To determine whether an association exists between partial depopulation of a flock and increased Campylobacter colonization in that flock. Data from 1737 flocks of two Dutch integrators were used. Flocks that experienced partial depopulation were defined as 'exposed' and those that did not as 'nonexposed'. Multivariable modelling was accomplished with, in addition to 'exposure', the independent variables 'age of broilers at slaughter' and 'season' to adjust for possible confounding. The response variable was 'Campylobacter colonization'. The odds ratio (OR) for partial depopulation for integrator A was 0.8 [95% CI (0.4, 1.8)]; for integrator B the OR = 0.8 [95% CI (0.5, 1.3)]. Age and season were confounders: the difference in Campylobacter status between exposed and nonexposed flocks of integrator A could be explained by both variables; for integrator B, only season was associated with Campylobacter status. We found no significant association between partial depopulation and an increased risk of Campylobacter colonization among broiler flocks at final depopulation. This study demonstrates that Campylobacter colonization in a broiler flock is not influenced by the partial depopulation of that flock.
Piirtola, Maarit; Kaprio, Jaakko; Kujala, Urho M; Heikkilä, Kauko; Koskenvuo, Markku; Svedberg, Pia; Silventoinen, Karri; Ropponen, Annina
2016-08-04
Education is associated with health related lifestyle choices including leisure-time physical inactivity. However, the longitudinal associations between education and inactivity merit further studies. We investigated the association between education and leisure-time physical inactivity over a 35-year follow-up with four time points controlling for multiple covariates including familial confounding. This study of the population-based Finnish Twin Cohort consisted of 5254 twin individuals born in 1945-1957 (59 % women), of which 1604 were complete same-sexed twin pairs. Data on leisure-time physical activity and multiple covariates was available from four surveys conducted in 1975, 1981, 1990 and 2011 (response rates 72 to 89 %). The association between years of education and leisure-time physical inactivity (<1.5 metabolic equivalent hours/day) was first analysed for each survey. Then, the role of education was investigated for 15-year and 35-year inactivity periods in the longitudinal analyses. The co-twin control design was used to analyse the potential familial confounding of the effects. All analyses were conducted with and without multiple covariates. Odds Ratios (OR) with 95 % Confidence Intervals (CI) were calculated using logistic and conditional (fixed-effects) regression models. Each additional year of education was associated with less inactivity (OR 0.94 to 0.95, 95 % CI 0.92, 0.99) in the cross-sectional age- and sex-adjusted analyses. The associations of education with inactivity in the 15- and 35-year follow-ups showed a similar trend: OR 0.97 (95 % CI 0.93, 1.00) and OR 0.94 (95 % CI 0.91, 0.98), respectively. In all co-twin control analyses, each year of higher education was associated with a reduced likelihood of inactivity suggesting direct effect (i.e. independent from familial confounding) of education on inactivity. However, the point estimates were lower than in the individual-level analyses. Adjustment for multiple covariates did not change these associations. Higher education is associated with lower odds of leisure-time physical inactivity during the three-decade follow-up. The association was found after adjusting for several confounders, including familial factors. Hence, the results point to the conclusion that education has an independent role in the development of long-term physical inactivity and tailored efforts to promote physical activity among lower educated people would be needed throughout adulthood.
Musculoskeletal pain and re-employment among unemployed job seekers: a three-year follow-up study.
Nwaru, Chioma A; Nygård, Clas-Håkan; Virtanen, Pekka
2016-07-08
Poor health is a potential risk factor for not finding employment among unemployed individuals. We investigated the associations between localized and multiple-site musculoskeletal pain and re-employment in a three-year follow-up of unemployed job seekers. Unemployed people (n = 539) from six localities in southern Finland who participated in various active labour market policy measures at baseline in 2002/2003 were recruited into a three-year health service intervention trial. A questionnaire was used to collect data on musculoskeletal health and background characteristics at baseline and on employment status at the end of the follow-up. We conducted a complete case (n = 284) and multiple imputation analyses using logistic regression to investigate the association between baseline musculoskeletal pain and re-employment after three years. Participants with severe pain in the lower back were less likely to become re-employed. This was independent of potential confounding variables. Pain in the hands/upper extremities, neck/shoulders, lower extremities, as well as multiple site were not determinants of re-employment. Our findings lend some support to the hypothesis that poor health can potentially cause health selection into employment. There is the need to disentangle health problems in order to clearly appreciate their putative impact on employment. This will allow for more targeted interventions for the unemployed.
African ancestry, lung function and the effect of genetics
Wehrmeister, Fernando C.; Hartwig, Fernando P.; Perez-Padilla, Rogelio; Gigante, Denise P.; Barros, Fernando C.; Oliveira, Isabel O.; Ferreira, Gustavo D.; Horta, Bernardo L.
2015-01-01
African-Americans have smaller lung function compared with European-Americans. The aim of this study was to disentangle the contribution of genetics from other variables on lung function. A cohort was followed from birth to 30 years of age in Brazil. Several variables were collected: genomic analysis based on DNA; forced expiratory volume in 1 s (FEV1) and forced vital capacity (FVC) obtained by spirometry; height measured by anthropometrists; and thorax circumference evaluated by photonic scanner. Crude and adjusted linear regression models were calculated according to African ancestry. The sample comprised 2869 participants out of 3701 members of the cohort. Males with higher African ancestry by DNA analysis had a smaller FEV1 (−0.13 L, 95% CI −0.23– −0.03 L) and FVC (−0.21 L, 95% CI −0.32– −0.09 L) compared with those with less African ancestry, having accounted for height, sitting to standing height ratio and other confounders. Similar effects were seen in females. After adjustment, ancestry remained significantly associated with lung function, but the large effect of adjustment for confounding among males (but not females) does not allow us to exclude the possibility that residual confounding may still account for these findings. PMID:25700383
Pandis, Nikolaos; Polychronopoulou, Argy; Madianos, Phoebus; Makou, Margarita; Eliades, Theodore
2011-06-01
The objective of this article was to record reporting characteristics related to study quality of research published in major specialty dental journals with the highest impact factor (Journal of Endodontics, Journal of Oral and Maxillofacial Surgery, American Journal of Orthodontics and Dentofacial Orthopedics; Pediatric Dentistry, Journal of Clinical Periodontology, and International Journal of Prosthetic Dentistry). The included articles were classified into the following 3 broad subject categories: (1) cross-sectional (snap-shot), (2) observational, and (3) interventional. Multinomial logistic regression was conducted for effect estimation using the journal as the response and randomization, sample calculation, confounding discussed, multivariate analysis, effect measurement, and confidence intervals as the explanatory variables. The results showed that cross-sectional studies were the dominant design (55%), whereas observational investigations accounted for 13%, and interventions/clinical trials for 32%. Reporting on quality characteristics was low for all variables: random allocation (15%), sample size calculation (7%), confounding issues/possible confounders (38%), effect measurements (16%), and multivariate analysis (21%). Eighty-four percent of the published articles reported a statistically significant main finding and only 13% presented confidence intervals. The Journal of Clinical Periodontology showed the highest probability of including quality characteristics in reporting results among all dental journals. Copyright © 2011 Elsevier Inc. All rights reserved.
The role of sex and body weight on the metabolic effects of high-fat diet in C57BL/6N mice
Ingvorsen, C; Karp, N A; Lelliott, C J
2017-01-01
Background: Metabolic disorders are commonly investigated using knockout and transgenic mouse models on the C57BL/6N genetic background due to its genetic susceptibility to the deleterious metabolic effects of high-fat diet (HFD). There is growing awareness of the need to consider sex in disease progression, but limited attention has been paid to sexual dimorphism in mouse models and its impact in metabolic phenotypes. We assessed the effect of HFD and the impact of sex on metabolic variables in this strain. Methods: We generated a reference data set encompassing glucose tolerance, body composition and plasma chemistry data from 586 C57BL/6N mice fed a standard chow and 733 fed a HFD collected as part of a high-throughput phenotyping pipeline. Linear mixed model regression analysis was used in a dual analysis to assess the effect of HFD as an absolute change in phenotype, but also as a relative change accounting for the potential confounding effect of body weight. Results: HFD had a significant impact on all variables tested with an average absolute effect size of 29%. For the majority of variables (78%), the treatment effect was modified by sex and this was dominated by male-specific or a male stronger effect. On average, there was a 13.2% difference in the effect size between the male and female mice for sexually dimorphic variables. HFD led to a significant body weight phenotype (24% increase), which acts as a confounding effect on the other analysed variables. For 79% of the variables, body weight was found to be a significant source of variation, but even after accounting for this confounding effect, similar HFD-induced phenotypic changes were found to when not accounting for weight. Conclusion: HFD and sex are powerful modifiers of metabolic parameters in C57BL/6N mice. We also demonstrate the value of considering body size as a covariate to obtain a richer understanding of metabolic phenotypes. PMID:28394359
A Comparison of Agent-Based Models and the Parametric G-Formula for Causal Inference.
Murray, Eleanor J; Robins, James M; Seage, George R; Freedberg, Kenneth A; Hernán, Miguel A
2017-07-15
Decision-making requires choosing from treatments on the basis of correctly estimated outcome distributions under each treatment. In the absence of randomized trials, 2 possible approaches are the parametric g-formula and agent-based models (ABMs). The g-formula has been used exclusively to estimate effects in the population from which data were collected, whereas ABMs are commonly used to estimate effects in multiple populations, necessitating stronger assumptions. Here, we describe potential biases that arise when ABM assumptions do not hold. To do so, we estimated 12-month mortality risk in simulated populations differing in prevalence of an unknown common cause of mortality and a time-varying confounder. The ABM and g-formula correctly estimated mortality and causal effects when all inputs were from the target population. However, whenever any inputs came from another population, the ABM gave biased estimates of mortality-and often of causal effects even when the true effect was null. In the absence of unmeasured confounding and model misspecification, both methods produce valid causal inferences for a given population when all inputs are from that population. However, ABMs may result in bias when extrapolated to populations that differ on the distribution of unmeasured outcome determinants, even when the causal network linking variables is identical. © The Author(s) 2017. Published by Oxford University Press on behalf of the Johns Hopkins Bloomberg School of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Association between zolpidem use and glaucoma risk: a Taiwanese population-based case-control study.
Ho, Yi-Hao; Chang, Yue-Cune; Huang, Wei-Cheng; Chen, Hsin-Yi; Lin, Che-Chen; Sung, Fung-Chang
2015-01-01
To date, the relationship between zolpidem use and subsequent risk of glaucoma in a Taiwanese population has not been assessed. We used data from the National Health Insurance system to investigate whether zolpidem use was related to glaucoma risk. A 1:4 matched case-control study was conducted. The cases were patients newly diagnosed with glaucoma from 2001 to 2010. The controls were randomly selected non-glaucoma subjects matched by sex and age (± 5 years). Zolpidem exposure and/or the average dosage of zolpidem used (mg/year) were evaluated. Medical comorbidities were considered as confounding factors. Multiple logistic regression models were used to evaluate the potential risk of zolpidem exposure on glaucoma with/without adjustment for the effects of confounding variables. The exposure rate of zolpidem use in the glaucoma group was significantly higher than that of the control group (2.8% vs. 2.0%, P < 0.0001). The adjusted odds ratio (OR) of the risk of glaucoma for those with zolpidem use vs. those without was 1.19 (95% confidence interval [CI], 1.02-1.38). Compared to non-zolpidem users, zolpidem users with an average dose of more than 200 mg/year had significantly increased risk of glaucoma (OR 1.31, 95% CI 1.03-1.68). This study suggests that the use of zolpidem might increase the risk of subsequent glaucoma. Further confirmatory studies are recommended to clarify this important issue.
Association Between Zolpidem Use and Glaucoma Risk: A Taiwanese Population-Based Case-Control Study
Ho, Yi-Hao; Chang, Yue-Cune; Huang, Wei-Cheng; Chen, Hsin-Yi; Lin, Che-Chen; Sung, Fung-Chang
2015-01-01
Background To date, the relationship between zolpidem use and subsequent risk of glaucoma in a Taiwanese population has not been assessed. Methods We used data from the National Health Insurance system to investigate whether zolpidem use was related to glaucoma risk. A 1:4 matched case-control study was conducted. The cases were patients newly diagnosed with glaucoma from 2001 to 2010. The controls were randomly selected non-glaucoma subjects matched by sex and age (±5 years). Zolpidem exposure and/or the average dosage of zolpidem used (mg/year) were evaluated. Medical comorbidities were considered as confounding factors. Multiple logistic regression models were used to evaluate the potential risk of zolpidem exposure on glaucoma with/without adjustment for the effects of confounding variables. Results The exposure rate of zolpidem use in the glaucoma group was significantly higher than that of the control group (2.8% vs. 2.0%, P < 0.0001). The adjusted odds ratio (OR) of the risk of glaucoma for those with zolpidem use vs. those without was 1.19 (95% confidence interval [CI], 1.02–1.38). Compared to non-zolpidem users, zolpidem users with an average dose of more than 200 mg/year had significantly increased risk of glaucoma (OR 1.31, 95% CI 1.03–1.68). Conclusions This study suggests that the use of zolpidem might increase the risk of subsequent glaucoma. Further confirmatory studies are recommended to clarify this important issue. PMID:25720944
Association Between Zolpidem Use and Glaucoma Risk: A Taiwanese Population-Based Case-Control Study.
Ho, Yi-Hao; Chang, Yue-Cune; Huang, Wei-Cheng; Chen, Hsin-Yi; Lin, Che-Chen; Sung, Fung-Chang
2014-08-23
Background: To date, the relationship between zolpidem use and subsequent risk of glaucoma in a Taiwanese population has not been assessed.Methods: We used data from the National Health Insurance system to investigate whether zolpidem use was related to glaucoma risk. A 1:4 matched case-control study was conducted. The cases were patients newly diagnosed with glaucoma from 2001 to 2010. The controls were randomly selected non-glaucoma subjects matched by sex and age (±5 years). Zolpidem exposure and/or the average dosage of zolpidem used (mg/year) were evaluated. Medical comorbidities were considered as confounding factors. Multiple logistic regression models were used to evaluate the potential risk of zolpidem exposure on glaucoma with/without adjustment for the effects of confounding variables.Results: The exposure rate of zolpidem use in the glaucoma group was significantly higher than that of the control group (2.8% vs. 2.0%, P < 0.0001). The adjusted odds ratio (OR) of the risk of glaucoma for those with zolpidem use vs. those without was 1.19 (95% confidence interval [CI], 1.02-1.38). Compared to non-zolpidem users, zolpidem users with an average dose of more than 200 mg/year had significantly increased risk of glaucoma (OR 1.31, 95% CI 1.03-1.68).Conclusions: This study suggests that the use of zolpidem might increase the risk of subsequent glaucoma. Further confirmatory studies are recommended to clarify this important issue.
Casale, Teodorico; Sacco, Carmina; Ricci, Serafino; Loreti, Beatrice; Pacchiarotti, Alessandro; Cupelli, Vincenzo; Arcangeli, Giulio; Mucci, Nicola; Antuono, Vittorio; De Marco, Federica; Tomei, Gianfranco; Tomei, Francesco; Rosati, Maria Valeria
2016-06-01
Few studies in the literature have examined the effects of benzene on blood cells. The aim of this study was to evaluate the possible correlation between the blood benzene levels and the blood cell counts. From a population of 2658 workers, we studied a group of 215 subjects. Each worker underwent blood sampling for the assessment of the blood benzene levels and the blood cell counts. The Mann-Whitney U test for two-mode variables and the Kruskal-Wallis test for more-than-two-mode variables were performed on all subjects. We estimated the Pearson correlation index between the variables in the total sample and the subgroups divided according to sex, the smoking habit, and job. After the main confounding factors were evaluated, multiple linear regression was performed on both the total sample and the subgroups. A significant inverse correlation was found among the blood benzene levels and the white blood cells, lymphocytes, and neutrophils in traffic policemen, motorcyclists, and other outdoor workers. We did not find any significant correlation with any other parameters of blood cell count. Our results, which must be considered preliminary, indicate that increased blood benzene levels in outdoor workers lead to decreased counts of white blood cells, neutrophils, and lymphocytes, because of possible immune effects. These are worth investigating in the future by specific immune tests. Copyright © 2016. Published by Elsevier Ltd.
Hayashi, Tatsuya; Saitoh, Satoshi; Takahashi, Junji; Tsuji, Yoshinori; Ikeda, Kenji; Kobayashi, Masahiro; Kawamura, Yusuke; Fujii, Takeshi; Inoue, Masafumi; Miyati, Tosiaki; Kumada, Hiromitsu
2017-04-01
The two-point Dixon method for magnetic resonance imaging (MRI) is commonly used to non-invasively measure fat deposition in the liver. The aim of the present study was to assess the usefulness of MRI-fat fraction (MRI-FF) using the two-point Dixon method based on the non-alcoholic fatty liver disease activity score. This retrospective study included 106 patients who underwent liver MRI and MR spectroscopy, and 201 patients who underwent liver MRI and histological assessment. The relationship between MRI-FF and MR spectroscopy-fat fraction was used to estimate the corrected MRI-FF for hepatic multi-peaks of fat. Then, a color FF map was generated with the corrected MRI-FF based on the non-alcoholic fatty liver disease activity score. We defined FF variability as the standard deviation of FF in regions of interest. Uniformity of hepatic fat was visually graded on a three-point scale using both gray-scale and color FF maps. Confounding effects of histology (iron, inflammation and fibrosis) on corrected MRI-FF were assessed by multiple linear regression. The linear correlations between MRI-FF and MR spectroscopy-fat fraction, and between corrected MRI-FF and histological steatosis were strong (R 2 = 0.90 and R 2 = 0.88, respectively). Liver fat variability significantly increased with visual fat uniformity grade using both of the maps (ρ = 0.67-0.69, both P < 0.001). Hepatic iron, inflammation and fibrosis had no significant confounding effects on the corrected MRI-FF (all P > 0.05). The two-point Dixon method and the gray-scale or color FF maps based on the non-alcoholic fatty liver disease activity score were useful for fat quantification in the liver of patients without severe iron deposition. © 2016 The Japan Society of Hepatology.
Marashinia, Farzad; Heshmat, Ramin; Motlagh, Mohammad-Esmaeil; Qorbani, Mostafa; Taslimi, Mahnaz; Nourbakhsh, Mohsen; Ardalan, Gelayol; Poursafa, Parinaz
2013-01-01
Introduction This study explores the associations of weight perceptions with actual body mass index (BMI) and attempts to lose weight in a nationally representative sample of a pediatric population. Material and methods Data were collected from school students of 27 provinces in Iran, as part of “the national survey of school student high risk behaviors”. We used t-test for continuous data and chi square test for categorical data. The correlation between categorical variables was assessed by Cramer's phi test. A multiple nominal logistic regression model was fitted to data to assess the association between perceived body weight and gender by adjusting for potential confounding variables. Results The study participants consisted of 5570 (2784 girls, 70% urban) students with mean age of 14.7 ±2.4 years. Overall, 17.3% of students were underweight, and 17.7% were overweight or obese. Nearly 25% and 50% of participants reported themselves as appropriate weight and very obese, respectively. In both genders, the strength of association between perceived weight and actual BMI was quite high (Cramer's phi coefficient = 0.5, p < 0.0001), and that of perceived body weight with trying to lose weight was moderate (Cramer's phi coefficient = 0.2, p < 0.0001). Overweight students were more likely than their obese peers to try to lose weight. After adjusting for possible confounders, the chance of perceiving oneself as very obese compared to perceiving oneself as very thin was 1.56-fold higher in girls than in boys, i.e. OR (95% CI): 1.56 (1.27-1.91). Conclusions This study revealed a considerably frequent “mismatch” between actual weight status and body shape dissatisfaction, which supports the necessity of increasing public awareness in this regard. PMID:23671430
Landau, Sabine; Emsley, Richard; Dunn, Graham
2018-06-01
Random allocation avoids confounding bias when estimating the average treatment effect. For continuous outcomes measured at post-treatment as well as prior to randomisation (baseline), analyses based on (A) post-treatment outcome alone, (B) change scores over the treatment phase or (C) conditioning on baseline values (analysis of covariance) provide unbiased estimators of the average treatment effect. The decision to include baseline values of the clinical outcome in the analysis is based on precision arguments, with analysis of covariance known to be most precise. Investigators increasingly carry out explanatory analyses to decompose total treatment effects into components that are mediated by an intermediate continuous outcome and a non-mediated part. Traditional mediation analysis might be performed based on (A) post-treatment values of the intermediate and clinical outcomes alone, (B) respective change scores or (C) conditioning on baseline measures of both intermediate and clinical outcomes. Using causal diagrams and Monte Carlo simulation, we investigated the performance of the three competing mediation approaches. We considered a data generating model that included three possible confounding processes involving baseline variables: The first two processes modelled baseline measures of the clinical variable or the intermediate variable as common causes of post-treatment measures of these two variables. The third process allowed the two baseline variables themselves to be correlated due to past common causes. We compared the analysis models implied by the competing mediation approaches with this data generating model to hypothesise likely biases in estimators, and tested these in a simulation study. We applied the methods to a randomised trial of pragmatic rehabilitation in patients with chronic fatigue syndrome, which examined the role of limiting activities as a mediator. Estimates of causal mediation effects derived by approach (A) will be biased if one of the three processes involving baseline measures of intermediate or clinical outcomes is operating. Necessary assumptions for the change score approach (B) to provide unbiased estimates under either process include the independence of baseline measures and change scores of the intermediate variable. Finally, estimates provided by the analysis of covariance approach (C) were found to be unbiased under all the three processes considered here. When applied to the example, there was evidence of mediation under all methods but the estimate of the indirect effect depended on the approach used with the proportion mediated varying from 57% to 86%. Trialists planning mediation analyses should measure baseline values of putative mediators as well as of continuous clinical outcomes. An analysis of covariance approach is recommended to avoid potential biases due to confounding processes involving baseline measures of intermediate or clinical outcomes, and not simply for increased precision.
Khan, Md Mobarak H; Kraemer, Alexander
2008-07-23
Worldwide one billion people are living in slum communities and experts projected that this number would double by 2030. Slum populations, which are increasing at an alarming rate in Bangladesh mainly due to rural-urban migration, are often neglected and characterized by poverty, poor housing, overcrowding, poor environment, and high prevalence of communicable diseases. Unfortunately, comparisons between women living in slums and those not living in slums are very limited in Bangladesh. The objectives of the study were to examine the association of living in slums (dichotomized as slum versus non-slum) with selected public health-related variables among women, first without adjusting for the influence of other factors and then in the presence of socio-economic variables. Secondary data was used in this study. 120 women living in slums (as cases) and 480 age-matched women living in other areas (as controls) were extracted from the Bangladesh Demographic and Health Survey 2004. Many socio-economic and demographic variables were analysed. SPSS was used to perform simple as well as multiple analyses. P-values based on t-test and Wald test were also reported to show the significance level. Unadjusted results indicated that a significantly higher percent of women living in slums came from country side, had a poorer status by household characteristics, had less access to mass media, and had less education than women not living in slums. Mean BMI, knowledge of AIDS indicated by ever heard about AIDS, knowledge of avoiding AIDS by condom use, receiving adequate antenatal visits (4 or more) during the last pregnancy, and safe delivery practices assisted by skilled sources were significantly lower among women living in slums than those women living in other areas. However, all the unadjusted significant associations with the variable slum were greatly attenuated and became insignificant (expect safe delivery practices) when some socio-economic variables namely childhood place of residence, a composite variable of household characteristics, a composite variable of mass media access, and education were inserted into the multiple regression models. Taken together, childhood place of residence, the composite variable of mass media access, and education were the strongest predictors for the health related outcomes. Reporting unadjusted findings of public health variables in women from slums versus non-slums can be misleading due to confounding factors. Our findings suggest that an association of childhood place of residence, mass media access and public health education should be considered before making any inference based on slum versus non-slum comparisons.
2012-01-01
Background Automated classification of histopathology involves identification of multiple classes, including benign, cancerous, and confounder categories. The confounder tissue classes can often mimic and share attributes with both the diseased and normal tissue classes, and can be particularly difficult to identify, both manually and by automated classifiers. In the case of prostate cancer, they may be several confounding tissue types present in a biopsy sample, posing as major sources of diagnostic error for pathologists. Two common multi-class approaches are one-shot classification (OSC), where all classes are identified simultaneously, and one-versus-all (OVA), where a “target” class is distinguished from all “non-target” classes. OSC is typically unable to handle discrimination of classes of varying similarity (e.g. with images of prostate atrophy and high grade cancer), while OVA forces several heterogeneous classes into a single “non-target” class. In this work, we present a cascaded (CAS) approach to classifying prostate biopsy tissue samples, where images from different classes are grouped to maximize intra-group homogeneity while maximizing inter-group heterogeneity. Results We apply the CAS approach to categorize 2000 tissue samples taken from 214 patient studies into seven classes: epithelium, stroma, atrophy, prostatic intraepithelial neoplasia (PIN), and prostate cancer Gleason grades 3, 4, and 5. A series of increasingly granular binary classifiers are used to split the different tissue classes until the images have been categorized into a single unique class. Our automatically-extracted image feature set includes architectural features based on location of the nuclei within the tissue sample as well as texture features extracted on a per-pixel level. The CAS strategy yields a positive predictive value (PPV) of 0.86 in classifying the 2000 tissue images into one of 7 classes, compared with the OVA (0.77 PPV) and OSC approaches (0.76 PPV). Conclusions Use of the CAS strategy increases the PPV for a multi-category classification system over two common alternative strategies. In classification problems such as histopathology, where multiple class groups exist with varying degrees of heterogeneity, the CAS system can intelligently assign class labels to objects by performing multiple binary classifications according to domain knowledge. PMID:23110677
Pingault, Jean Baptiste; Côté, Sylvana M; Petitclerc, Amélie; Vitaro, Frank; Tremblay, Richard E
2015-01-01
Parental educational expectations have been associated with children's educational attainment in a number of long-term longitudinal studies, but whether this relationship is causal has long been debated. The aims of this prospective study were twofold: 1) test whether low maternal educational expectations contributed to failure to graduate from high school; and 2) compare the results obtained using different strategies for accounting for confounding variables (i.e. multivariate regression and propensity score matching). The study sample included 1,279 participants from the Quebec Longitudinal Study of Kindergarten Children. Maternal educational expectations were assessed when the participants were aged 12 years. High school graduation—measuring educational attainment—was determined through the Quebec Ministry of Education when the participants were aged 22-23 years. Findings show that when using the most common statistical approach (i.e. multivariate regressions to adjust for a restricted set of potential confounders) the contribution of low maternal educational expectations to failure to graduate from high school was statistically significant. However, when using propensity score matching, the contribution of maternal expectations was reduced and remained statistically significant only for males. The results of this study are consistent with the possibility that the contribution of parental expectations to educational attainment is overestimated in the available literature. This may be explained by the use of a restricted range of potential confounding variables as well as the dearth of studies using appropriate statistical techniques and study designs in order to minimize confounding. Each of these techniques and designs, including propensity score matching, has its strengths and limitations: A more comprehensive understanding of the causal role of parental expectations will stem from a convergence of findings from studies using different techniques and designs.
Durbin, Richard; Winn, John
2010-01-01
Gene expression measurements are influenced by a wide range of factors, such as the state of the cell, experimental conditions and variants in the sequence of regulatory regions. To understand the effect of a variable of interest, such as the genotype of a locus, it is important to account for variation that is due to confounding causes. Here, we present VBQTL, a probabilistic approach for mapping expression quantitative trait loci (eQTLs) that jointly models contributions from genotype as well as known and hidden confounding factors. VBQTL is implemented within an efficient and flexible inference framework, making it fast and tractable on large-scale problems. We compare the performance of VBQTL with alternative methods for dealing with confounding variability on eQTL mapping datasets from simulations, yeast, mouse, and human. Employing Bayesian complexity control and joint modelling is shown to result in more precise estimates of the contribution of different confounding factors resulting in additional associations to measured transcript levels compared to alternative approaches. We present a threefold larger collection of cis eQTLs than previously found in a whole-genome eQTL scan of an outbred human population. Altogether, 27% of the tested probes show a significant genetic association in cis, and we validate that the additional eQTLs are likely to be real by replicating them in different sets of individuals. Our method is the next step in the analysis of high-dimensional phenotype data, and its application has revealed insights into genetic regulation of gene expression by demonstrating more abundant cis-acting eQTLs in human than previously shown. Our software is freely available online at http://www.sanger.ac.uk/resources/software/peer/. PMID:20463871
Kazazi, Leila; Foroughan, Mahshid; Nejati, Vahid; Shati, Mohsen
2018-04-01
Age associated cognitive decline or normal cognitive aging is related with lower levels of functioning in real life, and may interfere with maintaining independence and health related quality of life (HRQL). In this study, health related quality of life and cognitive function in community-dwelling older adults were evaluated with the aim of exploring the association between them by adjusting for potential confounders. This cross-sectional study, was implemented on 425 community-dwelling older adults aged 60 and over, between August 2016 and October 2016 in health centers of the municipality of Tehran, Iran, using Mini Mental State Examination (MMSE) to assess cognitive function and Short Form-36 scales (SF-36) to assess HRQL. The relation between HRQL and cognitive function was evaluated by Pearson's correlation coefficient, and the impact of cognitive function on HRQL adjusted for potential confounders was estimated by linear regression model. All analyses were done using SPSS, version 22.0. A positive significant correlation between cognitive function and quality of life (r=0.434; p<0.001) and its dimensions was observed. Two variables of educational level (B=2.704; 95% CI: 2.09 to 3.30; p<0.001) and depression (B=2.554; 95% CI: 2.00 to 3.10; p<0.001) were assumed as potential confounder by changing effect measure after entering the model. After adjusting for potential confounders in regression model, the association between MMSE scores and quality of life persisted (B=2.417; 95% CI: 1.86 to 2.96; p<0.001). The results indicate that cognitive function was associated with HRQL in older adults with age associated cognitive function. Two variables of educational level and depression can affect the relation between cognitive decline and HRQL.
Prevalence of Pterygia in Hawaii: Examining Cumulative Surfing Hours as a Risk Factor.
Lin, Alexander D; Miles, Ku'ulei; Brinks, Mitchel V
2016-08-01
To examine the association between surfing and pterygium prevalence in Hawaii. A convenience sampling was performed at four beaches on the island of Oahu, Hawaii. A total of 169 individuals were interviewed and underwent penlight examination to assess grade and extent of pterygium. Of 169 participants aged 18-80 years, 88 non-surfers, 41 occasional surfers, 15 recreational surfers and 25 surfing enthusiasts were identified based on their lifetime surfing hours. Overall, 19 participants were found to have pterygia (28 pterygia total) including two non-surfers (2.3%), five occasional surfers (12.2%), three recreational surfers (20.0%), and nine enthusiast surfers (36.0%). Variables associated with pterygium prevalence were lifetime surfing hours (p < 0.0001), outdoor occupation (p = 0.04), Hawaiian residence (p = 0.009), and Hawaiian/Pacific Islander ethnicity (p = 0.002). Multiple logistic regression with the outcome of pterygium was performed, along with multiple linear regression for the continuous outcomes of corneal extent, chord length, and apex-visual axis gap, with lifetime surfing hours as the primary explanatory variable. After adjustment for confounders, a significant linear relationship was observed between chord length and lifetime surfing hours (p = 0.01). Surfing was associated with an increased pterygium prevalence and trend towards an association with increased pterygium severity. Increased risk of exposure to wind, particle irritation, and ultraviolet (UV) radiation while surfing may contribute to pterygium development. Implications for public health include promoting UV protective eyewear during surfing, in addition to raising awareness about the association of pterygia and the sport of surfing.
DOE Office of Scientific and Technical Information (OSTI.GOV)
Hu, H.; Kim, Rokho; Korrick, S.
1996-12-31
In an earlier report based on participants in the Veterans Administration Normative Aging Study, we found a significant association between the risk of hypertension and lead levels in tibia. To examine the possible confounding effects of education and occupation, we considered in this study five levels of education and three levels of occupation as independent variables in the statistical model. Of 1,171 active subjects seen between August 1991 and December 1994, 563 provided complete data for this analysis. In the initial logistic regression model, acre and body mass index, family history of hypertension, and dietary sodium intake, but neither cumulativemore » smoking nor alcohol ingestion, conferred increased odds ratios for being hypertensive that were statistically significant. When the lead biomarkers were added separately to this initial logistic model, tibia lead and patella lead levels were associated with significantly elevated odds ratios for hypertension. In the final backward elimination logistic regression model that included categorical variables for education and occupation, the only variables retained were body mass index, family history of hypertension, and tibia lead level. We conclude that education and occupation variables were not confounding the association between the lead biomarkers and hypertension that we reported previously. 27 refs., 3 tabs.« less
Climate is a key driver regulating vegetation structure across rural ecosystems. In urban ecosystems, multiple interactions between humans and the environment can have homogenizing influences, confounding the relationship between vegetation structure and climate. In fact, vegetat...
Kwon, Jin-Woo; Choi, Jin A; La, Tae Yoon
2016-11-01
The aim of this article was to assess the associations of serum 25-hydroxyvitamin D [25(OH)D] and daily sun exposure time with myopia in Korean adults.This study is based on the Korea National Health and Nutrition Examination Survey (KNHANES) of Korean adults in 2010-2012; multiple logistic regression analyses were performed to examine the associations of serum 25(OH)D levels and daily sun exposure time with myopia, defined as spherical equivalent ≤-0.5D, after adjustment for age, sex, household income, body mass index (BMI), exercise, intraocular pressure (IOP), and education level. Also, multiple linear regression analyses were performed to examine the relationship between serum 25(OH)D levels with spherical equivalent after adjustment for daily sun exposure time in addition to the confounding factors above.Between the nonmyopic and myopic groups, spherical equivalent, age, IOP, BMI, waist circumference, education level, household income, and area of residence differed significantly (all P < 0.05). Compared with subjects with daily sun exposure time <2 hour, subjects with sun exposure time ≥2 to <5 hour, and those with sun exposure time ≥5 hour had significantly less myopia (P < 0.001). In addition, compared with subjects were categorized into quartiles of serum 25(OH)D, the higher quartiles had gradually lower prevalences of myopia after adjustment for confounding factors (P < 0.001). In multiple linear regression analyses, spherical equivalent was significantly associated with serum 25(OH)D concentration after adjustment for confounding factors (P = 0.002).Low serum 25(OH)D levels and shorter daily sun exposure time may be independently associated with a high prevalence of myopia in Korean adults. These data suggest a direct role for vitamin D in the development of myopia.
Szatkiewicz, Jin P; Wang, WeiBo; Sullivan, Patrick F; Wang, Wei; Sun, Wei
2013-02-01
Structural variation is an important class of genetic variation in mammals. High-throughput sequencing (HTS) technologies promise to revolutionize copy-number variation (CNV) detection but present substantial analytic challenges. Converging evidence suggests that multiple types of CNV-informative data (e.g. read-depth, read-pair, split-read) need be considered, and that sophisticated methods are needed for more accurate CNV detection. We observed that various sources of experimental biases in HTS confound read-depth estimation, and note that bias correction has not been adequately addressed by existing methods. We present a novel read-depth-based method, GENSENG, which uses a hidden Markov model and negative binomial regression framework to identify regions of discrete copy-number changes while simultaneously accounting for the effects of multiple confounders. Based on extensive calibration using multiple HTS data sets, we conclude that our method outperforms existing read-depth-based CNV detection algorithms. The concept of simultaneous bias correction and CNV detection can serve as a basis for combining read-depth with other types of information such as read-pair or split-read in a single analysis. A user-friendly and computationally efficient implementation of our method is freely available.
Cong, Xiaowei
2018-05-01
Outdoor air pollution may be associated with cancer risk at different sites. This study sought to investigate outdoor air pollution from waste gas emission effects on multiple cancer incidences in a retrospective population-based study in Shanghai, China. Trends in cancer incidence for males and females and trends in waste gas emissions for the total waste gas, industrial waste gas, other waste gas, SO 2 , and soot were investigated between 1983 and 2010 in Shanghai, China. Regression models after adjusting for confounding variables were constructed to estimate associations between waste gas emissions and multiple cancer incidences in the whole group and stratified by sex, Engel coefficient, life expectancy, and number of doctors per 10,000 populations to further explore whether changes of waste gas emissions were associated with multiple cancer incidences. More than 550,000 new cancer patients were enrolled and reviewed. Upward trends in multiple cancer incidences for males and females and in waste gas emissions were observed from 1983 to 2010 in Shanghai, China. Waste gas emissions came mainly from industrial waste gas. Waste gas emissions was significantly positively associated with cancer incidence of salivary gland, small intestine, colorectal, anus, gallbladder, thoracic organs, connective and soft tissue, prostate, kidney, bladder, thyroid, non-Hodgkin's lymphoma, lymphatic leukemia, myeloid leukemia, and other unspecified sites (all p < 0.05). Negative association between waste gas emissions and the esophagus cancer incidence was observed (p < 0.05). The results of the whole group were basically consistent with the results of the stratified analysis. The results from this retrospective population-based study suggest ambient air pollution from waste gas emissions was associated with multiple cancer incidences.
COMPOSITE SAMPLING FOR SOIL VOC ANALYSIS
Data published by numerous researchers over the last decade demonstrate that there is a high degree of spatial variability in the measurement of volatile organic compounds (VOCs) in soil at contaminated waste sites. This phenomenon is confounded by the use of a small sample aliqu...
Childhood Medical and Behavioral Consequences of Maternal Cocaine Use1
Singer, Lynn; Farkas, Kathleen; Kliegman, Robert
2014-01-01
Reviewed available studies of the impact of fetal cocaine exposure on child medical and developmental outcome, as well as the current status of clinical psychological interventions and research strategies. Current studies are inconclusive but suggest that prenatal exposure to crack-cocaine can have significant effects on the growth and neurological development of the infant, with the potential of later learning and behavioral disabilities. Social-environmental correlates of maternal cocaine use are confounding factors with known negative effects on child outcome. Large, population-based studies using multivariate analyses are needed to determine the independent effects of cocaine on child outcome relative to other confounding variables. PMID:1382125
Dettenborn, L; Tietze, A; Kirschbaum, C; Stalder, T
2012-11-01
To inform the future use of hair cortisol measurement, we have investigated influences of potential confounding variables (natural hair colour, frequency of hair washes, age, sex, oral contraceptive (OC) use and smoking status) on hair cortisol levels. The main study sample comprised 360 participants (172 women) covering a wide range of ages (1-91 years; mean = 25.95). In addition, to more closely examine influences of natural hair colour and young age on hair cortisol levels, two additional samples comprising 69 participants with natural blond or dark brown hair (hair colour sample) as well as 28 young children and 34 adults (young age sample) were recruited. Results revealed a lack of an effect for natural hair colour, OC use, and smoking status on hair cortisol levels (all p's >0.10). No influence of frequency of hair washes was seen for proximal hair segments (p = 0.335) but for the third hair segment indicating lower cortisol content (p = 0.008). We found elevated hair cortisol levels in young children and older adults (p < 0.001). Finally, men showed higher hair cortisol levels than women (p = 0.002). The present data indicate that hair cortisol measurement provides a useful tool in stress-related psychobiological research when applied with the consideration of possible confounders including age and sex.
Vandasova, Zdenka; Vencálek, Ondřej; Puklová, Vladimíra
2016-01-01
Noise is one of the most extensive environmental factors affecting the general population. The present study is focused on the association between discomfort caused by noise and the incidence of certain diseases (ischaemic heart disease, stroke and hypertension). This cross-sectional questionnaire study, conducted in 10 cities in the Czech Republic, comprises two stages with 3592 obtained questionnaires in the first phase and 762 in the second phase. Twelve variables describe subjective responses to noise from different sources at different times of day. The intensity of the associations between variables was measured by correlation coefficient. Logistic regression was used for fitting models of morbidity, and confounders such as age and socio-economic status were included. The hypotheses from the first phase were independently validated using data from the second phase. The general rates of noise annoyance/sleep disturbance had greater correlation with traffic noise variables than with neighbourhood noise variables. Factors significantly associated with diseases are: for hypertension - annoyance by traffic noise (the elderly, odds ratio (OR) 1.4) and sleep disturbance by traffic and neighbourhood noise (the elderly, OR 1.6); for ischaemic heart disease - the general rate of noise annoyance (all respondents, OR 1.5 and the adults 30-60 years, OR 1.8) and the general rate of annoyance and sleep disturbance (all respondents, OR 1.3); for stroke - annoyance and sleep disturbance by traffic and neighbourhood noise (all respondents, OR 1.8). Factors that include multiple sources of noise or non-specific noise are associated with the studied diseases more frequently than the source-specific factors.
Meta-Analyses of Seven of NIDA’s Principles of Drug Addiction Treatment
Pearson, Frank S.; Prendergast, Michael L.; Podus, Deborah; Vazan, Peter; Greenwell, Lisa; Hamilton, Zachary
2011-01-01
Seven of the 13 Principles of Drug Addiction Treatment disseminated by the National Institute on Drug Abuse (NIDA) were meta-analyzed as part of the Evidence-based Principles of Treatment (EPT) project. By averaging outcomes over the diverse programs included in EPT, we found that five of the NIDA principles examined are supported: matching treatment to the client’s needs; attending to the multiple needs of clients; behavioral counseling interventions; treatment plan reassessment; and counseling to reduce risk of HIV. Two of the NIDA principles are not supported: remaining in treatment for an adequate period of time and frequency of testing for drug use. These weak effects could be the result of the principles being stated too generally to apply to the diverse interventions and programs that exist or of unmeasured moderator variables being confounded with the moderators that measured the principles. Meta-analysis should be a standard tool for developing principles of effective treatment for substance use disorders. PMID:22119178
Influence of social support on health-related quality of life outcomes in head and neck cancer.
Karnell, Lucy Hynds; Christensen, Alan J; Rosenthal, Eben L; Magnuson, J Scott; Funk, Gerry F
2007-02-01
Evidence that social support influences health-related quality of life (HRQOL) in oncologic patients could be particularly important for head and neck cancer because this disease can affect speech, eating, and facial aesthetics. Multiple regression analyses were used in this prospective, observational study to determine the association between 394 patients' ratings of perceived post-treatment social support and HRQOL outcomes while controlling for possible confounding variables. Higher social support scores were significantly associated with higher scores in speech (p = .007), aesthetics (p = .015), social disruption (p = .045), and general mental health (p = .016) and with fewer depressive symptoms (p = .023) but not with general physical health (p = .191) or eating (p = .114). The magnitude of differences in the HRQOL outcomes for patients whose social support scores fell in the lowest and highest quartiles were clinically meaningful. Given the association between social support and HRQOL outcomes in this patient population, modification of perceived social support through clinical interventions could improve the survivorship of these patients.
Long-term consequences of adolescent gang membership for adult functioning.
Gilman, Amanda B; Hill, Karl G; Hawkins, J David
2014-05-01
We examined the possible public health consequences of adolescent gang membership for adult functioning. Data were drawn from the Seattle Social Development Project, a longitudinal study focusing on the development of positive and problem outcomes. Using propensity score matching and logistic regression analyses, we assessed the effects of adolescent gang membership on illegal behavior, educational and occupational attainment, and physical and mental health at the ages of 27, 30, and 33 years. In comparison with their nongang peers, who had been matched on 23 confounding risk variables known to be related to selection into gang membership, those who had joined a gang in adolescence had poorer outcomes in multiple areas of adult functioning, including higher rates of self-reported crime, receipt of illegal income, incarceration, drug abuse or dependence, poor general health, and welfare receipt and lower rates of high school graduation. The finding that adolescent gang membership has significant consequences in adulthood beyond criminal behavior indicates the public health importance of the development of effective gang prevention programs.
Abnormal cardiac autonomic control in sickle cell disease following transient hypoxia.
Sangkatumvong, Suvimol; Khoo, Michael C K; Coates, Thomas D
2008-01-01
Abnormalities in autonomic control in sickle cell anemia (SCA) patients have been reported by multiple researchers. However their potential causal association with sickle cell crisis remains unknown. We employed hypoxia, a known trigger to sickle cell crisis, to perturb the autonomic systems of the subjects. Cardiac autonomic control was non-invasively assessed by tracking the changes in heart rate variability (HRV) that occur following brief exposure to a hypoxia stimulus. Time varying spectral analysis of HRV was applied to estimate the cardiac autonomic response to the transient episode of hypoxia. The results demonstrate that cardiovascular autonomic response to hypoxia is substantially more sensitive in SCA than in normal controls. We also developed a model to compensate for the confounding effects of respiration on the HRV spectral indices by using the corresponding respiration signal to compensate for the respiratory correlated part of the HRV. This technique improved the resolution with which the effect of hypoxia on changes in HRV could be measured.
Data mining: childhood injury control and beyond.
Tepas, Joseph J
2009-08-01
Data mining is defined as the automatic extraction of useful, often previously unknown information from large databases or data sets. It has become a major part of modern life and is extensively used in industry, banking, government, and health care delivery. The process requires a data collection system that integrates input from multiple sources containing critical elements that define outcomes of interest. Appropriately designed data mining processes identify and adjust for confounding variables. The statistical modeling used to manipulate accumulated data may involve any number of techniques. As predicted results are periodically analyzed against those observed, the model is consistently refined to optimize precision and accuracy. Whether applying integrated sources of clinical data to inferential probabilistic prediction of risk of ventilator-associated pneumonia or population surveillance for signs of bioterrorism, it is essential that modern health care providers have at least a rudimentary understanding of what the concept means, how it basically works, and what it means to current and future health care.
Pearson, Frank S; Prendergast, Michael L; Podus, Deborah; Vazan, Peter; Greenwell, Lisa; Hamilton, Zachary
2012-07-01
Of the 13 principles of drug addiction treatment disseminated by the National Institute on Drug Abuse (NIDA), 7 were meta-analyzed as part of the Evidence-based Principles of Treatment (EPT) project. By averaging outcomes over the diverse programs included in the EPT, we found that 5 of the NIDA principles examined are supported: matching treatment to the client's needs, attending to the multiple needs of clients, behavioral counseling interventions, treatment plan reassessment, and counseling to reduce risk of HIV. Two of the NIDA principles are not supported: remaining in treatment for an adequate period and frequency of testing for drug use. These weak effects could be the result of the principles being stated too generally to apply to the diverse interventions and programs that exist or unmeasured moderator variables being confounded with the moderators that measured the principles. Meta-analysis should be a standard tool for developing principles of effective treatment for substance use disorders. Copyright © 2012 Elsevier Inc. All rights reserved.
Nested case-control studies: should one break the matching?
Borgan, Ørnulf; Keogh, Ruth
2015-10-01
In a nested case-control study, controls are selected for each case from the individuals who are at risk at the time at which the case occurs. We say that the controls are matched on study time. To adjust for possible confounding, it is common to match on other variables as well. The standard analysis of nested case-control data is based on a partial likelihood which compares the covariates of each case to those of its matched controls. It has been suggested that one may break the matching of nested case-control data and analyse them as case-cohort data using an inverse probability weighted (IPW) pseudo likelihood. Further, when some covariates are available for all individuals in the cohort, multiple imputation (MI) makes it possible to use all available data in the cohort. In the paper we review the standard method and the IPW and MI approaches, and compare their performance using simulations that cover a range of scenarios, including one and two endpoints.
Allen, Mark S; Laborde, Sylvain; Walter, Emma E
2017-03-01
This prospective study explored the potential mediating role of health-related behavior (alcohol involvement, diet, television viewing, and physical activity) in the association between personality and change in memory performance over 2 years. A nationally representative sample of 8,376 U.K. participants aged 55 years and older (4,572 women, 3,804 men) completed self-report measures of personality and health-related behavior in 2010, and completed a memory performance task in 2010 and 2012. After removing variance associated with potential confounding variables, neuroticism and agreeableness had negative associations, and openness and conscientiousness positive associations with change in memory performance. There were no moderation effects by age, sex, education level, or ethnicity. Multiple mediator models demonstrated that physical activity, television viewing, and alcohol intake mediated associations between personality and change in memory performance. These findings provide evidence that the association between personality and memory performance in older adults can be explained, in part, through health-related behavior.
Matsumoto, Masatoshi; Inoue, Kazuo; Noguchi, Satomi; Toyokawa, Satoshi; Kajii, Eiji
2009-02-18
In many countries, there is a surplus of physicians in some communities and a shortage in others. Population size is known to be correlated with the number of physicians in a community, and is conventionally considered to represent the power of communities to attract physicians. However, associations between other demographic/economic variables and the number of physicians in a community have not been fully evaluated. This study seeks other parameters that correlate with the physician population and show which characteristics of a community determine its "attractiveness" to physicians. Associations between the number of physicians and selected demographic/economic/life-related variables of all of Japan's 3132 municipalities were examined. In order to exclude the confounding effect of community size, correlations between the physician-to-population ratio and other variable-to-population ratios or variable-to-area ratios were evaluated with simple correlation and multiple regression analyses. The equity of physician distribution against each variable was evaluated by the orenz curve and Gini index. Among the 21 variables selected, the service industry workers-to-population ratio (0.543), commercial land price (0.527), sales of goods per person (0.472), and daytime population density (0.451) were better correlated with the physician-to-population ratio than was population density (0.409). Multiple regression analysis showed that the service industry worker-to-population ratio, the daytime population density, and the elderly rate were each independently correlated with the physician-to-population ratio (standardized regression coefficient 0.393, 0.355, 0.089 respectively; each p<0.001). Equity of physician distribution was higher against service industry population (Gini index=0.26) and daytime population (0.28) than against population (0.33). Daytime population and service industry population in a municipality are better parameters of community attractiveness to physicians than population. Because attractiveness is supposed to consist of medical demand and the amenities of urban life, the two parameters may represent the amount of medical demand and/or the extent of urban amenities of the community more precisely than population does. The conventional demand-supply analysis based solely on population as the demand parameter may overestimate the inequity of the physician distribution among communities.
Borgen, Nicolai T
2014-11-01
This paper addresses the recent discussion on confounding in the returns to college quality literature using the Norwegian case. The main advantage of studying Norway is the quality of the data. Norwegian administrative data provide information on college applications, family relations and a rich set of control variables for all Norwegian citizens applying to college between 1997 and 2004 (N = 141,319) and their succeeding wages between 2003 and 2010 (676,079 person-year observations). With these data, this paper uses a subset of the models that have rendered mixed findings in the literature in order to investigate to what extent confounding biases the returns to college quality. I compare estimates obtained using standard regression models to estimates obtained using the self-revelation model of Dale and Krueger (2002), a sibling fixed effects model and the instrumental variable model used by Long (2008). Using these methods, I consistently find increasing returns to college quality over the course of students' work careers, with positive returns only later in students' work careers. I conclude that the standard regression estimate provides a reasonable estimate of the returns to college quality. Copyright © 2014 Elsevier Inc. All rights reserved.
Hydrologic impacts of land disturbance and management can be confounded by rainfall variability. As a consequence, attempts to gauge and quantify these effects through streamflow monitoring are typically subject to uncertainties. This paper addresses the quantification and deline...
Brody, Gene H; Miller, Gregory E; Yu, Tianyi; Beach, Steven R H; Chen, Edith
2016-04-01
This study tested the hypothesis that supportive family environments during adolescence buffer exposure to racial discrimination, reducing its impact on biological weathering and its manifestation in cellular aging. Perceived racial discrimination, support in the family environment, and confounder variables were assessed for 3 consecutive years across adolescence in two independent cohorts of African American youth from rural Georgia. DNA was extracted from peripheral blood mononuclear cells collected during young adulthood. Patterns of methylation were used to index the epigenetic ages of these cells and the extent to which they differed from participants' chronological ages. Among youth in supportive family environments, exposure to higher levels of racial discrimination did not forecast greater epigenetic aging. Among youth in less supportive family environments, exposure to higher levels of racial discrimination did forecast greater epigenetic aging. The associations emerged independently of confounder variables, and the results were replicated across the two cohorts. © The Author(s) 2016.
Instrumental variable methods in comparative safety and effectiveness research†
Brookhart, M. Alan; Rassen, Jeremy A.; Schneeweiss, Sebastian
2010-01-01
Summary Instrumental variable (IV) methods have been proposed as a potential approach to the common problem of uncontrolled confounding in comparative studies of medical interventions, but IV methods are unfamiliar to many researchers. The goal of this article is to provide a non-technical, practical introduction to IV methods for comparative safety and effectiveness research. We outline the principles and basic assumptions necessary for valid IV estimation, discuss how to interpret the results of an IV study, provide a review of instruments that have been used in comparative effectiveness research, and suggest some minimal reporting standards for an IV analysis. Finally, we offer our perspective of the role of IV estimation vis-à-vis more traditional approaches based on statistical modeling of the exposure or outcome. We anticipate that IV methods will be often underpowered for drug safety studies of very rare outcomes, but may be potentially useful in studies of intended effects where uncontrolled confounding may be substantial. PMID:20354968
SELECTING LEAST-DISTURBED SURVEY SITES FOR GREAT PLAINS STREAMS AND RIVERS
True reference condition probably does not exist for streams in highly utilized regions such as the Great Plains. Selecting least-disturbed sites for large regions is confounded by the association between human uses and natural gradients, and by multiple kinds of disturbance. U...
Use of a quality trait index to increase the reliability of phenotypic evaluations in broccoli
USDA-ARS?s Scientific Manuscript database
Selection of superior broccoli hybrids involves multiple considerations, including optimization of head quality traits. Quality assessment of broccoli heads is often confounded by relatively subjective human preferences for optimal appearance of heads. To assist the selection process, we assessed fi...
Smartberries: Interpreting Erdrich's Love Medicine
ERIC Educational Resources Information Center
Treuer, David
2005-01-01
The structure of "Love Medicines" interpreted by Hertha D. Sweet Wong who claims that the book's "multiple narrators confound conventional Western expectations of an autonomous protagonist, a dominant narrative voice, and a consistently chronological narrative". "Love Medicine" is a brilliant use of the Western literary tactics that create the…
Autonomic imbalance is associated with reduced facial recognition in somatoform disorders.
Pollatos, Olga; Herbert, Beate M; Wankner, Sarah; Dietel, Anja; Wachsmuth, Cornelia; Henningsen, Peter; Sack, Martin
2011-10-01
Somatoform disorders are characterized by the presence of multiple somatic symptoms. While the accuracy of perceiving bodily signal (interoceptive awareness) is only sparely investigated in somatoform disorders, recent research has associated autonomic imbalance with cognitive and emotional difficulties in stress-related diseases. This study aimed to investigate how sympathovagal reactivity interacts with performance in recognizing emotions in faces (facial recognition task). Using a facial recognition and appraisal task, skin conductance levels (SCLs), heart rate (HR) and heart rate variability (HRV) were assessed in 26 somatoform patients and compared to healthy controls. Interoceptive awareness was assessed by a heartbeat detection task. We found evidence for a sympathovagal imbalance in somatoform disorders characterized by low parasympathetic reactivity during emotional tasks and increased sympathetic activation during baseline. Somatoform patients exhibited a reduced recognition performance for neutral and sad emotional expressions only. Possible confounding variables such as alexithymia, anxiety or depression were taken into account. Interoceptive awareness was reduced in somatoform patients. Our data demonstrate an imbalance in sympathovagal activation in somatoform disorders associated with decreased parasympathetic activation. This might account for difficulties in processing of sad and neutral facial expressions in somatoform patients which might be a pathogenic mechanism for increased everyday vulnerability. Copyright © 2011 Elsevier Inc. All rights reserved.
Harmonization of multi-site diffusion tensor imaging data.
Fortin, Jean-Philippe; Parker, Drew; Tunç, Birkan; Watanabe, Takanori; Elliott, Mark A; Ruparel, Kosha; Roalf, David R; Satterthwaite, Theodore D; Gur, Ruben C; Gur, Raquel E; Schultz, Robert T; Verma, Ragini; Shinohara, Russell T
2017-11-01
Diffusion tensor imaging (DTI) is a well-established magnetic resonance imaging (MRI) technique used for studying microstructural changes in the white matter. As with many other imaging modalities, DTI images suffer from technical between-scanner variation that hinders comparisons of images across imaging sites, scanners and over time. Using fractional anisotropy (FA) and mean diffusivity (MD) maps of 205 healthy participants acquired on two different scanners, we show that the DTI measurements are highly site-specific, highlighting the need of correcting for site effects before performing downstream statistical analyses. We first show evidence that combining DTI data from multiple sites, without harmonization, may be counter-productive and negatively impacts the inference. Then, we propose and compare several harmonization approaches for DTI data, and show that ComBat, a popular batch-effect correction tool used in genomics, performs best at modeling and removing the unwanted inter-site variability in FA and MD maps. Using age as a biological phenotype of interest, we show that ComBat both preserves biological variability and removes the unwanted variation introduced by site. Finally, we assess the different harmonization methods in the presence of different levels of confounding between site and age, in addition to test robustness to small sample size studies. Copyright © 2017 Elsevier Inc. All rights reserved.
Sibai, Tarek; Tohme, Rania A; Beydoun, Hind A; Kanaan, Nabil; Sibai, Abla Mehio
2009-03-01
Adolescent violence is a significant public health problem. The primary objective of this study is to assess the prevalence and correlates of violent behavior among adolescent students in Lebanon. A cross-sectional study was conducted among a representative sample of 827 secondary students enrolled in public and private schools in Beirut. Using a series of multiple logistic regression techniques, socio-demographic variables which significantly associated with violent behavior were included as potential confounders in building the models for risk behavior. Nearly 42 and 17% of adolescents reported being involved in physical fights and weapon carrying, respectively. Boys were significantly more likely to use violence than girls. Whereas associations with physical fights were stronger for socio-economic variables and perceived rank in class, weapon carrying was significantly associated with problem behaviors, such as unintentional injury, substance abuse and sexual activity, with effect measures being stronger than those estimated for physical fighting. Compared with other countries, the rates of violent behavior in Lebanon are relatively high. The results from this study are discussed in light of the political ecology of Lebanon which may contribute to a culture that perpetrates violent behavior and may have influenced the clustering pattern of risk behaviors.
McCurdy, Karen; Kisler, Tiffani; Gorman, Kathleen S.; Metallinos-Katsaras, Elizabeth
2015-01-01
Objective Examine how income-related challenges around food and health are associated with variation in self-reported maternal body weight among low-income mothers. Design Cross-sectional, correlational design. Convenience sample recruited from 7 daycare centers and a Supplemental Nutrition Assistance Program outreach project. Maternal self-report data collected between October 2009 and May 2011. Setting Two Northeastern cities. Participants Sample of 166 mothers; 67% overweight or obese, 55% Hispanic, 42% reporting household food insecurity (HFI). Main Outcome Measures Maternal self-reported height and weight to calculate Body Mass Index (BMI). Independent variables: food program participation, supermarket use, 8-item food shopping practices scale, HFI, maternal depressive symptoms, self-rated health (SRH). Analysis Hierarchical multiple regression analysis tested relationships between maternal BMI with the independent variables of interest, adjusting for demographic confounds. Results Shopping practices to stretch food dollars (P = .04), using community food assistance programs (P < .05), and HFI (P < .04) correlated with heavier maternal BMIs; higher SRH corresponded to lower BMIs (P =.004). Conclusions and Implications Some strategies low-income mothers use to manage food resources are associated with heavier BMIs. Nutrition educators, public health practitioners, and researchers need to collaboratively address the associations between these strategies, food insecurity, poor health, and unhealthy weight. PMID:25794991
American Visceral Leishmaniasis: Factors Associated with Lethality in the State of São Paulo, Brazil
Madalosso, Geraldine; Fortaleza, Carlos Magno; Ribeiro, Ana Freitas; Cruz, Lisete Lage; Nogueira, Péricles Alves; Lindoso, José Angelo Lauletta
2012-01-01
Objectives. To identify factors associated with death in visceral leishmaniasis (VL) cases. Patients and Methodology. We evaluated prognostic factors for death from VL in São Paulo state, Brazil, from 1999 to 2005. A prognostic study nested in a clinical cohort was carried out by data analysis of 376 medical files. A comparison between VL fatal cases and survivors was performed for clinical, laboratory, and biological features. Association between variables and death was assessed by univariate analysis, and the multiple logistic regression model was used to determine adjusted odds ratio for death, controlling confounding factors. Results. Data analysis identified 53 fatal cases out of 376 patients, between 1999 and 2005 in São Paulo state. Lethality was 14.1% (53/376), being higher in patients older than fifty years. The main causes of death were sepsis, bleeding, liver failure, and cardiotoxicity due to treatment. Variables significantly associated with death were severe anemia, bleeding, heart failure, jaundice, diarrhea, fever for more than sixty days, age older than fifty years, and antibiotic use. Conclusion. Educational health measures are needed for the general population and continuing education programs for health professionals working in the affected areas with the purpose of identifying and treating early cases, thus preventing the disease evolution towards death. PMID:23024661
McCurdy, Karen; Kisler, Tiffani; Gorman, Kathleen S; Metallinos-Katsaras, Elizabeth
2015-01-01
To examine how income-related challenges regarding food and health are associated with variation in self-reported maternal body weight among low-income mothers. Cross-sectional design. Two Northeastern cities. Seven day care centers and a Supplemental Nutrition Assistance Program outreach project. Sample of 166 mothers; 67% were overweight or obese, 55% were Hispanic, and 42% reported household food insecurity (HFI). Maternal self-reported height and weight to calculate body mass index (BMI). Independent variables were food program participation, supermarket use, 8-item food shopping practices scale, HFI, maternal depressive symptoms, and self-rated health. Hierarchical multiple regression analysis tested relationships between maternal BMI with the independent variables of interest, adjusting for demographic confounds. Shopping practices to stretch food dollars (P = .04), using community food assistance programs (P < .05), and HFI (P < .04) correlated with heavier maternal BMIs; higher self-rated health corresponded to lower BMIs (P = .004). Some strategies low-income mothers use to manage food resources are associated with heavier BMIs. Nutrition educators, public health practitioners, and researchers need to collaboratively address the associations between these strategies, food insecurity, poor health, and unhealthy weight. Copyright © 2015 Society for Nutrition Education and Behavior. Published by Elsevier Inc. All rights reserved.
Magalhães, Taís C A; Vieira, Sarah A; Priore, Silvia E; Ribeiro, Andréia Q; Lamounier, Joel A; Franceschini, Sylvia C C; Sant'Ana, Luciana F R
2012-01-01
To evaluate the effect of exclusive breastfeeding and consumption of other foods in the first six months of life in the nutritional status and body composition of children. A retrospective cohort study with 185 children aged from 4 to 7 years was monitored during the first months of life in a program of support to breastfeeding. We evaluated weight, height, waist circumference, and body composition by using DEXA. The nutritional status was assessed by the BMI/age index. The parameters of adiposity were classified by using as the cutoff point, the 85th percentile of the sample itself, according to gender and age. Confounding factors considered were variables related to maternal, pregnancy, birth, sociodemographic, health, lifestyle, and diet. Bivariate and multivariate analyses were performed, the latter by means of multiple logistic regression. The median exclusive breastfeeding was 3 months. Of the children, 42.7% received cow's milk and 35.7% received infant formula. Regarding nutritional status, 21.1% of the children showed changes. The variables of infant feeding were not independently associated with nutritional status and body composition of the children and there were no differences between the groups studied. Breastfeeding was not a protective factor to overweight and body fat in children.
Stenset, V; Hofoss, D; Johnsen, L; Skinningsrud, A; Berstad, A E; Negaard, A; Reinvang, I; Gjerstad, L; Fladby, T
2008-12-01
To identify possible associations between white matter lesions (WML) and cognition in patients with memory complaints, stratified in groups with normal and low cerebrospinal fluid (CSF) Abeta42 values. 215 consecutive patients with subjective memory complaints were retrospectively included. Patients were stratified into two groups with normal (n = 127) or low (n = 88) CSF Abeta42 levels (cut-off is 450 ng/l). Cognitive scores from the Mini-Mental State Examination (MMSE) and the Neurobehavioral Cognitive Status Examination (Cognistat) were used as continuous dependent variables in linear regression. WML load was used as a continuous independent variable and was scored with a visual rating scale. The regression model was corrected for possible confounding factors. WML were significantly associated with MMSE and all Cognistat subscores except language (repetition and naming) and attention in patients with normal CSF Abeta42 levels. No significant associations were observed in patients with low CSF Abeta42. WML were associated with affection of multiple cognitive domains, including delayed recall and executive functions, in patients with normal CSF Abeta42 levels. The lack of such associations for patients with low CSF Abeta42 (i.e. with evidence for amyloid deposition), suggests that amyloid pathology may obscure cognitive effects of WML.
Cavalcante, Milady Cutrim Vieira; Lamy, Fernando; França, Ana Karina Teixeira da Cunha; Lamy, Zeni Carvalho
2017-05-01
Several factors can interfere in the mother-child relationship. Studies about different maternal characteristics and this relationship are scarce; they mainly evaluate women with psychopathology and use simultaneous regression models with adjustment for multiple confounders. This study aimed to assess factors associated with losses in the mother-child relationship through a cohort of 3,215 mothers of children between 15 and 36 months of age. Losses in the mother-child relationship, assessed by the Postpartum Bonding Questionnaire, was the outcome variable and the explanatory variables were demographic, socioeconomic, reproductive health and mental health of mothers as well as the conditions of the birth of children. It used multivariate regression analysis with a hierarchical approach in which the hierarchical blocks were structured according to the influence on the mother-child relationship. The prevalence of losses in the mother-child relationship was high (12.6%) and associated risk factors to lower maternal education (RR = 1.64), having unplanned pregnancy (RR = 1.42), consumption of alcoholic beverages during pregnancy (RR = 1.42) and maternal stress symptoms (RR = 1.88) and depression (RR = 2.00). Education and elements related to mental health were risks for damage in the mother-child relationship.
[Maternal posture and its influence on birthweight].
Takito, Monica Yuri; Benício, Maria Helena D'Aquino; Latorre, Maria do Rosário Dias de Oliveira
2005-06-01
To analyze the relationship between maternal posture/physical activity and inadequate birthweight. Prospective cohort study involving 152 pregnant women from a public low-risk antenatal care facility. Three interviews evaluating the frequency of physical activity were administered to each pregnant woman during gestation. Birthweight (inadequate when <3,000 g and adequate when > or =3,000 g) was the dependent variable and the frequency of physical activity the independent variable. Statistical analysis was performed using logistic univariate analysis and multiple regression controlling for schooling, smoking, living with spouse, and baseline nutritional status. The practice of walking for at least 50 minutes during the first period of pregnancy was identified as a protective factor against inadequate birthweight (adjusted OR=0.44; 95% CI: 0.20-0.98). Standing for 2.5 hours or longer during the second semester of pregnancy was associated with increased risk (adjusted OR=3.23; 95% CI: 1.30-7.99). Dose-response relationships were identified for washing clothing by hand and cooking (p-value for linear trend <0.01 and 0.05, respectively). After confounder control, only washing clothing during the second trimester of gestation remained statistically significant. Our results show the importance of medical orientation regarding posture and physical activity during antenatal care, aiming at the reduction of inadequate birthweight.
Snehalatha, C; Ramachandran, A
2009-04-01
The aim of this study was to investigate the prevalence of cardiovascular risk factors in normoglycaemic Asian-Indians and its association with urbanisation and plasma glucose. Data on 5,517 normoglycaemic individuals aged >/=20 years (1,521, 1,851 and 2,145 inhabitants of a city, a town and villages, respectively), from a 2006 diabetes survey, were used. The presence of single and multiple abnormalities, including obesity, abdominal obesity, hypertension and dyslipidaemia, was analysed. Risk associations for hypertension and dyslipidaemia were analysed using multivariate analysis after adjusting for confounding variables. Both the prevalence and clustering of cardiovascular risk variables were higher in urban areas than in villages, and were higher in the city than in the town. Nearly half (47%) of the normoglycaemic individuals from villages had at least one risk factor. Age, male sex, BMI, waist circumference, lack of physical activity and living in an urban environment were associated with hypertension and dyslipidaemia; postprandial glucose level showed no association with hypertension or dyslipidaemia after adjusting for these risk factors. Fasting plasma glucose level showed an independent association with dyslipidaemia. There is a high prevalence of cardiovascular risk factors even among normoglycaemic individuals, and this is highest in urban areas. Fasting plasma glucose level showed an association with dyslipidaemia but not hypertension.
Magalhães, Taís C. A.; Vieira, Sarah A.; Priore, Silvia E.; Ribeiro, Andréia Q.; Lamounier, Joel A.; Franceschini, Sylvia C. C.; Sant'Ana, Luciana F. R.
2012-01-01
Objective. To evaluate the effect of exclusive breastfeeding and consumption of other foods in the first six months of life in the nutritional status and body composition of children. Methods. A retrospective cohort study with 185 children aged from 4 to 7 years was monitored during the first months of life in a program of support to breastfeeding. We evaluated weight, height, waist circumference, and body composition by using DEXA. The nutritional status was assessed by the BMI/age index. The parameters of adiposity were classified by using as the cutoff point, the 85th percentile of the sample itself, according to gender and age. Confounding factors considered were variables related to maternal, pregnancy, birth, sociodemographic, health, lifestyle, and diet. Bivariate and multivariate analyses were performed, the latter by means of multiple logistic regression. Results. The median exclusive breastfeeding was 3 months. Of the children, 42.7% received cow's milk and 35.7% received infant formula. Regarding nutritional status, 21.1% of the children showed changes. The variables of infant feeding were not independently associated with nutritional status and body composition of the children and there were no differences between the groups studied. Conclusion. Breastfeeding was not a protective factor to overweight and body fat in children. PMID:23193378
Linear mixed-effects modeling approach to FMRI group analysis
Chen, Gang; Saad, Ziad S.; Britton, Jennifer C.; Pine, Daniel S.; Cox, Robert W.
2013-01-01
Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance–covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins. Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance–covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details. PMID:23376789
Social roles and alcohol consumption: a study of 10 industrialised countries.
Kuntsche, Sandra; Knibbe, Ronald A; Gmel, Gerhard
2009-04-01
The empirical evidence as regards the precise associations between alcohol use and social roles, and these associations across genders and cultures is heterogeneous. The literature tends to focus on two central but conflicting theories. The first - classic role theory - assumes that a higher number of social roles is associated with a more structured life and thus fewer opportunities to drink heavily. The second - the multiple burden hypothesis - posits that the increasing complexity of multiple social roles leads to higher stress levels, and thus to increased alcohol use. Survey data on 25-54-year olds in 10 western industrialised countries which participate in the Gender, Alcohol and Culture: An International Study (GenACIS) project were used to test whether holding the three main social roles - partnership, parenthood, and paid labour - had a more protective or a more detrimental association with problematic alcohol use than holding fewer roles. Age and education were included as possible confounders, while the outcome variables were risky single occasion drinking (RSOD) and heavy-volume drinking. For both men and women and in almost all countries, the study found that those who had all three roles were least likely to drink heavily or engage in RSOD, thus supporting the assumptions of classic role theory. It also found that the protective effect of multiple roles was more consistent for RSOD. There were a few countries where a two-role model gave a better fit. Results for Germany (RSOD), Switzerland, and the Unites States (heavy-volume drinking) indicate that the role of paid labour appears to be particularly relevant for risky alcohol use among women. Despite some variability in the association between paid labour and heavy drinking or RSOD among women, in almost all countries the greater the number of roles a person held, the lower their risk of this type of alcohol use was.
Linear mixed-effects modeling approach to FMRI group analysis.
Chen, Gang; Saad, Ziad S; Britton, Jennifer C; Pine, Daniel S; Cox, Robert W
2013-06-01
Conventional group analysis is usually performed with Student-type t-test, regression, or standard AN(C)OVA in which the variance-covariance matrix is presumed to have a simple structure. Some correction approaches are adopted when assumptions about the covariance structure is violated. However, as experiments are designed with different degrees of sophistication, these traditional methods can become cumbersome, or even be unable to handle the situation at hand. For example, most current FMRI software packages have difficulty analyzing the following scenarios at group level: (1) taking within-subject variability into account when there are effect estimates from multiple runs or sessions; (2) continuous explanatory variables (covariates) modeling in the presence of a within-subject (repeated measures) factor, multiple subject-grouping (between-subjects) factors, or the mixture of both; (3) subject-specific adjustments in covariate modeling; (4) group analysis with estimation of hemodynamic response (HDR) function by multiple basis functions; (5) various cases of missing data in longitudinal studies; and (6) group studies involving family members or twins. Here we present a linear mixed-effects modeling (LME) methodology that extends the conventional group analysis approach to analyze many complicated cases, including the six prototypes delineated above, whose analyses would be otherwise either difficult or unfeasible under traditional frameworks such as AN(C)OVA and general linear model (GLM). In addition, the strength of the LME framework lies in its flexibility to model and estimate the variance-covariance structures for both random effects and residuals. The intraclass correlation (ICC) values can be easily obtained with an LME model with crossed random effects, even at the presence of confounding fixed effects. The simulations of one prototypical scenario indicate that the LME modeling keeps a balance between the control for false positives and the sensitivity for activation detection. The importance of hypothesis formulation is also illustrated in the simulations. Comparisons with alternative group analysis approaches and the limitations of LME are discussed in details. Published by Elsevier Inc.
NASA Astrophysics Data System (ADS)
Yong, Chin-Khian
2013-09-01
A partially confounded factorial conjoint choice experiments design was used to examine the monetary value of the willingness to pay for E-book Reader's attributes. Conjoint analysis is an efficient, cost-effective, and most widely used quantitative method in marketing research to understand consumer preferences and value trade-off. Value can be interpreted by customer or consumer as the received of multiple benefits from a price that was paid. The monetary value of willingness to pay for battery life, internal memory, external memory, screen size, text to Speech, touch screen, and converting handwriting to digital text of E-book reader were estimated in this study. Due to the significant interaction effect of the attributes with the price, the monetary values for the seven attributes were found to be different at different values of odds of purchasing versus not purchasing. The significant interactions effects were one of the main contribution of the partially confounded factorial conjoint choice experiment.
Kupek, Emil
2006-03-15
Structural equation modelling (SEM) has been increasingly used in medical statistics for solving a system of related regression equations. However, a great obstacle for its wider use has been its difficulty in handling categorical variables within the framework of generalised linear models. A large data set with a known structure among two related outcomes and three independent variables was generated to investigate the use of Yule's transformation of odds ratio (OR) into Q-metric by (OR-1)/(OR+1) to approximate Pearson's correlation coefficients between binary variables whose covariance structure can be further analysed by SEM. Percent of correctly classified events and non-events was compared with the classification obtained by logistic regression. The performance of SEM based on Q-metric was also checked on a small (N = 100) random sample of the data generated and on a real data set. SEM successfully recovered the generated model structure. SEM of real data suggested a significant influence of a latent confounding variable which would have not been detectable by standard logistic regression. SEM classification performance was broadly similar to that of the logistic regression. The analysis of binary data can be greatly enhanced by Yule's transformation of odds ratios into estimated correlation matrix that can be further analysed by SEM. The interpretation of results is aided by expressing them as odds ratios which are the most frequently used measure of effect in medical statistics.
Rothenberg, Stephen J; Rothenberg, Jesse C
2005-09-01
Statistical evaluation of the dose-response function in lead epidemiology is rarely attempted. Economic evaluation of health benefits of lead reduction usually assumes a linear dose-response function, regardless of the outcome measure used. We reanalyzed a previously published study, an international pooled data set combining data from seven prospective lead studies examining contemporaneous blood lead effect on IQ (intelligence quotient) of 7-year-old children (n = 1,333). We constructed alternative linear multiple regression models with linear blood lead terms (linear-linear dose response) and natural-log-transformed blood lead terms (log-linear dose response). We tested the two lead specifications for nonlinearity in the models, compared the two lead specifications for significantly better fit to the data, and examined the effects of possible residual confounding on the functional form of the dose-response relationship. We found that a log-linear lead-IQ relationship was a significantly better fit than was a linear-linear relationship for IQ (p = 0.009), with little evidence of residual confounding of included model variables. We substituted the log-linear lead-IQ effect in a previously published health benefits model and found that the economic savings due to U.S. population lead decrease between 1976 and 1999 (from 17.1 microg/dL to 2.0 microg/dL) was 2.2 times (319 billion dollars) that calculated using a linear-linear dose-response function (149 billion dollars). The Centers for Disease Control and Prevention action limit of 10 microg/dL for children fails to protect against most damage and economic cost attributable to lead exposure.
Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study
2009-01-01
OBJECTIVE—To examine associations of neonatal adiposity with maternal glucose levels and cord serum C-peptide in a multicenter multinational study, the Hyperglycemia and Adverse Pregnancy Outcome (HAPO) Study, thereby assessing the Pederson hypothesis linking maternal glycemia and fetal hyperinsulinemia to neonatal adiposity. RESEARCH DESIGN AND METHODS—Eligible pregnant women underwent a standard 75-g oral glucose tolerance test between 24 and 32 weeks gestation (as close to 28 weeks as possible). Neonatal anthropometrics and cord serum C-peptide were measured. Associations of maternal glucose and cord serum C-peptide with neonatal adiposity (sum of skin folds >90th percentile or percent body fat >90th percentile) were assessed using multiple logistic regression analyses, with adjustment for potential confounders, including maternal age, parity, BMI, mean arterial pressure, height, gestational age at delivery, and the baby's sex. RESULTS—Among 23,316 HAPO Study participants with glucose levels blinded to caregivers, cord serum C-peptide results were available for 19,885 babies and skin fold measurements for 19,389. For measures of neonatal adiposity, there were strong statistically significant gradients across increasing levels of maternal glucose and cord serum C-peptide, which persisted after adjustment for potential confounders. In fully adjusted continuous variable models, odds ratios ranged from 1.35 to 1.44 for the two measures of adiposity for fasting, 1-h, and 2-h plasma glucose higher by 1 SD. CONCLUSIONS—These findings confirm the link between maternal glucose and neonatal adiposity and suggest that the relationship is mediated by fetal insulin production and that the Pedersen hypothesis describes a basic biological relationship influencing fetal growth. PMID:19011170
Heijink, Richard; Koolman, Xander; Westert, Gert P
2013-06-01
Healthcare expenditures rise as a share of GDP in most countries, raising questions regarding the value of further spending increases. Against this backdrop, we assessed the value of healthcare spending growth in 14 western countries between 1996 and 2006. We estimated macro-level health production functions using avoidable mortality as outcome measure. Avoidable mortality comprises deaths from certain conditions "that should not occur in the presence of timely and effective healthcare". We investigated the relationship between total avoidable mortality and healthcare spending using descriptive analyses and multiple regression models, focussing on within-country variation and growth rates. We aimed to take into account the role of potential confounders and dynamic effects such as time lags. Additionally, we explored a method to estimate macro-level cost-effectiveness. We found an average yearly avoidable mortality decline of 2.6-5.3% across countries. Simultaneously, healthcare spending rose between 1.9 and 5.9% per year. Most countries with above-average spending growth demonstrated above-average reductions in avoidable mortality. The regression models showed a significant association between contemporaneous and lagged healthcare spending and avoidable mortality. The time-trend, representing an exogenous shift of the health production function, reduced the impact of healthcare spending. After controlling for this time-trend and other confounders, i.e. demographic and socioeconomic variables, a statistically significant relationship between healthcare spending and avoidable mortality remained. We tentatively conclude that macro-level healthcare spending increases provided value for money, at least for the disease groups, countries and years included in this study.
Jiang, Jing; Zhao, You-Jin; Hu, Xin-Yu; Du, Ming-Ying; Chen, Zi-Qi; Wu, Min; Li, Kai-Ming; Zhu, Hong-Yan; Kumar, Poornima; Gong, Qi-Yong
2017-01-01
Background Multiple meta-analyses of diffusion tensor imaging (DTI) studies have reported impaired white matter integrity in patients with major depressive disorder (MDD). However, owing to inclusion of medicated patients in these studies, it is difficult to conclude whether these reported alterations are associated with MDD or confounded by medication effects. A meta-analysis of DTI studies on medication-free (medication-naive and medication washout) patients with MDD would therefore be necessary to disentangle MDD-specific effects. Methods We analyzed white matter alterations between medication-free patients with MDD and healthy controls using anisotropic effect size–signed differential mapping (AES-SDM). We used DTI query software for fibre tracking. Results Both pooled and subgroup meta-analyses in medication washout patients showed robust fractional anisotropy (FA) reductions in white matter of the right cerebellum hemispheric lobule, body of the corpus callosum (CC) and bilateral superior longitudinal fasciculus III (SLF III), whereas FA reductions in the genu of the CC and right anterior thalamic projections were seen in only medication-naive patients. Fibre tracking showed that the main tracts with observed FA reductions included the right cerebellar tracts, body of the CC, bilateral SLF III and arcuate fascicle. Limitations The analytic techniques, patient characteristics and clinical variables of the included studies were heterogeneous; we could not exclude the effects of nondrug therapies owing to a lack of data. Conclusion By excluding the confounding influences of current medication status, findings from the present study may provide a better understanding of the underlying neuropathology of MDD. PMID:27780031
Emesis as a Screening Diagnostic for Low Dose Rate (LDR) Total Body Radiation Exposure.
Camarata, Andrew S; Switchenko, Jeffrey M; Demidenko, Eugene; Flood, Ann B; Swartz, Harold M; Ali, Arif N
2016-04-01
Current radiation disaster manuals list the time-to-emesis (TE) as the key triage indicator of radiation dose. The data used to support TE recommendations were derived primarily from nearly instantaneous, high dose-rate exposures as part of variable condition accident databases. To date, there has not been a systematic differentiation between triage dose estimates associated with high and low dose rate (LDR) exposures, even though it is likely that after a nuclear detonation or radiologic disaster, many surviving casualties would have received a significant portion of their total exposure from fallout (LDR exposure) rather than from the initial nuclear detonation or criticality event (high dose rate exposure). This commentary discusses the issues surrounding the use of emesis as a screening diagnostic for radiation dose after LDR exposure. As part of this discussion, previously published clinical data on emesis after LDR total body irradiation (TBI) is statistically re-analyzed as an illustration of the complexity of the issue and confounding factors. This previously published data includes 107 patients who underwent TBI up to 10.5 Gy in a single fraction delivered over several hours at 0.02 to 0.04 Gy min. Estimates based on these data for the sensitivity of emesis as a screening diagnostic for the low dose rate radiation exposure range from 57.1% to 76.6%, and the estimates for specificity range from 87.5% to 99.4%. Though the original data contain multiple confounding factors, the evidence regarding sensitivity suggests that emesis appears to be quite poor as a medical screening diagnostic for LDR exposures.
Menstrual cycle perturbation by organohalogens and elements in the Cree of James Bay, Canada.
Wainman, Bruce C; Kesner, James S; Martin, Ian D; Meadows, Juliana W; Krieg, Edward F; Nieboer, Evert; Tsuji, Leonard J
2016-04-01
Persistent organohalogens (POHs) and metals have been linked to alterations in menstrual cycle function and fertility in humans. The Cree First Nations people living near James Bay in Ontario and Quebec, Canada, have elevated levels of POHs, mercury and lead compared to other Canadians. The present study examines the interrelationships between selected POHs and elements on menstrual cycle function in these Cree women. Menstrual cycle characteristics were derived from structured daily diaries and endocrine measurements from daily urine samples collected during one cycle for 42 women age 19-42. We measured 31 POHs in blood plasma and 18 elements in whole blood, for 31 of the participants. POHs and elements detected in ≥ 70% of the participants were transformed by principal component (PC) analysis to reduce the contaminant exposure data to fewer, uncorrelated PCA variables. Multiple regression analysis revealed that, after adjusting for confounders, PC-3 values showed significant negative association with cycle length, after adjusting for confounders (p = 0.002). PC-3 accounted for 9.2% of the variance and shows positive loadings for cadmium, selenium, and PBDE congeners 47 and 153, and a negative loading for copper. Sensitivity analysis of the model to quantify likely effect sizes showed a range of menstrual cycle length from 25.3 to 28.3 days using the lower and upper 95% confidence limits of mean measured contaminant concentrations to predict cycle length. Our observations support the hypothesis that the menstrual cycle function of these women may be altered by exposure to POHs and elements from their environment. Copyright © 2015. Published by Elsevier Ltd.
Ferreira, Cesar Biselli; Bassi, Estevão; Lucena, Lucas; Carreta, Hernandez; Miranda, Leandro Costa; Tierno, Paulo Fernando Guimarães Mazorcchi; Amorim, Robson Luis; Zampieri, Fernando Godinho; Malbouisson, Luis Marcelo Sá
2015-01-01
Objective To assess the impact of intracranial pressure monitoring on the short-term outcomes of traumatic brain injury patients. Methods Retrospective observational study including 299 consecutive patients admitted due to traumatic brain injury from January 2011 through July 2012 at a Level 1 trauma center in São Paulo, Brazil. Patients were categorized in two groups according to the measurement of intracranial pressure (measured intracranial pressure and non-measured intracranial pressure groups). We applied a propensity-matched analysis to adjust for possible confounders (variables contained in the Crash Score prognostic algorithm). Results Global mortality at 14 days (16%) was equal to that observed in high-income countries in the CRASH Study and was better than expected based on the CRASH calculator score (20.6%), with a standardized mortality ratio of 0.77. A total of 28 patients received intracranial pressure monitoring (measured intracranial pressure group), of whom 26 were paired in a 1:1 fashion with patients from the non-measured intracranial pressure group. There was no improvement in the measured intracranial pressure group compared to the non-measured intracranial pressure group regarding hospital mortality, 14-day mortality, or combined hospital and chronic care facility mortality. Survival up to 14 days was also similar between groups. Conclusion Patients receiving intracranial pressure monitoring tend to have more severe traumatic brain injuries. However, after adjusting for multiple confounders using propensity scoring, no benefits in terms of survival were observed among intracranial pressure-monitored patients and those managed with a systematic clinical protocol. PMID:26761468
Ferreira, Cesar Biselli; Bassi, Estevão; Lucena, Lucas; Carreta, Hernandez; Miranda, Leandro Costa; Tierno, Paulo Fernando Guimarães Mazorcchi; Amorim, Robson Luis; Zampieri, Fernando Godinho; Malbouisson, Luis Marcelo Sá
2015-01-01
To assess the impact of intracranial pressure monitoring on the short-term outcomes of traumatic brain injury patients. Retrospective observational study including 299 consecutive patients admitted due to traumatic brain injury from January 2011 through July 2012 at a Level 1 trauma center in São Paulo, Brazil. Patients were categorized in two groups according to the measurement of intracranial pressure (measured intracranial pressure and non-measured intracranial pressure groups). We applied a propensity-matched analysis to adjust for possible confounders (variables contained in the Crash Score prognostic algorithm). Global mortality at 14 days (16%) was equal to that observed in high-income countries in the CRASH Study and was better than expected based on the CRASH calculator score (20.6%), with a standardized mortality ratio of 0.77. A total of 28 patients received intracranial pressure monitoring (measured intracranial pressure group), of whom 26 were paired in a 1:1 fashion with patients from the non-measured intracranial pressure group. There was no improvement in the measured intracranial pressure group compared to the non-measured intracranial pressure group regarding hospital mortality, 14-day mortality, or combined hospital and chronic care facility mortality. Survival up to 14 days was also similar between groups. Patients receiving intracranial pressure monitoring tend to have more severe traumatic brain injuries. However, after adjusting for multiple confounders using propensity scoring, no benefits in terms of survival were observed among intracranial pressure-monitored patients and those managed with a systematic clinical protocol.
Pfeiffer, Christine M.; Sternberg, Maya R.; Schleicher, Rosemary L.; Rybak, Michael E.
2016-01-01
Biochemical indicators of water-soluble vitamin (WSV) status have been measured in a nationally representative sample of the US population in NHANES 2003–2006. To examine whether demographic differentials in nutritional status were related to and confounded by certain variables, we assessed the association of sociodemographic (age, sex, race-ethnicity, education, income) and lifestyle variables (dietary supplement use, smoking, alcohol consumption, BMI, physical activity) with biomarkers of WSV status in adults (≥20 y): serum and RBC folate, serum pyridoxal-5′-phosphate (PLP), serum 4-pyridoxic acid, serum total cobalamin (B-12), plasma total homocysteine (tHcy), plasma methylmalonic acid (MMA), and serum ascorbic acid. Age (except for PLP) and smoking (except for MMA) were generally the strongest significant correlates of these biomarkers (|r| ≤0.43) and together with supplement use explained more of the variability as compared to the other covariates in bivariate analysis. In multiple regression models, sociodemographic and lifestyle variables together explained from 7% (B-12) to 29% (tHcy) of the biomarker variability. We observed significant associations for most biomarkers (≥6 out of 8) with age, sex, race-ethnicity, supplement use, smoking, and BMI; and for some biomarkers with PIR (5/8), education (1/8), alcohol consumption (4/8), and physical activity (5/8). We noted large estimated percent changes in biomarker concentrations between race-ethnic groups (from −24% to 20%), between supplement users and nonusers (from −12% to 104%), and between smokers and nonsmokers (from −28% to 8%). In summary, age, sex, and race-ethnic differentials in biomarker concentrations remained significant after adjusting for sociodemographic and lifestyle variables. Supplement use and smoking were important correlates of biomarkers of WSV status. PMID:23576641
Pfeiffer, Christine M; Sternberg, Maya R; Schleicher, Rosemary L; Rybak, Michael E
2013-06-01
Biochemical indicators of water-soluble vitamin (WSV) status were measured in a nationally representative sample of the U.S. population in NHANES 2003-2006. To examine whether demographic differentials in nutritional status were related to and confounded by certain variables, we assessed the association of sociodemographic (age, sex, race-ethnicity, education, income) and lifestyle (dietary supplement use, smoking, alcohol consumption, BMI, physical activity) variables with biomarkers of WSV status in adults (aged ≥ 20 y): serum and RBC folate, serum pyridoxal-5'-phosphate (PLP), serum 4-pyridoxic acid, serum total cobalamin (vitamin B-12), plasma total homocysteine (tHcy), plasma methylmalonic acid (MMA), and serum ascorbic acid. Age (except for PLP) and smoking (except for MMA) were generally the strongest significant correlates of these biomarkers (|r| ≤ 0.43) and together with supplement use explained more of the variability compared with the other covariates in bivariate analysis. In multiple regression models, sociodemographic and lifestyle variables together explained from 7 (vitamin B-12) to 29% (tHcy) of the biomarker variability. We observed significant associations for most biomarkers (≥ 6 of 8) with age, sex, race-ethnicity, supplement use, smoking, and BMI and for some biomarkers with PIR (5 of 8), education (1 of 8), alcohol consumption (4 of 8), and physical activity (5 of 8). We noted large estimated percentage changes in biomarker concentrations between race-ethnic groups (from -24 to 20%), between supplement users and nonusers (from -12 to 104%), and between smokers and nonsmokers (from -28 to 8%). In summary, age, sex, and race-ethnic differentials in biomarker concentrations remained significant after adjusting for sociodemographic and lifestyle variables. Supplement use and smoking were important correlates of biomarkers of WSV status.
On heart rate variability and autonomic activity in homeostasis and in systemic inflammation.
Scheff, Jeremy D; Griffel, Benjamin; Corbett, Siobhan A; Calvano, Steve E; Androulakis, Ioannis P
2014-06-01
Analysis of heart rate variability (HRV) is a promising diagnostic technique due to the noninvasive nature of the measurements involved and established correlations with disease severity, particularly in inflammation-linked disorders. However, the complexities underlying the interpretation of HRV complicate understanding the mechanisms that cause variability. Despite this, such interpretations are often found in literature. In this paper we explored mathematical modeling of the relationship between the autonomic nervous system and the heart, incorporating basic mechanisms such as perturbing mean values of oscillating autonomic activities and saturating signal transduction pathways to explore their impacts on HRV. We focused our analysis on human endotoxemia, a well-established, controlled experimental model of systemic inflammation that provokes changes in HRV representative of acute stress. By contrasting modeling results with published experimental data and analyses, we found that even a simple model linking the autonomic nervous system and the heart confound the interpretation of HRV changes in human endotoxemia. Multiple plausible alternative hypotheses, encoded in a model-based framework, equally reconciled experimental results. In total, our work illustrates how conventional assumptions about the relationships between autonomic activity and frequency-domain HRV metrics break down, even in a simple model. This underscores the need for further experimental work towards unraveling the underlying mechanisms of autonomic dysfunction and HRV changes in systemic inflammation. Understanding the extent of information encoded in HRV signals is critical in appropriately analyzing prior and future studies. Copyright © 2014 Elsevier Inc. All rights reserved.
On heart rate variability and autonomic activity in homeostasis and in systemic inflammation
Scheff, Jeremy D.; Griffel, Benjamin; Corbett, Siobhan A.; Calvano, Steve E.; Androulakis, Ioannis P.
2014-01-01
Analysis of heart rate variability (HRV) is a promising diagnostic technique due to the noninvasive nature of the measurements involved and established correlations with disease severity, particularly in inflammation-linked disorders. However, the complexities underlying the interpretation of HRV complicate understanding the mechanisms that cause variability. Despite this, such interpretations are often found in literature. In this paper we explored mathematical modeling of the relationship between the autonomic nervous system and the heart, incorporating basic mechanisms such as perturbing mean values of oscillating autonomic activities and saturating signal transduction pathways to explore their impacts on HRV. We focused our analysis on human endotoxemia, a well-established, controlled experimental model of systemic inflammation that provokes changes in HRV representative of acute stress. By contrasting modeling results with published experimental data and analyses, we found that even a simple model linking the autonomic nervous system and the heart confound the interpretation of HRV changes in human endotoxemia. Multiple plausible alternative hypotheses, encoded in a model-based framework, equally reconciled experimental results. In total, our work illustrates how conventional assumptions about the relationships between autonomic activity and frequency-domain HRV metrics break down, even in a simple model. This underscores the need for further experimental work towards unraveling the underlying mechanisms of autonomic dysfunction and HRV changes in systemic inflammation. Understanding the extent of information encoded in HRV signals is critical in appropriately analyzing prior and future studies. PMID:24680646
Scribner, Kim T.; Garner, G.W.; Amstrup, Steven C.; Cronin, M.A.; Dizon, Andrew E.; Chivers, Susan J.; Perrin, William F.
1997-01-01
A summary of existing population genetics literature is presented for polar bears (Ursus maritimus) and interpreted in the context of the species' life-history characteristics and regional heterogeneity in environmental regimes and movement patterns. Several nongenetic data sets including morphology, contaminant levels, geographic variation in reproductive characteristics, and the location and distribution of open-water foraging habitat suggest some degree of spatial structuring. Eleven populations are recognized by the IUCN Polar Bear Specialist Group. Few genetics studies exist for polar bears. Interpretation and generalizations of regional variation in intra- and interpopulation levels of genetic variability are confounded by the paucity of data from many regions and by the fact that no single informative genetic marker has been employed in multiple regions. Early allozyme studies revealed comparatively low levels of genetic variability and no compelling evidence of spatial structuring. Studies employing mitochondrial DNA (mtDNA) also found low levels of genetic variation, a lack of phylogenetic structure, and no significant evidence for spatial variation in haplotype frequency. In contrast, microsatellite variable number of tandem repeat (VNTR) loci have revealed significant heterogeneity in allele frequency among populations in the Canadian Arctic. These regions are characterized by archipelgic patterns of sea-ice movements. Further studies using highly polymorphic loci are needed in regions characterized by greater polar bear dependency on pelagic sea-ice movements and in regions for which no data currently exist (i.e., Laptev and Novaya Zemlya/Franz Josef).
Manipulating perceptual parameters in a continuous performance task.
Shalev, Nir; Humphreys, Glyn; Demeyere, Nele
2018-02-01
Sustained attention (SA) is among the most studied faculties of human cognition, and thought to be crucial for many aspects of behavior. Measuring SA often relies on performance on a continuous, low-demanding task. Such continuous performance tasks (CPTs) have many variations, and sustained attention is typically estimated based on variability in reaction times. While relying on reaction times may be useful in some cases, it can pose a challenge when working with clinical populations. To increase interpersonal variability in task parameters that do not rely on speed, researchers have increased demands for memory and response inhibition. These approaches, however, may be confounded when used to assess populations that suffer from multiple cognitive deficits. In the current study, we propose a new approach for increasing task variability by increasing the attentional demands. In order to do so, we created a new variation of a CPT - a masked version, where inattention is more likely to cause misidentifying a target. After establishing that masking indeed decreases target detection, we further investigated which task parameter may influence response biases. To do so, we contrasted two versions of the CPT with different target/distractor ratio. We then established how perceptual parameters can be controlled independently in a CPT. Following the experimental manipulations, we tested the MCCPT with aging controls and chronic stroke patients to assure the task can be used with target populations. The results confirm the MCCPT as a task providing high sensitivity without relying on reaction speed, and feasible for patients.
Mukabutera, Assumpta; Thomson, Dana R; Hedt-Gauthier, Bethany L; Atwood, Sidney; Basinga, Paulin; Nyirazinyoye, Laetitia; Savage, Kevin P; Habimana, Marcellin; Murray, Megan
2017-12-01
Public health interventions are often implemented at large scale, and their evaluation seems to be difficult because they are usually multiple and their pathways to effect are complex and subject to modification by contextual factors. We assessed whether controlling for rainfall-related variables altered estimates of the efficacy of a health programme in rural Rwanda and have a quantifiable effect on an intervention evaluation outcomes. We conducted a retrospective quasi-experimental study using previously collected cross-sectional data from the 2005 and 2010 Rwanda Demographic and Health Surveys (DHS), 2010 DHS oversampled data, monthly rainfall data collected from meteorological stations over the same period, and modelled output of long-term rainfall averages, soil moisture, and rain water run-off. Difference-in-difference models were used. Rainfall factors confounded the PIH intervention impact evaluation. When we adjusted our estimates of programme effect by controlling for a variety of rainfall variables, several effectiveness estimates changed by 10% or more. The analyses that did not adjust for rainfall-related variables underestimated the intervention effect on the prevalence of ARI by 14.3%, fever by 52.4% and stunting by 10.2%. Conversely, the unadjusted analysis overestimated the intervention's effect on diarrhoea by 56.5% and wasting by 80%. Rainfall-related patterns have a quantifiable effect on programme evaluation results and highlighted the importance and complexity of controlling for contextual factors in quasi-experimental design evaluations. © 2017 John Wiley & Sons Ltd.
Russo, Gianluca; Vita, Serena; Miglietta, Alessandro; Terrazzini, Nadia; Sannella, Alessandra; Vullo, Vincenzo
2016-06-01
Data on diseases' determinants and health status of asylum seekers (ASs) are limited. We performed a cross-sectional retrospective study in a large ASs centre in Italy. Data were collected during a 1-year period. Descriptive statistics were calculated. A χ(2) test was used to assess the association between socio-demographics characteristics of ASs and screening test results. A multiple logistic regression analysis was performed to identify diseases' predictors by using ICD-10 diagnoses classification as outcome variable, socio-demographic characteristics as independent variable and visits' number as confounding variable. Overall, data on 792 ASs (mean age 27 years, 80% males, 58% from Africa) were assessed, 43% underwent voluntary infectious diseases screening and 2843 diagnoses were recorded. The most frequent diagnoses were: respiratory diseases, symptoms/signs not elsewhere classified, digestive diseases and infectious diseases. Gender was the most frequent predictor of ICD-10 diagnoses, while African origin, civil status and education were, respectively, predictive of cardiovascular and infectious diseases, genitourinary diseases and pregnancy-related disorders. Higher mean age was associated with syphilis, HIV and HCV infection and African origin with HIV infection. Communicable diseases were not prevalent in the ASs population we analysed. A stronger cultural mediation support is needed to facilitate prevention, access and continuity of care for ASs. © The Author 2015. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Noise, stress, and annoyance in a pediatric intensive care unit.
Morrison, Wynne E; Haas, Ellen C; Shaffner, Donald H; Garrett, Elizabeth S; Fackler, James C
2003-01-01
To measure and describe hospital noise and determine whether noise can be correlated with nursing stress measured by questionnaire, salivary amylase, and heart rate. Cohort observational study. Tertiary care center pediatric intensive care unit. Registered nurses working in the unit. None. Eleven nurse volunteers were recruited. An audiogram, questionnaire data, salivary amylase, and heart rate were collected in a quiet room. Each nurse was observed for a 3-hr period during patient care. Heart rate and sound level were recorded continuously; saliva samples and stress/annoyance ratings were collected every 30 mins. Variables assessed as potential confounders were years of nursing experience, caffeine intake, patients' Pediatric Risk of Mortality Score, shift assignment, and room assignment. Data were analyzed by random effects multiple linear regression using Stata 6.0. The average daytime sound level was 61 dB(A), nighttime 59 dB(A). Higher average sound levels significantly predicted higher heart rates (p =.014). Other significant predictors of tachycardia were higher caffeine intake, less nursing experience, and daytime shift. Ninety percent of the variability in heart rate was explained by the regression equation. Amylase measurements showed a large variability and were not significantly affected by noise levels. Higher average sound levels were also predictive of greater subjective stress (p =.021) and annoyance (p =.016). In this small study, noise was shown to correlate with several measures of stress including tachycardia and annoyance ratings. Further studies of interventions to reduce noise are essential.
Emsley, Richard; Dunn, Graham; White, Ian R
2010-06-01
Complex intervention trials should be able to answer both pragmatic and explanatory questions in order to test the theories motivating the intervention and help understand the underlying nature of the clinical problem being tested. Key to this is the estimation of direct effects of treatment and indirect effects acting through intermediate variables which are measured post-randomisation. Using psychological treatment trials as an example of complex interventions, we review statistical methods which crucially evaluate both direct and indirect effects in the presence of hidden confounding between mediator and outcome. We review the historical literature on mediation and moderation of treatment effects. We introduce two methods from within the existing causal inference literature, principal stratification and structural mean models, and demonstrate how these can be applied in a mediation context before discussing approaches and assumptions necessary for attaining identifiability of key parameters of the basic causal model. Assuming that there is modification by baseline covariates of the effect of treatment (i.e. randomisation) on the mediator (i.e. covariate by treatment interactions), but no direct effect on the outcome of these treatment by covariate interactions leads to the use of instrumental variable methods. We describe how moderation can occur through post-randomisation variables, and extend the principal stratification approach to multiple group methods with explanatory models nested within the principal strata. We illustrate the new methodology with motivating examples of randomised trials from the mental health literature.
Tan, Chuen Seng; Støer, Nathalie C; Chen, Ying; Andersson, Marielle; Ning, Yilin; Wee, Hwee-Lin; Khoo, Eric Yin Hao; Tai, E-Shyong; Kao, Shih Ling; Reilly, Marie
2017-01-01
The control of confounding is an area of extensive epidemiological research, especially in the field of causal inference for observational studies. Matched cohort and case-control study designs are commonly implemented to control for confounding effects without specifying the functional form of the relationship between the outcome and confounders. This paper extends the commonly used regression models in matched designs for binary and survival outcomes (i.e. conditional logistic and stratified Cox proportional hazards) to studies of continuous outcomes through a novel interpretation and application of logit-based regression models from the econometrics and marketing research literature. We compare the performance of the maximum likelihood estimators using simulated data and propose a heuristic argument for obtaining the residuals for model diagnostics. We illustrate our proposed approach with two real data applications. Our simulation studies demonstrate that our stratification approach is robust to model misspecification and that the distribution of the estimated residuals provides a useful diagnostic when the strata are of moderate size. In our applications to real data, we demonstrate that parity and menopausal status are associated with percent mammographic density, and that the mean level and variability of inpatient blood glucose readings vary between medical and surgical wards within a national tertiary hospital. Our work highlights how the same class of regression models, available in most statistical software, can be used to adjust for confounding in the study of binary, time-to-event and continuous outcomes.
Farm-scale variation of soil quality indices and association with edaphic properties
USDA-ARS?s Scientific Manuscript database
Soil organisms are indicators of dynamic soil quality because their community structure and population density are sensitive to management changes. However, edaphic properties can also affect soil organisms and high spatial variability can confound their utility for soil evaluation. In the present...
Microcounseling Skill Discrimination Scale: A Methodological Note
ERIC Educational Resources Information Center
Stokes, Joseph; Romer, Daniel
1977-01-01
Absolute ratings on the Microcounseling Skill Discrimination Scale (MSDS) confound the individual's use of the rating scale and actual ability to discriminate effective and ineffective counselor behaviors. This note suggests methods of scoring the MSDS that will eliminate variability attributable to response language and improve the validity of…
D.A. Marquis
1991-01-01
Many studies have shown that initial tree diameter is closely correlated with subsequent tree growth. But initial tree diameter is actually a confounded variable, incorporating both competitive position (crown class) and age.
Hierarchical analysis of forest bird species-environment relationships in the Oregon Coast Range
Samuel A. Cushman; Kevin McGarigal
2004-01-01
Species in biological communities respond to environmental variation simultaneously across a range of organizational levels. Accordingly, it is important to quantify the effects of environmental factors at multiple levels on species distribution and abundance. Hierarchical methods that explicitly separate the independent and confounded influences of environmental...
Does vagotomy protect against multiple sclerosis?
Sundbøll, Jens; Horváth-Puhó, Erzsébet; Adelborg, Kasper; Svensson, Elisabeth
2017-07-01
To examine the association between vagotomy and multiple sclerosis. We conducted a matched cohort study of all patients who underwent truncal or super-selective vagotomy and a comparison cohort, by linking Danish population-based medical registries (1977-1995). Hazard ratios (HRs) for multiple sclerosis, adjusting for potential confounders were computed by means of Cox regression analysis. Median age of multiple sclerosis onset corresponded to late onset multiple sclerosis. No association with multiple sclerosis was observed for truncal vagotomy (0-37 year adjusted HR=0.91, 95% confidence interval [CI]: 0.48-1.74) or super-selective vagotomy (0-37 year adjusted HR=1.28, 95% CI: 0.79-2.09) compared with the general population. We found no association between vagotomy and later risk of late onset multiple sclerosis. Copyright © 2017 Elsevier B.V. All rights reserved.
Antidepressants and testicular cancer: cause versus association.
Andrade, Chittaranjan
2014-03-01
A data mining study that examined associations between 105 drugs and 55 cancer sites found significant associations between 2 selective serotonin reuptake inhibitors (fluoxetine and paroxetine) and testicular cancer. The study suggested several reasons why these associations merited further investigation. A later study tested specific relationships between 12 antidepressant drugs and testicular cancer and subtypes thereof; whereas significant relationships were again found, these disappeared after adjusting for confounding variables. These 2 studies are educative because they illustrate how false-positive results can easily arise in exploratory research and how confounding may be responsible for statistically significant relationships in study designs that are not randomized controlled trials. © Copyright 2014 Physicians Postgraduate Press, Inc.
Inoue, Tatsuro; Misu, Syogo; Tanaka, Toshiaki; Sakamoto, Hiroki; Iwata, Kentaro; Chuman, Yuki; Ono, Rei
2017-10-01
Malnutrition is common in patients with hip fractures, and elderly patients with hip fractures lose functional independence and often fail to recover previous functional status. The aim of this study was to determine whether pre-fracture nutritional status predicts functional status of patients with hip fracture at discharge from acute hospitals. In the present multicenter prospective cohort study, pre-fracture nutritional status was assessed using the Mini Nutritional Assessment Short-Form (MNA-SF). At discharge from acute hospitals, functional status was evaluated using a functional independent measurement instrument (FIM). Subsequently, multiple regression analyses were performed using FIM as the dependent variable and MNA-SF as the independent variable. Among the 204 patients analyzed in the present study, the mean length of hospital stay was 26.2 ± 12.6 days, and according to MNA-SF assessments, 51 (25.0%) patients were malnourished, 98 (48.0%) were at risk of malnutrition, and 55 (27.0%) were well-nourished before fracture. At discharge, FIM scores were higher in patients who were well-nourished than in those who were malnourished or were at risk of malnutrition (p < 0.01). After adjustment for confounding factors, multiple regression analyses showed that MNA-SF was a significant independent predictor for FIM at discharge (well-nourished vs. malnourished, β = -0.86, p < 0.01). Pre-fracture nutritional status was a significant independent predictor for functional status at discharge during the acute phase, warranting early assessment of nutritional status and early intervention for successful postoperative rehabilitation. Copyright © 2016 Elsevier Ltd and European Society for Clinical Nutrition and Metabolism. All rights reserved.
Mengistie, Zemenu; Woldeamanuel, Yimtubezinash; Asrat, Daniel; Adera, Addis
2014-11-20
Bacterial vaginosis is one of the most common genital tract infections among reproductive age group. The prevalence of bacterial vaginosis varies from country to country even in the same country it varies among populations of interest. Different social and sexual factors can contribute to the development of bacterial vaginosis. The aim of this study was to determine the prevalence of bacterial vaginosis and to identify the possible risk factors associated among pregnant women attending antenatal care in Tikur Anbessa University Hospital, Addis Ababa, Ethiopia. Randomly selected 57 symptomatic and 195 asymptomatic pregnant women aged between 18 and 40 years visiting obstetric and gynecological clinic from November 2011 to April 2012 screenedusing Gram stain Nugent scoring system. Statistical analysis like univariate analysis to calculate frequencies and proportions, bivariate analysis to see association of selected exposure variables with the outcome variable, and multivariate analysis to check the association of possible factors with bacterial vaginosis by adjusting potential confounding factors was calculated using SPSS (Version 16.0). The prevalence of bacterial vaginosis is 19.4% using Gram stain Nugent scoring system. In addition, prevalence of bacterial vaginosis is 31.6% and 15.9% among symptomatic and asymptomatic pregnant women respectively. A high percentage of bacterial vaginosis positive pregnant women were asymptomatic (63.3%). 36.7% bacterial vaginosis positive pregnant women reported abnormal vaginal discharge with or without unpleasant smell. Multiple lifetime sexual partner (OR: 8.6; 95% CI: 2.5, 29) and previous history of spontaneous abortion (OR: 5.9; 95% CI: 1.5, 23) had remained significantly associated with prevalence of bacterial vaginosis. The prevalence of bacterial vaginosis is higher among asymptomatic pregnant women and associated with the factors previous history of multiple lifetime sexual partner and spontaneous abortion.
Identifying Priorities for Mental Health Interventions in War-Affected Youth: A Longitudinal Study.
Betancourt, Theresa S; Gilman, Stephen E; Brennan, Robert T; Zahn, Ista; VanderWeele, Tyler J
2015-08-01
War-affected youth often suffer from multiple co-occurring mental health problems. These youth often live in low-resource settings where it may be infeasible to provide mental health services that simultaneously address all of these co-occurring mental health issues. It is therefore important to identify the areas where targeted interventions would do the most good. This analysis uses observational data from 3 waves of a longitudinal study on mental health in a sample of 529 war-affected youth (24.2% female; ages 10-17 at T1, 2002) in Sierra Leone. We regressed 4 mental health outcomes at T3 (2008) on internalizing (depression/anxiety) and externalizing (hostility/aggression) problems and prosocial attitudes/behaviors and community variables at T2 (2004) controlling for demographics, war exposures, and previous mental health scores at T1, allowing us to assess the relative impact of potential mental health intervention targets in shaping mental health outcomes over time. Controlling for baseline covariates at T1 and all other exposures/potential intervention targets at T2, we observed a significant association between internalizing problems at T2 and 3 of the 4 outcomes at T3: internalizing (β = 0.27, 95% confidence interval [CI]: 0.11-0.42), prosocial attitudes (β = -0.20, 95% CI: -0.33 to -0.07) and posttraumatic stress symptoms (β = 0.22, 95% CI: 0.02-0.43). No other potential intervention target had similar substantial effects. Reductions in internalizing may have multiple benefits for other mental health outcomes at a later point in time, even after controlling for confounding variables. Copyright © 2015 by the American Academy of Pediatrics.
Psychopathology Differences between Batterers and Nonbatterers: Psychosocial Modifiers.
ERIC Educational Resources Information Center
Hamberger, L. Kevin; Hastings, James E.
Research attention on men who batter their female partners has focused on psychological and personality characteristics of the men. However, occupational status, educational attainment, and abuse history may have been confounded with the personality variables studied. This study made within-group and between-group comparisons of batterers and…
Partial Identification of Treatment Effects: Applications to Generalizability
ERIC Educational Resources Information Center
Chan, Wendy
2016-01-01
Results from large-scale evaluation studies form the foundation of evidence-based policy. The randomized experiment is often considered the gold standard among study designs because the causal impact of a treatment or intervention can be assessed without threats of confounding from external variables. Policy-makers have become increasingly…
ERIC Educational Resources Information Center
Horwitz, Elaine K.
2000-01-01
Replies on behalf of the authors of a previous article to a critique of their study on foreign language anxiety. Argues that factors other than cognitive-linguistic deficits contribute to problems in foreign language achievement. (Author/VWL)
Multimedia Learning: Are We Still Asking the Wrong Questions?
ERIC Educational Resources Information Center
Rey, Gunter Daniel
2010-01-01
The article discusses problems that arise from comparing different kinds of presentation modes such as texts, pictures or animations with regard to learning outcome. These comparisons are confounded with or depend on other variables like quality of the instructional design, learning content, familiarity with the presentation mode as well as…
Pelle, Aline J; van den Broek, Krista C; Szabó, Balázs; Kupper, Nina
2010-11-05
Psychological factors, like Type D personality (i.e., the tendency to experience negative emotions and to inhibit emotional distress) have been linked to impaired health outcomes. Criticism on the role of psychological factors in cardiac disease has postulated that such constructs may be confounded by disease severity. Hence, we examined whether Type D personality is associated with brain natriuretic peptide (BNP), a sensitive marker of disease severity in chronic heart failure (CHF), in 202 consecutive CHF outpatients. No differences in logBNP levels were found between Type D and non-Type D patients (t(200) = -1.03, p = .30). After adjusting for demographic and clinical confounders, Type D personality remained unassociated with logBNP levels (β=.04, p = .55), whereas older age (β = .27, p<.001), being prescribed beta-blockers (β = .15, p = .02), lower left ventricular ejection fraction (β = -.38, p<.0001), and kidney dysfunction (β = .17, p = .01) were associated with higher logBNP. To conclude, Type D personality was not associated with BNP in CHF outpatients, whereas clinical variables were associated with BNP. These findings oppose the suggestion that Type D personality is confounded by indicators of disease severity. Copyright © 2009 Elsevier Ireland Ltd. All rights reserved.
Schempf, Ashley H; Kaufman, Jay S
2012-10-01
A common epidemiologic objective is to evaluate the contribution of residential context to individual-level disparities by race or socioeconomic position. We reviewed analytic strategies to account for the total (observed and unobserved factors) contribution of environmental context to health inequalities, including conventional fixed effects (FE) and hybrid FE implemented within a random effects (RE) or a marginal model. To illustrate results and limitations of the various analytic approaches of accounting for the total contextual component of health disparities, we used data on births nested within neighborhoods as an applied example of evaluating neighborhood confounding of racial disparities in gestational age at birth, including both a continuous and a binary outcome. Ordinary and RE models provided disparity estimates that can be substantially biased in the presence of neighborhood confounding. Both FE and hybrid FE models can account for cluster level confounding and provide disparity estimates unconfounded by neighborhood, with the latter having greater flexibility in allowing estimation of neighborhood-level effects and intercept/slope variability when implemented in a RE specification. Given the range of models that can be implemented in a hybrid approach and the frequent goal of accounting for contextual confounding, this approach should be used more often. Published by Elsevier Inc.
Riffle, Michael; Merrihew, Gennifer E; Jaschob, Daniel; Sharma, Vagisha; Davis, Trisha N; Noble, William S; MacCoss, Michael J
2015-11-01
Regulation of protein abundance is a critical aspect of cellular function, organism development, and aging. Alternative splicing may give rise to multiple possible proteoforms of gene products where the abundance of each proteoform is independently regulated. Understanding how the abundances of these distinct gene products change is essential to understanding the underlying mechanisms of many biological processes. Bottom-up proteomics mass spectrometry techniques may be used to estimate protein abundance indirectly by sequencing and quantifying peptides that are later mapped to proteins based on sequence. However, quantifying the abundance of distinct gene products is routinely confounded by peptides that map to multiple possible proteoforms. In this work, we describe a technique that may be used to help mitigate the effects of confounding ambiguous peptides and multiple proteoforms when quantifying proteins. We have applied this technique to visualize the distribution of distinct gene products for the whole proteome across 11 developmental stages of the model organism Caenorhabditis elegans. The result is a large multidimensional dataset for which web-based tools were developed for visualizing how translated gene products change during development and identifying possible proteoforms. The underlying instrument raw files and tandem mass spectra may also be downloaded. The data resource is freely available on the web at http://www.yeastrc.org/wormpes/ . Graphical Abstract ᅟ.
NASA Astrophysics Data System (ADS)
Riffle, Michael; Merrihew, Gennifer E.; Jaschob, Daniel; Sharma, Vagisha; Davis, Trisha N.; Noble, William S.; MacCoss, Michael J.
2015-11-01
Regulation of protein abundance is a critical aspect of cellular function, organism development, and aging. Alternative splicing may give rise to multiple possible proteoforms of gene products where the abundance of each proteoform is independently regulated. Understanding how the abundances of these distinct gene products change is essential to understanding the underlying mechanisms of many biological processes. Bottom-up proteomics mass spectrometry techniques may be used to estimate protein abundance indirectly by sequencing and quantifying peptides that are later mapped to proteins based on sequence. However, quantifying the abundance of distinct gene products is routinely confounded by peptides that map to multiple possible proteoforms. In this work, we describe a technique that may be used to help mitigate the effects of confounding ambiguous peptides and multiple proteoforms when quantifying proteins. We have applied this technique to visualize the distribution of distinct gene products for the whole proteome across 11 developmental stages of the model organism Caenorhabditis elegans. The result is a large multidimensional dataset for which web-based tools were developed for visualizing how translated gene products change during development and identifying possible proteoforms. The underlying instrument raw files and tandem mass spectra may also be downloaded. The data resource is freely available on the web at http://www.yeastrc.org/wormpes/.
Bellera, Carine; Proust-Lima, Cécile; Joseph, Lawrence; Richaud, Pierre; Taylor, Jeremy; Sandler, Howard; Hanley, James; Mathoulin-Pélissier, Simone
2018-04-01
Background Biomarker series can indicate disease progression and predict clinical endpoints. When a treatment is prescribed depending on the biomarker, confounding by indication might be introduced if the treatment modifies the marker profile and risk of failure. Objective Our aim was to highlight the flexibility of a two-stage model fitted within a Bayesian Markov Chain Monte Carlo framework. For this purpose, we monitored the prostate-specific antigens in prostate cancer patients treated with external beam radiation therapy. In the presence of rising prostate-specific antigens after external beam radiation therapy, salvage hormone therapy can be prescribed to reduce both the prostate-specific antigens concentration and the risk of clinical failure, an illustration of confounding by indication. We focused on the assessment of the prognostic value of hormone therapy and prostate-specific antigens trajectory on the risk of failure. Methods We used a two-stage model within a Bayesian framework to assess the role of the prostate-specific antigens profile on clinical failure while accounting for a secondary treatment prescribed by indication. We modeled prostate-specific antigens using a hierarchical piecewise linear trajectory with a random changepoint. Residual prostate-specific antigens variability was expressed as a function of prostate-specific antigens concentration. Covariates in the survival model included hormone therapy, baseline characteristics, and individual predictions of the prostate-specific antigens nadir and timing and prostate-specific antigens slopes before and after the nadir as provided by the longitudinal process. Results We showed positive associations between an increased prostate-specific antigens nadir, an earlier changepoint and a steeper post-nadir slope with an increased risk of failure. Importantly, we highlighted a significant benefit of hormone therapy, an effect that was not observed when the prostate-specific antigens trajectory was not accounted for in the survival model. Conclusion Our modeling strategy was particularly flexible and accounted for multiple complex features of longitudinal and survival data, including the presence of a random changepoint and a time-dependent covariate.
Idehen, Esther E; Koponen, Päivikki; Härkänen, Tommi; Kangasniemi, Mari; Pietilä, Anna-Maija; Korhonen, Tellervo
2018-05-04
Cervical cancer is currently ranked as the fourth commonly diagnosed cancer in women globally. A higher incidence has been reported in low- and-middle-income countries, and the disease poses significant public health challenges. Evidence suggests that this disease is preventable by means of regular screening using the Papanicolaou (Pap) test. However, limited knowledge exists about disparities in cervical screening participation among immigrants compared with non-immigrants, in countries with universal cervical screening programmes. We aimed to examine disparities in cervical screening participation among women of Russian, Somali, and Kurdish, origin in Finland, comparing them with the general Finnish population (Finns). We controlled for differences in several socio-demographic and health-related variables as potential confounders. We employed data from the Finnish Migrant Health and Well-being Study 2010-2012 and the National Health 2011 Survey. Data collection involved face-to-face interviews. Data on screening participation in the previous five years from women aged 29-60 were available from 537 immigrants (257 Russians, 113 Somalis, 167 Kurds) and from 436 Finns. For statistical analyses, we used multiple logistic regression. Age-adjusted screening participation rates were as follows: Russians 79% (95% CI 72.9-84.4), Somalis 41% (95% CI 31.4-50.1), and Kurds 64% (95% CI 57.2-70.8), compared with 94% (95% CI 91.4-95.9) among Finns. After additionally adjusting for socio-demographic and health-related confounders, all the immigrant groups showed a significantly lower likelihood of screening participation when compared with Finns. The Odds Ratios were as follows: Russians 0.32 (95% CI 0.18-0.58), Somalis 0.10 (95% CI 0.04-0.23), and Kurds 0.17 (95% CI 0.09-0.35). However, when additionally accounting for country of origin-confounder interactions, such differences were attenuated. Our results indicate disparities in screening participation among these immigrants and a lower likelihood of screening participation compared with the general Finnish population. To improve equity in cervical cancer screening participation, appropriate culturally tailored intervention programmes for each immigrant group might be beneficial.
What is the association between wealth and mental health?
Carter, K N; Blakely, T; Collings, S; Imlach Gunasekara, F; Richardson, K
2009-03-01
Socioeconomic inequalities in mental health have been shown in a number of populations. This study aims to investigate the association between asset wealth and psychological distress in New Zealand and whether it is independent of other socioeconomic measures and baseline health status. Data for this study were from the first three waves of the Survey of Families, Income and Employment (SoFIE) conducted in New Zealand (2002-2004/05) (n = 15 340). The Kessler-10 was used as a measure of psychological distress. The association of quintiles of wealth with psychological distress was investigated using logistic regression, controlling for confounders, socioeconomic variables and prior health status. The odds ratio (OR) of reporting high psychological distress were greater in the lowest wealth quintile compared with the highest (OR 3.06, 95% CI 2.68 to 3.50). Adjusting for age and sex did not alter the relationship; however, adjusting for income and area deprivation attenuated the OR to 1.73 (95% CI 1.48 to 2.04). Further controlling for baseline health status reduced the OR to 1.45 (95% CI 1.23 to 1.71), although the confidence interval still excluded the null. Inequalities in wealth are strongly associated with psychological distress, over and above other confounding demographic variables and baseline health status. Much, but not all, of that association is confounded by adult socioeconomic position. This suggests that policy measures to improve asset wealth, through savings and home ownership, may have positive health implications and help to reduce health inequalities.
Multisample adjusted U-statistics that account for confounding covariates.
Satten, Glen A; Kong, Maiying; Datta, Somnath
2018-06-19
Multisample U-statistics encompass a wide class of test statistics that allow the comparison of 2 or more distributions. U-statistics are especially powerful because they can be applied to both numeric and nonnumeric data, eg, ordinal and categorical data where a pairwise similarity or distance-like measure between categories is available. However, when comparing the distribution of a variable across 2 or more groups, observed differences may be due to confounding covariates. For example, in a case-control study, the distribution of exposure in cases may differ from that in controls entirely because of variables that are related to both exposure and case status and are distributed differently among case and control participants. We propose to use individually reweighted data (ie, using the stratification score for retrospective data or the propensity score for prospective data) to construct adjusted U-statistics that can test the equality of distributions across 2 (or more) groups in the presence of confounding covariates. Asymptotic normality of our adjusted U-statistics is established and a closed form expression of their asymptotic variance is presented. The utility of our approach is demonstrated through simulation studies, as well as in an analysis of data from a case-control study conducted among African-Americans, comparing whether the similarity in haplotypes (ie, sets of adjacent genetic loci inherited from the same parent) occurring in a case and a control participant differs from the similarity in haplotypes occurring in 2 control participants. Copyright © 2018 John Wiley & Sons, Ltd.
Nantes, Julia C; Proulx, Sébastien; Zhong, Jidan; Holmes, Scott A; Narayanan, Sridar; Brown, Robert A; Hoge, Richard D; Koski, Lisa
2017-08-15
Converging areas of research have implicated glutamate and γ-aminobutyric acid (GABA) as key players in neuronal signalling and other central functions. Further research is needed, however, to identify microstructural and behavioral links to regional variability in levels of these neurometabolites, particularly in the presence of demyelinating disease. Thus, we sought to investigate the extent to which regional glutamate and GABA levels are related to a neuroimaging marker of microstructural damage and to motor and cognitive performance. Twenty-one healthy volunteers and 47 people with multiple sclerosis (all right-handed) participated in this study. Motor and cognitive abilities were assessed with standard tests used in the study of multiple sclerosis. Proton magnetic resonance spectroscopy data were acquired from sensorimotor and parietal regions of the brains' left cerebral hemisphere using a MEGA-PRESS sequence. Our analysis protocol for the spectroscopy data was designed to account for confounding factors that could contaminate the measurement of neurometabolite levels due to disease, such as the macromolecule signal, partial volume effects, and relaxation effects. Glutamate levels in both regions of interest were lower in people with multiple sclerosis. In the sensorimotor (though not the parietal) region, GABA concentration was higher in the multiple sclerosis group compared to controls. Lower magnetization transfer ratio within grey and white matter regions from which spectroscopy data were acquired was linked to neurometabolite levels. When adjusting for age, normalized brain volume, MTR, total N-acetylaspartate level, and glutamate level, significant relationships were found between lower sensorimotor GABA level and worse performance on several tests, including one of upper limb motor function. This work highlights important methodological considerations relevant to analysis of spectroscopy data, particularly in the afflicted human brain. These findings support that regional neurotransmitter levels are linked to local microstructural integrity and specific behavioral abilities that can be affected in diseases such as multiple sclerosis. Copyright © 2017 Elsevier Inc. All rights reserved.
Roll-Tilt Perception Using a Somatosensory Bar Task
NASA Technical Reports Server (NTRS)
Black, F. O.; Wade, S. W.; Arshi, A.
1999-01-01
Visual estimates of roll-tilt perception during static roll-tilt are confounded by an offset due to the ocular counterroll that simultaneously occurs. An alternative, non-visual ('somatosensory') measure of roll-tilt perception was developed which is not contaminated by this offset. The aims of this study were to determine: 1) inter-subject variability of somatosensory settings across test session in normal subjects and patients with unilateral or bilateral vestibular loss and 2) intra-subject variability of settings across test session in normal subjects.
Effect of antenatal corticosteroids on fetal growth and gestational age at birth.
Murphy, Kellie E; Willan, Andrew R; Hannah, Mary E; Ohlsson, Arne; Kelly, Edmond N; Matthews, Stephen G; Saigal, Saroj; Asztalos, Elizabeth; Ross, Susan; Delisle, Marie-France; Amankwah, Kofi; Guselle, Patricia; Gafni, Amiram; Lee, Shoo K; Armson, B Anthony
2012-05-01
To estimate the effect of multiple courses of antenatal corticosteroids on neonatal size, controlling for gestational age at birth and other confounders, and to determine whether there was a dose-response relationship between number of courses of antenatal corticosteroids and neonatal size. This is a secondary analysis of the Multiple Courses of Antenatal Corticosteroids for Preterm Birth Study, a double-blind randomized controlled trial of single compared with multiple courses of antenatal corticosteroids in women at risk for preterm birth and in which fetuses administered multiple courses of antenatal corticosteroids weighed less, were shorter, and had smaller head circumferences at birth. All women (n=1,858) and children (n=2,304) enrolled in the Multiple Courses of Antenatal Corticosteroids for Preterm Birth Study were included in the current analysis. Multiple linear regression analyses were undertaken. Compared with placebo, neonates in the antenatal corticosteroids group were born earlier (estimated difference and confidence interval [CI]: -0.428 weeks, CI -0.10264 to -0.75336; P=.01). Controlling for gestational age at birth and confounding factors, multiple courses of antenatal corticosteroids were associated with a decrease in birth weight (-33.50 g, CI -66.27120 to -0.72880; P=.045), length (-0.339 cm, CI -0.6212 to -0.05676]; P=.019), and head circumference (-0.296 cm, -0.45672 to -0.13528; P<.001). For each additional course of antenatal corticosteroids, there was a trend toward an incremental decrease in birth weight, length, and head circumference. Fetuses exposed to multiple courses of antenatal corticosteroids were smaller at birth. The reduction in size was partially attributed to being born at an earlier gestational age but also was attributed to decreased fetal growth. Finally, a dose-response relationship exists between the number of corticosteroid courses and a decrease in fetal growth. The long-term effect of these findings is unknown. ClinicalTrials.gov, www.clinicaltrials.gov, NCT00187382. II.
Manheimer, Eric; van der Windt, Daniëlle; Cheng, Ke; Stafford, Kristen; Liu, Jianping; Tierney, Jayne; Lao, Lixing; Berman, Brian M.; Langenberg, Patricia; Bouter, Lex M.
2013-01-01
BACKGROUND Recent systematic reviews of adjuvant acupuncture for IVF have pooled heterogeneous trials, without examining variables that might explain the heterogeneity. The aims of our meta-analysis were to quantify the overall pooled effects of adjuvant acupuncture on IVF clinical pregnancy success rates, and evaluate whether study design-, treatment- and population-related factors influence effect estimates. METHODS We included randomized controlled trials that compared needle acupuncture administered within 1 day of embryo transfer, versus sham acupuncture or no adjuvant treatment. Our primary outcome was clinical pregnancy rates. We obtained from all investigators additional methodological details and outcome data not included in their original publications. We analysed sham-controlled and no adjuvant treatment-controlled trials separately, but since there were no large or significant differences between these two subsets, we pooled all trials for subgroup analyses. We prespecified 11 subgroup variables (5 clinical and 6 methodological) to investigate sources of heterogeneity, using single covariate meta-regressions. RESULTS Sixteen trials (4021 participants) were included in the meta-analyses. There was no statistically significant difference between acupuncture and controls when combining all trials [risk ratio (RR) 1.12, 95% confidence interval (CI), 0.96–1.31; I2 = 68%; 16 trials; 4021 participants], or when restricting to sham-controlled (RR 1.02, 0.83–1.26; I2 = 66%; 7 trials; 2044 participants) or no adjuvant treatment-controlled trials (RR 1.22, 0.97–1.52; I2 = 67%; 9 trials; 1977 participants). The type of control used did not significantly explain the statistical heterogeneity (interaction P = 0.27). Baseline pregnancy rate, measured as the observed rate of clinical pregnancy in the control group of each trial, was a statistically significant effect modifier (interaction P < 0.001), and this covariate explained most of the heterogeneity of the effects of adjuvant acupuncture across all trials (adjusted R2 = 93%; I2 residual = 9%). Trials with lower control group rates of clinical pregnancy showed larger effects of adjuvant acupuncture (RR 1.53, 1.28–1.84; 7 trials; 1732 participants) than trials with higher control group rates of clinical pregnancy (RR 0.90, 0.80–1.01; 9 trials; 2289 participants). The asymmetric funnel plot showed a tendency for the intervention effects to be more beneficial in smaller trials. CONCLUSIONS We found no pooled benefit of adjuvant acupuncture for IVF. The subgroup finding of a benefit in trials with lower, but not higher, baseline pregnancy rates (the only statistically significant subgroup finding in our earlier review) has been confirmed in this update, and was not explained by any confounding variables evaluated. However, this baseline pregnancy rate subgroup finding among published trials requires further confirmation and exploration in additional studies because of the multiple subgroup tests conducted, the risk of unidentified confounders, the multiple different factors that determine baseline rates, and the possibility of publication bias. PMID:23814102
Iqbal, M. Bilal; Robinson, Simon D.; Ding, Lillian; Fung, Anthony; Aymong, Eve; Chan, Albert W.; Hodge, Steven; Della Siega, Anthony; Nadra, Imad J.
2016-01-01
Background Cardiogenic shock complicating ST-elevation myocardial infarction (STEMI) is associated with significant morbidity and mortality. In the primary percutaneous coronary intervention (PPCI) era, randomized trials have not shown a survival benefit with intra-aortic balloon pump (IABP) therapy. This differs to observational data which show a detrimental effect, potentially reflecting bias and confounding. Without robust and valid risk adjustment, findings from non-randomized studies may remain biased. Methods We compared long-term mortality following IABP therapy in patients with cardiogenic shock undergoing PPCI during 2008–2013 from the British Columbia Cardiac Registry. We addressed measured and unmeasured confounding using propensity score and instrumental variable methods. Results A total of 12,105 patients with STEMI were treated with PPCI during the study period. Of these, 700 patients (5.8%) had cardiogenic shock. Of the patients with cardiogenic shock, 255 patients (36%) received IABP therapy. Multivariable analyses identified IABP therapy to be associated with increased mortality up to 3 years (HR = 1.67, 95% CI:1.20–2.67, p<0.001). This association was lost in propensity-matched analyses (HR = 1.23, 95% CI: 0.84–1.80, p = 0.288). When addressing measured and unmeasured confounders, instrumental variable analyses demonstrated that IABP therapy was not associated with mortality at 3 years (Δ = 16.7%, 95% CI: -12.7%, 46.1%, p = 0.281). Subgroup analyses demonstrated IABP was associated with increased mortality in non-diabetics; patients not undergoing multivessel intervention; patients without renal disease and patients not having received prior thrombolysis. Conclusions In this observational analysis of patients with STEMI and cardiogenic shock, when adjusting for confounding, IABP therapy had a neutral effect with no association with long-term mortality. These findings differ to previously reported observational studies, but are in keeping with randomized trial data. PMID:26870950
Pingault, Jean Baptiste; Côté, Sylvana M.; Petitclerc, Amélie; Vitaro, Frank; Tremblay, Richard E.
2015-01-01
Background Parental educational expectations have been associated with children’s educational attainment in a number of long-term longitudinal studies, but whether this relationship is causal has long been debated. The aims of this prospective study were twofold: 1) test whether low maternal educational expectations contributed to failure to graduate from high school; and 2) compare the results obtained using different strategies for accounting for confounding variables (i.e. multivariate regression and propensity score matching). Methodology/Principal Findings The study sample included 1,279 participants from the Quebec Longitudinal Study of Kindergarten Children. Maternal educational expectations were assessed when the participants were aged 12 years. High school graduation – measuring educational attainment – was determined through the Quebec Ministry of Education when the participants were aged 22–23 years. Findings show that when using the most common statistical approach (i.e. multivariate regressions to adjust for a restricted set of potential confounders) the contribution of low maternal educational expectations to failure to graduate from high school was statistically significant. However, when using propensity score matching, the contribution of maternal expectations was reduced and remained statistically significant only for males. Conclusions/Significance The results of this study are consistent with the possibility that the contribution of parental expectations to educational attainment is overestimated in the available literature. This may be explained by the use of a restricted range of potential confounding variables as well as the dearth of studies using appropriate statistical techniques and study designs in order to minimize confounding. Each of these techniques and designs, including propensity score matching, has its strengths and limitations: A more comprehensive understanding of the causal role of parental expectations will stem from a convergence of findings from studies using different techniques and designs. PMID:25803867
Pfeiffer, R M; Riedl, R
2015-08-15
We assess the asymptotic bias of estimates of exposure effects conditional on covariates when summary scores of confounders, instead of the confounders themselves, are used to analyze observational data. First, we study regression models for cohort data that are adjusted for summary scores. Second, we derive the asymptotic bias for case-control studies when cases and controls are matched on a summary score, and then analyzed either using conditional logistic regression or by unconditional logistic regression adjusted for the summary score. Two scores, the propensity score (PS) and the disease risk score (DRS) are studied in detail. For cohort analysis, when regression models are adjusted for the PS, the estimated conditional treatment effect is unbiased only for linear models, or at the null for non-linear models. Adjustment of cohort data for DRS yields unbiased estimates only for linear regression; all other estimates of exposure effects are biased. Matching cases and controls on DRS and analyzing them using conditional logistic regression yields unbiased estimates of exposure effect, whereas adjusting for the DRS in unconditional logistic regression yields biased estimates, even under the null hypothesis of no association. Matching cases and controls on the PS yield unbiased estimates only under the null for both conditional and unconditional logistic regression, adjusted for the PS. We study the bias for various confounding scenarios and compare our asymptotic results with those from simulations with limited sample sizes. To create realistic correlations among multiple confounders, we also based simulations on a real dataset. Copyright © 2015 John Wiley & Sons, Ltd.
Fatigue in Patients with Multiple Sclerosis: Is It Related to Pro- and Anti-Inflammatory Cytokines?
Malekzadeh, Arjan; Van de Geer-Peeters, Wietske; De Groot, Vincent; Elisabeth Teunissen, Charlotte; Beckerman, Heleen; TREFAMS-ACE Study Group
2015-01-01
Objective. To investigate the pathophysiological role of pro- and anti-inflammatory cytokines in primary multiple sclerosis-related fatigue. Methods. Fatigued and non-fatigued patients with multiple sclerosis (MS) were recruited and their cytokine profiles compared. Patients with secondary fatigue were excluded. Fatigue was assessed with the self-reported Checklist Individual Strength (CIS20r), subscale fatigue. A CIS20r fatigue cut-off score of 35 was applied to differentiate between non-fatigued (CIS20r fatigue ≤34) and fatigued (CIS20r fatigue ≥35) patients with MS. Blood was collected to determine the serum concentrations of pro-inflammatory cytokines (IL-1β, IL-2, IL-6, IL-8, IL-12p70, IL-17, TNFα, and IFN-γ) and anti-inflammatory cytokines (IL-4, IL-5, IL-10, and IL-13). We controlled for the confounding effect of age, gender, duration of MS, disease severity, type of MS, and use of immunomodulatory drugs. Results. Similar cytokine levels were observed between MS patients with (n = 21) and without fatigue (n = 14). Adjusted multiple regression analyses showed a single significant positive relationship, that of IL-6 with CIS20r fatigue score. The explained variance of the IL-6 model was 21.1%, once adjusted for the confounding effect of age. Conclusion. The pro-inflammatory cytokine interleukin-6 (IL-6) may play a role in the pathophysiology of primary fatigue in patients with MS. Trial Registrations. ISRCTN69520623, ISRCTN58583714, and ISRCTN82353628. PMID:25722532
Factors associated with the choice of general medicine as a career among Japanese medical students.
Kawamoto, Ryuichi; Ninomiya, Daisuke; Kasai, Yoshihisa; Kusunoki, Tomo; Ohtsuka, Nobuyuki; Kumagi, Teru; Abe, Masanori
2016-01-01
Background In Japan, there is a shortage of young physicians in various specialties; the present situation of general medicine or family medicine (GM/FM) in particular is risky. The factors influencing the career choice of Japanese medical students are poorly understood. This study aims to identify factors related to choosing GM/FM as a career. Methods The study was designed as a cross-sectional survey. Students at one medical school in Japan filled out a questionnaire. Students were asked to state their intended medical specialty, and they rated the importance of specific individual and occupational aspects using a 4-point likert scale. Factor analysis was performed on the variables. Reliability of the factor scores was estimated using Cronbach's alpha coefficients; biserial correlations between the factors and career choices were calculated. Furthermore, multiple linear regression analysis was performed using career choice (GM/FM vs. others) as the criterion variable and the factors plus demographic characteristics as confounding variables. Results Factor analysis produced six factors that explained future career plans. Medical students in this study had a positive and realistic idea about GM/FM, but only 18.8% of them chose GM/FM first as a career. The significant variables associated with choosing GM/FM first as a career were: 'Admission from hometown' (β=0.189, P=0.001), 'Student preparing for the entrance exam' (β=0.172; P=0.001), 'Intent for rural practice' (β=0.123, P=0.016), and 'Work-life balance' (β=0.126, P=0.013). While significant variables that were negatively associated with choosing GM/FM were 'Presence of medical relatives' (β=-0.107, P=0.037) and 'Scientific orientation' (β=-0.125, P=0.013). Conclusions Strategies have been suggested, such as recruiting medical students with significant variables that were associated with choosing GM/FM first as a career. By engaging students early in their choice of career, we may be able to increase enthusiasm for this specialty.
Factors associated with the choice of general medicine as a career among Japanese medical students.
Kawamoto, Ryuichi; Ninomiya, Daisuke; Kasai, Yoshihisa; Kusunoki, Tomo; Ohtsuka, Nobuyuki; Kumagi, Teru; Abe, Masanori
2016-01-01
In Japan, there is a shortage of young physicians in various specialties; the present situation of general medicine or family medicine (GM/FM) in particular is risky. The factors influencing the career choice of Japanese medical students are poorly understood. This study aims to identify factors related to choosing GM/FM as a career. The study was designed as a cross-sectional survey. Students at one medical school in Japan filled out a questionnaire. Students were asked to state their intended medical specialty, and they rated the importance of specific individual and occupational aspects using a 4-point likert scale. Factor analysis was performed on the variables. Reliability of the factor scores was estimated using Cronbach's alpha coefficients; biserial correlations between the factors and career choices were calculated. Furthermore, multiple linear regression analysis was performed using career choice (GM/FM vs. others) as the criterion variable and the factors plus demographic characteristics as confounding variables. Factor analysis produced six factors that explained future career plans. Medical students in this study had a positive and realistic idea about GM/FM, but only 18.8% of them chose GM/FM first as a career. The significant variables associated with choosing GM/FM first as a career were: 'Admission from hometown' (β=0.189, P=0.001), 'Student preparing for the entrance exam' (β=0.172; P=0.001), 'Intent for rural practice' (β=0.123, P=0.016), and 'Work-life balance' (β=0.126, P=0.013). While significant variables that were negatively associated with choosing GM/FM were 'Presence of medical relatives' (β=-0.107, P=0.037) and 'Scientific orientation' (β=-0.125, P=0.013). Strategies have been suggested, such as recruiting medical students with significant variables that were associated with choosing GM/FM first as a career. By engaging students early in their choice of career, we may be able to increase enthusiasm for this specialty.
Supervised normalization of microarrays
Mecham, Brigham H.; Nelson, Peter S.; Storey, John D.
2010-01-01
Motivation: A major challenge in utilizing microarray technologies to measure nucleic acid abundances is ‘normalization’, the goal of which is to separate biologically meaningful signal from other confounding sources of signal, often due to unavoidable technical factors. It is intuitively clear that true biological signal and confounding factors need to be simultaneously considered when performing normalization. However, the most popular normalization approaches do not utilize what is known about the study, both in terms of the biological variables of interest and the known technical factors in the study, such as batch or array processing date. Results: We show here that failing to include all study-specific biological and technical variables when performing normalization leads to biased downstream analyses. We propose a general normalization framework that fits a study-specific model employing every known variable that is relevant to the expression study. The proposed method is generally applicable to the full range of existing probe designs, as well as to both single-channel and dual-channel arrays. We show through real and simulated examples that the method has favorable operating characteristics in comparison to some of the most highly used normalization methods. Availability: An R package called snm implementing the methodology will be made available from Bioconductor (http://bioconductor.org). Contact: jstorey@princeton.edu Supplementary information: Supplementary data are available at Bioinformatics online. PMID:20363728
Concepts and considerations of tooth wear: Part II--The mechanical component.
Raigrodski, Ariel J; Dogan, Sami
2008-05-01
Identification of etiological factors is essential for successful management of tooth wear. In many cases, the diagnosis may be complicated because of the multiple etiologic factors which may confound the clinical appearance of tooth wear. Comprehensively addressing all of these factors is paramount to the long-term success of patient management.
An Evidence-Based Approach to an Adolescent with Emotional and Behavioral Dysregulation
ERIC Educational Resources Information Center
McClellan, Jon M.; Hamilton, John D.
2006-01-01
Children and adolescents in community mental health settings often present with multiple overlapping syndromes, as well as confounding issues such as family turmoil, abuse and neglect, and involvement with social welfare and juvenile justice systems. Most interventions proven to have efficacy in randomized, controlled trials have nevertheless not…
Learning and Motivational Benefits of Acronym Use in Introductory Psychology
ERIC Educational Resources Information Center
Stalder, Daniel R.
2005-01-01
Study 1 assessed students' use and perceptions of acronyms at 3 different exam times in 2 sections of Introduction to Psychology. Acronym use consistently predicted higher performance on acronym related exam items, and I partially discounted 2 possible confounds. Students rated acronyms as helpful in multiple ways, including increasing motivation…
An empirical study of rape in the context of multiple murder.
DeLisi, Matt
2014-03-01
In recent years, multiple homicide offending has received increased research attention from criminologists; however, there is mixed evidence about the role of rape toward the perpetration of multiple murder. Drawing on criminal career data from a nonprobability sample of 618 confined male homicide offenders selected from eight U.S. states, the current study examines the role of rape as a predictor of multiple homicide offending. Bivariate analyses indicated a significant association between rape and murder charges. Multivariate path regression models indicated that rape had a significant and robust association with multiple murder. This relationship withstood the confounding effects of kidnapping, prior prison confinement, and prior murder, rape, and kidnapping. These results provide evidence that rape potentially serves as a gateway to multiple murder for some serious offenders. Suggestions for future research are proffered.
Pickersgill, C H; Marr, C M; Reid, S W
2001-01-01
A quantitative investigation of the variation that can occur during the course of ultrasonography of the equine superficial digital flexor tendons (SDFT) was undertaken. The aim of this investigation was to use an objective measure, namely the measurement of CSA, to quantify the variability occurring during the course of the ultrasonographic assessment of the equine SDFT. The effects of 3 variables on the CSA measurements were determined. 1) Image acquisition operator (IAc): two different operators undertaking the ultrasonographic examination; 2) image analysis operator (IAn): two different operators undertaking the calculation of CSA values from previously stored images; and 3) analytical equipment (used during CSA measurement) (IEq): the use of 2 different sets of equipment during calculation of CSA values. Tendon cross-sectional area (CSA) measurements were used as the comparative variable of 3 potential sources: interoperator, during image acquisition; interoperator, during CSA measurement; and intraoperator, when using different analytical equipment. Two operators obtained transverse ultrasonographic images from the forelimb SDFTs of 16 National Hunt (NH) Thoroughbred (TB) racehorses, each undertaking analysis of their own and the other operator's images. One operator undertook analysis of their images using 2 sets of equipment. There was no statistically significant difference in the results obtained when different operators undertook image acquisition (P>0.05). At all but the most distal level, there was no significant difference when different equipment was used during analysis (P>0.05). A significant difference (P<0.01) was reported when different operators undertook image analysis, one operator consistently returning larger measurements. Different operators undertaking different stages of an examination can result in significant variability. To reduce confounding during ultrasonographic investigations involving multiple persons, one operator should undertake image analysis, although different operators may undertake image acquisition.
Yee, Lydia T S
2017-01-01
Words are frequently used as stimuli in cognitive psychology experiments, for example, in recognition memory studies. In these experiments, it is often desirable to control for the words' psycholinguistic properties because differences in such properties across experimental conditions might introduce undesirable confounds. In order to avoid confounds, studies typically check to see if various affective and lexico-semantic properties are matched across experimental conditions, and so databases that contain values for these properties are needed. While word ratings for these variables exist in English and other European languages, ratings for Chinese words are not comprehensive. In particular, while ratings for single characters exist, ratings for two-character words-which often have different meanings than their constituent characters, are scarce. In this study, ratings for 292 two-character Chinese nouns were obtained from Cantonese speakers in Hong Kong. Affective variables, including valence and arousal, and lexico-semantic variables, including familiarity, concreteness, and imageability, were rated in the study. The words were selected from a film subtitle database containing word frequency information that could be extracted and listed alongside the resulting ratings. Overall, the subjective ratings showed good reliability across all rated dimensions, as well as good reliability within and between the different groups of participants who each rated a subset of the words. Moreover, several well-established relationships between the variables found consistently in other languages were also observed in this study, demonstrating that the ratings are valid. The resulting word database can be used in studies where control for the above psycholinguistic variables is critical to the research design.
NASA Astrophysics Data System (ADS)
Abdussalam, Auwal; Monaghan, Andrew; Dukic, Vanja; Hayden, Mary; Hopson, Thomas; Leckebusch, Gregor
2013-04-01
Northwest Nigeria is a region with high risk of bacterial meningitis. Since the first documented epidemic of meningitis in Nigeria in 1905, the disease has been endemic in the northern part of the country, with epidemics occurring regularly. In this study we examine the influence of climate on the interannual variability of meningitis incidence and epidemics. Monthly aggregate counts of clinically confirmed hospital-reported cases of meningitis were collected in northwest Nigeria for the 22-year period spanning 1990-2011. Several generalized linear statistical models were fit to the monthly meningitis counts, including generalized additive models. Explanatory variables included monthly records of temperatures, humidity, rainfall, wind speed, sunshine and dustiness from weather stations nearest to the hospitals, and a time series of polysaccharide vaccination efficacy. The effects of other confounding factors -- i.e., mainly non-climatic factors for which records were not available -- were estimated as a smooth, monthly-varying function of time in the generalized additive models. Results reveal that the most important explanatory climatic variables are mean maximum monthly temperature, relative humidity and dustiness. Accounting for confounding factors (e.g., social processes) in the generalized additive models explains more of the year-to-year variation of meningococcal disease compared to those generalized linear models that do not account for such factors. Promising results from several models that included only explanatory variables that preceded the meningitis case data by 1-month suggest there may be potential for prediction of meningitis in northwest Nigeria to aid decision makers on this time scale.
Urinary Biomarkers at Early ADPKD Disease Stage
Petzold, Katja; Poster, Diane; Krauer, Fabienne; Spanaus, Katharina; Andreisek, Gustav; Nguyen-Kim, Thi Dan Linh; Pavik, Ivana; Ho, Thien Anh; Serra, Andreas L.; Rotar, Laura
2015-01-01
Background Autosomal dominant polycystic kidney disease (ADPKD) is characterized by a decline in renal function at late disease stage when the majority of functional renal parenchyma is replaced by cystic tissue. Thus, kidney function, assessed by estimated glomerular filtration rate (eGFR) does not well represent disease burden in early disease. Here, we investigated various urinary markers for tubular injury and their association with disease burden in ADPKD patients at early disease course. Methods ADPKD patients between 18 and 40 years with an eGFR greater or equal to 70 ml per min per 1.73m2 were eligible for this cross-sectional study. Urinary Neutrophil Gelatinase-Associated Lipocalin (NGAL), Kidney Injury Molecule-1 (KIM-1), and Uromodulin (UMOD) were investigated by Enzyme-Linked Immunosorbent Assay. Clara Cell Protein 16 (CC16) was investigated by Latex Immuno Assay. Cryoscopy was performed to assess urine osmolality and Urinary Albumin-to-Creatinine Ratio (UACR) was calculated. The association and the predictive properties of the markers on eGFR and height adjusted total kidney volume (htTKV) was evaluated using multiple regression analysis, incorporating different control variables for adjustment. Internal bootstrapping validated the obtained results. Results In 139 ADPKD patients (age 31 ±7 years, mean eGFR of 93 ± 19 ml per min per 1.73 m2) the total kidney volume was negatively correlated with eGFR and UMOD and positive associated with age, UACR, KIM-1 and urine osmolality after adjustment for possible confounders. Urine osmolality and htTKV were also associated with eGFR, whereas no association of CC16, NGAL and UMOD with eGFR or htTKV was found. Conclusion UACR and urinary KIM-1 are independently associated with kidney size but not with renal function in our study population. Urine osmolality was associated with eGFR and kidney volume following adjustment for multiple confounders. Despite statistical significance, the clinical value of our results is not yet conceivable. Further studies are needed to evaluate the property of the aforementioned biomarkers to assess disease state at early ADPKD stage. PMID:25875363
Using ecological propensity score to adjust for missing confounders in small area studies.
Wang, Yingbo; Pirani, Monica; Hansell, Anna L; Richardson, Sylvia; Blangiardo, Marta
2017-11-09
Small area ecological studies are commonly used in epidemiology to assess the impact of area level risk factors on health outcomes when data are only available in an aggregated form. However, the resulting estimates are often biased due to unmeasured confounders, which typically are not available from the standard administrative registries used for these studies. Extra information on confounders can be provided through external data sets such as surveys or cohorts, where the data are available at the individual level rather than at the area level; however, such data typically lack the geographical coverage of administrative registries. We develop a framework of analysis which combines ecological and individual level data from different sources to provide an adjusted estimate of area level risk factors which is less biased. Our method (i) summarizes all available individual level confounders into an area level scalar variable, which we call ecological propensity score (EPS), (ii) implements a hierarchical structured approach to impute the values of EPS whenever they are missing, and (iii) includes the estimated and imputed EPS into the ecological regression linking the risk factors to the health outcome. Through a simulation study, we show that integrating individual level data into small area analyses via EPS is a promising method to reduce the bias intrinsic in ecological studies due to unmeasured confounders; we also apply the method to a real case study to evaluate the effect of air pollution on coronary heart disease hospital admissions in Greater London. © The Author 2017. Published by Oxford University Press.
Carroll, Robert; Metcalfe, Chris; Steeg, Sarah; Davies, Neil M; Cooper, Jayne; Kapur, Nav; Gunnell, David
2016-01-01
Clinical guidelines have recommended psychosocial assessment of self-harm patients for years, yet estimates of its impact on the risk of repeat self-harm vary. Assessing the association of psychosocial assessment with risk of repeat self-harm is challenging due to the effects of confounding by indication. We analysed data from a cohort study of 15,113 patients presenting to the emergency departments of three UK hospitals to investigate the association of psychosocial assessment with risk of repeat hospital presentation for self-harm. Time of day of hospital presentation was used as an instrument for psychosocial assessment, attempting to control for confounding by indication. Conventional regression analysis suggested psychosocial assessment was not associated with risk of repeat self-harm within 12 months (Risk Difference (RD) 0.00 95% confidence interval (95%CI) -0.01 to 0.02). In contrast, IV analysis suggested risk of repeat self-harm was reduced by 18% (RD -0.18, 95%CI -0.32 to -0.03) in those patients receiving a psychosocial assessment. However, the instrument of time of day did not remove all potential effects of confounding by indication, suggesting the IV effect estimate may be biased. We found that psychosocial assessments reduce risk of repeat self-harm. This is in-line with other non-randomised studies based on populations in which allocation to assessment was less subject to confounding by indication. However, as our instrument did not fully balance important confounders across time of day, the IV effect estimate should be interpreted with caution.
Valera, Beatriz; Dewailly, Eric; Poirier, Paul
2013-01-01
Epidemiological evidence suggests a negative impact of methylmercury (MeHg) on cardiovascular risk factors such as blood pressure (BP) and heart rate (HR). This issue is of concern in Arctic populations such as in the Inuit of Nunavik since this contaminant is accumulated in fish and marine mammals, which still represent the subsistence diet of this population. We examined the associations between MeHg and BP and resting HR among Inuit adults. The "Santé Quebec" health survey was conducted in 1992 in the 14 villages of Nunavik and a complete set of data was obtained for 313 Inuit adults≥18 years. Blood samples were collected in order to determine total mercury, lead, total polychlorinated biphenyls (PCBs), n-3 polyunsaturated fatty acids (PUFAs), fasting glucose and lipid profile while socio-demographic variables were obtained through questionnaires. Anthropometric measurements as well as BP and resting HR were obtained using standardised protocols. Pulse pressure (PP: systolic BP minus diastolic BP) was also calculated. Multiple linear regression was used in order to determine the change in the dependent variables associated with the quartiles of MeHg concentration, taking the quartile 1 as reference. The mean age of the participants was 38±14 years and the sample was composed of 132 men (42.2%) and 181 women (57.8%). MeHg geometric mean was 15.4 μg/L (95%CI: 13.9-17.0) and levels ranged from 0.8 to 112.0 μg/L. Resting HR increased linearly across quartiles of blood MeHg concentration after adjusting for confounders (p for trend=0.02). An increase of 6.9 beats per minute (bpm) between the 4th and 1st quartile was observed after adjusting for confounders. No significant association was observed between blood MeHg and systolic BP, diastolic BP or PP. MeHg was associated with increasing resting HR after considering traditional risk factors as well as other contaminants (lead and total PCBs) and n-3 PUFAs. In contrast, no significant association with blood pressure was observed in this study. Copyright © 2012 Elsevier Inc. All rights reserved.
Small and Large Number Processing in Infants and Toddlers with Williams Syndrome
ERIC Educational Resources Information Center
Van Herwegen, Jo; Ansari, Daniel; Xu, Fei; Karmiloff-Smith, Annette
2008-01-01
Previous studies have suggested that typically developing 6-month-old infants are able to discriminate between small and large numerosities. However, discrimination between small numerosities in young infants is only possible when variables continuous with number (e.g. area or circumference) are confounded. In contrast, large number discrimination…
Using Learning Analytics to Characterize Student Experimentation Strategies in Engineering Design
ERIC Educational Resources Information Center
Vieira, Camilo; Goldstein, Molly Hathaway; Purzer, Senay; Magana, Alejandra J.
2016-01-01
Engineering design is a complex process both for students to participate in and for instructors to assess. Informed designers use the key strategy of conducting experiments as they test ideas to inform next steps. Conversely, beginning designers experiment less, often with confounding variables. These behaviours are not easy to assess in…
Growth and physiology of aspen supplied with different fertilizer addition rates
Mark D. Coleman; Richard E. Dickson; J.G. Isebrands
1998-01-01
Variable internal plant nutrient content may confound plant response to environmental stress. Plant nutrient content may be controlled with relative addition rate techniques in solution culture. However, because raising large numbers of plants in flowing solution culture is difficult. we investieated the feasibility of raisine plants in soil mix using relative...
Neighborhood Frequency Effect in Chinese Word Recognition: Evidence from Naming and Lexical Decision
ERIC Educational Resources Information Center
Li, Meng-Feng; Gao, Xin-Yu; Chou, Tai-Li; Wu, Jei-Tun
2017-01-01
Neighborhood frequency is a crucial variable to know the nature of word recognition. Different from alphabetic scripts, neighborhood frequency in Chinese is usually confounded by component character frequency and neighborhood size. Three experiments were designed to explore the role of the neighborhood frequency effect in Chinese and the stimuli…
Impact of Trauma Work on Social Work Clinicians: Empirical Findings.
ERIC Educational Resources Information Center
Cunningham, Maddy
2003-01-01
Describes a study of social work clinicians working with two types of trauma: sexual abuse and cancer. The effect of clinician's cognitive schemas and the confounding variables of personal history of abuse and years' experience are described. Clinicians who worked primarily with clients who were sexually abused reported more disruptions in…
Maternal Age at Childbirth and Offspring Disruptive Behaviors: Testing the Causal Hypothesis
ERIC Educational Resources Information Center
D'Onofrio, Brian M.; Goodnight, Jackson A.; Van Hulle, Carol A.; Rodgers, Joseph Lee; Rathouz, Paul J.; Waldman, Irwin D.; Lahey, Benjamin B.
2009-01-01
Background: Recent studies suggest that the association between maternal age at childbearing (MAC) and children's disruptive behaviors is the result of family factors that are confounded with both variables, rather than a casual effect of environmental factors specifically related to MAC. These studies, however, relied on restricted samples and…
Long-Term Positive Associations between Music Lessons and IQ
ERIC Educational Resources Information Center
Schellenberg, E. Glenn
2006-01-01
In Study 1 (N = 147), duration of music lessons was correlated positively with IQ and with academic ability among 6-to 11-year-olds, even when potential confounding variables (i.e., family income, parents' education, involvement in nonmusical activities) were held constant. In Study 2 (N = 150), similar but weaker associations between playing…
Summary. Measures of health outcomes are collinear with meteorology and air quality, making analysis of connections between human health and air quality difficult. The purpose of this analysis was to determine time scales and periods shared by the variables of interest (and...
Influences of Phonological Context on Tense Marking in Spanish-English Dual Language Learners
ERIC Educational Resources Information Center
Combiths, Philip N.; Barlow, Jessica A.; Potapova, Irina; Pruitt-Lord, Sonja
2017-01-01
Purpose: The emergence of tense-morpheme marking during language acquisition is highly variable, which confounds the use of tense marking as a diagnostic indicator of language impairment in linguistically diverse populations. In this study, we seek to better understand tense-marking patterns in young bilingual children by comparing phonological…
International Business, Gender, and Training: Preparing for the Global Economy.
ERIC Educational Resources Information Center
Hugenberg, Lawrence W.; And Others
This paper deals with how to train employees in transnational companies to be more "global" in their understanding and awareness of different cultures. The paper defines as "target culture" where the employee or his family will be working and considers as a confounding variable in the international marketplace the role to be…
The Sensitivity of the Right Hemisphere to Contextual Information in Sentences
ERIC Educational Resources Information Center
Gouldthorp, Bethanie; Coney, Jeffrey
2009-01-01
One explanation for the inconsistencies in research examining the sentence comprehension abilities of the right hemisphere (RH) is the presence of confounding variables that have generally served to disadvantage the processing capacities of the RH. As such, the present study aimed to investigate hemispheric differences in the use of message-level…
Learning at the Speed of Light: Deep Learning and Accelerated Online Graduate Courses
ERIC Educational Resources Information Center
Trekles, Anastasia M.
2013-01-01
An increasing number of university programs, particularly at the graduate level, are moving to an accelerated, time-compressed model for online degree offerings. However, the literature revealed that research in distance education effectiveness is still confounded by many variables, including course design and student approach to learning.…
Brasic, J R; Barnett, J Y; Kaplan, D; Sheitman, B B; Aisemberg, P; Lafargue, R T; Kowalik, S; Clark, A; Tsaltas, M O; Young, J G
1994-07-01
In an open, nonblind clinical trial, clomipramine reduced adventitious movements and compulsions in five previously medicated prepubertal boys with autistic disorder and severe mental retardation. Poorly adapted rating scales, interrater variability, subject heterogeneity, different treatment histories, and environmental stresses confounded the assessment of treatment effects.
Wingenfeld, Katja; Schulz, Michael; Damkroeger, Annika; Philippsen, Christine; Rose, Matthias; Driessen, Martin
2010-09-01
In psychoneuroendocrinology research, salivary measures have become increasingly important. While several studies focus on determinants of salivary cortisol such as age, gender, and gynaecological variables, less research has focused on confounding variables of salivary alpha-amylase (sAA). In a large sample of nurses (N=215) we analyzed the impact of age, gender, intake of oral contraceptives, smoking, coffee consumption as well as psychological parameters, such as work stress and burnout, on basal diurnal sAA release. Saliva was collected at 07:00 h, 11:30 h, 17:30 h, and 20:00 h on a working day during early shift. Only gender could be identified to have an impact on sAA, with females having a more pronounced sAA increase over the course of the day. Whereas depression, anxiety, work stress and burnout were not associated with sAA, a small negative correlation between social difficulties, measured with the Chronic Stress Screening Scale, and sAA could be identified. Copyright (c) 2010 Elsevier B.V. All rights reserved.
Age at immigration and crime in Stockholm using sibling comparisons.
Beckley, Amber L
2015-09-01
Past Swedish research has shown that immigrants arriving in the receiving country at an older age are less likely to commit crime than immigrants arriving at a younger age. Segmented assimilation theory argues that the family and neighborhood may be important factors affecting how age at immigration and crime are related to one another. This study used population-based register data on foreign-background males from Stockholm to test the effect of age at immigration on crime. Potential confounding from the family and neighborhood was addressed using variables and modeling strategies. Initial results, using variables to control for confounding, showed that people who immigrated around age 4 were the most likely to be suspected of a crime. When controlling for unmeasured family characteristics, it seemed that a later age at immigration was tied to a lower likelihood of crime, which does not corroborate past research findings. The effect of age at immigration, however, was not statistically significant. The results imply that future research on entire families may be a worthwhile endeavor. Copyright © 2015 Elsevier Inc. All rights reserved.
Portugal, Liana Catarina L; Pereira, Mirtes Garcia; Alves, Rita de Cássia S; Tavares, Gisella; Lobo, Isabela; Rocha-Rego, Vanessa; Marques-Portella, Carla; Mendlowicz, Mauro V; Coutinho, Evandro S; Fiszman, Adriana; Volchan, Eliane; Figueira, Ivan; Oliveira, Letícia de
2012-03-01
Tonic immobility is a defensive reaction occurring under extreme life threats. Patients with posttraumatic stress disorder (PTSD) reporting peritraumatic tonic immobility show the most severe symptoms and a poorer response to treatment. This study investigated the predictive value of tonic immobility for posttraumatic stress symptoms in a non-clinical sample. One hundred and ninety-eight college students exposed to various life threatening events were selected to participate. The Posttraumatic Stress Disorder Checklist - Civilian Version (PCL-C) and tonic immobility questions were used. Linear regression models were fitted to investigate the association between peritraumatic tonic immobility and PCL-C scores. Peritraumatic dissociation, peritraumatic panic reactions, negative affect, gender, type of trauma, and time since trauma were considered as confounding variables. We found significant association between peritraumatic tonic immobility and PTSD symptoms in a non-clinical sample exposed to various traumas, even after regression controlled for confounding variables (β = 1.99, p = 0.017). This automatic reaction under extreme life threatening stress, although adaptive for defense, may have pathological consequences as implied by its association with PTSD symptoms.
Maternal and Paternal Height and the Risk of Preeclampsia.
Lee, Yunsung; Magnus, Per
2018-04-01
The etiology of preeclampsia is unknown. Tall women have been found to have lower incidence of preeclampsia. This points to a possible biological causal effect but may be because of socioeconomic confounding. We used paternal height as an unexposed control to examine confounding. The MoBa (Norwegian Mother and Child Cohort Study) was used to extract data on parental heights, maternal prepregnancy weight, other background factors, and pregnancy outcomes for 99 968 singleton births. Multiple logistic regression was used to estimate odds ratios for preeclampsia according to parental height. The adjusted odds ratio for preeclampsia was 0.74 (95% CI, 0.66-0.82) for women >172 cm as compared with women <164 cm. The adjusted odds ratio for preeclampsia for men >186 cm was 1.03 (95% CI, 0.93-1.15) compared with men <178 cm. The association between maternal height and preeclampsia is unlikely to be because of confounding by familial, socioeconomic factors or by fetal genes related to height. The observed association between maternal height and preeclampsia merits further investigation. © 2018 American Heart Association, Inc.
Iwakawa, Hiromi; Nakamura, Yasuyuki; Fukui, Tomiho; Fukuwatari, Tsutomu; Ugi, Satoshi; Maegawa, Hiroshi; Doi, Yukio; Shibata, Katsumi
2016-01-01
We examined the concentrations of water-soluble vitamins in blood and urinary excretion of 22 patients with type 2 diabetes mellitus (type 2DM) and 20 healthy control participants. Macronutrient and vitamin intakes of type 2DM subjects were measured using a weighed food record method. Control participants consumed a semipurified diet for eight days. Multiple linear regression models were used to determine whether significant differences existed in vitamin concentrations in blood independent of age, sex, and other confounding factors. Concentrations of vitamins B2, B6, C, niacin, and folate in blood were significantly lower in type 2DM subjects than in controls, independent of confounding factors. Renal clearances of vitamins B6, C, niacin, and folate were significantly higher in type 2DM subjects than in controls. In conclusion, concentrations of vitamins B2, B6, C, niacin, and folate in blood were significantly lower in type 2DM subjects than in controls, independent of confounding factors; based on the evidence of increased urinary clearance of these vitamins, the lower levels were likely due to impaired reabsorption processes.
Wen, Xiaopiao; Liang, Yuan; Zhu, Jiang; Wu, Tangchun
2013-01-01
The aim of the current study was to explore the relationship between housework and health of retired older adults, and to provide new evidences and clues for the effects of housework on health. The data came from the baseline survey of the Tongji-Dongfeng Cohort Study with 25,501 participants. The independent variable was housework including child care, elder care, grocery shopping & cooking, and washing clothes & cleaning house. The dependent variable was self-reported two-week illness. Age, education, marital status, smoking and drinking were used as potential confounding variables. There are gender differences in the four types of housework, with higher prevalence among women than among men. The prevalence of two-week illness in women was higher than that in men. After adjusting for potential confounding variables, the four types of housework had almost no significant effects on two-week illness, except for washing clothes & cleaning house with a negative effect for women (OR = 1.17, 95% CI 1.06-1.29). The association of housework with health, especially of child care, elder care, and grocery shopping & cooking without significance and of washing clothes & cleaning house with significance for women, would provide a valuable clue for in-depth study of housework, namely the effects of housework on health would be subdivided into the dimensions of psychological and social activity besides physical activity, and it may not be suitable to simply classify housework as a kind of exercise.
Comparison of methods for the analysis of relatively simple mediation models.
Rijnhart, Judith J M; Twisk, Jos W R; Chinapaw, Mai J M; de Boer, Michiel R; Heymans, Martijn W
2017-09-01
Statistical mediation analysis is an often used method in trials, to unravel the pathways underlying the effect of an intervention on a particular outcome variable. Throughout the years, several methods have been proposed, such as ordinary least square (OLS) regression, structural equation modeling (SEM), and the potential outcomes framework. Most applied researchers do not know that these methods are mathematically equivalent when applied to mediation models with a continuous mediator and outcome variable. Therefore, the aim of this paper was to demonstrate the similarities between OLS regression, SEM, and the potential outcomes framework in three mediation models: 1) a crude model, 2) a confounder-adjusted model, and 3) a model with an interaction term for exposure-mediator interaction. Secondary data analysis of a randomized controlled trial that included 546 schoolchildren. In our data example, the mediator and outcome variable were both continuous. We compared the estimates of the total, direct and indirect effects, proportion mediated, and 95% confidence intervals (CIs) for the indirect effect across OLS regression, SEM, and the potential outcomes framework. OLS regression, SEM, and the potential outcomes framework yielded the same effect estimates in the crude mediation model, the confounder-adjusted mediation model, and the mediation model with an interaction term for exposure-mediator interaction. Since OLS regression, SEM, and the potential outcomes framework yield the same results in three mediation models with a continuous mediator and outcome variable, researchers can continue using the method that is most convenient to them.
NASA Technical Reports Server (NTRS)
Seeley, M. W.; Ruschy, D. L.; Linden, D. R.
1983-01-01
A cooperative research project was initiated in 1982 to study differences in thematic mapper spectral characteristics caused by variable tillage and crop residue practices. Initial evaluations of radiometric data suggest that spectral separability of variably tilled soils can be confounded by moisture and weathering effects. Separability of bare tilled soils from those with significant amounts of corn residue is enhanced by wet conditions, but still possible under dry conditions when recent tillage operations have occurred. In addition, thematic mapper data may provide an alternative method to study the radiant energy balance at the soil surface in conjunction with variable tillage systems.
Physical Activity and Bone Health in Schoolchildren: The Mediating Role of Fitness and Body Fat
Torres-Costoso, Ana; Gracia-Marco, Luis; Sánchez-López, Mairena; Notario-Pacheco, Blanca; Arias-Palencia, Natalia; Martínez-Vizcaíno, Vicente
2015-01-01
Background The relationship between physical activity (PA) and bone health is well known, although the role of percent body fat (%BF) and fitness as confounders or mediators in this relationship remains uncertain. Objective To examine whether the association between PA and bone mineral content (BMC) is mediated by %BF and cardiorespiratory fitness (CRF). Methods In this cross sectional study, BMC, total %BF (by DXA), vigorous PA (VPA), CRF, age and height were measured in 132 schoolchildren (62 boys, aged 8–11 years). ANCOVA was used to test differences in BMC by %BF, CRF and VPA, controlling for different sets of confounders. Simple mediation analyses and serial multiple mediation analyses were fitted to examine whether the relationship between PA and BMC is mediated by %BF and fitness. Results Children with high %BF had higher total body BMC than their peers after controlling for all sets of confounders. Children with good CRF or VPA had significantly less total body BMC after controlling for age and sex but in children with good CRF this inverse relation disappeared after adjusting by %BF. %BF and CRF both act as a full mediator in the association between VPA and BMC, after inclusion of the potential confounders in the models. Conclusion Fitness and %BF seem to have a mediator role on the relationship between physical activity and bone mass. PMID:25915941
Morsi, Rami Z; Safa, Rawan; Baroud, Serge F; Fawaz, Cherine N; Farha, Jad I; El-Jardali, Fadi; Chaaya, Monique
2017-04-11
Since July 2015, Lebanon has been experiencing a waste management crisis. Dumpsites in inhabited areas and waste burning have emerged due to the waste accumulation, further adding to the gravity of the situation. However, the association between the crisis and health of the population has not been scientifically reported. A comparative cross-sectional study was conducted to assess whether exposure to open dumpsites and waste burning is associated with acute health symptoms. The study sample included 221 male workers between the ages of 18-60 years selected from two areas chosen based on their proximity to a garbage dumpsite and waste burning. 110 workers were exposed to a garbage dumpsite and waste burning, and 111 workers were not. Data were collected via a face-to-face interview using a newly developed validated structured questionnaire. Chi-square tests were used to check for statistically significant differences between exposure and covariates. Multivariable analyses using multiple logistic regression were used to compare health symptoms between exposed and unexposed workers adjusting for potential confounders. The prevalence of acute health symptoms was greater among the exposed workers than the non-exposed workers, including gastrointestinal, respiratory, dermatological and constitutional symptoms. Controlling for confounding variables, such as age, insurance, family support, residence near dumpsite, work site, and smoking, a minimum odds ratio (OR) of 4.30 was obtained when comparing the exposed population to those non-exposed. The strong association between improper waste management and physical health calls for immediate attention by the government, stakeholders and community members to find optimal solutions for this waste management crisis and set immediate priority interventions such as regular waste collection, volume reduction and recycling performance improvement. However, the long recall period may have underestimated our results.
Wang, Hongyu; Dumont, Xavier; Haufroid, Vincent; Bernard, Alfred
2017-09-12
Recent studies in children have reported associations of urinary cadmium (U-Cd), used as biomarker of Cd body burden, with renal dysfunction, retarded growth and impaired cognitive development in children. Little is known, however, about factors influencing U-Cd in children and likely to act as confounders. In a cross-sectional study involving 249 schoolchildren (mean age, 5.72 years; 138 boys), we measured the urine concentrations of cadmium, zinc, lead, albumin, alpha 1 -microglobulin (A1M), retinol-binding protein, β 2 -microglobulin and club cell protein (CC16). Determinants of U-Cd expressed per creatinine or adjusted to specific gravity were identified by multiple regression analyses. Girls and boys had similar median concentrations of U-Cd (0.22 and 0.24 μg/L, 0.33 and 0.35 μg/g creatinine, respectively). When models were run without including creatinine or specific gravity among independent variables, urinary zinc, urinary A1M and age emerged as the strongest predictors of U-Cd expressed per g creatinine or adjusted to SG. When adding creatinine among predictors, urinary creatinine emerged as an additional strong predictor correlating negatively with U-Cd per g creatinine. This strong residual influence of diuresis, not seen when adding specific gravity among predictors, linked U-Cd to U-A1M or U-CC16 through secondary associations mimicking those induced by Cd nephrotoxity. In young children U-Cd largely varies with diuresis, zinc metabolism and urinary A1M. These physiological determinants, unrelated to Cd body burden, may confound the child renal and developmental outcomes associated with low-level U-Cd.
Munger, R.; Isacson, P.; Hu, S.; Burns, T.; Hanson, J.; Lynch, C.F.; Cherryholmes, K.; Van Dorpe, P.; Hausler, W.J.
1997-01-01
In a statewide survey of 856 Iowa municipal drinking water supplies in 1986-1987 the Rathbun rural water system was found to contain elevated levels of triazine herbicides. Rates of low birth weight, prematurity, and intrauterine growth retardation (IUGR) in live singleton births during the period 1984-1990 by women living in 13 communities served by the Rathbun water system were compared to other communities of similar size in the same Iowa counties. The Rathbun communities had a greater risk of IUGR than southern Iowa communities with other surface sources of drinking water (relative risk = 1.8; 95% CI = 1.3, 2.7). Multiple linear regression analyses revealed that levels of the herbicides atrazine, metolachlor, and cyanazine were each significant predictors of community IUGR rates in southern Iowa after controlling for several potentially confounding factors including maternal smoking and socioeconomic variables. The association with IUGR was strongest for atrazine, but all three herbicides were intercorrelated and the independent contributions of each to IUGR risk could not be determined. We conclude that communities in southern Iowa with drinking water supplies contaminated with herbicides have elevated rates of IUGR compared to neighboring communities with different water supplies. Because of the limitations of the ecologic design of this study, including aggregate rather than individual measures of exposure and limited ability to control for confounding factors related to source of drinking water and risk of IUGR, a strong causal relationship between any specific water contaminant and risk of IUGR cannot yet be inferred. The association between the water supplied to the Rathbun communities and the increased risk of IUGR should be considered a preliminary finding that needs to be verified by more detailed epidemiologic studies.
Rothenberg, Stephen J.; Rothenberg, Jesse C.
2005-01-01
Statistical evaluation of the dose–response function in lead epidemiology is rarely attempted. Economic evaluation of health benefits of lead reduction usually assumes a linear dose–response function, regardless of the outcome measure used. We reanalyzed a previously published study, an international pooled data set combining data from seven prospective lead studies examining contemporaneous blood lead effect on IQ (intelligence quotient) of 7-year-old children (n = 1,333). We constructed alternative linear multiple regression models with linear blood lead terms (linear–linear dose response) and natural-log–transformed blood lead terms (log-linear dose response). We tested the two lead specifications for nonlinearity in the models, compared the two lead specifications for significantly better fit to the data, and examined the effects of possible residual confounding on the functional form of the dose–response relationship. We found that a log-linear lead–IQ relationship was a significantly better fit than was a linear–linear relationship for IQ (p = 0.009), with little evidence of residual confounding of included model variables. We substituted the log-linear lead–IQ effect in a previously published health benefits model and found that the economic savings due to U.S. population lead decrease between 1976 and 1999 (from 17.1 μg/dL to 2.0 μg/dL) was 2.2 times ($319 billion) that calculated using a linear–linear dose–response function ($149 billion). The Centers for Disease Control and Prevention action limit of 10 μg/dL for children fails to protect against most damage and economic cost attributable to lead exposure. PMID:16140626
Pregnancy context and women's health-related quality of life.
Gariepy, Aileen; Lundsberg, Lisbet S; Vilardo, Nicole; Stanwood, Nancy; Yonkers, Kimberly; Schwarz, Eleanor B
2017-05-01
The objective was to quantify the association of pregnancy context and health-related quality of life (HRQoL). English- or Spanish-speaking women, aged 16-44, with pregnancies <24 weeks' gestation were enrolled in this cross-sectional study between June 2014 and June 2015. Participants completed self-assessments of pregnancy "context," including timing, intention, wantedness, desirability, happiness, and planning (measured with the London Measure of Unplanned Pregnancy). HRQoL was measured using the Patient Reported Outcomes Measurement Information System Global Short Form. Associations between measures of pregnancy context and HRQoL scores in the lowest tertile were examined using multivariable logistic regression to adjust for potential confounding variables. We enrolled 161 participants (mean age=27.2±6.6 years). Only 14% self-identified as White, non-Hispanic; 42% Hispanic; 37% Black, non-Hispanic; and 7% multiracial. Most (79%) participants were unmarried, and 75% were parenting. Mean gestational age was 9±4.6 weeks. In unadjusted models, women reporting mixed feelings about wanting to have a baby, an undesired pregnancy or feeling unhappy about learning of their pregnancy more frequently had low mental and physical HRQoL compared to women reporting wanted, desired, happy pregnancies. Women with an unplanned pregnancy or pregnancy occurring at the wrong time also had lower physical HRQoL than women reporting pregnancies that were planned or happened at the right time. However, after multivariate adjustment, including history of depression, pregnancy contexts were not associated with low mental or physical HRQoL. After adjusting for multiple confounders, pregnancy context is not significantly associated with HRQoL. The focus on pregnancy intention in public health programs may not sufficiently assess multidimensional aspects of pregnancy context and may not align with patient-centered outcomes such as HRQoL. Copyright © 2017 Elsevier Inc. All rights reserved.
Caqueo-Urízar, Alejandra; Fond, Guillaume; Urzúa, Alfonso; Boyer, Laurent; Williams, David R
2016-12-01
The aim of the present study was (i) to assess the prevalence of Violent Behavior in Schizophrenia (VBS) in a sample of community-dwelling outpatients in three middle-income countries of Latin America and (ii) to determine the clinical and socio-demographical risk factors associated with VBS and aggression level. The study included 253 stabilized outpatients with schizophrenia and their principal caregivers from 3 public ambulatory psychiatric care centers in Bolivia (N=83), Chile (N=85), and Peru (N=85). VBS was defined according to the Overt Aggression Scale (OAS) score and the aggression level was measured by the aggression subscore of the Agitated Behavior Scale of Corrigan. We collected socio-demographic information and clinical data. Multiple linear and logistic regressions were performed to determine which variables were associated with VBS and aggression level. The prevalence of VBS differed statistically between the three countries (p<0.001) with 3.5% in Chile, 14.6% in Peru and 55.4% in Bolivia. After adjustment for confounding factors, VBS was associated with a younger age, a more severe psychotic symptomatology, a lower family income and unemployment. After adjustment for confounding factors, aggression level was associated with a more severe psychotic symptomatology, a lower family income, a younger age at illness onset and higher number of hospitalizations in the last 3years. These results may guide future health policies to specifically provide social support and rehabilitation care to VBS patients in middle-income countries, including psychoeducation and a more integrated work between the treating medical team and the social workers. Copyright © 2016 Elsevier B.V. All rights reserved.
Ceresini, Graziano; Marina, Michela; Lauretani, Fulvio; Maggio, Marcello; Bandinelli, Stefania; Ceda, Gian Paolo; Ferrucci, Luigi
2015-01-01
Objectives Thyroid dysfunction in the elderly is associated with adverse clinical outcomes, with mortality being associated with low TSH. However, it is still unknown whether variability of thyroid function test within the reference range is associated with mortality in older adults. We studied the association between plasma levels of TSH, free T3 (FT3), and free T4 (FT4), and all-cause mortality in older adults who had all three hormones within the normal range. Design Longitudinal study Setting Community-based Participants Total of 815 euthyroid participants of the InCHIANTI study, aged 65 years or older Measurements All subjects had TSH, FT3, and FT4 within the reference range at baseline. Plasma TSH, FT3 and FT4 were predictors and 9-year all-cause mortality was the outcome. Cox proportional hazards models adjusted for confounders were used to examine the relationship between quartiles of TSH, FT3, and FT4 and all-cause mortality over 9 years of follow-up. Results During the follow-up (mean persons-years 8643.74 [min-max, 35.36-16985.00]), 181 deaths occurred (22.2%). Participants with TSH in the lower quartile had higher mortality than the rest of the population. After adjusting for multiple confounders, participants with TSH in the lowest quartile (Hazard Ratio: 2.22; 95% Confidence Interval: 1.19–4.22) had significantly higher all-cause mortality than those with TSH in the highest quartile. Neither FT3 nor FT4 were associated with mortality. Conclusions In euthyroid elderly subjects, normal-low TSH represents an independent risk factor for all-cause mortality. PMID:27000328
Inoue, Akiomi; Kawakami, Norito; Eguchi, Hisashi; Tsutsumi, Akizumi
2018-05-01
We examined the interaction effect of job insecurity (JI) and role ambiguity (RA) on psychological distress in Japanese employees. Overall, 2184 male and 805 female employees from two factories of a manufacturing company in Japan completed a self-administered questionnaire comprising the scales measuring JI (Job Content Questionnaire), RA (National Institute for Occupational Safety and Health Generic Job Stress Questionnaire), psychological distress (K6 scale), and potential confounders (i.e., age, education, family size, occupational class, and work shift). Taking psychological distress as a dependent variable, hierarchical multiple regression analyses were conducted by gender and employment status (i.e., permanent and non-permanent employees). An interaction term of JI × RA was included in the model. After adjusting for potential confounders, the main effects of JI and RA on psychological distress were significant regardless of gender or employment status. Furthermore, the significant interaction effect of JI × RA on psychological distress was observed among permanent male employees (β = 0.053, p = 0.010). Post hoc simple slope analyses showed that the simple slope of JI was greater at higher levels of RA (i.e., one standard deviation [SD] above the mean) (β = 0.300, p < 0.001) compared to lower levels of RA (i.e., one SD below the mean) (β = 0.212, p < 0.001). On the other hand, the interaction effect of JI × RA was not significant among permanent or non-permanent female employees. The present study suggests that higher levels of RA strengthen the association of JI with psychological distress, at least among Japanese permanent male employees.
Metabolomic characterization of laborers exposed to welding fumes.
Wang, Kuo-Ching; Kuo, Ching-Hua; Tian, Tze-Feng; Tsai, Mong-Hsun; Chiung, Yin-Mei; Hsiech, Chun-Ming; Tsai, Sung-Jeng; Wang, San-Yuan; Tsai, Dong-Ming; Huang, Chiang-Ching; Tseng, Y Jane
2012-03-19
The complex composition of welding fumes, multiplicity of molecular targets, diverse cellular effects, and lifestyles associated with laborers vastly complicate the assessment of welding fume exposure. The urinary metabolomic profiles of 35 male welders and 16 male office workers at a Taiwanese shipyard were characterized via (1)H NMR spectroscopy and pattern recognition methods. Blood samples for the same 51 individuals were also collected, and the expression levels of the cytokines and other inflammatory markers were examined. This study dichotomized the welding exposure variable into high (welders) versus low (office workers) exposures to examine the differences of continuous outcome markers-metabolites and inflammatory markers-between the two groups. Fume particle assessments showed that welders were exposed to different concentrations of chromium, nickel, and manganese particles. Multivariate statistical analysis of urinary metabolomic patterns showed higher levels of glycine, taurine, betaine/TMAO, serine, S-sulfocysteine, hippurate, gluconate, creatinine, and acetone and lower levels of creatine among welders, while only TNF-α was significantly associated with welding fume exposure among all cytokines and other inflammatory markers measured. Of the identified metabolites, the higher levels of glycine, taurine, and betaine among welders were suspected to play some roles in modulating inflammatory and oxidative tissue injury processes. In this metabolomics experiment, we also discovered that the association of the identified metabolites with welding exposure was confounded by smoking, but not with drinking, which is a finding consistent with known modified response of inflammatory markers among smokers. Our results correspond with prior studies that utilized nonmetabolomic analytical techniques and suggest that the metabolomic profiling is an efficient method to characterize the overall effect of welding fume exposure and other confounders. © 2012 American Chemical Society
Vinagre, Catarina; Madeira, Diana; Mendonça, Vanessa; Dias, Marta; Roma, Joana; Diniz, Mário S
2014-06-01
Oxidative stress biomarkers have been widely used in the development of ecological indices and in the assessment of exposure of aquatic organisms to contaminants from agricultural, industrial and urban pollution. However, temperature is known to also have a significant effect on oxidative stress biomarkers. This way, temperature is a confounding factor that may result in difficulties in the interpretation of oxidative stress biomarkers response patterns. Since climate change is expected to result in more frequent and intense heat wave events it is pertinent to investigate the effect of increasing temperature in the oxidative stress response of common aquatic organisms. It is also important to assess the differential response of different body tissues, given that they are differently exposed to temperature depending on their location and physiological function. This study investigates the effect of increasing temperature (20 °C-34 °C) in the response of multiple biomarkers of oxidative stress: lipid peroxidation, glutathione-S-transferase, superoxide dismutase and catalase activities, in the muscle, liver and gills of a common coastal fish, the Rock goby, Gobius paganellus. The response of the oxidative stress biomarkers analysed were always higher in the gills than in the other tissues. Muscle generally presented the lower levels of any of the biomarkers tested when compared to other tissues. Nevertheless, muscle tissue always responded significantly to temperature, as did the liver, while the gills were unresponsive in terms of lipid peroxidation and glutathione-S-transferase. Unresponsive tissues to temperature may be particularly interesting as indicators of pollution, given that temperature will not be a confounding variable in their oxidative stress response. Copyright © 2014 Elsevier Ltd. All rights reserved.
Wlaźlak, Edyta; Surkont, Grzegorz; Shek, Ka L; Dietz, Hans P
2015-10-01
It has been claimed that urethral hypermobility and resting urethral pressure can largely explain stress incontinence in women. In this study we tried to replicate these findings in an unselected cohort of women seen for urodynamic testing, including as many potential confounders as possible. This study is a retrospective analysis of data obtained from 341 women. They attended for urodynamic testing due to symptoms of pelvic floor dysfunction. We excluded from the analysis women with a history of previous anti-incontinence and prolapse surgery. All patients had a standardised clinical assessment, 4D transperineal pelvic floor ultrasound and multichannel urodynamic testing. Urodynamic stress incontinence (USI) was diagnosed by multichannel urodynamic testing. Its severity was subjectively graded as mild, moderate and severe. Candidate variables were: age, BMI, symptoms of prolapse, vaginal parity, significant prolapse (compartment-specific), levator avulsion, levator hiatal area, Oxford grading, midurethral mobility, maximum urethral pressure (MUP), maximum cough pressure and maximum Valsalva pressure reached. On binary logistic regression, the following parameters were statistically significant in predicting urodynamic stress incontinence: age (P=0.03), significant rectocele (P=0.02), max. abdominal pressure reached (negatively, P<0.0001), midurethral mobility (P=0.0004) and MUP (negatively, P<0.0001). On multivariate analysis, accounting for multiple interdependencies, the following predictors remained significant: max. abdominal pressure reached (negatively, P<0.0001), cough pressure (P=0.006), midurethral mobility (P=0.003) and MUP (negatively, P<0.0001), giving an R(2) of 0.24. Mid-urethral mobility and MUP are the main predictors of USI. Demographic and clinical data are at best weak predictors. Our results suggest the presence of major unrecognised confounders. Copyright © 2015. Published by Elsevier Ireland Ltd.
Effect of Exposure to Air Pollution on Placental Weight in Isfahan-Iran.
Ghasemi-Tehrani, Hatav; Fallah, Setare; Mozafarian, Nafiseh; Miranzadeh, Sareh; Sadeghi, Shokooh; Azidhak, Azam
2017-06-01
Objective: To determine the effect of Air Quality Index (AQI) in the first trimester of pregnancy on birth weight, placental weight, and the ratio of placental weight to the birth weight (pw-bw) in Isfahan. Materials and methods: This cross-sectional study was done on 312 consecutive pregnant women in Beheshti Hospital in Isfahan city in 2013. Information on air pollution was received from the Environmental department of Isfahan. Average exposure to air pollution in the first trimester of pregnancy was calculated for eachpregnant woman. In order to compare quantitative and qualitative variables, analysis of variance (ANOVA), and chi-square were applied. After that, the multiple linear regression analysis was used to assess the association the Air Quality Index (AQI) on birth weight, placental weight and the ratio of pw-bw. Potential confounders including age, baby gender, smoking of husband, maternal BMI, maternal occupation, and education and mother's residential area were considered. A statistical significant association were considered for P-value less than 0.05. Results: The findings showed that there is inverse relationship between exposure to air pollution and placental weight in the first trimester of pregnancy after controlling potential confounders (β = -2.57, p-value = 0.008). The inverse relationship between air pollution and the ratio of pw-bw was found. (β = -0.001, p-value = 0.002). Conclusion: The results of this study suggest that air pollution is associated with newborns' health which in turn is a warning alarm for considering some actions in both sides of reducing the air pollution and teaching the pregnant women about the adverse effects of air pollution on the pregnancy outcomes.
Estimation of health effects of prenatal methylmercury exposure using structural equation models.
Budtz-Jørgensen, Esben; Keiding, Niels; Grandjean, Philippe; Weihe, Pal
2002-10-14
Observational studies in epidemiology always involve concerns regarding validity, especially measurement error, confounding, missing data, and other problems that may affect the study outcomes. Widely used standard statistical techniques, such as multiple regression analysis, may to some extent adjust for these shortcomings. However, structural equations may incorporate most of these considerations, thereby providing overall adjusted estimations of associations. This approach was used in a large epidemiological data set from a prospective study of developmental methyl-mercury toxicity. Structural equation models were developed for assessment of the association between biomarkers of prenatal mercury exposure and neuropsychological test scores in 7 year old children. Eleven neurobehavioral outcomes were grouped into motor function and verbally mediated function. Adjustment for local dependence and item bias was necessary for a satisfactory fit of the model, but had little impact on the estimated mercury effects. The mercury effect on the two latent neurobehavioral functions was similar to the strongest effects seen for individual test scores of motor function and verbal skills. Adjustment for contaminant exposure to poly chlorinated biphenyls (PCBs) changed the estimates only marginally, but the mercury effect could be reduced to non-significance by assuming a large measurement error for the PCB biomarker. The structural equation analysis allows correction for measurement error in exposure variables, incorporation of multiple outcomes and incomplete cases. This approach therefore deserves to be applied more frequently in the analysis of complex epidemiological data sets.
Scherer, P; Penner, I K; Rohr, A; Boldt, H; Ringel, I; Wilke-Burger, H; Burger-Deinerth, E; Isakowitsch, K; Zimmermann, M; Zahrnt, S; Hauser, R; Hilbert, K; Tiel-Wilck, K; Anvari, K; Behringer, A; Peglau, I; Friedrich, H; Plenio, A; Benesch, G; Ehret, R; Nippert, I; Finke, G; Schulz, I; Bergtholdt, B; Breitkopf, S; Kaskel, P; Reischies, F; Kugler, J
2007-04-01
Reliable, language-independent, short screening instruments to test for cognitive function in patients with multiple sclerosis (MS) remain rare, despite the high number of patients affected by cognitive decline. We developed a new, short screening instrument, the Faces Symbol Test (FST), and compared its diagnostic test characteristics with a composite of the Digit Symbol Substitution Test (DSST) and the Paced Auditory Serial Addition Test (PASAT), in 108 MS patients and 33 healthy controls. An Informant-Report Questionnaire, a Self-Report Questionnaire, and a neurologist's estimation of the Every Day Life Cognitive Status were also applied to the MS patients. The statistical analyses comprised of a receiver operating characteristic analysis for test accuracy and for confounding variables. The PASAT and DSST composite score estimated that 36.5% of the MS patients had cognitive impairment. The FST estimated that 40.7% of the MS patients were cognitively impaired (sensitivity 84%; specificity 85%). The FST, DSST and PASAT results were significantly correlated with the patients' physical impairment, as measured by the Expanded Disability Status Scale (EDSS). The results suggest that the FST might be a culture-free, sensitive, and practical short screening instrument for the detection of cognitive decline in patients with MS, including those in the early stages.
The affect of tissue depth variation on craniofacial reconstructions.
Starbuck, John M; Ward, Richard E
2007-10-25
We examined the affect of tissue depth variation on the reconstruction of facial form, through the application of the American method, utilizing published tissue depth measurements for emaciated, normal, and obese faces. In this preliminary study, three reconstructions were created on reproductions of the same skull for each set of tissue depth measurements. The resulting morphological variation was measured quantitatively using the anthropometric craniofacial variability index (CVI). This method employs 16 standard craniofacial anthropometric measurements and the results reflect "pattern variation" or facial harmony. We report no appreciable variation in the quantitative measure of the pattern facial form obtained from the three different sets of tissue depths. Facial similarity was assessed qualitatively utilizing surveys of photographs of the three reconstructions. Surveys indicated that subjects frequently perceived the reconstructions as representing different individuals. This disagreement indicates that size of the face may blind observers to similarities in facial form. This research is significant because it illustrates the confounding effect that normal human variation contributes in the successful recognition of individuals from a representational three-dimensional facial reconstruction. Research results suggest that successful identification could be increased if multiple reconstructions were created which reflect a wide range of possible outcomes for facial form. The creation of multiple facial images, from a single skull, will be facilitated as computerized versions of facial reconstruction are further developed and refined.
Anagnostouli, Maria; Katsavos, Serafeim; Kyrozis, Andreas; Gontika, Maria; Voumvourakis, Konstantinos I; Kapaki, Elisabeth
2016-08-01
Mechanisms of angiogenesis regulate multiple sclerosis (MS) lesions' evolution, displaying both neuroprotective and harmful effects. Factors traditionally considered as purely angiogenic, like vascular endothelial growth factor (VEGF), exert complex heterogenous actions on both neural and vascular malformation-derived tissues. Aim of this retrospective study was to examine, for the first time, potential associations between the presence of common vascular malformations, like vertebral hemangiomas (VHs), and several clinico-radiological MS parameters. 236 MS patients who were followed in our Outpatient Clinic were recruited in this study. Outcome measures concerned demographics, disease-derived variables, and MS-lesions' distribution in VHs - positive and negative patients. All data were collected retrospectively. Potential correlations were assessed with univariate statistical analyses (p = 0.05), followed by multivariate regression models, for purposes of confounder-effects elimination. VH presence showed significant negative correlations with presence of MS lesions in the thoracic (p = 0.005 for thoracic VHs), but not the cervical cord. Trends towards negative associations of VH presence with subtentorial MS lesions and positive family history for MS were also observed. Our observations suggest that VH presence may reduce the risk of thoracic demyelinating lesions in MS patients. They could be explained as part of a multifaceted angiogenic process, concomitantly enhancing neural repair and abnormal hemangioma vascularization.
Disruptions in Energy Balance: Does Nature overcome Nurture?
Fernández, José R.; Casazza, Krista; Divers, Jasmin; López-Alarcón, Mardya
2008-01-01
Fat accumulation, in general, is the result of a breakdown in the homeostatic regulation of energy balance. Although, the specific factors influencing the disruption of energy balance and why these factors affect individuals differently are not completely understood, numerous studies have identified multiple contributors. Environmental components influence food acquisition, eating, and lifestyle habits. However, the variability in obesity-related outcomes observed among individuals placed in similar controlled environments support the notion that genetic components also wield some control. Multiple genetic regions have been associated with measures related to energy balance; however, the replication of these genetic contributors to energy intake and energy expenditure in humans is relatively small perhaps because of the heterogeneity of human populations. Genetic tools such as genetic admixture account for individual’s genetic background in gene association studies, reducing the confounding effect of population stratification, and promise to be a relevant tool on the identification of genetic contributions to energy balance, particularly among individuals of diverse racial/ethnic backgrounds. Although it has been recognized that genes are expressed according to environmental influences, the search toward the understanding of nature and nurture in obesity will require the detailed study of the effect of genes under diverse physiologic and behavioral environments. It is evident that more research is needed to elucidate the methodological and statistical issues that underlie the interactions between genes and environments in obesity and its related comorbidities. PMID:18096193
Doughty, Michael J; Blades, Kenneth A; Ibrahim, Nassim
2002-03-01
The goal of this study was both to assess the frequency of self-reported eye symptoms for employees in a non-air-conditioned office building complex, and to specifically assess whether confounding variables (contact lens wear, medicines use, and provoking stimuli) had a discernible impact on the characteristics of this reporting. A questionnaire was distributed, in the winter months, to office employees in a building complex within a city-located University site, consisting of non-air-conditioned rooms with some mechanical ventilation. The questionnaire asked respondents to identify any eye symptoms experienced (foreign body sensation, soreness, dryness, grittiness, itchiness, stinging, intense burning), the frequency of their occurrence, details of any conditions provoking eye symptoms (such as changes in the microenvironment), their medication use and if they were contact lens wearers. The proportion of respondents for various categories was calculated, along with the total number of symptoms. If 'eye irritation' is considered as reporting any symptoms experienced at any time, then 63% of 292 respondents (median age 42.5 years) reported an average of 1.83 symptoms. However, the identification of contact lens wearers, medication use (especially for seasonal allergies) and provoking stimuli reduced this prevalence to just 29% (with the average number of symptoms being 1.51). If the frequency of experiencing symptoms was also taken into account (never, sometimes, often, and constantly), then only 12.3% of the respondents indicated symptoms often or constantly, i.e. could be considered to have significant symptoms. In all categories, men were more likely to report eye symptoms compared with women. The results confirm that office employees are likely to have some ocular symptoms, but that even a simple consideration of possible confounding variables can greatly change the reported prevalence of eye irritation that might be specifically associated with the workplace environment.
Godart, Nathalie; Berthoz, Sylvie; Rein, Zoé; Perdereau, Fabienne; Lang, François; Venisse, Jean-Luc; Halfon, Olivier; Bizouard, Pierre; Loas, Gwénolé; Corcos, Maurice; Jeammet, Philippe; Flament, Martine; Curt, Florence
2006-12-01
The objective of the present work is to determine whether the prevalence of depressive and anxiety disorders varies in subgroups of eating disorders (ED) according to age, ED duration, mode of care provision, and body mass index (BMI). Using the Mini International Neuropsychiatric Interview (MINI), the frequency of anxiety and depressive disorders was evaluated in 271 ED participants. Their prevalence was compared in subgroups of anorexics (AN-R and AN-BN) and bulimics (BN), both before and after controlling for potential confounding variables. Current or lifetime comorbidity of anxiety and depressive disorders did not differ between AN-R and AN-BN groups. Social phobia, panic disorders, and obsessive-compulsive disorder (OCD) were significantly more frequent in AN-BN and AN-R groups. Panic disorder was more frequent in the BN group. Several confounding factors, in particular those identified in the present study, may explain previous conflicting results on the frequency of anxiety and depressive disorders in ED. Nevertheless, the study confirmed that OCD is more frequent in AN, even after controlling for confounding factors. Copyright 2006 by Wiley Periodicals, Inc.
Reproductive and developmental effects of disinfection by-products in drinking water.
Reif, J S; Hatch, M C; Bracken, M; Holmes, L B; Schwetz, B A; Singer, P C
1996-01-01
Recent epidemiologic studies have reported associations between the consumption of chlorinated drinking water and reproductive and developmental effects. Here we review the available epidemiologic data, assess the hazard potential posed by exposure to disinfection by-products, identify critical data gaps, and offer recommendations for further research. The epidemiologic evidence supporting associations between exposure to water disinfection by-products (DBPs) and adverse pregnancy outcomes is sparse, and positive findings should be interpreted cautiously. The methods used during the early stages of research in this area have been diverse. Variability in exposure assessment and endpoints makes it difficult to synthesize or combine the available data. Exposure misclassification and unmeasured confounding may have lead to bias in risk estimation. Future studies of reproductive outcome and exposure to chlorinated water should use improved methods for exposure assessment to 1) assure selection of appropriate exposure markers, 2) assess seasonal and annual fluctuations in DBPs, 3) assess variability within the distribution system, and 4) assess exposure through multiple routes such as bathing and showering, as well as consumption. Population-based studies should be conducted to evaluate male and female fertility, conception delay, growth retardation, and specific birth defects. The reproductive and developmental effects of exposure to DBPs could be efficiently explored in ongoing investigations by incorporating valid exposure markers and relevant questionnaire information. Future studies should make use of naturally occurring variability in the concentrations of DBPs and may incorporate biomarkers of exposure and effect in their design. Epidemiologic investigations should be conducted in parallel with laboratory-based and animal studies in a coordinated, multidisciplinary approach. PMID:8930546
Byrd, Darrin; Christopfel, Rebecca; Arabasz, Grae; Catana, Ciprian; Karp, Joel; Lodge, Martin A; Laymon, Charles; Moros, Eduardo G; Budzevich, Mikalai; Nehmeh, Sadek; Scheuermann, Joshua; Sunderland, John; Zhang, Jun; Kinahan, Paul
2018-01-01
Positron emission tomography (PET) is a quantitative imaging modality, but the computation of standardized uptake values (SUVs) requires several instruments to be correctly calibrated. Variability in the calibration process may lead to unreliable quantitation. Sealed source kits containing traceable amounts of [Formula: see text] were used to measure signal stability for 19 PET scanners at nine hospitals in the National Cancer Institute's Quantitative Imaging Network. Repeated measurements of the sources were performed on PET scanners and in dose calibrators. The measured scanner and dose calibrator signal biases were used to compute the bias in SUVs at multiple time points for each site over a 14-month period. Estimation of absolute SUV accuracy was confounded by bias from the solid phantoms' physical properties. On average, the intrascanner coefficient of variation for SUV measurements was 3.5%. Over the entire length of the study, single-scanner SUV values varied over a range of 11%. Dose calibrator bias was not correlated with scanner bias. Calibration factors from the image metadata were nearly as variable as scanner signal, and were correlated with signal for many scanners. SUVs often showed low intrascanner variability between successive measurements but were also prone to shifts in apparent bias, possibly in part due to scanner recalibrations that are part of regular scanner quality control. Biases of key factors in the computation of SUVs were not correlated and their temporal variations did not cancel out of the computation. Long-lived sources and image metadata may provide a check on the recalibration process.
Liang, Xiaoyun; Vaughan, David N; Connelly, Alan; Calamante, Fernando
2018-05-01
The conventional way to estimate functional networks is primarily based on Pearson correlation along with classic Fisher Z test. In general, networks are usually calculated at the individual-level and subsequently aggregated to obtain group-level networks. However, such estimated networks are inevitably affected by the inherent large inter-subject variability. A joint graphical model with Stability Selection (JGMSS) method was recently shown to effectively reduce inter-subject variability, mainly caused by confounding variations, by simultaneously estimating individual-level networks from a group. However, its benefits might be compromised when two groups are being compared, given that JGMSS is blinded to other groups when it is applied to estimate networks from a given group. We propose a novel method for robustly estimating networks from two groups by using group-fused multiple graphical-lasso combined with stability selection, named GMGLASS. Specifically, by simultaneously estimating similar within-group networks and between-group difference, it is possible to address inter-subject variability of estimated individual networks inherently related with existing methods such as Fisher Z test, and issues related to JGMSS ignoring between-group information in group comparisons. To evaluate the performance of GMGLASS in terms of a few key network metrics, as well as to compare with JGMSS and Fisher Z test, they are applied to both simulated and in vivo data. As a method aiming for group comparison studies, our study involves two groups for each case, i.e., normal control and patient groups; for in vivo data, we focus on a group of patients with right mesial temporal lobe epilepsy.
Dehydroepiandrosterone-sulfate is associated with cardiovascular reactivity to stress in women.
Hirokawa, Kumi; Ohira, Tetsuya; Nagayoshi, Mako; Kajiura, Mitsugu; Imano, Hironori; Kitamura, Akihiko; Kiyama, Masahiko; Okada, Takeo; Iso, Hiroyasu
2016-07-01
Cardiovascular stress reactivity is a predictor of atherosclerosis and cardiac events. Dehydroepiandrosterone (DHEA) protects against cardiovascular diseases, but results among previous studies have been inconsistent. We investigated the association between dehydroepiandrosterone-sulfate (DHEA-s) and cardiovascular stress reactivity in Japanese women and men. Among 979 healthy Japanese subjects (641 women and 338 men), serum levels of DHEA-s, systolic blood pressure (SBP) and diastolic blood pressure (DBP), heart rate, heart rate variability, and peripheral blood flow were measured under rest and two types of task. Mean differences in measured variables during tasks and a post-task period were calculated as changes in stress reactivity. Variables of stress reactivity were adjusted for multiple potential confounding factors. In women, DHEA-s levels showed positive associations with changes in SBP and DBP (standardized beta=0.12, p=0.020; 0.17, 0.002, respectively). Stratification by menopausal status and other lifestyle factors (e.g., smoking status, alcohol consumption) were conducted. Significant positive associations remained in pre-menopausal (standardized beta=0.13, p=0.037; 0.18, 0.005), non-smoking (0.12, 0.010; 0.18, <0.001), and non-drinking women (0.14, 0.021; 0.21, 0.001), and women without a medical history (0.15, 0.020; 0.20, 0.001). In men, there was no significant association between DHEA-s levels and changes in stress reactivity. DHEA-s levels were positively associated with high blood-pressure reactivity to stress in women, and being menopausal, smoking, and alcohol consumption modified this association. Copyright © 2016 Elsevier Ltd. All rights reserved.
Correlation of Vitamin D status and orthodontic-induced external apical root resorption.
Tehranchi, Azita; Sadighnia, Azin; Younessian, Farnaz; Abdi, Amir H; Shirvani, Armin
2017-01-01
Adequate Vitamin D is essential for dental and skeletal health in children and adult. The purpose of this study was to assess the correlation of serum Vitamin D level with external-induced apical root resorption (EARR) following fixed orthodontic treatment. In this cross-sectional study, the prevalence of Vitamin D deficiency (defined by25-hydroxyvitamin-D) was determined in 34 patients (23.5% male; age range 12-23 years; mean age 16.63 ± 2.84) treated with fixed orthodontic treatment. Root resorption of four maxillary incisors was measured using before and after periapical radiographs (136 measured teeth) by means of a design-to-purpose software to optimize data collection. Teeth with a maximum percentage of root resorption (%EARR) were indicated as representative root resorption for each patient. A multiple linear regression model and Pearson correlation coefficient were used to assess the association of Vitamin D status and observed EARR. P < 0.05 was considered statistically significant. The Pearson coefficient between these two variables was determined about 0.15 ( P = 0.38). Regression analysis revealed that Vitamin D status of the patients demonstrated no significant statistical correlation with EARR, after adjustment of confounding variables using linear regression model ( P > 0.05). This study suggests that Vitamin D level is not among the clinical variables that are potential contributors for EARR. The prevalence of Vitamin D deficiency does not differ in patients with higher EARR. These data suggest the possibility that Vitamin D insufficiency may not contribute to the development of more apical root resorption although this remains to be confirmed by further longitudinal cohort studies.
Genetic variation in alpha2-adrenoreceptors and heart rate recovery after exercise
Kohli, Utkarsh; Diedrich, André; Kannankeril, Prince J.; Muszkat, Mordechai; Sofowora, Gbenga G.; Hahn, Maureen K.; English, Brett A.; Blakely, Randy D.; Stein, C. Michael
2015-01-01
Heart rate recovery (HRR) after exercise is an independent predictor of adverse cardiovascular outcomes. HRR is mediated by both parasympathetic reactivation and sympathetic withdrawal and is highly heritable. We examined whether common genetic variants in adrenergic and cholinergic receptors and transporters affect HRR. In our study 126 healthy subjects (66 Caucasians, 56 African Americans) performed an 8 min step-wise bicycle exercise test with continuous computerized ECG recordings. We fitted an exponential curve to the postexercise R-R intervals for each subject to calculate the recovery constant (kr) as primary outcome. Secondary outcome was the root mean square residuals averaged over 1 min (RMS1min), a marker of parasympathetic tone. We used multiple linear regressions to determine the effect of functional candidate genetic variants in autonomic pathways (6 ADRA2A, 1 ADRA2B, 4 ADRA2C, 2 ADRB1, 3 ADRB2, 2 NET, 2 CHT, and 1 GRK5) on the outcomes before and after adjustment for potential confounders. Recovery constant was lower (indicating slower HRR) in ADRA2B 301–303 deletion carriers (n = 54, P = 0.01), explaining 3.6% of the interindividual variability in HRR. ADRA2A Asn251Lys, ADRA2C rs13118771, and ADRB1 Ser49Gly genotypes were associated with RMS1min. Genetic variability in adrenergic receptors may be associated with HRR after exercise. However, most of the interindividual variability in HRR remained unexplained by the variants examined. Noncandidate gene-driven approaches to study genetic contributions to HRR in larger cohorts will be of interest. PMID:26058836
Abe, Toshi; Furui, Shigeru; Sasaki, Hiroshi; Sakamoto, Yasuo; Suzuki, Shigeru; Ishitake, Tatsuya; Terasaki, Kinuyo; Kohtake, Hiroshi; Norbash, Alexander M; Behrman, Richard H; Hayabuchi, Naofumi
2013-03-01
To evaluate low-dose X-ray radiation effects on the eye by measuring the amount of light scattering in specific regions of the lens, we compared exposed subjects (interventional radiologists) with unexposed subjects (employees of medical service companies), as a pilot study. According to numerous exclusionary rules, subjects with confounding variables contributing to cataract formation were excluded. Left eye examinations were performed on 68 exposed subjects and 171 unexposed subjects. The eye examinations consisted of an initial screening examination, followed by Scheimpflug imaging of the lens using an anterior eye segment analysis system. The subjects were assessed for the quantity of light scattering intensities found in each of the six layers of the lens. Multiple stepwise regression analyses were performed with the stepwise regression for six variables: age, radiation exposure, smoking, drinking, wearing glasses and workplace. In addition, an age-matched comparison between exposed and unexposed subjects was performed. Minimal increased light scattering intensity in the posterior subcapsular region showed statistical significance. Our results indicate that occupational radiation exposure in interventional radiologists may affect the posterior subcapsular region of the lens. Since by its very nature this retrospective study had many limitations, further well-designed studies concerning minimal radiation-related lens changes should be carried out in a low-dose exposure group.
Fast and robust group-wise eQTL mapping using sparse graphical models.
Cheng, Wei; Shi, Yu; Zhang, Xiang; Wang, Wei
2015-01-16
Genome-wide expression quantitative trait loci (eQTL) studies have emerged as a powerful tool to understand the genetic basis of gene expression and complex traits. The traditional eQTL methods focus on testing the associations between individual single-nucleotide polymorphisms (SNPs) and gene expression traits. A major drawback of this approach is that it cannot model the joint effect of a set of SNPs on a set of genes, which may correspond to hidden biological pathways. We introduce a new approach to identify novel group-wise associations between sets of SNPs and sets of genes. Such associations are captured by hidden variables connecting SNPs and genes. Our model is a linear-Gaussian model and uses two types of hidden variables. One captures the set associations between SNPs and genes, and the other captures confounders. We develop an efficient optimization procedure which makes this approach suitable for large scale studies. Extensive experimental evaluations on both simulated and real datasets demonstrate that the proposed methods can effectively capture both individual and group-wise signals that cannot be identified by the state-of-the-art eQTL mapping methods. Considering group-wise associations significantly improves the accuracy of eQTL mapping, and the successful multi-layer regression model opens a new approach to understand how multiple SNPs interact with each other to jointly affect the expression level of a group of genes.
Emerging Cardiovascular Risk Research: Impact of Pets on Cardiovascular Risk Prevention
Schreiner, Pamela J.
2016-01-01
Animals interact with humans in multiple ways, including as therapy and service animals, commercially as livestock, as wildlife, and in zoos. But the most common interaction is as companion animals in our homes, with an estimated 180 million cats and dogs living in US households. While pet ownership has been reported to have many health benefits, the findings are inconsistent. Cardiovascular risk factors such as lipids, glucose, obesity, and heart rate variability have improved, worsened, or remained the same in the limited number of studies considering companion animals. Physical activity increases have more consistently been linked with dog ownership, although whether this reflects antecedent motivation or direct benefit from the dog is unclear. Allergies and asthma also are variably linked to pet ownership and are confounded by family history of atopy and timing of exposure to pet dander. The benefits of companion animals are most likely to be through reduction in depression, anxiety, and social isolation, but these studies have been largely cross-sectional and may depend on degree of bonding of the owner with the animal. Positive relationships show measurably higher oxytocin with lower cortisol and alpha-amylase levels. Finally, pet ownership is also a marker of better socioeconomic status and family stability, and if companion animals are to provide cardiovascular risk benefit, the route should perhaps be through improved education and opportunity for ownership. PMID:27547289
Driving Safety after Spinal Surgery: A Systematic Review
Alkhalili, Kenan; Hannallah, Jack; Ibeche, Bashar; Bajammal, Sohail; Baco, Abdul Moeen
2017-01-01
This study aimed to assess driving reaction times (DRTs) after spinal surgery to establish a timeframe for safe resumption of driving by the patient postoperatively. The MEDLINE and Google Scholar databases were analyzed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) Statement for clinical studies that investigated changes in DRTs following cervical and lumbar spinal surgery. Changes in DRTs and patients' clinical presentation, pathology, anatomical level affected, number of spinal levels involved, type of intervention, pain level, and driving skills were assessed. The literature search identified 12 studies that investigated postoperative DRTs. Six studies met the inclusion criteria; five studies assessed changes in DRT after lumbar spine surgery and two studies after cervical spina surgery. The spinal procedures were selective nerve root block, anterior cervical discectomy and fusion, and lumbar fusion and/ordecompression. DRTs exhibited variable responses to spinal surgery and depended on the patients' clinical presentation, spinal level involved, and type of procedure performed. The evidence regarding the patients' ability to resume safe driving after spinal surgery is scarce. Normalization of DRT or a return of DRT to pre-spinal intervention level is a widely accepted indicator for safe driving, with variable levels of statistical significance owing to multiple confounding factors. Considerations of the type of spinal intervention, pain level, opioid consumption, and cognitive function should be factored in the assessment of a patient's ability to safely resume driving. PMID:28443178
Driving Safety after Spinal Surgery: A Systematic Review.
Alhammoud, Abduljabbar; Alkhalili, Kenan; Hannallah, Jack; Ibeche, Bashar; Bajammal, Sohail; Baco, Abdul Moeen
2017-04-01
This study aimed to assess driving reaction times (DRTs) after spinal surgery to establish a timeframe for safe resumption of driving by the patient postoperatively. The MEDLINE and Google Scholar databases were analyzed according to the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analysis) Statement for clinical studies that investigated changes in DRTs following cervical and lumbar spinal surgery. Changes in DRTs and patients' clinical presentation, pathology, anatomical level affected, number of spinal levels involved, type of intervention, pain level, and driving skills were assessed. The literature search identified 12 studies that investigated postoperative DRTs. Six studies met the inclusion criteria; five studies assessed changes in DRT after lumbar spine surgery and two studies after cervical spina surgery. The spinal procedures were selective nerve root block, anterior cervical discectomy and fusion, and lumbar fusion and/ordecompression. DRTs exhibited variable responses to spinal surgery and depended on the patients' clinical presentation, spinal level involved, and type of procedure performed. The evidence regarding the patients' ability to resume safe driving after spinal surgery is scarce. Normalization of DRT or a return of DRT to pre-spinal intervention level is a widely accepted indicator for safe driving, with variable levels of statistical significance owing to multiple confounding factors. Considerations of the type of spinal intervention, pain level, opioid consumption, and cognitive function should be factored in the assessment of a patient's ability to safely resume driving.
Garcia, Ana Maria.; Alexander, Richard B.; Arnold, Jeffrey G.; Norfleet, Lee; White, Michael J.; Robertson, Dale M.; Schwarz, Gregory E.
2016-01-01
Despite progress in the implementation of conservation practices, related improvements in water quality have been challenging to measure in larger river systems. In this paper we quantify these downstream effects by applying the empirical U.S. Geological Survey water-quality model SPARROW to investigate whether spatial differences in conservation intensity were statistically correlated with variations in nutrient loads. In contrast to other forms of water quality data analysis, the application of SPARROW controls for confounding factors such as hydrologic variability, multiple sources and environmental processes. A measure of conservation intensity was derived from the USDA-CEAP regional assessment of the Upper Mississippi River and used as an explanatory variable in a model of the Upper Midwest. The spatial pattern of conservation intensity was negatively correlated (p = 0.003) with the total nitrogen loads in streams in the basin. Total phosphorus loads were weakly negatively correlated with conservation (p = 0.25). Regional nitrogen reductions were estimated to range from 5 to 34% and phosphorus reductions from 1 to 10% in major river basins of the Upper Mississippi region. The statistical associations between conservation and nutrient loads are consistent with hydrological and biogeochemical processes such as denitrification. The results provide empirical evidence at the regional scale that conservation practices have had a larger statistically detectable effect on nitrogen than on phosphorus loadings in streams and rivers of the Upper Mississippi Basin.
Schöffl, I; Kemmler, W; Kladny, B; Vonstengel, S; Kalender, W A; Engelke, K
2008-01-01
The objective of this study was an integrated cross-sectional investigation for answering the question whether differences in bone mineral density in elderly postmenopausal women are associated with differences in habitual physical activity and unspecific exercise levels. Two hundred and ninety nine elderly women (69-/+3 years), without diseases or medication affecting bone metabolism were investigated. The influence of weight, body composition and physical activity on BMD was measured at multiple sites using different techniques (DXA, QCT, and QUS). Physical activity and exercise level were assessed by questionnaire, maximum strength of the legs and aerobic capacity. Variations in physical activity or habitual exercise had no effect on bone. The only significant univariate relation between strength/VO(2)max and BMD/BMC that remained after adjusting for confounding variables was between arm BMD (DXA) and hand-grip strength. The most important variable for explaining BMD was weight and for cortical BMC of the femur (QCT) lean body mass. Weight and lean body mass emerge as predominant predictors of BMD in normal elderly women, whereas the isolated effect of habitual physical activity, unspecific exercise participation, and muscle strength on bone parameters is negligible. Thus, an increase in the amount of habitual physical activity will probably have no beneficial impact on bone.
Moreno-Martínez, F Javier; Goñi-Imízcoz, Miguel; Spitznagel, Mary Beth
2011-10-01
Category specific semantic impairment (e.g. living versus nonliving things) has been reported in association with various pathologies, including herpes simplex encephalitis and semantic dementia. However, evidence is inconsistent regarding whether this effect exists in diseases progressively impacting diverse cortical regions, such as Alzheimer's disease (AD). Ceiling effects producing non-Gaussian distributions and poor control for confounds such as nuisance variables (e.g. familiarity) may contribute to this discrepancy. Fourteen AD patients were longitudinally studied examining category effects on three semantic tasks (picture naming, naming to description and word to picture matching) matched across domain on all known nuisance variables (NV). To address non-Gaussian distributions, we run bootstrap analyses to determine whether NV, semantic domain or control performance best predicted AD patient performance. Multiple hierarchical regression analyses revealed that, whilst NV accounted for most of the explained variance in patients in the three tasks, the influence of semantic domain was substantially lower. Individual logistic regression demonstrated a significant category effect in only a few patients and healthy controls. No significant qualitative changes were observed in patients over time. Our results confirm the importance of NVs as predictors of AD patient performance, suggesting that the role of semantic domain is not a useful predictor of the progressive deterioration in AD. Copyright © 2011 Elsevier Inc. All rights reserved.
García, Ana María; Alexander, Richard B; Arnold, Jeffrey G; Norfleet, Lee; White, Michael J; Robertson, Dale M; Schwarz, Gregory
2016-07-05
Despite progress in the implementation of conservation practices, related improvements in water quality have been challenging to measure in larger river systems. In this paper we quantify these downstream effects by applying the empirical U.S. Geological Survey water-quality model SPARROW to investigate whether spatial differences in conservation intensity were statistically correlated with variations in nutrient loads. In contrast to other forms of water quality data analysis, the application of SPARROW controls for confounding factors such as hydrologic variability, multiple sources and environmental processes. A measure of conservation intensity was derived from the USDA-CEAP regional assessment of the Upper Mississippi River and used as an explanatory variable in a model of the Upper Midwest. The spatial pattern of conservation intensity was negatively correlated (p = 0.003) with the total nitrogen loads in streams in the basin. Total phosphorus loads were weakly negatively correlated with conservation (p = 0.25). Regional nitrogen reductions were estimated to range from 5 to 34% and phosphorus reductions from 1 to 10% in major river basins of the Upper Mississippi region. The statistical associations between conservation and nutrient loads are consistent with hydrological and biogeochemical processes such as denitrification. The results provide empirical evidence at the regional scale that conservation practices have had a larger statistically detectable effect on nitrogen than on phosphorus loadings in streams and rivers of the Upper Mississippi Basin.
Islam, S K M Azizul; Kim, Dongkyu; Lee, Young-Sil; Moon, Seong-Su
2018-06-01
This study evaluated the association of Heart rate variability (HRV) measurements with diabetic peripheral neuropathy (DPN) in patients with type 2 diabetes. This study included 102 Korean subjects with type 2 diabetes. The Michigan neuropathy screening instrument (MNSI) questionnaire score, the MNSI examination score (MNSIES) and the total symptom score were examined for DPN evaluation. Noninvasive HRV measurements were performed using photoelectric plethysmography. Patients with a MNSIES > 2 were considered to have DPN. The MNSIES showed significant negative associations with the high frequency (HF) (r = -0.212, p = 0.033) and low frequency (LF) (r = -0.286, p = 0.004) powers. Multiple linear regression analysis revealed that only HF power maintained a significant negative association with the MNSIES (β = -0.184; 95% CI -0.365 to -0.003; p = 0.047), after controlling for significant related confounders, with HRV parameters in male patients with type 2 diabetes. The HF (p = 0.010) and LF (p = 0.025) powers differed significantly between male patients without and those with DPN according to the MNSIES. This study revealed a negative association of DPN, as assessed by the MNSIES, with HF power in male patients with type 2 diabetes. DPN defined by foot examination was predictive of cardiac autonomic neuropathy. Copyright © 2018 Elsevier B.V. All rights reserved.
Emerging Cardiovascular Risk Research: Impact of Pets on Cardiovascular Risk Prevention.
Schreiner, Pamela J
2016-02-01
Animals interact with humans in multiple ways, including as therapy and service animals, commercially as livestock, as wildlife, and in zoos. But the most common interaction is as companion animals in our homes, with an estimated 180 million cats and dogs living in US households. While pet ownership has been reported to have many health benefits, the findings are inconsistent. Cardiovascular risk factors such as lipids, glucose, obesity, and heart rate variability have improved, worsened, or remained the same in the limited number of studies considering companion animals. Physical activity increases have more consistently been linked with dog ownership, although whether this reflects antecedent motivation or direct benefit from the dog is unclear. Allergies and asthma also are variably linked to pet ownership and are confounded by family history of atopy and timing of exposure to pet dander. The benefits of companion animals are most likely to be through reduction in depression, anxiety, and social isolation, but these studies have been largely cross-sectional and may depend on degree of bonding of the owner with the animal. Positive relationships show measurably higher oxytocin with lower cortisol and alpha-amylase levels. Finally, pet ownership is also a marker of better socioeconomic status and family stability, and if companion animals are to provide cardiovascular risk benefit, the route should perhaps be through improved education and opportunity for ownership.
Purves-Smith, Fennigje M; Sgarioto, Nicolas; Hepple, Russell T
2014-04-01
It is accepted widely that fast-twitch muscle fibers are preferentially impacted in aging muscle, yet we hypothesize that this is not valid when aging muscle atrophy becomes severe. In this review, we summarize the evidence of fiber type-specific effect in aging muscle and the potential confounding roles of fibers coexpressing multiple myosin heavy-chain isoforms and their histochemical identification.
McGuire, Connor; Kristman, Vicki L; Shaw, William; Williams-Whitt, Kelly; Reguly, Paula; Soklaridis, Sophie
2015-09-01
To determine the association between supervisors' leadership style and autonomy and supervisors' likelihood of supporting job accommodations for back-injured workers. A cross-sectional study of supervisors from Canadian and US employers was conducted using a web-based, self-report questionnaire that included a case vignette of a back-injured worker. Autonomy and two dimensions of leadership style (considerate and initiating structure) were included as exposures. The outcome, supervisors' likeliness to support job accommodation, was measured with the Job Accommodation Scale (JAS). We conducted univariate analyses of all variables and bivariate analyses of the JAS score with each exposure and potential confounding factor. We used multivariable generalized linear models to control for confounding factors. A total of 796 supervisors participated. Considerate leadership style (β = .012; 95% CI .009-.016) and autonomy (β = .066; 95% CI .025-.11) were positively associated with supervisors' likelihood to accommodate after adjusting for appropriate confounding factors. An initiating structure leadership style was not significantly associated with supervisors' likelihood to accommodate (β = .0018; 95% CI -.0026 to .0061) after adjusting for appropriate confounders. Autonomy and a considerate leadership style were positively associated with supervisors' likelihood to accommodate a back-injured worker. Providing supervisors with more autonomy over decisions of accommodation and developing their considerate leadership style may aid in increasing work accommodation for back-injured workers and preventing prolonged work disability.
Loeys, Tom; Talloen, Wouter; Goubert, Liesbet; Moerkerke, Beatrijs; Vansteelandt, Stijn
2016-11-01
It is well known from the mediation analysis literature that the identification of direct and indirect effects relies on strong no unmeasured confounding assumptions of no unmeasured confounding. Even in randomized studies the mediator may still be correlated with unobserved prognostic variables that affect the outcome, in which case the mediator's role in the causal process may not be inferred without bias. In the behavioural and social science literature very little attention has been given so far to the causal assumptions required for moderated mediation analysis. In this paper we focus on the index for moderated mediation, which measures by how much the mediated effect is larger or smaller for varying levels of the moderator. We show that in linear models this index can be estimated without bias in the presence of unmeasured common causes of the moderator, mediator and outcome under certain conditions. Importantly, one can thus use the test for moderated mediation to support evidence for mediation under less stringent confounding conditions. We illustrate our findings with data from a randomized experiment assessing the impact of being primed with social deception upon observer responses to others' pain, and from an observational study of individuals who ended a romantic relationship assessing the effect of attachment anxiety during the relationship on mental distress 2 years after the break-up. © 2016 The British Psychological Society.
Chronic malnutrition and Trypanosoma cruzi infection in children.
de Andrade, A L; Zicker, F
1995-04-01
The distribution of T. cruzi infection overlaps regions with high prevalence of child malnutrition. We examined the possible association between T. cruzi infection and chronic malnutrition. In a cross-sectional survey conducted among 1900 7-12 year-old schoolchildren in 60 village schools in central Brazil, anthropometric measurements (NCHS) taken from 153 children with at least two positive serological tests for antibodies against T. cruzi (IIF, ELISA, IHA) were compared to two age and sex seronegative matched classmates. Information on children's medical history and socio-economic status (SES) were collected from parents of the participants. Seropositive children had a 2.4-fold risk (95% CI 1.4-4.0) of being stunted (z-score < 2.0 of height-for-age) when compared to uninfected children even after adjusting for confounding variables. Being underweight (z-score < -2.0 of weight-for-age) was also statistically associated with seropositivity to T. cruzi (OR = 2.8, 95% CI 1.4-5.6). No statistical evidence of multiplicative interaction between nutritional status and SES was detected. Further studies on nutrition and metabolism are required to look into a possible physiopathological mechanism for this association.
Kabir, Yearul; Zafar, Tasleem A; Waslien, Carol
2013-01-01
The associations between body image and attitudes toward obesity and thinness and their associations with measured body mass index (BMI) among female students of Kuwait University (n = 137) was examined in 2008. The body image perceptions were assessed using nine female silhouettes figures. The difference between current perceived body image (PBI) and ideal body image (IBI) was used as a measure of body image dissatisfaction (BID). Students tended to have a bigger PBI and smaller IBI than would be expected from their BMI category, leading to high levels of BID in each BMI category. PBI, IBI, BID, RBI were highly correlated with each other, and BMI was significantly correlated with each of them. The coefficients of these associations were not significantly altered in multiple regression analysis by the addition of potential confounding variables, such as age, marital status, physical activity, dieting behavior, parental education, and family size. These results suggest that PBI and a desire to be thinner were strongly related to BID and that thinness is becoming more desired in Kuwaiti society than the plump body image of the past.
Measurement of testosterone in human sexuality research: methodological considerations.
van Anders, Sari M; Goldey, Katherine L; Bell, Sarah N
2014-02-01
Testosterone (T) and other androgens are incorporated into an increasingly wide array of human sexuality research, but there are a number of issues that can affect or confound research outcomes. This review addresses various methodological issues relevant to research design in human studies with T; unaddressed, these issues may introduce unwanted noise, error, or conceptual barriers to interpreting results. Topics covered are (1) social and demographic factors (gender and sex; sexual orientations and sexual diversity; social/familial connections and processes; social location variables), (2) biological rhythms (diurnal variation; seasonality; menstrual cycles; aging and menopause), (3) sample collection, handling, and storage (saliva vs. blood; sialogogues, saliva, and tubes; sampling frequency, timing, and context; shipping samples), (4) health, medical issues, and the body (hormonal contraceptives; medications and nicotine; health conditions and stress; body composition, weight, and exercise), and (5) incorporating multiple hormones. Detailing a comprehensive set of important issues and relevant empirical evidence, this review provides a starting point for best practices in human sexuality research with T and other androgens that may be especially useful for those new to hormone research.
Job stress as a risk factor for absences among manual workers: a 12-month follow-up study
HEO, Yong-Seok; LEEM, Jong-Han; PARK, Shin-Goo; JUNG, Dal-Young; KIM, Hwan-Cheol
2015-01-01
This study was conducted to evaluate the impact of job stress on absence from work caused by illnesses and accidents through a prospective research design. A total of 2,349 manual workers were included in this analysis. In the first survey, job stress was determined using the Korean Occupational Stress Scale-Short Form. In the second survey, information on absence due to accidents or illnesses during the past one year was obtained through a questionnaire. The relationship was analyzed using a logistic regression model with multiple imputation. After adjusting for confounding variables for males, absence due to accidents was statistically associated with high job demand, insufficient job control, inadequate social support, and organizational injustice. In addition, high job demands and organizational injustice were related to increased absence due to illnesses in both genders. A lack of reward was associated with increased absence due to illnesses among female workers. We found that job stress was associated with a higher risk of absence caused by accidents or illnesses of manual workers. PMID:26212413
Job stress as a risk factor for absences among manual workers: a 12-month follow-up study.
Heo, Yong-Seok; Leem, Jong-Han; Park, Shin-Goo; Jung, Dal-Young; Kim, Hwan-Cheol
2015-01-01
This study was conducted to evaluate the impact of job stress on absence from work caused by illnesses and accidents through a prospective research design. A total of 2,349 manual workers were included in this analysis. In the first survey, job stress was determined using the Korean Occupational Stress Scale-Short Form. In the second survey, information on absence due to accidents or illnesses during the past one year was obtained through a questionnaire. The relationship was analyzed using a logistic regression model with multiple imputation. After adjusting for confounding variables for males, absence due to accidents was statistically associated with high job demand, insufficient job control, inadequate social support, and organizational injustice. In addition, high job demands and organizational injustice were related to increased absence due to illnesses in both genders. A lack of reward was associated with increased absence due to illnesses among female workers. We found that job stress was associated with a higher risk of absence caused by accidents or illnesses of manual workers.
Prediction versus aetiology: common pitfalls and how to avoid them.
van Diepen, Merel; Ramspek, Chava L; Jager, Kitty J; Zoccali, Carmine; Dekker, Friedo W
2017-04-01
Prediction research is a distinct field of epidemiologic research, which should be clearly separated from aetiological research. Both prediction and aetiology make use of multivariable modelling, but the underlying research aim and interpretation of results are very different. Aetiology aims at uncovering the causal effect of a specific risk factor on an outcome, adjusting for confounding factors that are selected based on pre-existing knowledge of causal relations. In contrast, prediction aims at accurately predicting the risk of an outcome using multiple predictors collectively, where the final prediction model is usually based on statistically significant, but not necessarily causal, associations in the data at hand.In both scientific and clinical practice, however, the two are often confused, resulting in poor-quality publications with limited interpretability and applicability. A major problem is the frequently encountered aetiological interpretation of prediction results, where individual variables in a prediction model are attributed causal meaning. This article stresses the differences in use and interpretation of aetiological and prediction studies, and gives examples of common pitfalls. © The Author 2017. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
Metabolic risk and television time in adolescent females.
Machado-Rodrigues, Aristides M; Leite, Neiva; Coelho-e-Silva, Manuel J; Enes, Fernando; Fernandes, Rômulo; Mascarenhas, Luís P G; Boguszewski, Margaret C S; Malina, Robert M
2015-02-01
A sedentary lifestyle is increasingly implicated in a negative metabolic health profile among youth. The present study examined relationships between clustered metabolic risk factors and TV viewing in female adolescents. The sample comprised 262 girls 14-17 years. Height, weight, fasting glucose, insulin, HDL cholesterol, triglycerides, and blood pressure were measured. Body mass index (BMI) was calculated. TV viewing time and moderate-to-vigorous physical activity (MVPA) were estimated from a 3-day diary. Outcome variables were normalized and expressed as Z scores which were summed into a metabolic risk score. Multiple linear regression analysis was used. TV viewing was independently associated with increased prevalence of clustered metabolic risk in girls after adjustment for several confounders (i.e., chronological age, BMI, MVPA, and parental education). The final model also indicated that lower levels of MVPA, higher BMI, and lower mother education were associated with higher metabolic risk. Increased TV viewing had an adverse effect on metabolic health of adolescent girls. The findings highlight the potential importance of preventive actions to ameliorate metabolic risk in youth which target both sedentary and physically active behaviors.
Huybrechts, Krista F; Gerhard, Tobias; Franklin, Jessica M; Levin, Raisa; Crystal, Stephen; Schneeweiss, Sebastian
2014-08-01
Nursing home residents are of particular interest for comparative effectiveness research given their susceptibility to adverse treatment effects and systematic exclusion from trials. However, the risk of residual confounding because of unmeasured markers of declining health using conventional analytic methods is high. We evaluated the validity of instrumental variable (IV) methods based on nursing home prescribing preference to mitigate such confounding, using psychotropic medications to manage behavioral problems in dementia as a case study. A cohort using linked data from Medicaid, Medicare, Minimum Data Set, and Online Survey, Certification and Reporting for 2001-2004 was established. Dual-eligible patients ≥65 years who initiated psychotropic medication use after admission were selected. Nursing home prescribing preference was characterized using mixed-effects logistic regression models. The plausibility of IV assumptions was explored, and the association between psychotropic medication class and 180-day mortality was estimated. High-prescribing and low-prescribing nursing homes differed by a factor of 2. Each preference-based IV measure described a substantial proportion of variation in psychotropic medication choice (β(IV → treatment): 0.22-0.36). Measured patient characteristics were well balanced across patient groups based on instrument status (52% average reduction in Mahalanobis distance). There was no evidence that instrument status was associated with markers of nursing home quality of care. Findings indicate that IV analyses using nursing home prescribing preference may be a useful approach in comparative effectiveness studies, and should extend naturally to analyses including untreated comparison groups, which are of great scientific interest but subject to even stronger confounding. Copyright © 2014 John Wiley & Sons, Ltd.
Liu, Xiaolei; Huang, Meng; Fan, Bin; Buckler, Edward S.; Zhang, Zhiwu
2016-01-01
False positives in a Genome-Wide Association Study (GWAS) can be effectively controlled by a fixed effect and random effect Mixed Linear Model (MLM) that incorporates population structure and kinship among individuals to adjust association tests on markers; however, the adjustment also compromises true positives. The modified MLM method, Multiple Loci Linear Mixed Model (MLMM), incorporates multiple markers simultaneously as covariates in a stepwise MLM to partially remove the confounding between testing markers and kinship. To completely eliminate the confounding, we divided MLMM into two parts: Fixed Effect Model (FEM) and a Random Effect Model (REM) and use them iteratively. FEM contains testing markers, one at a time, and multiple associated markers as covariates to control false positives. To avoid model over-fitting problem in FEM, the associated markers are estimated in REM by using them to define kinship. The P values of testing markers and the associated markers are unified at each iteration. We named the new method as Fixed and random model Circulating Probability Unification (FarmCPU). Both real and simulated data analyses demonstrated that FarmCPU improves statistical power compared to current methods. Additional benefits include an efficient computing time that is linear to both number of individuals and number of markers. Now, a dataset with half million individuals and half million markers can be analyzed within three days. PMID:26828793
Effects of psychosocial work factors on lifestyle changes: a cohort study.
Allard, Karin Olofsson; Thomsen, Jane Frølund; Mikkelsen, Sigurd; Rugulies, Reiner; Mors, Ole; Kærgaard, Anette; Kolstad, Henrik A; Kaerlev, Linda; Andersen, Johan Hviid; Hansen, Ase Marie; Bonde, Jens Peter
2011-12-01
To evaluate the effect of the demand-control-support model, the effort-reward imbalance model, and emotional demands on smoking, alcohol consumption, physical activity, and body mass index. This is a 2-year prospective cohort study of 3224 public sector employees. Measures were assessed with questionnaires. Multiple regression analyses were used to predict changes in lifestyle factors. Low reward predicted smoking, low-decision latitude predicted being inactive, and high demands predicted high-alcohol consumption but only for men at follow-up even after controlling for potential confounders. There were no other significant findings in the expected direction except for some of the confounders. We found only limited and inconsistent support for the hypothesis that a poor psychosocial work environment is associated with an adverse lifestyle.
Temperature variability during delirium in ICU patients: an observational study.
van der Kooi, Arendina W; Kappen, Teus H; Raijmakers, Rosa J; Zaal, Irene J; Slooter, Arjen J C
2013-01-01
Delirium is an acute disturbance of consciousness and cognition. It is a common disorder in the intensive care unit (ICU) and associated with impaired long-term outcome. Despite its frequency and impact, delirium is poorly recognized by ICU-physicians and -nurses using delirium screening tools. A completely new approach to detect delirium is to use monitoring of physiological alterations. Temperature variability, a measure for temperature regulation, could be an interesting component to monitor delirium, but whether temperature regulation is different during ICU delirium has not yet been investigated. The aim of this study was to investigate whether ICU delirium is related to temperature variability. Furthermore, we investigated whether ICU delirium is related to absolute body temperature. We included patients who experienced both delirium and delirium free days during ICU stay, based on the Confusion Assessment method for the ICU conducted by a research- physician or -nurse, in combination with inspection of medical records. We excluded patients with conditions affecting thermal regulation or therapies affecting body temperature. Daily temperature variability was determined by computing the mean absolute second derivative of the temperature signal. Temperature variability (primary outcome) and absolute body temperature (secondary outcome) were compared between delirium- and non-delirium days with a linear mixed model and adjusted for daily mean Richmond Agitation and Sedation Scale scores and daily maximum Sequential Organ Failure Assessment scores. Temperature variability was increased during delirium-days compared to days without delirium (β(unadjuste)d=0.007, 95% confidence interval (CI)=0.004 to 0.011, p<0.001). Adjustment for confounders did not alter this result (β(adjusted)=0.005, 95% CI=0.002 to 0.008, p<0.001). Delirium was not associated with absolute body temperature (β(unadjusted)=-0.03, 95% CI=-0.17 to 0.10, p=0.61). This did not change after adjusting for confounders (β(adjusted)=-0.03, 95% CI=-0.17 to 0.10, p=0.63). Our study suggests that temperature variability is increased during ICU delirium.
ERIC Educational Resources Information Center
Caroll, Edward
1983-01-01
Reviews a study which found that symptoms associated with posttraumatic stress disorder (PTSD) are neither unique to or common among Vietnam veteran psychiatric patients or their Korean war counterparts. Questions choice of study sample, confounding variables, and source of data. A rebuttal by the original author is included. (WAS)
Root disease and other unforeseen variables that confound restoration efforts
William J. Otrosina; Shi-Jean S. Sung; Stanley J. Zarnoch; Brian T. Sullivan; Charles H. Walkinshaw
2002-01-01
Unanticipated disease problems thwarting restoration efforts can emerge in forest ecosystems. An example is the longleaf pine (Pinus palustris Mill.) ecosystem. This species once occupied nearly 30 million ha but now its range is reduced to approximately 1.5 million ha. Restoring longleaf pine to many sites in its former range is an important goal...
Gender Bias in the Diagnosis of a Geriatric Standardized Patient: A Potential Confounding Variable
ERIC Educational Resources Information Center
Lewis, Roya; Lamdan, Ruth M.; Wald, David; Curtis, Michael
2006-01-01
Background: Gender bias has been reported in the diagnosis and treatment of patients with a variety of illnesses. In the context of our 10-station fourth year Objective Structured Clinical Evaluation, we queried whether this could influence diagnosis in a geriatric case. Case writers hypothesized that, due to this bias, the female standardized…
Typing Speed as a Confounding Variable and the Measurement of Quality in Divergent Thinking
ERIC Educational Resources Information Center
Forthmann, Boris; Holling, Heinz; Çelik, Pinar; Storme, Martin; Lubart, Todd
2017-01-01
The need to control for writing or typing speed when assessing divergent-thinking performance has been recognized since the early '90s. An even longer tradition in divergent-thinking research has the issue of scoring the responses for quality. This research addressed both issues within structural equation modeling. Three dimensions of…
Implementing the undergraduate mini-CEX: a tailored approach at Southampton University.
Hill, Faith; Kendall, Kathleen; Galbraith, Kevin; Crossley, Jim
2009-04-01
The mini-clinical evaluation exercise (mini-CEX) is widely used in the UK to assess clinical competence, but there is little evidence regarding its implementation in the undergraduate setting. This study aimed to estimate the validity and reliability of the undergraduate mini-CEX and discuss the challenges involved in its implementation. A total of 3499 mini-CEX forms were completed. Validity was assessed by estimating associations between mini-CEX score and a number of external variables, examining the internal structure of the instrument, checking competency domain response rates and profiles against expectations, and by qualitative evaluation of stakeholder interviews. Reliability was evaluated by overall reliability coefficient (R), estimation of the standard error of measurement (SEM), and from stakeholders' perceptions. Variance component analysis examined the contribution of relevant factors to students' scores. Validity was threatened by various confounding variables, including: examiner status; case complexity; attachment specialty; patient gender, and case focus. Factor analysis suggested that competency domains reflect a single latent variable. Maximum reliability can be achieved by aggregating scores over 15 encounters (R = 0.73; 95% confidence interval [CI] +/- 0.28 based on a 6-point assessment scale). Examiner stringency contributed 29% of score variation and student attachment aptitude 13%. Stakeholder interviews revealed staff development needs but the majority perceived the mini-CEX as more reliable and valid than the previous long case. The mini-CEX has good overall utility for assessing aspects of the clinical encounter in an undergraduate setting. Strengths include fidelity, wide sampling, perceived validity, and formative observation and feedback. Reliability is limited by variable examiner stringency, and validity by confounding variables, but these should be viewed within the context of overall assessment strategies.
Craig, Marlies H; Sharp, Brian L; Mabaso, Musawenkosi LH; Kleinschmidt, Immo
2007-01-01
Background Several malaria risk maps have been developed in recent years, many from the prevalence of infection data collated by the MARA (Mapping Malaria Risk in Africa) project, and using various environmental data sets as predictors. Variable selection is a major obstacle due to analytical problems caused by over-fitting, confounding and non-independence in the data. Testing and comparing every combination of explanatory variables in a Bayesian spatial framework remains unfeasible for most researchers. The aim of this study was to develop a malaria risk map using a systematic and practicable variable selection process for spatial analysis and mapping of historical malaria risk in Botswana. Results Of 50 potential explanatory variables from eight environmental data themes, 42 were significantly associated with malaria prevalence in univariate logistic regression and were ranked by the Akaike Information Criterion. Those correlated with higher-ranking relatives of the same environmental theme, were temporarily excluded. The remaining 14 candidates were ranked by selection frequency after running automated step-wise selection procedures on 1000 bootstrap samples drawn from the data. A non-spatial multiple-variable model was developed through step-wise inclusion in order of selection frequency. Previously excluded variables were then re-evaluated for inclusion, using further step-wise bootstrap procedures, resulting in the exclusion of another variable. Finally a Bayesian geo-statistical model using Markov Chain Monte Carlo simulation was fitted to the data, resulting in a final model of three predictor variables, namely summer rainfall, mean annual temperature and altitude. Each was independently and significantly associated with malaria prevalence after allowing for spatial correlation. This model was used to predict malaria prevalence at unobserved locations, producing a smooth risk map for the whole country. Conclusion We have produced a highly plausible and parsimonious model of historical malaria risk for Botswana from point-referenced data from a 1961/2 prevalence survey of malaria infection in 1–14 year old children. After starting with a list of 50 potential variables we ended with three highly plausible predictors, by applying a systematic and repeatable staged variable selection procedure that included a spatial analysis, which has application for other environmentally determined infectious diseases. All this was accomplished using general-purpose statistical software. PMID:17892584
Cox, Emily; Martin, Bradley C; Van Staa, Tjeerd; Garbe, Edeltraut; Siebert, Uwe; Johnson, Michael L
2009-01-01
The goal of comparative effectiveness analysis is to examine the relationship between two variables, treatment, or exposure and effectiveness or outcome. Unlike data obtained through randomized controlled trials, researchers face greater challenges with causal inference with observational studies. Recognizing these challenges, a task force was formed to develop a guidance document on methodological approaches to addresses these biases. The task force was commissioned and a Chair was selected by the International Society for Pharmacoeconomics and Outcomes Research Board of Directors in October 2007. This report, the second of three reported in this issue of the Journal, discusses the inherent biases when using secondary data sources for comparative effectiveness analysis and provides methodological recommendations to help mitigate these biases. The task force report provides recommendations and tools for researchers to mitigate threats to validity from bias and confounding in measurement of exposure and outcome. Recommendations on design of study included: the need for data analysis plan with causal diagrams; detailed attention to classification bias in definition of exposure and clinical outcome; careful and appropriate use of restriction; extreme care to identify and control for confounding factors, including time-dependent confounding. Design of nonrandomized studies of comparative effectiveness face several daunting issues, including measurement of exposure and outcome challenged by misclassification and confounding. Use of causal diagrams and restriction are two techniques that can improve the theoretical basis for analyzing treatment effects in study populations of more homogeneity, with reduced loss of generalizability.
Chao, Linda L; Abadjian, Linda R; Esparza, Iva L; Reeb, Rosemary
2016-09-01
Despite the fact that sleep disturbances are common in veterans with Gulf War Illness (GWI), there has been a paucity of published sleep studies in this veteran population to date. Therefore, the present study examined subjective sleep quality (assessed with the Pittsburgh Sleep Quality Index), insomnia severity (assessed with the Insomnia Severity Index), and risk for obstructive sleep apnea (assessed with the STOP questionnaire) in 98 Gulf War veterans. Veterans with GWI, defined either by the Kansas or Centers for Disease Control and Prevention criteria, had greater risk for obstructive sleep apnea (i.e., higher STOP scores) than veterans without GWI. This difference persisted even after accounting for potentially confounding demographic (e.g., age, gender) and clinical variables. Veterans with GWI, defined by either the Kansas or Centers for Disease Control and Prevention criteria, also had significantly greater insomnia severity and poorer sleep quality than veterans without GWI (p < 0.05), even after accounting for potentially confounding variables. Furthermore, there were significant, positive correlations between insomnia severity, subjective sleep quality, and GWI symptom severity (p ≤ 0.01). In stepwise linear regression models, insomnia severity significantly predicted GWI status over and above demographic and clinical variables. Together these findings provide good rationale for treating sleep disturbances in the management of GWI. Reprint & Copyright © 2016 Association of Military Surgeons of the U.S.
Covariate selection with group lasso and doubly robust estimation of causal effects
Koch, Brandon; Vock, David M.; Wolfson, Julian
2017-01-01
Summary The efficiency of doubly robust estimators of the average causal effect (ACE) of a treatment can be improved by including in the treatment and outcome models only those covariates which are related to both treatment and outcome (i.e., confounders) or related only to the outcome. However, it is often challenging to identify such covariates among the large number that may be measured in a given study. In this paper, we propose GLiDeR (Group Lasso and Doubly Robust Estimation), a novel variable selection technique for identifying confounders and predictors of outcome using an adaptive group lasso approach that simultaneously performs coefficient selection, regularization, and estimation across the treatment and outcome models. The selected variables and corresponding coefficient estimates are used in a standard doubly robust ACE estimator. We provide asymptotic results showing that, for a broad class of data generating mechanisms, GLiDeR yields a consistent estimator of the ACE when either the outcome or treatment model is correctly specified. A comprehensive simulation study shows that GLiDeR is more efficient than doubly robust methods using standard variable selection techniques and has substantial computational advantages over a recently proposed doubly robust Bayesian model averaging method. We illustrate our method by estimating the causal treatment effect of bilateral versus single-lung transplant on forced expiratory volume in one year after transplant using an observational registry. PMID:28636276
Covariate selection with group lasso and doubly robust estimation of causal effects.
Koch, Brandon; Vock, David M; Wolfson, Julian
2018-03-01
The efficiency of doubly robust estimators of the average causal effect (ACE) of a treatment can be improved by including in the treatment and outcome models only those covariates which are related to both treatment and outcome (i.e., confounders) or related only to the outcome. However, it is often challenging to identify such covariates among the large number that may be measured in a given study. In this article, we propose GLiDeR (Group Lasso and Doubly Robust Estimation), a novel variable selection technique for identifying confounders and predictors of outcome using an adaptive group lasso approach that simultaneously performs coefficient selection, regularization, and estimation across the treatment and outcome models. The selected variables and corresponding coefficient estimates are used in a standard doubly robust ACE estimator. We provide asymptotic results showing that, for a broad class of data generating mechanisms, GLiDeR yields a consistent estimator of the ACE when either the outcome or treatment model is correctly specified. A comprehensive simulation study shows that GLiDeR is more efficient than doubly robust methods using standard variable selection techniques and has substantial computational advantages over a recently proposed doubly robust Bayesian model averaging method. We illustrate our method by estimating the causal treatment effect of bilateral versus single-lung transplant on forced expiratory volume in one year after transplant using an observational registry. © 2017, The International Biometric Society.
Complex Causal Process Diagrams for Analyzing the Health Impacts of Policy Interventions
Joffe, Michael; Mindell, Jennifer
2006-01-01
Causal diagrams are rigorous tools for controlling confounding. They also can be used to describe complex causal systems, which is done routinely in communicable disease epidemiology. The use of change diagrams has advantages over static diagrams, because change diagrams are more tractable, relate better to interventions, and have clearer interpretations. Causal diagrams are a useful basis for modeling. They make assumptions explicit, provide a framework for analysis, generate testable predictions, explore the effects of interventions, and identify data gaps. Causal diagrams can be used to integrate different types of information and to facilitate communication both among public health experts and between public health experts and experts in other fields. Causal diagrams allow the use of instrumental variables, which can help control confounding and reverse causation. PMID:16449586
Quantile regression models of animal habitat relationships
Cade, Brian S.
2003-01-01
Typically, all factors that limit an organism are not measured and included in statistical models used to investigate relationships with their environment. If important unmeasured variables interact multiplicatively with the measured variables, the statistical models often will have heterogeneous response distributions with unequal variances. Quantile regression is an approach for estimating the conditional quantiles of a response variable distribution in the linear model, providing a more complete view of possible causal relationships between variables in ecological processes. Chapter 1 introduces quantile regression and discusses the ordering characteristics, interval nature, sampling variation, weighting, and interpretation of estimates for homogeneous and heterogeneous regression models. Chapter 2 evaluates performance of quantile rankscore tests used for hypothesis testing and constructing confidence intervals for linear quantile regression estimates (0 ≤ τ ≤ 1). A permutation F test maintained better Type I errors than the Chi-square T test for models with smaller n, greater number of parameters p, and more extreme quantiles τ. Both versions of the test required weighting to maintain correct Type I errors when there was heterogeneity under the alternative model. An example application related trout densities to stream channel width:depth. Chapter 3 evaluates a drop in dispersion, F-ratio like permutation test for hypothesis testing and constructing confidence intervals for linear quantile regression estimates (0 ≤ τ ≤ 1). Chapter 4 simulates from a large (N = 10,000) finite population representing grid areas on a landscape to demonstrate various forms of hidden bias that might occur when the effect of a measured habitat variable on some animal was confounded with the effect of another unmeasured variable (spatially and not spatially structured). Depending on whether interactions of the measured habitat and unmeasured variable were negative (interference interactions) or positive (facilitation interactions), either upper (τ > 0.5) or lower (τ < 0.5) quantile regression parameters were less biased than mean rate parameters. Sampling (n = 20 - 300) simulations demonstrated that confidence intervals constructed by inverting rankscore tests provided valid coverage of these biased parameters. Quantile regression was used to estimate effects of physical habitat resources on a bivalve mussel (Macomona liliana) in a New Zealand harbor by modeling the spatial trend surface as a cubic polynomial of location coordinates.
2006-01-01
enabling technologies such as built-in-test, advanced health monitoring algorithms, reliability and component aging models, prognostics methods, and...deployment and acceptance. This framework and vision is consistent with the onboard PHM ( Prognostic and Health Management) as well as advanced... monitored . In addition to the prognostic forecasting capabilities provided by monitoring system power, multiple confounding errors by electronic
ERIC Educational Resources Information Center
Matlock, Ki Lynn; Turner, Ronna
2016-01-01
When constructing multiple test forms, the number of items and the total test difficulty are often equivalent. Not all test developers match the number of items and/or average item difficulty within subcontent areas. In this simulation study, six test forms were constructed having an equal number of items and average item difficulty overall.…
[Effect of social desirability on dietary intake estimated from a food questionnaire].
Barros, Renata; Moreira, Pedro; Oliveira, Bruno
2005-01-01
Self-report of dietary intake could be biased by social thus affecting risk estimates in epidemiological studies. The objective of study was to assess the effect of social desirability on dietary intake from a food frequency questionnaire (FFQ). A convenience sample of 483 Portuguese university students was recruited. Subjects were invited to complete a two-part self-administered questionnaire: the first part included the Marlowe-Crowne Social Desirability Scale (M-CSDS), a physical activity questionnaire and self-reported height and weight; the second part, included a semi-quantitative FFQ validated for Portuguese adults, that should be returned after fulfillment. All subjects completed the first part of the questionnaire and 40.4% returned the FFQ fairly completed. In multiple regression analysis, after adjustment for energy and confounders, social desirability produced a significant positive effect in the estimates of dietary fibre, vitamin C, vitamin E, magnesium and potassium, in both genders. In multiple regression, after adjustment for energy and confounders, social desirability had a significant positive effect in the estimates of vegetable consumption, for both genders, and a negative effect in white bread and beer, for women. Social desirability affected nutritional and food intake estimated from a food frequency questionnaire.
ClonEvol: clonal ordering and visualization in cancer sequencing.
Dang, H X; White, B S; Foltz, S M; Miller, C A; Luo, J; Fields, R C; Maher, C A
2017-12-01
Reconstruction of clonal evolution is critical for understanding tumor progression and implementing personalized therapies. This is often done by clustering somatic variants based on their cellular prevalence estimated via bulk tumor sequencing of multiple samples. The clusters, consisting of the clonal marker variants, are then ordered based on their estimated cellular prevalence to reconstruct clonal evolution trees, a process referred to as 'clonal ordering'. However, cellular prevalence estimate is confounded by statistical variability and errors in sequencing/data analysis, and therefore inhibits accurate reconstruction of the clonal evolution. This problem is further complicated by intra- and inter-tumor heterogeneity. Furthermore, the field lacks a comprehensive visualization tool to facilitate the interpretation of complex clonal relationships. To address these challenges we developed ClonEvol, a unified software tool for clonal ordering, visualization, and interpretation. ClonEvol uses a bootstrap resampling technique to estimate the cellular fraction of the clones and probabilistically models the clonal ordering constraints to account for statistical variability. The bootstrapping allows identification of the sample founding- and sub-clones, thus enabling interpretation of clonal seeding. ClonEvol automates the generation of multiple widely used visualizations for reconstructing and interpreting clonal evolution. ClonEvol outperformed three of the state of the art tools (LICHeE, Canopy and PhyloWGS) for clonal evolution inference, showing more robust error tolerance and producing more accurate trees in a simulation. Building upon multiple recent publications that utilized ClonEvol to study metastasis and drug resistance in solid cancers, here we show that ClonEvol rediscovered relapsed subclones in two published acute myeloid leukemia patients. Furthermore, we demonstrated that through noninvasive monitoring ClonEvol recapitulated the emerging subclones throughout metastatic progression observed in the tumors of a published breast cancer patient. ClonEvol has broad applicability for longitudinal monitoring of clonal populations in tumor biopsies, or noninvasively, to guide precision medicine. ClonEvol is written in R and is available at https://github.com/ChrisMaherLab/ClonEvol. © The Author 2017. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: journals.permissions@oup.com.
The Association Between School Bonding and Smoking Amongst Chilean Adolescents.
Gaete, Jorge; Montgomery, Alan; Araya, Ricardo
2015-01-01
The objective of the research was to study the association between school bonding dimensions (school commitment and school attachment) and current adolescent smoking in Chile, controlling for confounding variables using the fifth Chilean School Population National Substance Use Survey, 2003 (CHSS-2003) data set. The CHSS-2003 is a stratified cross-sectional survey that gathers information about personal, familial, peer, and school factors and cigarette use using a self-reported questionnaire. Complete data from 21,956 adolescent students for all the variables of interest were used in the analyses. An exploratory factor analysis (EFA) was performed in order to explore the construct validity of the questionnaire and create the main exposure and potential confounding variables. Multivariable logistic regression analyses were undertaken to study the association between school bonding and smoking. The construct validity of the school attachment and school commitment scales was mainly supported by the EFA. Multivariable analyses showed strong evidence that, after adjusting for factors from different domains, school commitment (student's good grades and school attendance) appears to have a clear inverse association with current smoking (odds ratio [OR]=0.46, 95% confidence interval [CI]: 0.38-0.56). On the other hand, school attachment (their feelings towards their school and their teachers) was not associated with adolescent smoking (OR=1.16, 95% CI: 0.88-1.53). School commitment was strongly associated with current smoking. It is important to further study this variable with the aim of ascertaining whether or not interventions that improve school commitment may prevent or reduce smoking amongst adolescent students.
NASA Astrophysics Data System (ADS)
Zeng, Chuiqing; Richardson, Murray; King, Douglas J.
2017-08-01
Remote sensing methods to study spatial and temporal changes in water quality using satellite or aerial imagery are limited by the inherently low reflectance signal of water across the visible and near infrared spectrum, as well as environmental variables such as surface scattering effects (sun glint), substrate and aquatic vegetation reflectance, and atmospheric effects. This study exploits the low altitude, high-resolution remote sensing capabilities of unmanned aerial vehicle (UAV) platforms to examine the major environmental variables that affect water reflectance acquisition, without the confounding influence of atmospheric effects typical of higher-altitude platforms. After observation and analysis, we found: (1) multiple water spectra measured at the same location had a standard deviation of 10.4%; (2) water spectra changes associated with increasing altitude from 20 m to 100 m were negligible; (3) the difference between mean reflectance at three off-shore locations in an urban water body reached 29.9%; (4) water bottom visibility increased water reflectance by 20.1% in near shore areas compared to deep water spectra in a clear water lake; (5) emergent plants caused the water spectra to shift towards a shape that is characteristic of vegetation, whereas submerged vegetation showed limited effect on water spectra in the studied lake; (6) cloud and sun glint had major effects and caused water spectra to change abruptly; while glint and shadow effects on spectra may balance each other under certain conditions, the water reflectance can also be unpredictable at times due to wave effects and their effects on lines-of-site to calm water; (7) water spectra collected under a variety of different conditions (e.g. multiple locations, waves) resulted in weaker regression models compared to spectra collected under ideal conditions (e.g. single location, no wave), although the resulting model coefficients were relatively stable. The methods and results from this study contribute to better understanding of water reflectance acquisition using remote sensing, and can be applied in UAV-based water quality assessment or to aid in validation of higher altitude imagery.
Bekkhus, Mona; Lee, Yunsung; Nordhagen, Rannveig; Magnus, Per; Samuelsen, Sven O; Borge, Anne I H
2018-02-01
Prenatal exposure to maternal anxiety has been associated with child emotional difficulties in a number of epidemiological studies. One key concern, however, is that this link is vulnerable to confounding by pleiotropic genes or environmental family factors. Data on 82 383 mothers and children from the population-based Mother and Child Cohort Study and data on 21 980 siblings were used in this study. Mothers filled out questionnaires for each unique pregnancy, for infant difficulties at 6 months and for emotional difficulties at 36 months. The link between prenatal maternal anxiety and child difficulties were examined using logistic regression analyses and multiple linear regression analyses for the full study sample and the sibling sample. In the conventional full-cohort analyses, prenatal exposure to maternal anxiety was associated with child difficulties at both 6 months [odds ratio (OR) = 2.1 (1.94-2.27)] and 36 months [OR = 2.72 (2.47-2.99)]. The findings were essentially the same whether we examined difficulties at 6 months or at 36 months. However, these associations were no longer present once we controlled for potential social and genetic confounders in the sibling comparison analyses, either at 6 months [OR = 1.32 (0.91-1.90)] or at 36 months [OR = 1.28 (0.63-2.60)]. Findings from multiple regression analyses with continuous measures were essentially the same. Our finding lends little support for there being an independent prenatal effect on child emotional difficulties; rather, our findings suggest that the link between prenatal maternal anxiety and child difficulties could be confounded by pleiotropic genes or environmental family factors. © The Author 2017; all rights reserved. Published by Oxford University Press on behalf of the International Epidemiological Association
Heine, Martin; van den Akker, Lizanne Eva; Blikman, Lyan; Hoekstra, Trynke; van Munster, Erik; Verschuren, Olaf; Visser-Meily, Anne; Kwakkel, Gert
2016-11-01
(1) To assess real-time patterns of fatigue; (2) to assess the association between a real-time fatigue score and 3 commonly used questionnaires (Checklist Individual Strength [CIS] fatigue subscale, Modified Fatigue Impact Scale (MFIS), and Fatigue Severity Scale [FSS]); and (3) to establish factors that confound the association between the real-time fatigue score and the conventional fatigue questionnaires in patients with multiple sclerosis (MS). Cross-sectional study. MS-specialized outpatient facility. Ambulant patients with MS (N=165) experiencing severe self-reported fatigue. Not applicable. A real-time fatigue score was assessed by sending participants 4 text messages on a particular day (How fatigued do you feel at this moment?; score range, 0-10). Latent class growth mixed modeling was used to determine diurnal patterns of fatigue. Regression analyses were used to assess the association between the mean real-time fatigue score and the CIS fatigue subscale, MFIS, and FSS. Significant associations were tested for candidate confounders (eg, disease severity, work status, sleepiness). Four significantly different fatigue profiles were identified by the real-time fatigue score, namely a stable high (n=79), increasing (n=57), stable low (n=16), and decreasing (n=13). The conventional questionnaires correlated poorly (r<.300) with the real-time fatigue score. The Epworth Sleepiness Scale significantly reduced the regression coefficient between the real-time fatigue score and conventional questionnaires, ranging from 15.4% to 35%. Perceived fatigue showed 4 different diurnal patterns in patients with MS. Severity of sleepiness is an important confounder to take into account in the assessment of fatigue. Copyright © 2016 American Congress of Rehabilitation Medicine. Published by Elsevier Inc. All rights reserved.
Stability and change of body mass index as a predictor of disability pension.
Ropponen, Annina; Silventoinen, Karri; Koskenvuo, Markku; Svedberg, Pia; Kaprio, Jaakko
2016-06-01
To investigate whether stability or change in body mass index (BMI) predict disability pension (DP) due to musculoskeletal diagnosis (MSD) when controlling for familial confounding. Our study cohort consisted of 17,169 Finnish twins born before 1958. Data on BMI and multiple covariates from questionnaires in 1975 and 1981 were included and DPs were collected from the national pension registers until the end of 2004. Cox proportional hazards regression models with Hazard Ratios (HR) and 95% Confidence Intervals (CI) were used for statistical analyses. General DP was granted to 2853 individuals and DP due to MSD to 1143 individuals during the 23-year follow-up. A one-unit increase in BMI in both 1975 (HR 1.08, 95% CI 1.05, 1.10) and 1981 (HR 1.06, 95% CI 1.04, 1.07), as well as the stability of and change in BMI from 1975 to 1981 were all associated with an increased risk of DP. These associations held in the analyses controlling for multiple covariates (age, sex, socioeconomic status, education, marital status, leisure-time physical activity, and musculoskeletal pain), and mainly also familial confounding, that is, genetics and shared environment. HR for stable obesity was 2.28 (95% CI 1.69, 3.08) for DP due to MSD, and 1.91 (95% CI 1.56, 2.34) for general DP in the fully adjusted models. CONCLUSIONS BMI IS AN EARLY PREDICTOR OF GENERAL DP AND ALSO OF DP DUE TO MSD OWING TO THE INDEPENDENCY OF VARIOUS COVARIATES AND POTENTIALLY ALSO FAMILIAL CONFOUNDING, BMI MAY POSSIBLY HAVE A DIRECT EFFECT ON THE RISK OF DP. © 2016 the Nordic Societies of Public Health.
Bark analysis as a guide to cassava nutrition in Sierra Leone
DOE Office of Scientific and Technical Information (OSTI.GOV)
Godfrey-Sam-Aggrey, W.; Garber, M.J.
1979-01-01
Cassava main stem barks from two experiments in which similar fertilizers were applied directly in a 2/sup 5/ confounded factorial design were analyzed and the bark nutrients used as a guide to cassava nutrition. The application of multiple regression analysis to the respective root yields and bark nutrient concentrations enable nutrient levels and optimum adjusted root yields to be derived. Differences in bark nutrient concentrations reflected soil fertility levels. Bark analysis and the application of multiple regression analysis to root yields and bark nutrients appear to be useful tools for predicting fertilizer recommendations for cassava production.
Beyond greening and browning: the need for an integrated understanding of Arctic change
NASA Astrophysics Data System (ADS)
Gamon, J. A.; Huemmrich, K. F.; Hmimina, G.; Yu, R.
2017-12-01
Satellite records and field observations povide contradictory evidence for "greening" or "browning" of Arctic tundra. Large-scale observations of apparent greening have been based on satellite vegetation indices (e.g NDVI). However, a clear interpretation of these trends are confounded by changing snow cover and surface hydrology, both of which influence NDVI and are known to be changing independently of any direct vegetation response. Field studies have demonstrated greening in some areas, but not others, and have also documented changing permafrost depth, surface hydrology and snow cover. Together, these confounding factors can explain some of the contradictory evidence based regarding greening and browning. Given the multiple influences on Arctic NDVI, simple conclusions regarding greening and browning from satellite data alone can be incorrect; when these confounding factors are taken into account, some areas that show apparent greening in the satellite record appear to be undergoing productivity declines due to surface drying. These contradictory interpretations have profound implications for our understanding of changing surface energy balance, biogeochemistry, and surface-atmosphere feedbacks. To better address Arctic ecosystem responses to a changing climate, an integrated, multi-scale, multivariate approach that considers hydrology, permafrost, snow cover and vegetation is needed.
Halliday, Drew W R; Stawski, Robert S; MacDonald, Stuart W S
2017-02-01
Response time inconsistency (RTI) in cognitive performance predicts deleterious health outcomes in late-life; however, RTI estimates are often confounded by additional influences (e.g., individual differences in learning). Finger tapping is a basic sensorimotor measure largely independent of higher-order cognition that may circumvent such confounds of RTI estimates. We examined the within-person coupling of finger-tapping mean and RTI on working memory, and the moderation of these associations by cognitive status. A total of 262 older adults were recruited and classified as controls, cognitively-impaired-not-demented (CIND) unstable or CIND stable. Participants completed finger-tapping and working-memory tasks during multiple weekly assessments, repeated annually for 4 years. Within-person coupling estimates from multilevel models indicated that on occasions when RTI was greater, working-memory response latency was slower for the CIND-stable, but not for the CIND-unstable or control individuals. The finger-tapping task shows potential for minimizing confounds on RTI estimates, and for yielding RTI estimates sensitive to central nervous system function and cognitive status. © The Author 2016. Published by Oxford University Press. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com.
Truthmann, Julia; Mensink, Gert B M; Bosy-Westphal, Anja; Hapke, Ulfert; Scheidt-Nave, Christa; Schienkiewitz, Anja
2017-06-10
This study examined sex-specific differences in physical health-related quality of life (HRQoL) across subgroups of metabolic health and obesity. We specifically asked whether (1) obesity is related to lower HRQoL independent of metabolic health status and potential confounders, and (2) whether associations are similar in men and women. We used cross-sectional data from the German Health Interview and Examination Survey 2008-11. Physical HRQoL was measured using the Short Form-36 version 2 physical component summary (PCS) score. Based on harmonized ATPIII criteria for the definition of the metabolic health and a body mass index ≥ 30 kg/m 2 to define obesity, individuals were classified as metabolically healthy non-obese (MHNO), metabolically unhealthy non-obese (MUNO), metabolically healthy obese (MHO), and metabolically unhealthy obese (MUO). Sex-specific analyses including multivariable linear regression analyses were based on PCS as the dependent variable, metabolic health and obesity category as the independent variable with three categories and MHNO as the reference, and age, education, lifestyle and comorbidities as confounders. This study included 6860 participants (3298 men, 3562 women). Compared to MHNO, all other metabolic health and obesity categories had significantly lower PCS in both sexes. As reflected by the beta coefficients [95% confidence interval] from bivariable linear regression models, a significant inverse association with PCS was strongest for MUO (men: -7.0 [-8.2; -5.8]; women: -9.0 [-10.2; -7.9]), intermediate for MUNO (men: -4.2 [-5.3; -3.1]; women: -5.6 [-6.8; -4.4]) and least pronounced for MHO (men: -2.2 [-3.6; -0.8]; women -3.9 [-5.4; -2.5]). Differences in relation to MHNO remained statistically significant for all groups after adjusting for confounders, but decreased in particular for MUNO (men:-1.3 [-2.3; -0.3]; women: -1.5 [-2.7; -0.3]. Obesity was significantly related to lower physical HRQoL, independent of metabolic health status. Potential confounders including age, educational status, health-related behaviors, and comorbidities explained parts of the inverse relationship. Associations were evident in both sexes and consistently more pronounced among women than men.
ERIC Educational Resources Information Center
Soh, Kay Cheng
2012-01-01
University ranking has become ritualistic in higher education. Ranking results are taken as bona fide by rank users. Ranking systems usually use large data sets from highly heterogeneous universities of varied backgrounds. This poses the problem of Simpson's Paradox and the lurking variables causing it. Using QS 2011-2012 Ranking data, the dual…
ERIC Educational Resources Information Center
Van Lieshout, Ryan J.; Schmidt, Louis A.; Robinson, Monique; Niccols, Alison; Boyle, Michael H.
2013-01-01
Recent research suggests that fetal exposure to increased maternal body mass index (BMI) during pregnancy may be associated with psychopathology later in life. When this link first emerges, and if it is due to intrauterine exposures or confounding variables is not known. We therefore assessed associations between maternal pre-pregnancy BMI and:…
The Effects of Age on Second Language Grammar and Speech Production
ERIC Educational Resources Information Center
Huang, Becky H.
2014-01-01
The current study examined the age of learning effect on second language (L2) acquisition. The research goals of the study were twofold: to test whether there is an independent age effect controlling for other potentially confounding variables, and to clarify the age effect across L2 grammar and speech production domains. The study included 118…
Reduced streamflow lowers dry-season growth of rainbow trout in a small stream
Bret C. Harvey; Rodney J. Nakamoto; Jason L. White
2006-01-01
A wide variety of resource management activities can affect surface discharge in small streams. Often, the effects of variation in streamflow on fish survival and growth can be difficult to estimate because of possible confounding with the effects of other variables, such as water temperature and fish density. We measured the effect of streamflow on survival and growth...
John R. Donnelly
1971-01-01
Softwood stem cuttings from three mature sugar maple trees were treated with several types and concentrations of growth regulators. Lack of statistical significance was due to extreme variability in tree response: low levels of auxin stimulated rooting in two study trees, while auxins inhibited rooting in the other tree. It is postulated that variations in rooting...
ERIC Educational Resources Information Center
Licht, Carolyn; Chabot, David
2006-01-01
Bowen's theory (1978) of differentiation is the personality variable most critical to mature development and the attainment of psychological health. The study of psychological differentiation has been confounded by confusion regarding the constructs being studied and the measurement tools used to assess them. The purpose of this review is to…
Measuring the wetting angle and perimeter of single wood pulp fibers : a modified method
John H. Klungness
1981-01-01
In pulp processing development it is often necessary to measure the effect of a process variable on individual pulp fiber wettability. Such processes would include drying of market pulps, recycling of secondary fibers, and surface modification of fibers as in sizing. However, if wettability is measured on a fiber sheet surface, the results are confounded by...
ERIC Educational Resources Information Center
Hevel, Michael S.; Martin, Georgianna L.; Pascarella, Ernest T.
2014-01-01
We use a longitudinal national dataset to explore the effects of fraternity and sorority membership on students' socially responsible leadership in the fourth year of college. Controlling for a variety of potentially confounding variables, we find no net effect. These findings run counter to an earlier study based on the same students that found…
ERIC Educational Resources Information Center
Boniface, Russell; Protheroe, Nancy
Class-size reduction (CSR) has been a complex and contentious issue for the last quarter century. Although the small-class concept was adopted because it appealed to common sense, research over time has revealed a mix of confounding variables, instead of a definitive conclusion. Some CSR efforts, such as Tennessee's Project STAR and Wisconsin's…
Belek, I
2000-02-01
The aim of this study is to determine the separate effects of social class, income, education and area of residence on psychological distress. The study also assesses whether the association between prevalence of high score on the 12-item General Health Questionnaire (GHQ 12) and social class is independent of other variables. Psychological distress was assessed by means of the GHQ 12. The study covered 1,092 adults aged 15 years or more living in two different quarters of Antalya. Social class status was defined by occupational position, with income, education and area of residence treated as confounders. Chi-square and logistic regression analyses were used to evaluate the data. Large inequalities in psychological distress by all variables were observed. Psychological distress was significantly associated with class status, after adjusting for income, education, area of residence and other potential confounders (age, sex and marital status). Class inequalities in psychological distress were observed between blue-collar workers/unqualified employees and bourgeoisie. These findings support the view that the recent widening of inequalities among social classes in Turkey pose a substantial threat to health.
Santos, Daniela M; Lage, Laís V; Jabur, Eleonora K; Kaziyama, Helena H S; Iosifescu, Dan V; De Lucia, Mara Cristina S; Fráguas, Renério
2017-01-01
We developed this study to investigate the association of fibromyalgia with personality traits, controlling for depression and other potential confounders. We assessed personality traits using the Cloninger's Temperament and Character Inventory (TCI) in 78 female patients with fibromyalgia and in a control group of 78 subjects without fibromyalgia. The Mini-International Neuropsychiatric Interview was used to assess depression and anxiety diagnoses. To investigate the association between fibromyalgia and the Cloninger's Temperament and Character Inventory we performed unadjusted and adjusted analyses of covariance, using the TCI score as dependent variable and adjusting the model for depression, anxiety and for clinical and socio-demographic variables. We used a backward selection method to choose the final model. In the unadjusted analysis, fibromyalgia was associated with all personality traits, except persistency. After adjusting for depression and anxiety, patients with fibromyalgia presented decreased novelty seeking compared to controls; the differences in other personality traits were no longer significant. Novelty seeking was also correlated with the length of history of fibromyalgia and pain intensity. Decreased novelty seeking may be a personality trait associated with fibromyalgia. Depression and anxiety should be considered potential confounders in the evaluation of personality traits in this population.
Lee, Sunmin; Zhai, Shumenghui; Zhang, Guo (Yolanda); Ma, Xiang S; Lu, Xiaoxiao; Tan, Yin; Siu, Philip; Seals, Brenda; Ma, Grace X
2015-01-01
BACKGROUND Hepatitis C virus (HCV) is a major cause of chronic liver disease and cancer. Vietnamese Americans are at high risk of HCV infection, with men having the highest US incidence of liver cancer. This study examines an intervention to improve HCV knowledge among Vietnamese Americans. STUDY Seven Vietnamese community-based organizations in Pennsylvania and New Jersey recruited a total of 306 Vietnamese participants from 2010 to 2011. RESULTS Average knowledge scores for pretest and posttest were 3.32 and 5.88, respectively (maximum 10). After adjusting for confounding variables, age and higher education were positively associated with higher pretest scores and having a physician who spoke English or Vietnamese was negatively associated with higher pretest scores. Additionally, after adjusting for confounding variables, household income, education, and having an HCV-infected family member significantly increased knowledge scores. CONCLUSIONS Promotion and development of HCV educational programs can increase HCV knowledge among race and ethnic groups, such as Vietnamese Americans. Giving timely information to at-risk groups provides the opportunity to correct misconceptions, decrease HCV risk behaviors, and encourage testing that might improve timely HCV diagnosis and treatment. PMID:26561280
Genetic Instrumental Variable Studies of Effects of Prenatal Risk Factors
von Hinke Kessler Scholder, Stephanie
2013-01-01
Identifying the effects of maternal risk factors during pregnancy on infant and child health is an area of tremendous research interest. However, of interest to policy makers is unraveling the causal effects of prenatal risk factors, not their associations with child health, which may be confounded by several unobserved factors. In this paper, we evaluate the utility of genetic variants in three genes that have unequivocal evidence of being related to three major risk factors – CHRNA3 for smoking, ADH1B for alcohol use, and FTO for obesity – as instrumental variables for identifying the causal effects of such factors during pregnancy. Using two independent datasets, we find that these variants are overall predictive of the risk factors and are not systematically related to observed confounders, suggesting that they may be useful instruments. We also find some suggestive evidence that genetic effects are stronger during than before pregnancy. We provide an empirical example illustrating the use of these genetic variants as instruments to evaluate the effects of risk factors on birth weight. Finally, we offer suggestions for researchers contemplating the use of these variants as instruments. PMID:23701534
McGuire, Connor; Kristman, Vicki L; Williams-Whitt, Kelly; Reguly, Paula; Shaw, William; Soklaridis, Sophie
2015-01-01
PURPOSE To determine the association between supervisors’ leadership style and autonomy and supervisors’ likelihood of supporting job accommodations for back-injured workers. METHODS A cross-sectional study of supervisors from Canadian and US employers was conducted using a web-based, self-report questionnaire that included a case vignette of a back-injured worker. Autonomy and two dimensions of leadership style (considerate and initiating structure) were included as exposures. The outcome, supervisors’ likeliness to support job accommodation, was measured with the Job Accommodation Scale. We conducted univariate analyses of all variables and bivariate analyses of the JAS score with each exposure and potential confounding factor. We used multivariable generalized linear models to control for confounding factors. RESULTS A total of 796 supervisors participated. Considerate leadership style (β= .012; 95% CI: .009–.016) and autonomy (β= .066; 95% CI: .025–.11) were positively associated with supervisors’ likelihood to accommodate after adjusting for appropriate confounding factors. An initiating structure leadership style was not significantly associated with supervisors’ likelihood to accommodate (β = .0018; 95% CI: −.0026–.0061) after adjusting for appropriate confounders. CONCLUSIONS Autonomy and a considerate leadership style were positively associated with supervisors’ likelihood to accommodate a back-injured worker. Providing supervisors with more autonomy over decisions of accommodation and developing their considerate leadership style may aid in increasing work accommodation for back-injured workers and preventing prolonged work disability. PMID:25595332
Reporting of methodological features in observational studies of pre-harvest food safety.
Sargeant, Jan M; O'Connor, Annette M; Renter, David G; Kelton, David F; Snedeker, Kate; Wisener, Lee V; Leonard, Erin K; Guthrie, Alessia D; Faires, Meredith
2011-02-01
Observational studies in pre-harvest food safety may be useful for identifying risk factors and for evaluating potential mitigation strategies to reduce foodborne pathogens. However, there are no structured reporting guidelines for these types of study designs in livestock species. Our objective was to evaluate the reporting of observational studies in the pre-harvest food safety literature using guidelines modified from the human healthcare literature. We identified 100 pre-harvest food safety studies published between 1999 and 2009. Each study was evaluated independently by two reviewers using a structured checklist. Of the 38 studies that explicitly stated the observational study design, 27 were described as cross-sectional studies, eight as case-control studies, and three as cohort studies. Study features reported in over 75% of the selected studies included: description of the geographic location of the studies, definitions and sources of data for outcomes, organizational level and source of data for independent variables, description of statistical methods and results, number of herds enrolled in the study and included in the analysis, and sources of study funding. However, other features were not consistently reported, including details related to eligibility criteria for groups (such as barn, room, or pen) and individuals, numbers of groups and individuals included in various stages of the study, identification of primary outcomes, the distinction between putative risk factors and confounding variables, the identification of a primary exposure variable, the referent level for evaluation of categorical variable associations, methods of controlling confounding variables and missing variables, model fit, details of subset analysis, demographic information at the sampling unit level, and generalizability of the study results. Improvement in reporting of observational studies of pre-harvest food safety will aid research readers and reviewers in interpreting and evaluating the results of such studies. Copyright © 2010 Elsevier B.V. All rights reserved.
Do, D. Phuong; Wang, Lu; Elliott, Michael R.
2013-01-01
Extant observational studies generally support the existence of a link between neighborhood context and health. However, estimating the causal impact of neighborhood effects from observational data has proven to be a challenge. Omission of relevant factors may lead to overestimating the effects of neighborhoods on health while inclusion of time-varying confounders that may also be mediators (e.g., income, labor force status) may lead to underestimation. Using longitudinal data from the 1990 to 2007 years of the Panel Study of Income Dynamics, this study investigates the link between neighborhood poverty and overall mortality risk. A marginal structural modeling strategy is employed to appropriately adjust for simultaneous mediating and confounding factors. To address the issue of possible upward bias from the omission of key variables, sensitivity analysis to assess the robustness of results against unobserved confounding is conducted. We examine two continuous measures of neighborhood poverty – single-point and a running average. Both were specified as piece-wise linear splines with a knot at 20 percent. We found no evidence from the traditional naïve strategy that neighborhood context influences mortality risk. In contrast, for both the single-point and running average neighborhood poverty specifications, the marginal structural model estimates indicated a statistically significant increase in mortality risk with increasing neighborhood poverty above the 20 percent threshold. For example, below 20 percent neighborhood poverty, no association was found. However, after the 20 percent poverty threshold is reached, each 10 percentage point increase in running average neighborhood poverty was found to increase the odds for mortality by 89 percent [95% CI = 1.22, 2.91]. Sensitivity analysis indicated that estimates were moderately robust to omitted variable bias. PMID:23849239
A community study on emotional distress among Arab and Jewish Israelis over the age of 60.
Shemesh, Annarosa Anat; Kohn, Robert; Blumstein, Tzvia; Geraisy, Nabil; Novikov, Ilya; Levav, Itzhak
2006-01-01
The elderly constitute a vulnerable group for psychopathology, yet research on their mental health among both Arab and Jews in Israel remains limited. The same is the case in Arab countries. This paper reports on the contrasting distribution of the mean emotional distress (ED) scores and rates of suspected clinical cases, and their related risk factors, among community residents over the age of 60. Several national agencies conducted a survey on 5,055 elderly individuals to investigate their health status, including ED. The interview included socio-demographic and behavioral health items, as well as a modified 12 item-GHQ as a measure of ED. Total ED scores and prevalence rates for suspected psychopathology were calculated. Their respective risk factors were examined using univariate and multivariate methods. The ED scores were highest among Muslim Arabs (4.9), followed by Christian Arabs (4.2), Jews (3.1) and Druzes (2.8). Their estimated prevalence rates were 43.4%, 37.0%, 21.4%, and 17.0%, respectively. The gradient of these results remained unchanged in the multivariate analysis for ED scores adjusting for confounding variables. In contrast, logistic regression analysis controlling for confounding variables did not find a differential risk for suspected psychopathology between Arabs and Jews. Conceivably, the higher demoralization scores among elderly Arabs are associated with their minority status affiliation, as well as with the rapid social changes that have taken place in their midst. A cultural response style may be entertained as a possible explanation. However, these factors do not impact the risk for suspected psychopathology where no differential risk was noted after adjustments for confounders. Copyright 2005 John Wiley & Sons, Ltd.
Shakiba, Maryam; Mansournia, Mohammad Ali; Salari, Arsalan; Soori, Hamid; Mansournia, Nasrin; Kaufman, Jay S
2018-06-01
In longitudinal studies, standard analysis may yield biased estimates of exposure effect in the presence of time-varying confounders that are also intermediate variables. We aimed to quantify the relationship between obesity and coronary heart disease (CHD) by appropriately adjusting for time-varying confounders. This study was performed in a subset of participants from the Atherosclerosis Risk in Communities (ARIC) Study (1987-2010), a US study designed to investigate risk factors for atherosclerosis. General obesity was defined as body mass index (weight (kg)/height (m)2) ≥30, and abdominal obesity (AOB) was defined according to either waist circumference (≥102 cm in men and ≥88 cm in women) or waist:hip ratio (≥0.9 in men and ≥0.85 in women). The association of obesity with CHD was estimated by G-estimation and compared with results from accelerated failure-time models using 3 specifications. The first model, which adjusted for baseline covariates, excluding metabolic mediators of obesity, showed increased risk of CHD for all obesity measures. Further adjustment for metabolic mediators in the second model and time-varying variables in the third model produced negligible changes in the hazard ratios. The hazard ratios estimated by G-estimation were 1.15 (95% confidence interval (CI): 0.83, 1.47) for general obesity, 1.65 (95% CI: 1.35, 1.92) for AOB based on waist circumference, and 1.38 (95% CI: 1.13, 1.99) for AOB based on waist:hip ratio, suggesting that AOB increased the risk of CHD. The G-estimated hazard ratios for both measures were further from the null than those derived from standard models.
Harvey, Samuel B; Sellahewa, Dilan A; Wang, Min-Jung; Milligan-Saville, Josie; Bryan, Bridget T; Henderson, Max; Hatch, Stephani L; Mykletun, Arnstein
2018-06-01
Long-standing concerns exist about reverse causation and residual confounding in the prospective association between job strain and risk of future common mental disorders. We aimed to address these concerns through analysis of data collected in the UK National Child Development Study, a large British cohort study. Data from the National Child Development Study (n=6870) were analysed by use of multivariate logistic regression to investigate the prospective association between job strain variables at age 45 years and risk of future common mental disorders at age 50 years, controlling for lifetime psychiatric history and a range of other possible confounding variables across the lifecourse. Population attributable fractions were calculated to estimate the public health effect of job strain on midlife mental health. In the final model, adjusted for all measured confounders, high job demands (odds ratio 1·70, 95% CI 1·25-2·32; p=0·0008), low job control (1·89, 1·29-2·77; p=0·0010), and high job strain (2·22, 1·59-3·09; p<0·0001) remained significant independent predictors of future onset of common mental disorder. If causality is assumed, our findings suggest that 14% of new cases of common mental disorder could have been prevented through elimination of high job strain (population attributable fraction 0·14, 0·06-0·20). High job strain appears to independently affect the risk of future common mental disorders in midlife. These findings suggest that modifiable work-related risk factors might be an important target in efforts to reduce the prevalence of common mental disorders. iCare Foundation and Mental Health Branch, NSW Health. Copyright © 2018 Elsevier Ltd. All rights reserved.
Knowlden, Adam P; Burns, Maranda; Harcrow, Andy; Shewmake, Meghan E
2016-03-16
Poor sleep quality is a significant public health problem. The role of nutrition in predicting sleep quality is a relatively unexplored area of inquiry. The purpose of this study was to evaluate the capacity of 10 food choice categories, sleep confounding beverages, and psychological distress to predict the sleep quality of college students. A logistic regression model comprising 10 food choice variables (healthy proteins, unhealthy proteins, healthy dairy, unhealthy dairy, healthy grains, unhealthy grains, healthy fruits and vegetables, unhealthy empty calories, healthy beverages, unhealthy beverages), sleep confounding beverages (caffeinated/alcoholic beverages), as well as psychological distress (low, moderate, serious distress) was computed to determine the capacity of the variables to predict sleep quality (good/poor). The odds of poor sleep quality were 32.4% lower for each unit of increased frequency of healthy proteins consumed (p<0.001; OR=0.676), 14.1% lower for each unit of increased frequency of healthy dairy food choices consumed (p=0.024; OR=0.859), 13.1% higher for each unit of increased frequency of empty calorie food choices consumed (p=0.003; OR=1.131), and 107.3% higher for those classified in the moderate psychological distress (p=0.016; OR=2.073). Collectively, healthy proteins, healthy dairy, unhealthy empty calories, and moderate psychological distress were moderately predictive of sleep quality in the sample (Nagelkerke R2=23.8%). Results of the study suggested higher frequency of consumption of healthy protein and healthy dairy food choices reduced the odds of poor sleep quality, while higher consumption of empty calories and moderate psychological distress increased the odds of poor sleep quality.
Staffa, Steven J; Zurakowski, David
2018-01-09
In clinical research, the gold standard level of evidence is the randomized controlled trial (RCT). The availability of nonrandomized retrospective data is growing; however, a primary concern of analyzing such data is comparability of the treatment groups with respect to confounding variables. Propensity score matching (PSM) aims to equate treatment groups with respect to measured baseline covariates to achieve a comparison with reduced selection bias. It is a valuable statistical methodology that mimics the RCT, and it may create an "apples to apples" comparison while reducing bias due to confounding. PSM can improve the quality of anesthesia research and broaden the range of research opportunities. PSM is not necessarily a magic bullet for poor-quality data, but rather may allow the researcher to achieve balanced treatment groups similar to a RCT when high-quality observational data are available. PSM may be more appealing than the common approach of including confounders in a regression model because it allows for a more intuitive analysis of a treatment effect between 2 comparable groups.We present 5 steps that anesthesiologists can use to successfully implement PSM in their research with an example from the 2015 Pediatric National Surgical Quality Improvement Program: a validated, annually updated surgery and anesthesia pediatric database. The first step of PSM is to identify its feasibility with regard to the data at hand and ensure availability of data on any potential confounders. The second step is to obtain the set of propensity scores from a logistic regression model with treatment group as the outcome and the balancing factors as predictors. The third step is to match patients in the 2 treatment groups with similar propensity scores, balancing all factors. The fourth step is to assess the success of the matching with balance diagnostics, graphically or analytically. The fifth step is to apply appropriate statistical methodology using the propensity-matched data to compare outcomes among treatment groups.PSM is becoming an increasingly more popular statistical methodology in medical research. It often allows for improved evaluation of a treatment effect that may otherwise be invalid due to a lack of balance between the 2 treatment groups with regard to confounding variables. PSM may increase the level of evidence of a study and in turn increases the strength and generalizability of its results. Our step-by-step approach provides a useful strategy for anesthesiologists to implement PSM in their future research.
Park, Soojin; Steiner, Peter M; Kaplan, David
2018-06-01
Considering that causal mechanisms unfold over time, it is important to investigate the mechanisms over time, taking into account the time-varying features of treatments and mediators. However, identification of the average causal mediation effect in the presence of time-varying treatments and mediators is often complicated by time-varying confounding. This article aims to provide a novel approach to uncovering causal mechanisms in time-varying treatments and mediators in the presence of time-varying confounding. We provide different strategies for identification and sensitivity analysis under homogeneous and heterogeneous effects. Homogeneous effects are those in which each individual experiences the same effect, and heterogeneous effects are those in which the effects vary over individuals. Most importantly, we provide an alternative definition of average causal mediation effects that evaluates a partial mediation effect; the effect that is mediated by paths other than through an intermediate confounding variable. We argue that this alternative definition allows us to better assess at least a part of the mediated effect and provides meaningful and unique interpretations. A case study using ECLS-K data that evaluates kindergarten retention policy is offered to illustrate our proposed approach.
Wong, Shane Shucheng; Zhou, Bo; Goebert, Deborah; Hishinuma, Earl S
2013-10-01
Substance use is associated with suicidal ideation, planning and attempts among adolescents, but it is unclear how this association varies across different types and number of substances. This study examined the association between patterns of substance use and suicidality among a nationally representative sample of high school students in the United States during the last decade. Data from the 2001 to 2009 Youth Risk Behavior Survey including 73,183 high school students were analyzed. Logistic regression analyses examined the association between lifetime use of ten common substances of abuse (alcohol, cocaine, ecstasy, hallucinogens, heroin, inhalants, marijuana, methamphetamines, steroids, and tobacco) and four measures of suicidality over the last year (suicidal ideation, suicide plan, suicide attempt, and severe suicide attempt requiring medical attention), controlling for potential confounders (socio-demographic variables, interpersonal violence, sexual intercourse, and symptoms of depression and eating disorder). Among the ten substances, univariate analysis demonstrates that adolescents reporting a history of heroin use have the strongest association with suicidal ideation, suicide plan, suicide attempts and severe suicide attempts in the last year (odds ratio = 5.0, 5.9, 12.0, and 23.6 compared to non-users), followed by users of methamphetamines (OR = 4.3-13.1) and steroids (OR = 3.7-11.8). Cocaine, ecstasy, hallucinogens and inhalants had a moderate association with suicidality (OR = 3.1-10.8). Users of marijuana, alcohol and tobacco also had an increased odds ratio of suicidality (OR = 1.9-5.2). The association between each of ten substances and the four measures of suicidality remained significant with multivariate analysis controlling for multiple confounders (p < 0.05), except for the association between alcohol use and severe suicide attempts. The seven illicit substances had a stronger association with severe suicide attempts as compared to all other confounding risk factors except depression. The number of substances used had a graded relationship to suicidality. Substance abuse is a strong risk factor for suicidal thoughts and behaviors among American high school students, with the strength of this relationship dramatically increasing with particular illicit drugs and a higher number of substances. The findings reinforce the importance of routine screening for substance abuse in the assessment of adolescent suicide risk.
An appraisal of statistical procedures used in derivation of reference intervals.
Ichihara, Kiyoshi; Boyd, James C
2010-11-01
When conducting studies to derive reference intervals (RIs), various statistical procedures are commonly applied at each step, from the planning stages to final computation of RIs. Determination of the necessary sample size is an important consideration, and evaluation of at least 400 individuals in each subgroup has been recommended to establish reliable common RIs in multicenter studies. Multiple regression analysis allows identification of the most important factors contributing to variation in test results, while accounting for possible confounding relationships among these factors. Of the various approaches proposed for judging the necessity of partitioning reference values, nested analysis of variance (ANOVA) is the likely method of choice owing to its ability to handle multiple groups and being able to adjust for multiple factors. Box-Cox power transformation often has been used to transform data to a Gaussian distribution for parametric computation of RIs. However, this transformation occasionally fails. Therefore, the non-parametric method based on determination of the 2.5 and 97.5 percentiles following sorting of the data, has been recommended for general use. The performance of the Box-Cox transformation can be improved by introducing an additional parameter representing the origin of transformation. In simulations, the confidence intervals (CIs) of reference limits (RLs) calculated by the parametric method were narrower than those calculated by the non-parametric approach. However, the margin of difference was rather small owing to additional variability in parametrically-determined RLs introduced by estimation of parameters for the Box-Cox transformation. The parametric calculation method may have an advantage over the non-parametric method in allowing identification and exclusion of extreme values during RI computation.
Selection of higher order regression models in the analysis of multi-factorial transcription data.
Prazeres da Costa, Olivia; Hoffman, Arthur; Rey, Johannes W; Mansmann, Ulrich; Buch, Thorsten; Tresch, Achim
2014-01-01
Many studies examine gene expression data that has been obtained under the influence of multiple factors, such as genetic background, environmental conditions, or exposure to diseases. The interplay of multiple factors may lead to effect modification and confounding. Higher order linear regression models can account for these effects. We present a new methodology for linear model selection and apply it to microarray data of bone marrow-derived macrophages. This experiment investigates the influence of three variable factors: the genetic background of the mice from which the macrophages were obtained, Yersinia enterocolitica infection (two strains, and a mock control), and treatment/non-treatment with interferon-γ. We set up four different linear regression models in a hierarchical order. We introduce the eruption plot as a new practical tool for model selection complementary to global testing. It visually compares the size and significance of effect estimates between two nested models. Using this methodology we were able to select the most appropriate model by keeping only relevant factors showing additional explanatory power. Application to experimental data allowed us to qualify the interaction of factors as either neutral (no interaction), alleviating (co-occurring effects are weaker than expected from the single effects), or aggravating (stronger than expected). We find a biologically meaningful gene cluster of putative C2TA target genes that appear to be co-regulated with MHC class II genes. We introduced the eruption plot as a tool for visual model comparison to identify relevant higher order interactions in the analysis of expression data obtained under the influence of multiple factors. We conclude that model selection in higher order linear regression models should generally be performed for the analysis of multi-factorial microarray data.
Toward a clearer portrayal of confounding bias in instrumental variable applications.
Jackson, John W; Swanson, Sonja A
2015-07-01
Recommendations for reporting instrumental variable analyses often include presenting the balance of covariates across levels of the proposed instrument and levels of the treatment. However, such presentation can be misleading as relatively small imbalances among covariates across levels of the instrument can result in greater bias because of bias amplification. We introduce bias plots and bias component plots as alternative tools for understanding biases in instrumental variable analyses. Using previously published data on proposed preference-based, geography-based, and distance-based instruments, we demonstrate why presenting covariate balance alone can be problematic, and how bias component plots can provide more accurate context for bias from omitting a covariate from an instrumental variable versus non-instrumental variable analysis. These plots can also provide relevant comparisons of different proposed instruments considered in the same data. Adaptable code is provided for creating the plots.
Prevalence and factors associated with sedentary behavior in adolescents
Dias, Paula Jaudy Pedroso; Domingos, Isabela Prado; Ferreira, Márcia Gonçalves; Muraro, Ana Paula; Sichieri, Rosely; Gonçalves-Silva, Regina Maria Veras
2014-01-01
OBJECTIVE To analyze the prevalence of sedentary behavior and associated factors in adolescents. METHODS A cross-sectional study with adolescents aged 10 to 17 years, of both sexes, belonging to a 1994-1999 birth cohort in the city of Cuiabá, MT, Central Western Brazil. Data were collected using a questionnaire containing sociodemographic, economic, lifestyle and anthropometric variables. Sedentary behavior was determined as using television and/or computer/video games for a time greater than or equal to 4 hours/day. Associations with sedentary behavior were evaluated using body mass index in childhood and adolescence and sociodemographic and behavioral variables using hierarchical logistic regression. RESULTS The overall prevalence of sedentary behavior was 58.1%. Of the 1,716 adolescents evaluated, 50.7% (n = 870) were male. In multivariate analysis, after adjustment for confounding factors, the variables that remained associated with sedentary behavior were: age (14 and over) (OR = 3.51, 95%CI 2.19;5.60); higher socioeconomic class (OR = 3.83, 95%CI 2.10;7.01), higher level of maternal education (OR = 1.81, 95%CI 1.09;3.01); living in the country (OR = 0.49, 95%CI 0.30;0.81); insufficient physical activity (OR = 1.25, 95%CI 1.02;1.53); experimentation with alcoholic beverages (OR = 1.34, 95%CI 1.08;1.66) and being overweight in adolescence (OR = 1.33, 95%CI 1.06;1.68). CONCLUSIONS The high proportion of adolescents in sedentary activities and the lack of association with being overweight in childhood, indicates the need for educational initiatives to reduce multiple risk behaviors. Encouraging physical activity in young people as a way of reducing sedentary behavior and, consequently, being overweight is fundamental. PMID:24897048
Self-Reported bruxism and associated factors in Israeli adolescents.
Emodi Perlman, A; Lobbezoo, F; Zar, A; Friedman Rubin, P; van Selms, M K A; Winocur, E
2016-06-01
Little is known about the epidemiological characteristics of sleep and awake bruxism (SB and AB) in adolescents. The aims of the study were: to assess the prevalence rates of self-reported SB and AB in Israeli adolescents; to determine the associations between SB/AB and several demographical, exogenous and psychosocial factors in Israeli adolescents; and to investigate the possible concordance between SB and AB. The study made use of a questionnaire. The study population included 1000 students from different high schools in the centre of Israel. Prevalence of self-reported SB and AB in the Israeli adolescents studied was 9·2% and 19·2%, respectively. No gender difference was found regarding the prevalence of SB and AB. Multiple variable regression analysis revealed that the following predicting variables were related to SB: temporomandibular joint sounds (P = 0·002) and feeling stressed (P = 0·001). The following predicting variables were related to AB: age (P = 0·018), temporomandibular joint sounds (P = 0·002), oro-facial pain (P = 0·006), and feeling stressed (P = 0·002) or sad (P = 0·006). A significant association was found between SB and AB; that is, an individual reporting SB had a higher probability of reporting AB compared with an individual who did not report SB (odds ratio = 5·099). Chewing gum was the most common parafunction reported by adolescents. The results of this study demonstrate that self-reports of AB and SB are common in the Israeli adolescents population studied and are not related to gender. The significant correlation found between SB and AB may be a confounding bias that affects proper diagnosis of bruxism through self-reported questionnaires only. © 2016 John Wiley & Sons Ltd.
Family structure and park use among parents.
Fan, Yingling; French, Simone A; Das, Kirti V
2012-11-01
Despite the increasingly diversified family structure in the U.S., little research examines differences in park use between nontraditional and traditional family structures. This study examines family-structure differences in parent park use. It was hypothesized that working single parents and dual-worker parents have lower levels of park use than parents in two-parent, single-worker families. Data from a 2010 park-use survey in three urban neighborhoods in Minneapolis MN (N=261 parents) were analyzed in 2012. Multiple variables of park use were developed, including recalled measures over the past 3 days and over the past year. Family-structure differences in these variables were examined using multivariate regression analyses. After controlling for spatial clustering effects and confounding factors, working single parents reported 32.6% (p<0.10) fewer park visits and 62.0% (p<0.05) less time spent in parks in the past 3 days than parents in two-parent, single-worker families. Dual-worker parents did not report fewer park visits in the past 3 days than parents in two-parent, single-worker families, yet the length of time they spent in parks during these visits was 41.5% (p<0.10) less. Family-structure differences in past-year park-use measures were not significant. This research shows the importance of including family-structure variables and both recent and longer-term recalled measures of park use in park-use studies. Greater attention to the recreation needs of working single parents and dual-worker parents is needed in descriptive and intervention research aiming to promote park use among families with children. Copyright © 2012 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.
Correlation of Vitamin D status and orthodontic-induced external apical root resorption
Tehranchi, Azita; Sadighnia, Azin; Younessian, Farnaz; Abdi, Amir H.; Shirvani, Armin
2017-01-01
Background: Adequate Vitamin D is essential for dental and skeletal health in children and adult. The purpose of this study was to assess the correlation of serum Vitamin D level with external-induced apical root resorption (EARR) following fixed orthodontic treatment. Materials and Methods: In this cross-sectional study, the prevalence of Vitamin D deficiency (defined by25-hydroxyvitamin-D) was determined in 34 patients (23.5% male; age range 12–23 years; mean age 16.63 ± 2.84) treated with fixed orthodontic treatment. Root resorption of four maxillary incisors was measured using before and after periapical radiographs (136 measured teeth) by means of a design-to-purpose software to optimize data collection. Teeth with a maximum percentage of root resorption (%EARR) were indicated as representative root resorption for each patient. A multiple linear regression model and Pearson correlation coefficient were used to assess the association of Vitamin D status and observed EARR. P < 0.05 was considered statistically significant. Results: The Pearson coefficient between these two variables was determined about 0.15 (P = 0.38). Regression analysis revealed that Vitamin D status of the patients demonstrated no significant statistical correlation with EARR, after adjustment of confounding variables using linear regression model (P > 0.05). Conclusion: This study suggests that Vitamin D level is not among the clinical variables that are potential contributors for EARR. The prevalence of Vitamin D deficiency does not differ in patients with higher EARR. These data suggest the possibility that Vitamin D insufficiency may not contribute to the development of more apical root resorption although this remains to be confirmed by further longitudinal cohort studies. PMID:29238379