Bayesian methods for correcting misclassification: an example from birth defects epidemiology

Bayesian estimation of the measurement of interactions in epidemiological studies

Background

Interaction identification is important in epidemiological studies and can be detected by including a product term in the model. However, as Rothman noted, a product term in exponential models may be regarded as multiplicative rather than additive to better reflect biological interactions. Currently, the additive interaction is largely measured by the relative excess risk due to interaction (RERI), the attributable proportion due to interaction (AP), and the synergy index (S), and confidence intervals are developed via frequentist approaches. However, few studies have focused on the same issue from a Bayesian perspective. The present study aims to provide a Bayesian view of the estimation and credible intervals of the additive interaction measures.

Methods

Bayesian logistic regression was employed, and estimates and credible intervals were calculated from posterior samples of the RERI, AP and S. Since Bayesian inference depends only on posterior samples, it is very easy to apply this method to preventive factors. The validity of the proposed method was verified by comparing the Bayesian method with the delta and bootstrap approaches in simulation studies with example data.

Results

In all the simulation studies, the Bayesian estimates were very close to the corresponding true values. Due to the skewness of the interaction measures, compared with the confidence intervals of the delta method, the credible intervals of the Bayesian approach were more balanced and matched the nominal 95% level. Compared with the bootstrap method, the Bayesian method appeared to be a competitive alternative and fared better when small sample sizes were used.

Conclusions

The proposed Bayesian method is a competitive alternative to other methods. This approach can assist epidemiologists in detecting additive-scale interactions.

Self-reported tick exposure as an indicator of Lyme disease risk in an endemic region of Quebec, Canada

BACKGROUND

Lyme disease (LD) and other tick-borne diseases are emerging across Canada. Spatial and temporal LD risk is typically estimated using acarological surveillance and reported human cases, the former not considering human behavior leading to tick exposure and the latter occurring after infection.

OBJECTIVES

The primary objective was to explore, at the census subdivision level (CSD), the associations of self-reported tick exposure, alternative risk indicators (predicted tick density, eTick submissions, public health risk level), and ecological variables (Ixodes scapularis habitat suitability index and cumulative degree days > 0 °C) with incidence proportion of LD. A secondary objective was to explore which of these predictor variables were associated with self-reported tick exposure at the CSD level.

METHODS

Self-reported tick exposure was measured in a cross-sectional populational health survey conducted in 2018, among 10,790 respondents living in 116 CSDs of the Estrie region, Quebec, Canada. The number of reported LD cases per CSD in 2018 was obtained from the public health department. Generalized linear mixed-effets models accounting for spatial autocorrelation were built to fulfill the objectives.

RESULTS

Self-reported tick exposure ranged from 0.0 % to 61.5 % (median 8.9 %) and reported LD incidence rates ranged from 0 to 324 cases per 100,000 person-years, per CSD. A positive association was found between self-reported tick exposure and LD incidence proportion (ß = 0.08, CI = 0.04,0.11, p < 0.0001). The best-fit model included public health risk level (AIC: 144.2), followed by predicted tick density, ecological variables, self-reported tick exposure and eTick submissions (AIC: 158.4, 158.4, 160.4 and 170.1 respectively). Predicted tick density was the only significant predictor of self-reported tick exposure (ß = 0.83, CI = 0.16,1.50, p = 0.02).

DISCUSSION

This proof-of-concept study explores self-reported tick exposure as a potential indicator of LD risk using populational survey data. This approach may offer a low-cost and simple tool for evaluating LD risk and deserves further evaluation.

Bias analyses to investigate the impact of differential participation: Application to a birth defects case-control study

BACKGROUND

Certain associations observed in the National Birth Defects Prevention Study (NBDPS) contrasted with other research or were from areas with mixed findings, including no decrease in odds of spina bifida with periconceptional folic acid supplementation, moderately increased cleft palate odds with ondansetron use and reduced hypospadias odds with maternal smoking.

OBJECTIVES

To investigate the plausibility and extent of differential participation to produce effect estimates observed in NBDPS.

METHODS

We searched the literature for factors related to these exposures and participation and conducted deterministic quantitative bias analyses. We estimated case-control participation and expected exposure prevalence based on internal and external reports, respectively. For the folic acid-spina bifida and ondansetron-cleft palate analyses, we hypothesized the true odds ratio (OR) based on prior studies and quantified the degree of exposure over- (or under-) representation to produce the crude OR (cOR) in NBDPS. For the smoking-hypospadias analysis, we estimated the extent of selection bias needed to nullify the association as well as the maximum potential harmful OR.

RESULTS

Under our assumptions (participation, exposure prevalence, true OR), there was overrepresentation of folic acid use and underrepresentation of ondansetron use and smoking among participants. Folic acid-exposed spina bifida cases would need to have been ≥1.2× more likely to participate than exposed controls to yield the observed null cOR. Ondansetron-exposed cleft palate cases would need to have been 1.6× more likely to participate than exposed controls if the true OR is null. Smoking-exposed hypospadias cases would need to have been ≥1.2 times less likely to participate than exposed controls for the association to falsely appear protective (upper bound of selection bias adjusted smoking-hypospadias OR = 2.02).

CONCLUSIONS

Differential participation could partly explain certain associations observed in NBDPS, but questions remain about why. Potential impacts of other systematic errors (e.g. exposure misclassification) could be informed by additional research.

Effect of alcohol consumption on breast cancer: probabilistic bias analysis for adjustment of exposure misclassification bias and confounders

PURPOSE

This study was conducted to evaluate the effect of alcohol consumption on breast cancer, adjusting for alcohol consumption misclassification bias and confounders.

METHODS

This was a case-control study of 932 women with breast cancer and 1000 healthy control. Using probabilistic bias analysis method, the association between alcohol consumption and breast cancer was adjusted for the misclassification bias of alcohol consumption as well as a minimally sufficient set of adjustment of confounders derived from a causal directed acyclic graph. Population attributable fraction was estimated using the Miettinen's Formula.

RESULTS

Based on the conventional logistic regression model, the odds ratio estimate between alcohol consumption and breast cancer was 1.05 (95% CI: 0.57, 1.91). However, the adjusted estimates of odds ratio based on the probabilistic bias analysis ranged from 1.82 to 2.29 for non-differential and from 1.93 to 5.67 for differential misclassification. Population attributable fraction ranged from 1.51 to 2.57% using non-differential bias analysis and 1.54-3.56% based on differential bias analysis.

CONCLUSION

A marked measurement error was in self-reported alcohol consumption so after correcting misclassification bias, no evidence against independence between alcohol consumption and breast cancer changed to a substantial positive association.

Noninterventional studies in the COVID-19 era: methodological considerations for study design and analysis

The global COVID-19 pandemic has generated enormous morbidity and mortality, as well as large health system disruptions including changes in use of prescription medications, outpatient encounters, emergency department admissions, and hospitalizations. These pandemic-related disruptions are reflected in real-world data derived from electronic medical records, administrative claims, disease or medication registries, and mobile devices. We discuss how pandemic-related disruptions in healthcare utilization may impact the conduct of noninterventional studies designed to characterize the utilization and estimate the effects of medical interventions on health-related outcomes. Using hypothetical studies, we highlight consequences that the pandemic may have on study design elements including participant selection and ascertainment of exposures, outcomes, and covariates. We discuss the implications of these pandemic-related disruptions on possible threats to external validity (participant selection) and internal validity (for example, confounding, selection bias, missing data bias). These concerns may be amplified in populations disproportionately impacted by COVID-19, such as racial/ethnic minorities, rural residents, or people experiencing poverty. We propose a general framework for researchers to carefully consider during the design and analysis of noninterventional studies that use real-world data from the COVID-19 era.

Maternal major depression disorder misclassification errors: Remedies for valid individual- and population-level inference

Individual and population level inference about risk and burden of MDD, particularly maternal MDD, is often made using case-finding tools that are imperfect and prone to misclassification error (i.e. false positives and negatives). These errors or biases are rarely accounted for and lead to inappropriate clinical decisions, inefficient allocation of scarce resources, and poor planning of maternal MDD prevention and treatment interventions. The argument that the use of existing maternal MDD case-finding instruments results in misclassification errors is not new; in fact, it has been argued for decades, but by and large its implications and particularly how to correct for these errors for valid inference is unexplored. Correction of the estimates of maternal MDD prevalence, case-finding tool sensitivity and specificity is possible and should be done to inform valid individual and population-level inferences.

Invited Commentary: Dealing With the Inevitable Deficiencies of Bias Analysis-and All Analyses

Lash et al. (Am J Epidemiol. 2021;190(8):1604-1612) have presented detailed critiques of 3 bias analyses that they identify as "suboptimal." This identification raises the question of what "optimal" means for bias analysis, because it is practically impossible to do statistically optimal analyses of typical population studies-with or without bias analysis. At best the analysis can only attempt to satisfy practice guidelines and account for available information both within and outside the study. One should not expect a full accounting for all sources of uncertainty; hence, interval estimates and distributions for causal effects should never be treated as valid uncertainty assessments-they are instead only example analyses that follow from collections of often questionable assumptions. These observations reinforce those of Lash et al. and point to the need for more development of methods for judging bias-parameter distributions and utilization of available information.

STRATOS guidance document on measurement error and misclassification of variables in observational epidemiology: Part 2-More complex methods of adjustment and advanced topics

We continue our review of issues related to measurement error and misclassification in epidemiology. We further describe methods of adjusting for biased estimation caused by measurement error in continuous covariates, covering likelihood methods, Bayesian methods, moment reconstruction, moment-adjusted imputation, and multiple imputation. We then describe which methods can also be used with misclassification of categorical covariates. Methods of adjusting estimation of distributions of continuous variables for measurement error are then reviewed. Illustrative examples are provided throughout these sections. We provide lists of available software for implementing these methods and also provide the code for implementing our examples in the Supporting Information. Next, we present several advanced topics, including data subject to both classical and Berkson error, modeling continuous exposures with measurement error, and categorical exposures with misclassification in the same model, variable selection when some of the variables are measured with error, adjusting analyses or design for error in an outcome variable, and categorizing continuous variables measured with error. Finally, we provide some advice for the often met situations where variables are known to be measured with substantial error, but there is only an external reference standard or partial (or no) information about the type or magnitude of the error.

Effect of Smoking on Breast Cancer by Adjusting for Smoking Misclassification Bias and Confounders Using a Probabilistic Bias Analysis Method

Purpose

The aim of this study was to determine the association between smoking and breast cancer after adjusting for smoking misclassification bias and confounders.

Methods

In this case–control study, 1000 women with breast cancer and 1000 healthy controls were selected. Using a probabilistic bias analysis method, the association between smoking and breast cancer was adjusted for the bias resulting from misclassification of smoking secondary to self-reporting as well as a minimally sufficient adjustment set of confounders derived from a causal directed acyclic graph (cDAG). Population attributable fraction (PAF) for smoking was calculated using Miettinen’s formula.

Results

While the odds ratio (OR) from the conventional logistic regression model between smoking and breast cancer was 0.64 (95% CI: 0.36–1.13), the adjusted ORs from the probabilistic bias analysis were in the ranges of 2.63–2.69 and 1.73–2.83 for non-differential and differential misclassification, respectively. PAF ranges obtained were 1.36–1.72% and 0.62–2.01% using the non-differential bias analysis and differential bias analysis, respectively.

Conclusion

After misclassification correction for smoking, the non-significant negative-adjusted association between smoking and breast cancer changed to a significant positive-adjusted association.

Bayesian Methods for Exposure Misclassification Adjustment in a Mediation Analysis: Hemoglobin and Malnutrition in the Association Between Ascaris and IQ

BACKGROUND

Soil-transmitted helminth infections have been found to be associated with child development. The objective was to investigate hemoglobin levels and malnutrition as mediators of the association between Ascaris infection and intelligence quotient (IQ) scores in children.

METHODS

We conducted a longitudinal cohort study in Iquitos, Peru, between September 2011 and July 2016. A total of 1760 children were recruited at 1 year of age and followed up annually to 5 years. We measured Ascaris infection and malnutrition at each study visit, and hemoglobin levels were measured as of age 3. The exposure was defined as the number of detected Ascaris infections between age 1 and 5. We measured IQ scores at age 5 and used Bayesian models to correct exposure misclassification.

RESULTS

We included a sample of 781 children in the analysis. In results adjusted for Ascaris misclassification, mean hemoglobin levels mediated the association between Ascaris infection and IQ scores. The natural direct effects (not mediated by hemoglobin) (95% CrI) and natural indirect effects (mediated by hemoglobin) (95% CrI) were compared with no or one infection: -0.9 (-4.6, 2.8) and -4.3 (-6.9, -1.6) for the effect of two infections; -1.4 (-3.8, 1.0) and -1.2 (-2.0, -0.4) for three infections; and -0.4 (-3.2, 2.4) and -2.7 (-4.3, -1.0) for four or five infections.

CONCLUSION

Our results are consistent with the hypothesis that hemoglobin levels mediate the association between Ascaris infection and IQ scores. Additional research investigating the effect of including iron supplements in STH control programs is warranted.

Risk of gastroschisis with maternal genitourinary infections: the US National birth defects prevention study 1997-2011

OBJECTIVE

To assess the association between occurrence and timing of maternal self-reported genitourinary tract infection (urinary tract infections [UTIs] and/or sexually transmitted infection [STI]) and risk for gastroschisis in the offspring.

DESIGN

Population-based case-control study.

SETTING

National Birth Defects Prevention Study, a multisite study in the USA.

PARTICIPANTS

Mothers of 1366 gastroschisis cases and 11 238 healthy controls.

MAIN OUTCOME MEASURES

Crude and adjusted ORs (aORs) with 95% CIs.

RESULTS

Genitourinary infections were frequent in case (19.3%) and control women (9.9%) during the periconceptional period (defined as 3 months prior to 3 months after conception). UTI and/or STI in the periconceptional period were associated with similarly increased risks for gastroschisis (aOR 1.5, 95% CI 1.3 to 1.8; aOR 1.6, 95% CI 1.2 to 2.3, respectively). The risk was increased with a UTI before (aOR 2.5; 95% CI 1.4 to 4.5) or after (aOR 1.7; 95% CI 1.1 to 2.6) conception only among women ≥25 years of age. The risk was highest among women <20 years of age with an STI before conception (aOR 3.6; 95% CI 1.5 to 8.4) and in women ≥25 years of age, the risk was similar for before (aOR 2.9; 95% CI 1.0 to 8.5) and after (aOR 2.8; 95% CI 1.3 to 6.1) conception. A specific STI pathogen was reported in 89.3% (50/56) of cases and 84.3% (162/191) of controls with Chlamydia trachomatis the most common (25/50 cases, 50%; 58/162 controls, 36%) and highest among women <20 years of age (16/25 cases, 64%; 22/33 controls, 67%).

CONCLUSIONS

UTI and/or STI were associated with an increased risk for gastroschisis, with the strength of the association varying by maternal age and timing of infection.

The Impact of Nondifferential Exposure Misclassification on the Performance of Propensity Scores for Continuous and Binary Outcomes: A Simulation Study

PURPOSE

To investigate the ability of the propensity score (PS) to reduce confounding bias in the presence of nondifferential misclassification of treatment, using simulations.

METHODS

Using an example from the pregnancy medication safety literature, we carried out simulations to quantify the effect of nondifferential misclassification of treatment under varying scenarios of sensitivity and specificity, exposure prevalence (10%, 50%), outcome type (continuous and binary), true outcome (null and increased risk), confounding direction, and different PS applications (matching, stratification, weighting, regression), and obtained measures of bias and 95% confidence interval coverage.

RESULTS

All methods were subject to substantial bias toward the null due to nondifferential exposure misclassification (range: 0%-47% for 50% exposure prevalence and 0%-80% for 10% exposure prevalence), particularly if specificity was low (<97%). PS stratification produced the least biased effect estimates. We observed that the impact of sensitivity and specificity on the bias and coverage for each adjustment method is strongly related to prevalence of exposure: as exposure prevalence decreases and/or outcomes are continuous rather than categorical, the effect of misclassification is magnified, producing larger biases and loss of coverage of 95% confidence intervals. PS matching resulted in unpredictably biased effect estimates.

CONCLUSIONS

The results of this study underline the importance of assessing exposure misclassification in observational studies in the context of PS methods. Although PS methods reduce confounding bias, bias owing to nondifferential misclassification is of potentially greater concern.

Epidemiologic analyses with error-prone exposures: review of current practice and recommendations

PURPOSE

Variables in observational studies are commonly subject to measurement error, but the impact of such errors is frequently ignored. As part of the STRengthening Analytical Thinking for Observational Studies Initiative, a task group on measurement error and misclassification seeks to describe the current practice for acknowledging and addressing measurement error.

METHODS

Task group on measurement error and misclassification conducted a literature survey of four types of research studies that are typically impacted by exposure measurement error: (1) dietary intake cohort studies, (2) dietary intake population surveys, (3) physical activity cohort studies, and (4) air pollution cohort studies.

RESULTS

The survey revealed that while researchers were generally aware that measurement error affected their studies, very few adjusted their analysis for the error. Most articles provided incomplete discussion of the potential effects of measurement error on their results. Regression calibration was the most widely used method of adjustment.

CONCLUSIONS

Methods to correct for measurement error are available but require additional data regarding the error structure. There is a great need to incorporate such data collection within study designs and improve the analytical approach. Increased efforts by investigators, editors, and reviewers are needed to improve presentation of research when data are subject to error.

The impact of maternal smoking during pregnancy on childhood asthma: adjusted for exposure misclassification; results from the National Health and Nutrition Examination Survey, 2011-2012

PURPOSE

We sought to examine the association between childhood asthma and self-reported maternal smoking during pregnancy (MSDP) after adjusting for a range of exposure misclassification scenarios using a Bayesian approach that incorporated exposure misclassification probability estimates from the literature.

METHODS

Self-reported MSDP and asthma data were extracted from National Health and Nutrition Examination Survey 2011-2012. The association between self-reported MSDP and asthma was adjusted for exposure misclassification using a Bayesian bias model approach.

RESULTS

We included 3074 subjects who were 1-15 years of age, including 492 asthma cases. The mean (SD) of age of the participants was 8.5 (4.1) and 7.1 (4.2) years and the number (percentage) of female was 205 (42%) and 1314 (51%) among asthmatic and nonasthmatic groups, respectively. The odds ratio (OR) for the association between self-reported MSDP and asthma in logistic regression adjusted for confounders was 1.28 (95% confidence interval: 0.92, 1.77). In a Bayesian analysis that adjusted for exposure misclassification using external data, we found different ORs between MSDP and asthma by applying different priors (posterior ORs 0.90 [95% credible interval {CRI}: 0.47, 1.60] to 3.05 [95% CRI: 1.73, 5.53] in differential and 1.22 [CRI 95%: 0.62, 2.25] to 1.60 CRI: 1.18, 2.19) in nondifferential misclassification settings.

CONCLUSIONS

Given the assumptions and the accuracy of the bias model, the estimated effect of MSDP on asthma after adjusting for misclassification was strengthened in many scenarios.

A Bayesian Approach to Modeling Risk of Hospital Admissions Associated With Schizophrenia Accounting for Underdiagnosis of the Disorder in Administrative Records

Schizophrenia is a debilitating serious mental illness characterized by a complex array of symptoms with varying severity and duration. Patients may seek treatment only intermittently, contributing to challenges diagnosing the disorder. A misdiagnosis may potentially bias and reduce study validity. Thus we developed a statistical model to assess the risk of 1-year hospitalization for patients diagnosed with schizophrenia, accounting for when schizophrenia is underreported in administrative databases. A retrospective study design identified patients seeking care during 2010 within an integrated health care system from the Health Maintenance Organization Research Network located in the southwestern United States. Bayesian analysis addressed the problem of underdiagnosed schizophrenia with a statistical measurement error model assuming varying rates of underreporting. Results were then compared to classical multivariable logistic regression. Assuming no underreporting, there was an 87% greater relative odds of hospitalization associated with schizophrenia, OR = 1.87, CI [1.08, 3.23]. Effect sizes and interval estimates representing the association between hospitalization and schizophrenia were reduced with the Bayesian approach accounting for underdiagnosis, suggesting that less severe patients may be underrepresented in studies of schizophrenia. The analytical approach has useful applications in other contexts where the identification of patients with a given condition may be underreported in administrative records.

Sensitivity Analyses for Misclassification of Cause of Death in the Parametric G-Formula

Cause-specific mortality is an important outcome in studies of interventions to improve survival, yet causes of death can be misclassified. Here, we present an approach to performing sensitivity analyses for misclassification of cause of death in the parametric g-formula. The g-formula is a useful method to estimate effects of interventions in epidemiologic research because it appropriately accounts for time-varying confounding affected by prior treatment and can estimate risk under dynamic treatment plans. We illustrate our approach using an example comparing acquired immune deficiency syndrome (AIDS)-related mortality under immediate and delayed treatment strategies in a cohort of therapy-naive adults entering care for human immunodeficiency virus infection in the United States. In the standard g-formula approach, 10-year risk of AIDS-related mortality under delayed treatment was 1.73 (95% CI: 1.17, 2.54) times the risk under immediate treatment. In a sensitivity analysis assuming that AIDS-related death was measured with sensitivity of 95% and specificity of 90%, the 10-year risk ratio comparing AIDS-related mortality between treatment plans was 1.89 (95% CI: 1.13, 3.14). When sensitivity and specificity are unknown, this approach can be used to estimate the effects of dynamic treatment plans under a range of plausible values of sensitivity and specificity of the recorded event type.

The effect of cumulative soil-transmitted helminth infections over time on child development: a 4-year longitudinal cohort study in preschool children using Bayesian methods to adjust for exposure misclassification

Background

Limited research has documented an association between soil-transmitted helminth (STH) infections and child development. This has recently been identified as an important knowledge gap.

Methods

A longitudinal cohort study was conducted in Iquitos, Peru, between September 2011 and July 2016. A cohort of 880 children, recruited at 1 year of age, was followed up to 5 years. STH infection was measured annually and child development was measured with the Wechsler Preschool and Primary Scale of Intelligence III (WPPSI-III) at 5 years. Linear-regression models were used to investigate the effect of the number of detected STH infections between 1 and 5 years of age on WPPSI-III scores at 5 years of age. Bayesian latent class analysis was used to adjust for exposure misclassification.

Results

A total of 781 (88.8%) children were included in the analysis. In multivariable analysis, adjusted for STH misclassification, increasing numbers of Ascaris, Trichuris, hookworm and any STH infections were associated with lower WPPSI-III scores. Among the largest observed effects were those for the effect of Ascaris infection on verbal IQ scores [difference in IQ (95% CrI) for two, three, and four or five detected infections compared with zero or one infection: -8.27 (-13.85, -3.10), -6.69 (-12.05, -2.05) and -5.06 (-10.75, 0.05), respectively]. Misclassification of STH infection generally led to a bias towards the null.

Conclusions

These results document an association between STH infection and child development. The results highlight the importance of adjusting for STH misclassification; however, future research is needed to accurately determine the sensitivity of STH diagnostic techniques. STH control in preschool children may contribute to lowering the disease burden associated with poor child development.

A longitudinal cohort study of soil-transmitted helminth infections during the second year of life and associations with reduced long-term cognitive and verbal abilities

BACKGROUND

Soil-transmitted helminth (STH) infection leads to malnutrition and anemia, and has been linked to impaired child development. Previous research on this topic is limited and mostly conducted in school-age children. The goal of this study was to determine the effect of the number of detected STH infections between one and two years of age on subsequent cognitive and verbal abilities, in a cohort of preschool children.

METHODOLOGY/PRINCIPAL FINDINGS

A longitudinal cohort study was conducted in 880 children in Iquitos, Peru between September 2011 and July 2016. Children were recruited at one year of age and followed up at 18 months and then annually between two and five years of age. STH infection was measured with the Kato-Katz technique or the direct smear technique. Child development was measured with the Bayley Scales of Infant and Toddler Development-III at the one to three-year visits and with the Wechsler Preschool and Primary Scale of Intelligence-III at the four and five-year visits. Hierarchical multivariable linear regression models were used to account for the repeated outcome measures for each child and Bayesian latent class analysis was used to adjust for STH misclassification. Children found infected with any STH infection between one and two years of age had lower cognitive scores between two and five years of age (between group score differences (95% credible intervals) for infected once, and infected two or three times, compared to never infected: -4.31 (-10.64, -0.14) and -3.70 (-10.11, -0.11), respectively). Similar results were found for Ascaris infection and for verbal scores.

CONCLUSIONS/SIGNIFICANCE

An association was found between having been infected with Ascaris or any STH between one and two years of age and lower cognitive and verbal abilities later in childhood. These results suggest that targeting children for STH control as of one year of age is particularly important.

Bias due to differential and non-differential disease- and exposure misclassification in studies of vaccine effectiveness

BACKGROUND

Studies of vaccine effectiveness (VE) rely on accurate identification of vaccination and cases of vaccine-preventable disease. In practice, diagnostic tests, clinical case definitions and vaccination records often present inaccuracies, leading to biased VE estimates. Previous studies investigated the impact of non-differential disease misclassification on VE estimation.

METHODS

We explored, through simulation, the impact of non-differential and differential disease- and exposure misclassification when estimating VE using cohort, case-control, test-negative case-control and case-cohort designs. The impact of misclassification on the estimated VE is demonstrated for VE studies on childhood seasonal influenza and pertussis vaccination. We additionally developed a web-application graphically presenting bias for user-selected parameters.

RESULTS

Depending on the scenario, the misclassification parameters had differing impacts. Decreased exposure specificity had greatest impact for influenza VE estimation when vaccination coverage was low. Decreased exposure sensitivity had greatest impact for pertussis VE estimation for which high vaccination coverage is typically achieved. The impact of the exposure misclassification parameters was found to be more noticeable than that of the disease misclassification parameters. When misclassification is limited, all study designs perform equally. In case of substantial (differential) disease misclassification, the test-negative design performs worse.

CONCLUSIONS

Misclassification can lead to significant bias in VE estimates and its impact strongly depends on the scenario. We developed a web-application for assessing the potential (joint) impact of possibly differential disease- and exposure misclassification that can be modified by users to their own study scenario. Our results and the simulation tool may be used to guide better design, conduct and interpretation of future VE studies.

Environmental exposure mixtures: questions and methods to address them

PURPOSE OF THIS REVIEW

This review provides a summary of statistical approaches that researchers can use to study environmental exposure mixtures. Two primary considerations are the form of the research question and the statistical tools best suited to address that question. Because the choice of statistical tools is not rigid, we make recommendations about when each tool may be most useful.

RECENT FINDINGS

When dimensionality is relatively low, some statistical tools yield easily interpretable estimates of effect (e.g., risk ratio, odds ratio) or intervention impacts. When dimensionality increases, it is often necessary to compromise this interpretablity in favor of identifying interesting statistical signals from noise; this requires applying statistical tools that are oriented more heavily towards dimension reduction via shrinkage and/or variable selection.

SUMMARY

The study of complex exposure mixtures has prompted development of novel statistical methods. We suggest that further validation work would aid practicing researchers in choosing among existing and emerging statistical tools for studying exposure mixtures.

Hierarchical Semi-Bayes Methods for Misclassification in Perinatal Epidemiology

BACKGROUND

Validation data are used to estimate the extent of misclassification in epidemiologic studies. In the Penn MOMS cohort, prepregnancy body mass index is subject to misclassification, and validation data are available to estimate the extent of misclassification. We use these data to estimate the association between maternal prepregnancy body mass index and early preterm (<32 weeks) birth using a semi-Bayes hierarchical model, allowing for more flexible adjustment for misclassification.

METHODS

We propose a two-stage model that first fits a Bayesian hierarchical model for the bias parameters in the validation study. This model shrinks bias parameters in different groups toward one another in an effort to gain precision and improve mean squared error. In the second stage, we draw random samples from the posterior distribution of the bias parameters to implement a probabilistic bias analysis adjusting for exposure misclassification in a frequentist outcome model.

RESULTS

Bias parameters from the hierarchical model were often more substantively reasonable and often had smaller variance. Adjusting results for misclassification generally attenuated the strength of the unadjusted associations. After adjusting for misclassification, underweight mothers were not at increased risk of early preterm birth relative to normal weight mothers. Severely obese mothers had an increased risk of early preterm birth relative to normal weight mothers.

CONCLUSIONS

The two-stage semi-Bayesian hierarchical model borrowed strength between group-specific bias parameters to adjust for exposure misclassification. Model results support evidence of an increased risk of early preterm birth among severely obese mothers, relative to normal weight mothers.

Administrative Claims Data Versus Augmented Pregnancy Data for the Study of Pharmaceutical Treatments in Pregnancy

Purpose of Review

Administrative claims databases, which collect reimbursement-related information generated from healthcare encounters, are increasingly used to evaluate medication safety in pregnancy. We reviewed the strengths and limitations of claims-only databases and how other data sources may be used to improve the accuracy and completeness of information critical for studying medication safety in pregnancy.

Recent Findings

Research on medication safety in pregnancy requires information on pregnancy episodes, mother-infant linkage, medication exposure, gestational age, maternal and birth outcomes, confounding factors, and (in some studies) long-term follow-up data. Claims data reliably identifies live births and possibly other pregnancies. It allows mother-infant linkage and has prospectively collected prescription medication information. Its diagnosis and procedure information allows estimation of gestational age. It captures maternal medical conditions but generally has incomplete data on reproductive and lifestyle factors. It has information on certain, typically short-term maternal and infant outcomes that may require chart review confirmation. Other data sources including electronic health records and birth registries can augment claims data or be analyzed alone. Interviews, surveys, or biological samples provide additional information. Nationwide and regional birth and pregnancy registries, such as those in several European and North American countries, generally contain more complete information essential for pregnancy research compared to claims-only databases.

Summary

Claims data offers several advantages in medication safety in pregnancy research. Its limitations can be partially addressed by linking it with other data sources or supplementing with primary data collection. Rigorous assessment of data quality and completeness is recommended regardless of data sources.

Measurement Error and Environmental Epidemiology: a Policy Perspective

PURPOSE OF REVIEW

Measurement error threatens public health by producing bias in estimates of the population impact of environmental exposures. Quantitative methods to account for measurement bias can improve public health decision making.

RECENT FINDINGS

We summarize traditional and emerging methods to improve inference under a standard perspective, in which the investigator estimates an exposure-response function, and a policy perspective, in which the investigator directly estimates population impact of a proposed intervention. Under a policy perspective, the analyst must be sensitive to errors in measurement of factors that modify the effect of exposure on outcome, must consider whether policies operate on the true or measured exposures, and may increasingly need to account for potentially dependent measurement error of two or more exposures affected by the same policy or intervention. Incorporating approaches to account for measurement error into such a policy perspective will increase the impact of environmental epidemiology.

Maternal active smoking and risk of oral clefts: a meta-analysis

OBJECTIVE

To assess the association between maternal active cigarette smoking and the risk of oral clefts in the offspring.

STUDY DESIGN

Oral clefts are divided into three subgroups: total clefts, cleft lip with or without cleft palate (CL ± P), and cleft palate only (CP). Data from studies on different levels of smoking were gathered to examine the dose-response effect.

RESULTS

The present meta-analysis included 29 case-control and cohort studies through Cochrane, PubMed, and Ovid Medline searches. A modest but statistically significant association was found between maternal active smoking and CL ± P (odds ratio [OR] 1.368; 95% confidence interval [CI] 1.259-1.486) as well as CP (OR 1.241; 95% CI 1.117-1.378). Half the studies showed positive dose-response effect for each subgroup (test for linear trend, P < .05).

CONCLUSIONS

There is a moderate risk for having a child with a CL ± P or CP in women who smoke during pregnancy. We could not confirm whether there was a positive dose-response effect between maternal smoking and clefts.

Multiple Imputation to Account for Measurement Error in Marginal Structural Models

BACKGROUND

Marginal structural models are an important tool for observational studies. These models typically assume that variables are measured without error. We describe a method to account for differential and nondifferential measurement error in a marginal structural model.

METHODS

We illustrate the method estimating the joint effects of antiretroviral therapy initiation and current smoking on all-cause mortality in a United States cohort of 12,290 patients with HIV followed for up to 5 years between 1998 and 2011. Smoking status was likely measured with error, but a subset of 3,686 patients who reported smoking status on separate questionnaires composed an internal validation subgroup. We compared a standard joint marginal structural model fit using inverse probability weights to a model that also accounted for misclassification of smoking status using multiple imputation.

RESULTS

In the standard analysis, current smoking was not associated with increased risk of mortality. After accounting for misclassification, current smoking without therapy was associated with increased mortality (hazard ratio [HR]: 1.2 [95% confidence interval [CI] = 0.6, 2.3]). The HR for current smoking and therapy [0.4 (95% CI = 0.2, 0.7)] was similar to the HR for no smoking and therapy (0.4; 95% CI = 0.2, 0.6).

CONCLUSIONS

Multiple imputation can be used to account for measurement error in concert with methods for causal inference to strengthen results from observational studies.

Reliability of Reported Maternal Smoking: Comparing the Birth Certificate to Maternal Worksheets and Prenatal and Hospital Medical Records, New York City and Vermont, 2009

Maternal smoking is captured on the 2003 US Standard Birth Certificate based on self-reported tobacco use before and during pregnancy collected on post-delivery maternal worksheets. Study objectives were to compare smoking reported on the birth certificate to maternal worksheets and prenatal and hospital medical records. The authors analyzed a sample of New York City (NYC) and Vermont women (n = 1,037) with a live birth from January to August 2009 whose responses to the Pregnancy Risk Assessment Monitoring System survey were linked with birth certificates and abstracted medical records and maternal worksheets. We calculated smoking prevalence and agreement (kappa) between sources overall and by maternal and hospital characteristics. Smoking before and during pregnancy was 13.7 and 10.4% using birth certificates, 15.2 and 10.7% using maternal worksheets, 18.1 and 14.1% using medical records, and 20.5 and 15.0% using either maternal worksheets or medical records. Birth certificates had "almost perfect" agreement with maternal worksheets for smoking before and during pregnancy (κ = 0.92 and 0.89) and "substantial" agreement with medical records (κ = 0.70 and 0.74), with variation by education, insurance, and parity. Smoking information on NYC and Vermont birth certificates closely agreed with maternal worksheets but was underestimated compared with medical records, with variation by select maternal characteristics. Opportunities exist to improve birth certificate smoking data, such as reducing the stigma of smoking, and improving the collection, transcription, and source of information.

The lag-time approach improved drug-outcome association estimates in presence of protopathic bias

OBJECTIVES

Protopathic bias is a systematic error which occurs when measured exposure status may be affected by the latent onset of the target outcome. In this article, we aimed to discuss the benefits and drawbacks of the lag-time approach to address this type of bias.

STUDY DESIGN AND SETTING

The lag-time approach consists in excluding from exposure assessment the period immediately preceding the outcome detection date. With the help of simple causal diagrams, we illustrate the rationale and limitations of such strategy. The lag-time approach was illustrated in a case-crossover study, based on the health care utilization databases of the Italian Lombardy Region, on the real-world effectiveness of some respiratory drugs (exposure) in preventing asthma exacerbations (outcome).

RESULTS

A total of 7,300 of patients who were admitted to an emergency department (ED) for asthma during 2010-2012 (cases) were included. Use (vs. nonuse) of short-acting beta-agonists (SABAs, an asthma reliever medication) during the 90 days before the ED admission date was associated with an increased risk of the outcome [odds ratio (OR): 1.95; 95% confidence interval (CI): 1.72, 2.22]. This paradoxical finding may be explained by protopathic bias, as SABA use prior the ED admission may be affected by preceding respiratory distress. Indeed, when a 120-day period preceding the ED admission was ignored from drug exposure assessment (lag time), SABAs were found to be associated with a reduced risk of the outcome (OR: 0.81; 95% CI: 0.84, 0.92), as expected.

CONCLUSIONS

The lag-time approach can be a useful strategy to circumvent protopathic bias in observational studies.

Are all biases missing data problems?

Estimating causal effects is a frequent goal of epidemiologic studies. Traditionally, there have been three established systematic threats to consistent estimation of causal effects. These three threats are bias due to confounders, selection, and measurement error. Confounding, selection, and measurement bias have typically been characterized as distinct types of biases. However, each of these biases can also be characterized as missing data problems that can be addressed with missing data solutions. Here we describe how the aforementioned systematic threats arise from missing data as well as review methods and their related assumptions for reducing each bias type. We also link the assumptions made by the reviewed methods to the missing completely at random (MCAR) and missing at random (MAR) assumptions made in the missing data framework that allow for valid inferences to be made based on the observed, incomplete data.

Misclassification in administrative claims data: quantifying the impact on treatment effect estimates

Misclassification is present in nearly every epidemiologic study, yet is rarely quantified in analysis in favor of a focus on random error. In this review, we discuss past and present wisdom on misclassification and what measures should be taken to quantify this influential bias, with a focus on bias in pharmacoepidemiologic studies. To date, pharmacoepidemiology primarily utilizes data obtained from administrative claims, a rich source of prescription data but susceptible to bias from unobservable factors including medication sample use, medications filled but not taken, health conditions that are not reported in the administrative billing data, and inadequate capture of confounders. Due to the increasing focus on comparative effectiveness research, we provide a discussion of misclassification in the context of an active comparator, including a demonstration of treatment effects biased away from the null in the presence of nondifferential misclassification. Finally, we highlight recently developed methods to quantify bias and offer these methods as potential options for strengthening the validity and quantifying uncertainty of results obtained from pharmacoepidemiologic research.

Potential sensitivity of bias analysis results to incorrect assumptions of nondifferential or differential binary exposure misclassification

BACKGROUND

Results of bias analyses for exposure misclassification are dependent on assumptions made during analysis. We describe how adjustment for misclassification is affected by incorrect assumptions about whether sensitivity and specificity are the same (nondifferential) or different (differential) for cases and noncases.

METHODS

We adjusted for exposure misclassification using probabilistic bias analysis, under correct and incorrect assumptions about whether exposure misclassification was differential or not. First, we used simulated data sets in which nondifferential and differential misclassification were introduced. Then, we used data on obesity and diabetes from the National Health and Nutrition Examination Survey (NHANES) in which both self-reported (misclassified) and measured (true) obesity were available, using literature estimates of sensitivity and specificity to adjust for bias. The ratio of odds ratio (ROR; observed odds ratio divided by true odds ratio) was used to quantify magnitude of bias, with ROR = 1 signifying no bias.

RESULTS

In the simulated data sets, under incorrect assumptions (eg, assuming nondifferential misclassification when it was truly differential), results were biased, with RORs ranging from 0.18 to 2.46. In NHANES, results adjusted based on incorrect assumptions also produced biased results, with RORs ranging from 1.26 to 1.55; results were more biased when making these adjustments than when using the misclassified exposure values (ROR = 0.91).

CONCLUSIONS

Making an incorrect assumption about nondifferential or differential exposure misclassification in bias analyses can lead to more biased results than if no adjustment is performed. In our analyses, incorporating uncertainty using probabilistic bias analysis was not sufficient to overcome this problem.

The National Birth Defects Prevention Study: A review of the methods

BACKGROUND

The National Birth Defects Prevention Study (NBDPS) is a large population-based multicenter case-control study of major birth defects in the United States.

METHODS

Data collection took place from 1998 through 2013 on pregnancies ending between October 1997 and December 2011. Cases could be live born, stillborn, or induced terminations, and were identified from birth defects surveillance programs in Arkansas, California, Georgia, Iowa, Massachusetts, New Jersey, New York, North Carolina, Texas, and Utah. Controls were live born infants without major birth defects identified from the same geographical regions and time periods as cases by means of either vital records or birth hospitals. Computer-assisted telephone interviews were completed with women between 6 weeks and 24 months after the estimated date of delivery. After completion of interviews, families received buccal cell collection kits for the mother, father, and infant (if living).

RESULTS

There were 47,832 eligible cases and 18,272 eligible controls. Among these, 32,187 (67%) and 11,814 (65%), respectively, provided interview information about their pregnancies. Buccal cell collection kits with a cytobrush for at least one family member were returned by 19,065 case and 6,211 control families (65% and 59% of those who were sent a kit). More than 500 projects have been proposed by the collaborators and over 200 manuscripts published using data from the NBDPS through December 2014.

CONCLUSION

The NBDPS has made substantial contributions to the field of birth defects epidemiology through its rigorous design, including case classification, detailed questionnaire and specimen collection, large study population, and collaborative activities across Centers.

Misclassification in administrative claims data: quantifying the impact on treatment effect estimates

Misclassification is present in nearly every epidemiologic study, yet is rarely quantified in analysis in favor of a focus on random error. In this review, we discuss past and present wisdom on misclassification and what measures should be taken to quantify this influential bias, with a focus on bias in pharmacoepidemiologic studies. To date, pharmacoepidemiology primarily utilizes data obtained from administrative claims, a rich source of prescription data but susceptible to bias from unobservable factors including medication sample use, medications filled but not taken, health conditions that are not reported in the administrative billing data, and inadequate capture of confounders. Due to the increasing focus on comparative effectiveness research, we provide a discussion of misclassification in the context of an active comparator, including a demonstration of treatment effects biased away from the null in the presence of nondifferential misclassification. Finally, we highlight recently developed methods to quantify bias and offer these methods as potential options for strengthening the validity and quantifying uncertainty of results obtained from pharmacoepidemiologic research.

Using bayesian models to assess the effects of under-reporting of cannabis use on the association with birth defects, national birth defects prevention study, 1997-2005

BACKGROUND

Studies on associations between periconceptional cannabis exposure and birth defects have mainly relied on self-reported exposure. Therefore, the results may be biased due to under-reporting of the exposure. The aim of this study was to quantify the potential effects of this form of exposure misclassification.

METHODS

Using multivariable logistic regression, we re-analysed associations between periconceptional cannabis use and 20 specific birth defects using data from the National Birth Defects Prevention Study from 1997-2005 for 13 859 case infants and 6556 control infants. For seven birth defects, we implemented four Bayesian models based on various assumptions concerning the sensitivity of self-reported cannabis use to estimate odds ratios (ORs), adjusted for confounding and under-reporting of the exposure. We used information on sensitivity of self-reported cannabis use from the literature for prior assumptions.

RESULTS

The results unadjusted for under-reporting of the exposure showed an association between cannabis use and anencephaly (posterior OR 1.9 [95% credible interval (CRI) 1.1, 3.2]) which persisted after adjustment for potential exposure misclassification. Initially, no statistically significant associations were observed between cannabis use and the other birth defect categories studied. Although adjustment for under-reporting did not notably change these effect estimates, cannabis use was associated with esophageal atresia (posterior OR 1.7 [95% CRI 1.0, 2.9]), diaphragmatic hernia (posterior OR 1.8 [95% CRI 1.1, 3.0]), and gastroschisis (posterior OR 1.7 [95% CRI 1.2, 2.3]) after correction for exposure misclassification.

CONCLUSIONS

Under-reporting of the exposure may have obscured some cannabis-birth defect associations in previous studies. However, the resulting bias is likely to be limited.

Prevention of orofacial clefts caused by smoking: implications of the Surgeon General's report

BACKGROUND

According to the 2014 Surgeon General's Report, smoking in early pregnancy can cause orofacial clefts. We sought to examine the implications of this causal link for the potential prevention of orofacial clefts in the United States.

METHODS

Using published data on the strength of the association between orofacial clefts and smoking in early pregnancy and the prevalence of smoking at the start of pregnancy, we estimated the attributable fraction for smoking as a cause of orofacial clefts. We then used the prevalence of orofacial clefts in the United States to estimate the number of orofacial clefts that could be prevented in the United States each year by eliminating exposure to smoking during early pregnancy. We also estimated the financial impact of preventing orofacial clefts caused by maternal smoking based on a published estimate of attributable healthcare costs through age 10 for orofacial clefts.

RESULTS

The estimated attributable fraction of orofacial clefts caused by smoking in early pregnancy was 6.1% (95% uncertainty interval 4.4%, 7.7%). Complete elimination of smoking in early pregnancy could prevent orofacial clefts in approximately 430 infants per year in the United States, and could save an estimated $40.4 million in discounted healthcare costs through age 10 for each birth cohort.

CONCLUSION

Understanding the magnitude of the preventable burden of orofacial clefts related to maternal smoking could help focus smoking cessation efforts on women who might become pregnant.

Autism spectrum disorder, flea and tick medication, and adjustments for exposure misclassification: the CHARGE (CHildhood Autism Risks from Genetics and Environment) case-control study

BACKGROUND

The environmental contribution to autism spectrum disorders (ASD) is largely unknown, but household pesticides are receiving increased attention. We examined associations between ASD and maternally-reported use of imidacloprid, a common flea and tick treatment for pets.

METHODS

Bayesian logistic models were used to estimate the association between ASD and imidacloprid and to correct for potential differential exposure misclassification due to recall in a case control study of ASD.

RESULTS

Our analytic dataset included complete information for 262 typically developing controls and 407 children with ASD. Compared with exposure among controls, the odds of prenatal imidacloprid exposure among children with ASD were slightly higher, with an odds ratio (OR) of 1.3 (95% Credible Interval [CrI] 0.78, 2.2). A susceptibility window analysis yielded higher ORs for exposures during pregnancy than for early life exposures, whereas limiting to frequent users of imidacloprid, the OR increased to 2.0 (95% CI 1.0, 3.9).

CONCLUSIONS

Within plausible estimates of sensitivity and specificity, the association could result from exposure misclassification alone. The association between imidacloprid exposure and ASD warrants further investigation, and this work highlights the need for validation studies regarding prenatal exposures in ASD.

Bayesian sensitivity analysis methods to evaluate bias due to misclassification and missing data using informative priors and external validation data

BACKGROUND

Recent research suggests that the Bayesian paradigm may be useful for modeling biases in epidemiological studies, such as those due to misclassification and missing data. We used Bayesian methods to perform sensitivity analyses for assessing the robustness of study findings to the potential effect of these two important sources of bias.

METHODS

We used data from a study of the joint associations of radiotherapy and smoking with primary lung cancer among breast cancer survivors. We used Bayesian methods to provide an operational way to combine both validation data and expert opinion to account for misclassification of the two risk factors and missing data. For comparative purposes we considered a "full model" that allowed for both misclassification and missing data, along with alternative models that considered only misclassification or missing data, and the naïve model that ignored both sources of bias.

RESULTS

We identified noticeable differences between the four models with respect to the posterior distributions of the odds ratios that described the joint associations of radiotherapy and smoking with primary lung cancer. Despite those differences we found that the general conclusions regarding the pattern of associations were the same regardless of the model used. Overall our results indicate a nonsignificantly decreased lung cancer risk due to radiotherapy among nonsmokers, and a mildly increased risk among smokers.

CONCLUSIONS

We described easy to implement Bayesian methods to perform sensitivity analyses for assessing the robustness of study findings to misclassification and missing data.

Correcting for exposure misclassification using survival analysis with a time-varying exposure

PURPOSE

Survival analysis is increasingly being used in perinatal epidemiology to assess time-varying risk factors for various pregnancy outcomes. Here we show how quantitative correction for exposure misclassification can be applied to a Cox regression model with a time-varying dichotomous exposure.

METHODS

We evaluated influenza vaccination during pregnancy in relation to preterm birth among 2267 non-malformed infants whose mothers were interviewed as part of the Slone Birth Defects Study during 2006 through 2011. The hazard of preterm birth was modeled using a time-varying exposure Cox regression model with gestational age as the time-scale. The effect of exposure misclassification was then modeled using a probabilistic bias analysis that incorporated vaccination date assignment. The parameters for the bias analysis were derived from both internal and external validation data.

RESULTS

Correction for misclassification of prenatal influenza vaccination resulted in an adjusted hazard ratio (AHR) slightly higher and less precise than the conventional analysis: Bias-corrected AHR 1.04 (95% simulation interval, 0.70-1.52); conventional AHR, 1.00 (95% confidence interval, 0.71-1.41).

CONCLUSIONS

Probabilistic bias analysis allows epidemiologists to assess quantitatively the possible confounder-adjusted effect of misclassification of a time-varying exposure, in contrast with a speculative approach to understanding information bias.

Maternal occupational exposure to polycyclic aromatic hydrocarbons: effects on gastroschisis among offspring in the National Birth Defects Prevention Study

BACKGROUND

Exposure to polycyclic aromatic hydrocarbons (PAHs) occurs in many occupational settings. There is evidence in animal models that maternal exposure to PAHs during pregnancy is associated with gastroschisis in offspring; however, to our knowledge, no human studies examining this association have been conducted.

OBJECTIVE

Our goal was to conduct a case-control study assessing the association between estimated maternal occupational exposure to PAHs and gastroschisis in offspring.

METHODS

Data from gastroschisis cases and control infants were obtained from the population-based National Birth Defects Prevention Study for the period 1997-2002. Exposure to PAHs was assigned by industrial hygienist consensus, based on self-reported maternal occupational histories from 1 month before conception through the third month of pregnancy. Logistic regression was used to determine the association between estimated occupational PAH exposure and gastroschisis among children whose mothers were employed for at least 1 month during the month before conception through the third month of pregnancy.

RESULTS

The prevalence of estimated occupational PAH exposure was 9.0% in case mothers (27 of 299) and 3.6% in control mothers (107 of 2,993). Logistic regression analyses indicated a significant association between occupational PAHs and gastroschisis among mothers ≥ 20 years of age [odds ratio (OR) = 2.53; 95% confidence interval (CI): 1.27, 5.04] after adjusting for maternal body mass index, education, gestational diabetes, and smoking. This association was not seen in mothers < 20 years (OR = 1.14; 95% CI: 0.55, 2.33), which is notable because although young maternal age is the strongest known risk factor for gastroschisis, most cases are born to mothers ≥ 20 years.

CONCLUSION

Our findings indicate an association between occupational exposure to PAHs among mothers who are ≥ 20 years and gastroschisis. These results contribute to a body of evidence that PAHs may be teratogenic.

Is probabilistic bias analysis approximately Bayesian?

Case-control studies are particularly susceptible to differential exposure misclassification when exposure status is determined following incident case status. Probabilistic bias analysis methods have been developed as ways to adjust standard effect estimates based on the sensitivity and specificity of exposure misclassification. The iterative sampling method advocated in probabilistic bias analysis bears a distinct resemblance to a Bayesian adjustment; however, it is not identical. Furthermore, without a formal theoretical framework (Bayesian or frequentist), the results of a probabilistic bias analysis remain somewhat difficult to interpret. We describe, both theoretically and empirically, the extent to which probabilistic bias analysis can be viewed as approximately Bayesian. Although the differences between probabilistic bias analysis and Bayesian approaches to misclassification can be substantial, these situations often involve unrealistic prior specifications and are relatively easy to detect. Outside of these special cases, probabilistic bias analysis and Bayesian approaches to exposure misclassification in case-control studies appear to perform equally well.

GENES AS INSTRUMENTS FOR STUDYING RISK BEHAVIOR EFFECTS: AN APPLICATION TO MATERNAL SMOKING AND OROFACIAL CLEFTS

This study uses instrumental variable (IV) models with genetic instruments to assess the effects of maternal smoking on the child's risk of orofacial clefts (OFC), a common birth defect. The study uses genotypic variants in neurotransmitter and detoxification genes relateded to smoking as instruments for cigarette smoking before and during pregnancy. Conditional maximum likelihood and two-stage IV probit models are used to estimate the IV model. The data are from a population-level sample of affected and unaffected children in Norway. The selected genetic instruments generally fit the IV assumptions but may be considered "weak" in predicting cigarette smoking. We find that smoking before and during pregnancy increases OFC risk substantially under the IV model (by about 4-5 times at the sample average smoking rate). This effect is greater than that found with classical analytic models. This may be because the usual models are not able to consider self-selection into smoking based on unobserved confounders, or it may to some degree reflect limitations of the instruments. Inference based on weak-instrument robust confidence bounds is consistent with standard inference. Genetic instruments may provide a valuable approach to estimate the "causal" effects of risk behaviors with genetic-predisposing factors (such as smoking) on health and socioeconomic outcomes.

Maternal smoking and congenital heart defects in the Baltimore-Washington Infant Study

OBJECTIVE

We investigated associations between maternal cigarette smoking during the first trimester and the risk of congenital heart defects (CHDs) among the infants.

METHODS

The Baltimore-Washington Infant Study was the first population-based case-control study of CHDs conducted in the United States. Case and control infants were enrolled during the period 1981-1989. We excluded mothers with overt pregestational diabetes and case mothers whose infants had noncardiac anomalies (with the exception of atrioventricular septal defects with Down syndrome) from the analysis, which resulted in 2525 case and 3435 control infants. Self-reported first-trimester maternal cigarette consumption was ascertained via an in-person interview after delivery. Associations for 26 different groups of CHDs with maternal cigarette consumption were estimated by using logistic regression models. Odds ratios (ORs) corresponded to a 20-cigarette-per-day increase in consumption.

RESULTS

We observed statistically significant positive associations between self-reported first-trimester maternal cigarette consumption and the risk of secundum-type atrial septal defects (OR: 1.36 [95% confidence interval (CI): 1.04-1.78]), right ventricular outflow tract defects (OR: 1.32 [95% CI: 1.06-1.65]), pulmonary valve stenosis (OR: 1.35 [95% CI: 1.05-1.74]), truncus arteriosus (OR: 1.90 [95% CI: 1.04-3.45]), and levo-transposition of the great arteries (OR: 1.79 [95% CI: 1.04-3.10]). A suggestive association was observed for atrioventricular septal defects among infants without Down syndrome (OR: 1.50 [95% CI: 0.99-2.29]).

CONCLUSIONS

These findings add to the existing body of evidence that implicates first-trimester maternal cigarette smoking as a modest risk factor for select CHD phenotypes.

Bayesian bias adjustments of the lung cancer SMR in a cohort of German carbon black production workers

BACKGROUND

A German cohort study on 1,528 carbon black production workers estimated an elevated lung cancer SMR ranging from 1.8-2.2 depending on the reference population. No positive trends with carbon black exposures were noted in the analyses. A nested case control study, however, identified smoking and previous exposures to known carcinogens, such as crystalline silica, received prior to work in the carbon black industry as important risk factors.We used a Bayesian procedure to adjust the SMR, based on a prior of seven independent parameter distributions describing smoking behaviour and crystalline silica dust exposure (as indicator of a group of correlated carcinogen exposures received previously) in the cohort and population as well as the strength of the relationship of these factors with lung cancer mortality. We implemented the approach by Markov Chain Monte Carlo Methods (MCMC) programmed in R, a statistical computing system freely available on the internet, and we provide the program code.

RESULTS

When putting a flat prior to the SMR a Markov chain of length 1,000,000 returned a median posterior SMR estimate (that is, the adjusted SMR) in the range between 1.32 (95% posterior interval: 0.7, 2.1) and 1.00 (0.2, 3.3) depending on the method of assessing previous exposures.

CONCLUSIONS

Bayesian bias adjustment is an excellent tool to effectively combine data about confounders from different sources. The usually calculated lung cancer SMR statistic in a cohort of carbon black workers overestimated effect and precision when compared with the Bayesian results. Quantitative bias adjustment should become a regular tool in occupational epidemiology to address narrative discussions of potential distortions.

Specifying exposure classification parameters for sensitivity analysis: family breast cancer history

One of the challenges to implementing sensitivity analysis for exposure misclassification is the process of specifying the classification proportions (eg, sensitivity and specificity). The specification of these assignments is guided by three sources of information: estimates from validation studies, expert judgment, and numerical constraints given the data. The purpose of this teaching paper is to describe the process of using validation data and expert judgment to adjust a breast cancer odds ratio for misclassification of family breast cancer history. The parameterization of various point estimates and prior distributions for sensitivity and specificity were guided by external validation data and expert judgment. We used both nonprobabilistic and probabilistic sensitivity analyses to investigate the dependence of the odds ratio estimate on the classification error. With our assumptions, a wider range of odds ratios adjusted for family breast cancer history misclassification resulted than portrayed in the conventional frequentist confidence interval.