Search

Ensuring that patients with opioid use disorder (OUD) have access to optimal medication therapies is a critical challenge in substance use epidemiology. Rudolph et al. (Am J Epidemiol. 2023;XXX(X):XXXX-XXXX) demonstrated that sophisticated data-adaptive statistical techniques can be used to learn optimal, individualized treatment rules that can aid providers in choosing a medication treatment modality for a particular patient with OUD. This important work also highlights the effects of the mathematization of epidemiologic research. Here, we define mathematization and demonstrate how it operates in the context of effectiveness research on medications for OUD using the paper by Rudolph et al. as a springboard. In particular, we address the normative dimension of mathematization and how it tends to resolve a fundamental tension in epidemiologic practice between technical sophistication and public health considerations in favor of more technical solutions. The process of mathematization is a fundamental part of epidemiology; we argue not for eliminating it but for balancing mathematization and technical demands equally with practical and community-centric public health needs. [ABSTRACT FROM AUTHOR]

In this paper, we consider methods for generating draws of a binary random variable whose expectation conditional on covariates follows a logistic regression model with known covariate coefficients. We examine approximations for finding a "balancing intercept," that is, a value for the intercept of the logistic model that leads to a desired marginal expectation for the binary random variable. We show that a recently proposed analytical approximation can produce inaccurate results, especially when targeting more extreme marginal expectations or when the linear predictor of the regression model has high variance. We then formulate the balancing intercept as a solution to an integral equation, implement a numerical approximation for solving the equation based on Monte Carlo methods, and show that the approximation works well in practice. Our approach to the basic problem of the balancing intercept provides an example of a broadly applicable strategy for formulating and solving problems that arise in the design of simulation studies used to evaluate or teach epidemiologic methods. [ABSTRACT FROM AUTHOR]

In diagnostic medicine, the true disease status of a patient is often represented on an ordinal scale—for example, cancer stage (0, I, II, III, or IV) or coronary artery disease severity measured using the Coronary Artery Disease Reporting and Data System (CAD-RADS) scale (none, minimal, mild, moderate, severe, or occluded). With advances in quantitation of diagnostic images and in artificial intelligence (AI), both supervised and unsupervised algorithms are being developed to help physicians correctly grade disease. Most of the diagnostic accuracy literature deals with binary disease status (disease present or absent); however, tests diagnosing ordinal-scaled diseases should not be reduced to a binary status just to simplify diagnostic accuracy testing. In this paper, we propose different characterizations of ordinal-scale accuracy for different clinical use scenarios, along with methods for comparing tests. In the simplest scenario, just the proportion of correct grades is considered; other scenarios address the magnitude and direction of misgrading; and at the other extreme, a weighted accuracy measure with weights based on the relative costs of different types of misgrading is presented. The various scenarios are illustrated using a coronary artery disease example where the accuracy of AI algorithms in providing patients with the correct CAD-RADS grade is assessed. [ABSTRACT FROM AUTHOR]

In his 1976 paper "Estimability and Estimation in Case-Referent Studies" (Am J Epidemiol. 1976;103(2):226-235), Miettinen weaved together a patchwork of new ideas into a coherent view of case-control studies. His article spurred theoretical development in epidemiologic methods and became a platform for teaching about some key concepts in epidemiologic study design. [ABSTRACT FROM AUTHOR]

Measurement error is pervasive in epidemiologic research. Epidemiologists often assume that mismeasurement of study variables is nondifferential with respect to other analytical variables and then rely on the heuristic that "nondifferential misclassification will bias estimates towards the null." However, there are many exceptions to the heuristic for which bias towards the null cannot be assumed. In this paper, we compile and characterize 7 exceptions to this rule and encourage analysts to take a more critical and nuanced approach to evaluating and discussing bias from nondifferential mismeasurement. [ABSTRACT FROM AUTHOR]

Simulation methods are a powerful set of tools that can allow researchers to better characterize phenomena from the real world. As such, the ability to simulate data represents a critical set of skills that epidemiologists should use to better understand epidemiologic concepts and ensure that they have the tools to continue to self-teach even when their formal instruction ends. Simulation methods are not always taught in epidemiology methods courses, whereas causal directed acyclic graphs (DAGs) often are. Therefore, this paper details an approach to building simulations from DAGs and provides examples and code for learning to perform simulations. We recommend using very simple DAGs to learn the procedures and code necessary to set up a simulation that builds on key concepts frequently of interest to epidemiologists (e.g. mediation, confounding bias, M bias). We believe that following this approach will allow epidemiologists to gain confidence with a critical skill set that may in turn have a positive impact on how they conduct future epidemiologic studies. [ABSTRACT FROM AUTHOR]

Data collected from a validation substudy permit calculation of a bias-adjusted estimate of effect that is expected to equal the estimate that would have been observed had the gold standard measurement been available for the entire study population. In this paper, we develop and apply a framework for adaptive validation to determine when sufficient validation data have been collected to yield a bias-adjusted effect estimate with a prespecified level of precision. Prespecified levels of precision are decided a priori by the investigator, based on the precision of the conventional estimate and allowing for wider confidence intervals that would still be substantively meaningful. We further present an applied example of the use of this method to address exposure misclassification in a study of transmasculine/transfeminine youth and self-harm. Our method provides a novel approach to effective and efficient estimation of classification parameters as validation data accrue, with emphasis on the precision of the bias-adjusted estimate. This method can be applied within the context of any parent epidemiologic study design in which validation data will be collected and modified to meet alternative criteria given specific study or validation study objectives. [ABSTRACT FROM AUTHOR]

An increasing number of recent studies have suggested that doubly robust estimators with cross-fitting should be used when estimating causal effects with machine learning methods. However, not all existing programs that implement doubly robust estimators support machine learning methods and cross-fitting, or provide estimates on multiplicative scales. To address these needs, we developed AIPW , a software package implementing augmented inverse probability weighting (AIPW) estimation of average causal effects in R (R Foundation for Statistical Computing, Vienna, Austria). Key features of the AIPW package include cross-fitting and flexible covariate adjustment for observational studies and randomized controlled trials (RCTs). In this paper, we use a simulated RCT to illustrate implementation of the AIPW estimator. We also perform a simulation study to evaluate the performance of the AIPW package compared with other doubly robust implementations, including CausalGAM , npcausal , tmle , and tmle3. Our simulation showed that the AIPW package yields performance comparable to that of other programs. Furthermore, we also found that cross-fitting substantively decreases the bias and improves the confidence interval coverage for doubly robust estimators fitted with machine learning algorithms. Our findings suggest that the AIPW package can be a useful tool for estimating average causal effects with machine learning methods in RCTs and observational studies. [ABSTRACT FROM AUTHOR]

Laboratory-based animal research has revealed a number of exposures with multigenerational effects—ones that affect the children and grandchildren of those directly exposed. An important task for epidemiology is to investigate these relationships in human populations. Without the relative control achieved in laboratory settings, however, population-based studies of multigenerational associations have had to use a broader range of study designs. Current strategies to obtain multigenerational data include exploiting birth registries and existing cohort studies, ascertaining exposures within them, and measuring outcomes across multiple generations. In this paper, we describe the methodological challenges inherent to multigenerational studies in human populations. After outlining standard taxonomy to facilitate discussion of study designs and target exposure associations, we highlight the methodological issues, focusing on the interplay between study design, analysis strategy, and the fact that outcomes may be related to family size. In a simulation study, we show that different multigenerational designs lead to estimates of different exposure associations with distinct scientific interpretations. Nevertheless, target associations can be recovered by incorporating (possibly) auxiliary information, and we provide insights into choosing an appropriate target association. Finally, we identify areas requiring further methodological development. [ABSTRACT FROM AUTHOR]

A conventional Mendelian randomization analysis assesses the causal effect of a risk factor on an outcome by using genetic variants that are solely associated with the risk factor of interest as instrumental variables. However, in some cases, such as the case of triglyceride level as a risk factor for cardiovascular disease, it may be difficult to find a relevant genetic variant that is not also associated with related risk factors, such as other lipid fractions. Such a variant is known as pleiotropic. In this paper, we propose an extension of Mendelian randomization that uses multiple genetic variants associated with several measured risk factors to simultaneously estimate the causal effect of each of the risk factors on the outcome. This “multivariable Mendelian randomization” approach is similar to the simultaneous assessment of several treatments in a factorial randomized trial. In this paper, methods for estimating the causal effects are presented and compared using real and simulated data, and the assumptions necessary for a valid multivariable Mendelian randomization analysis are discussed. Subject to these assumptions, we demonstrate that triglyceride-related pathways have a causal effect on the risk of coronary heart disease independent of the effects of low-density lipoprotein cholesterol and high-density lipoprotein cholesterol. [ABSTRACT FROM PUBLISHER]

The field of epidemiology has been defined as the study of the spread and control of disease. However, epidemiology frequently focuses on studies of etiology and distribution of disease at the cost of understanding the best ways to control disease. Moreover, only a small fraction of scientific discoveries are translated into public health practice, and the process from discovery to translation is exceedingly slow. Given the importance of translational science, the future of epidemiologic training should include competency in implementation science, whose goal is to rapidly move evidence into practice. Our purpose in this paper is to provide epidemiologists with a primer in implementation science, which includes dissemination research and implementation research as defined by the National Institutes of Health. We describe the basic principles of implementation science, highlight key components for conducting research, provide examples of implementation studies that encompass epidemiology, and offer resources and opportunities for continued learning. There is a clear need for greater speed, relevance, and application of evidence into practice, programs, and policies and an opportunity to enable epidemiologists to conduct research that not only will inform practitioners and policymakers of risk but also will enhance the likelihood that evidence will be implemented. [ABSTRACT FROM AUTHOR]

Longitudinal data sources offer new opportunities for the evaluation of sequential interventions. To adjust for time-dependent confounding in these settings, longitudinal targeted maximum likelihood based estimation (TMLE), a doubly robust method that can be coupled with machine learning, has been proposed. This paper provides a tutorial in applying longitudinal TMLE, in contrast to inverse probability of treatment weighting and g-computation based on iterative conditional expectations. We apply these methods to estimate the causal effect of nutritional interventions on clinical outcomes among critically ill children in a United Kingdom study (Control of Hyperglycemia in Paediatric Intensive Care, 2008-2011). We estimate the probability of a child's being discharged alive from the pediatric intensive care unit by a given day, under a range of static and dynamic feeding regimes. We find that before adjustment, patients who follow the static regime "never feed" are discharged by the end of the fifth day with a probability of 0.88 (95% confidence interval: 0.87, 0.90), while for the patients who follow the regime "feed from day 3," the probability of discharge is 0.64 (95% confidence interval: 0.62, 0.66). After adjustment for time-dependent confounding, most of this difference disappears, and the statistical methods produce similar results. TMLE offers a flexible estimation approach; hence, we provide practical guidance on implementation to encourage its wider use. [ABSTRACT FROM AUTHOR]

Distributed networks of health-care data sources are increasingly being utilized to conduct pharmacoepidemio-logic database studies. Such networks may contain data that are not physically pooled but instead are distributed horizontally (separate patients within each data source) or vertically (separate measures within each data source) in order to preserve patient privacy. While multivariable methods for the analysis of horizontally distributed data are frequently employed, few practical approaches have been put forth to deal with vertically distributed healthcare databases. In this paper, we propose 2 propensity score-based approaches to vertically distributed data analysis and test their performance using 5 example studies. We found that these approaches produced point estimates close to what could be achieved without partitioning. We further found a performance benefit (i.e., lower mean squared error) for sequentially passing a propensity score through each data domain (called the "sequential approach") as compared with fitting separate domain-specific propensity scores (called the "parallel approach"). These results were validated in a small simulation study. This proof-of-concept study suggests a new multivariable analysis approach to vertically distributed health-care databases that is practical, preserves patient privacy, and warrants further investigation for use in clinical research applications that rely on health-care databases. [ABSTRACT FROM AUTHOR]

Influenza viruses undergo frequent antigenic changes. As a result, the viruses circulating change within and between seasons, and the composition of the influenza vaccine is updated annually. Thus, estimation of the vaccine's effectiveness is not constant across seasons. In order to provide annual estimates of the influenza vaccine's effectiveness, health departments have increasingly adopted the "test-negative design," using enhanced data from routine surveillance systems. In this design, patients presenting to participating general practitioners with influenza-like illness are swabbed for laboratory testing; those testing positive for influenza virus are defined as cases, and those testing negative form the comparison group. Data on patients' vaccination histories and confounder profiles are also collected. Vaccine effectiveness is estimated from the odds ratio comparing the odds of testing positive for influenza among vaccinated patients and unvaccinated patients, adjusting for confounders. The test-negative design is purported to reduce bias associated with confounding by health-care-seeking behavior and misclassification of cases. In this paper, we use directed acyclic graphs to characterize potential biases in studies of influenza vaccine effectiveness using the test-negative design. We show how studies using this design can avoid or minimize bias and where bias may be introduced with particular study design variations. [ABSTRACT FROM AUTHOR]

The idea that epidemiologic studies should start from first exposure onward has been advocated in the past few years. The study of incident exposures is contrasted with studies of prevalent exposures in which follow-up may commence after first exposure. The former approach is seen as a hallmark of a good study and necessary for causal inference. We argue that studying incident exposures may be necessary in some situations, but it is not always necessary and is not the preferred option in many instances. Conducting a study involves decisions as to which person-time experience should be included. Although studies of prevalent exposures involve left truncation (missingness on the left), studies of incident exposures may involve right censoring (missingness on the right) and therefore may not be able to assess the long-term effects of exposure. These considerations have consequences for studies of dynamic (open) populations that involve a mixture of prevalent and incident exposures. We argue that studies with prevalent exposures will remain a necessity for epidemiology. The purpose of this paper is to restore the balance between the emphasis on first exposure cohorts and the richness of epidemiologic information obtained when studying prevalent exposures. [ABSTRACT FROM AUTHOR]

Researchers conducting observational studies need to consider 3 types of biases: selection bias, information bias, and confounding bias. A whole arsenal of statistical tools can be used to deal with information and confounding biases. However, methods for addressing selection bias and unmeasured confounding are less developed. In this paper, we propose general bounding formulas for bias, including selection bias and unmeasured confounding. This should help researchers make more prudent interpretations of their (potentially biased) results. [ABSTRACT FROM AUTHOR]

In cohort mortality studies, there often is interest in associations between an exposure of primary interest and mortality due to a range of different causes. A standard approach to such analyses involves fitting a separate regression model for each type of outcome. However, the statistical precision of some estimated associations may be poor because of sparse data. In this paper, we describe a hierarchical regression model for estimation of parameters describing outcome-specific relative rate functions and associated credible intervals. The proposed model uses background stratification to provide flexible control for the outcome-specific associations of potential confounders, and it employs a hierarchical “shrinkage” approach to stabilize estimates of an exposure’s associations with mortality due to different causes of death. The approach is illustrated in analyses of cancer mortality in 2 cohorts: a cohort of dioxin-exposed US chemical workers and a cohort of radiation-exposed Japanese atomic bomb survivors. Compared with standard regression estimates of associations, hierarchical regression yielded estimates with improved precision that tended to have less extreme values. The hierarchical regression approach also allowed the fitting of models with effect-measure modification. The proposed hierarchical approach can yield estimates of association that are more precise than conventional estimates when one wishes to estimate associations with multiple outcomes. [ABSTRACT FROM AUTHOR]

To detect gene-environment interactions, a logistic regression model is typically fitted to a set of case-control data, and the focus is on testing of the cross-product terms (gene × environment) in the model. A significant result is indicative of a gene-environment interaction under a multiplicative model for disease odds. Based on the sufficient-cause model for rates, in this paper we put forward a general approach to testing for sufficient-cause gene-environment interactions in case-control studies. The proposed tests can be tailored to detect a particular type of sufficient-cause gene-environment interaction with greater sensitivity. These tests include testing for autosomal dominant, autosomal recessive, and gene-dosage interactions. The tests can also detect trend interactions (e.g., a larger gene-environment interaction with a higher level of environmental exposure) and threshold interactions (e.g., gene-environment interaction occurs only when environmental exposure reaches a certain threshold level). Two assumptions are necessary for the validity of the tests: 1) the rare-disease assumption and 2) the no-redundancy assumption. Another 2 assumptions are optional but, if imposed correctly, can boost the statistical powers of the tests: 3) the gene-environment independence assumption and 4) the Hardy-Weinberg equilibrium assumption. SAS code (SAS Institute, Inc., Cary, North Carolina) for implementing the methods is provided. [ABSTRACT FROM AUTHOR]

Instrumental variable (IV) methods are increasingly being used in comparative effectiveness research. Studies using these methods often compare 2 particular treatments, and the researchers perform their IV analyses conditional on patients’ receiving this subset of treatments (while ignoring the third option of “neither treatment”). The ensuing selection bias that occurs due to this restriction has gone relatively unnoticed in interpretations and discussions of these studies’ results. In this paper we describe the structure of this selection bias with examples drawn from commonly proposed instruments such as calendar time and preference, illustrate the bias with causal diagrams, and estimate the magnitude and direction of possible bias using simulations. A noncausal association between the proposed instrument and the outcome can occur in analyses restricted to patients receiving a subset of the possible treatments. This results in bias in the numerator for the standard IV estimator; the bias is amplified in the treatment effect estimate. The direction and magnitude of the bias in the treatment effect estimate are functions of the distribution of and relationships between the proposed instrument, treatment values, unmeasured confounders, and outcome. IV methods used to compare a subset of treatment options are prone to substantial biases, even when the proposed instrument appears relatively strong. [ABSTRACT FROM AUTHOR]

Cohort studies are often enriched for a primary exposure of interest to improve cost-effectiveness, which presents analytical challenges not commonly discussed in epidemiology. In this paper, we use causal diagrams to represent exposure-enriched cohort studies, illustrate a scenario wherein the risk ratio for the effect of a secondary exposure on an outcome is biased, and propose an analytical method for correcting for such bias. In our motivating example, maternal smoking (Z) is a cause of fetal growth restriction (X), which subsequently affects preterm birth (Y) (i.e., Z → X → Y); strong positive associations exist between both Z, X and X, Y; and enrichment for X increases its prevalence from 10% to 50%. In the X-enriched cohort, unadjusted and X-adjusted analyses lead to bias in the risk ratio for the total effect of Z on Y. After application of inverse probability weights, the bias is corrected, with a small loss of efficiency in comparison with a same-sized study without X-enrichment. With increasing interest in conducting secondary analyses to reduce research costs, caution should be employed when analyzing studies that have already been enriched, intentionally or unintentionally, for a primary exposure of interest. Causal diagrams can help identify scenarios in which secondary analyses may be biased. Inverse probability weights can be used to remove the bias. [ABSTRACT FROM AUTHOR]

Medical devices play a vital role in diagnosing, treating, and preventing diseases and are an integral part of the health-care system. Many devices, including implantable medical devices, enter the market through a regulatory pathway that was not designed to assure safety and effectiveness. Several recent studies and high-profile device recalls have demonstrated the need for well-designed, valid postmarketing studies of medical devices. Medical device epidemiology is a relatively new field compared with pharmacoepidemiology, which for decades has been developed to assess the safety and effectiveness of medications. Many methodological considerations in pharmacoepidemiology apply to medical device epidemiology. Fundamental differences in mechanisms of action and use and in how exposure data are captured mean that comparative effectiveness studies of medical devices often necessitate additional and different considerations. In this paper, we discuss some of the most salient issues encountered in conducting comparative effectiveness research on implantable devices. We discuss special methodological considerations regarding the use of data sources, exposure and outcome definitions, timing of exposure, and sources of bias. [ABSTRACT FROM PUBLISHER]

Inverse probability weighting (IPW), a well-established method of controlling for confounding in observational studies with binary exposures, has been extended to analyses with continuous exposures. Methods developed for continuous exposures may not apply when the exposure is quasicontinuous because of irregular exposure distributions that violate key assumptions. We used simulations and cluster-randomized clinical trial data to assess 4 approaches developed for continuous exposures—ordinary least squares (OLS), covariate balancing generalized propensity scores (CBGPS), nonparametric covariate balancing generalized propensity scores (npCBGPS), and quantile binning (QB)—and a novel method, a cumulative probability model (CPM), in quasicontinuous exposure settings. We compared IPW stability, covariate balance, bias, mean squared error, and standard error estimation across 3,000 simulations with 6 different quasicontinuous exposures, varying in skewness and granularity. In general, CBGPS and npCBGPS resulted in excellent covariate balance, and npCBGPS was the least biased but the most variable. The QB and CPM approaches had the lowest mean squared error, particularly with marginally skewed exposures. We then successfully applied the IPW approaches, together with missing-data techniques, to assess how session attendance (out of a possible 15) in a partners-based clustered intervention among pregnant couples living with human immunodeficiency virus in Mozambique (2017–2022) influenced postpartum contraceptive uptake. [ABSTRACT FROM AUTHOR]

A covariate-adjusted estimate of an exposure-outcome association may be biased if the exposure variable suffers measurement error. We propose an approach to correct for exposure measurement error in a covariate-adjusted estimate of the association between a continuous exposure variable and outcome of interest. Our proposed approach requires data for a reference population in which the exposure was a priori set to some known level (e.g. 0, and is therefore unexposed); however, our approach does not require an exposure validation study or replicate measures of exposure, which are typically needed when addressing bias due to exposure measurement error. A key condition for this method, which we refer to as "partial population exchangeability," requires that the association between a measured covariate and outcome in the reference population equals the association between that covariate and outcome in the target population in the absence of exposure. We illustrate the approach using simulations and an example. [ABSTRACT FROM AUTHOR]

Dengue is a serious public health concern in Brazil and globally. In the absence of a universal vaccine or specific treatments, prevention relies on vector control and disease surveillance. Accurate and early forecasts can help reduce the spread of the disease. In this study, we developed a model for predicting monthly dengue cases in Brazilian cities 1 month ahead, using data from 2007–2019. We compared different machine learning algorithms and feature selection methods using epidemiologic and meteorological variables. We found that different models worked best in different cities, and a random forests model trained on monthly dengue cases performed best overall. It produced lower errors than a seasonal naive baseline model, gradient boosting regression, a feed-forward neural network, or support vector regression. For each city, we computed the mean absolute error between predictions and true monthly numbers of dengue cases on the test data set. The median error across all cities was 12.2 cases. This error was reduced to 11.9 when selecting the optimal combination of algorithm and input features for each city individually. Machine learning and especially decision tree ensemble models may contribute to dengue surveillance in Brazil, as they produce low out-of-sample prediction errors for a geographically diverse set of cities. [ABSTRACT FROM AUTHOR]

There is growing acknowledgement of the legacy of White supremacy and racism in the discipline of epidemiology. Our department in the University of Washington School of Public Health undertook a systematic effort to begin addressing institutionalized racism and inclusive teaching in our courses. In July 2020, we introduced a new tool (the "Course Development Plan" (CDP)) to advance our curriculum. The CDP includes 2 components: 1) a guideline document that provides strategies on how to modify curricula and classroom teaching to incorporate antiracism and principles of equity, diversity, and inclusion (EDI); and 2) a structured worksheet for instructors to share EDI and antiracism practices they already incorporate and practices they plan to incorporate into their classes. Worksheets for each class are submitted prior to the beginning of the quarter and are reviewed by a peer faculty member and at least 1 epidemiology student; reviewers provide written feedback on the CDP worksheet. Further evaluation to assess the impact of the CDP process on classroom climate is ongoing. In this commentary, we discuss our department's efforts, the challenges we faced, and our hopes for next steps. [ABSTRACT FROM AUTHOR]

In this article, I provide a perspective on the tobacco epidemic and epidemiology, describing the impact of the tobacco-caused disease epidemic on the field of epidemiology. Although there is an enormous body of epidemiologic evidence on the associations of smoking with health, little systematic attention has been given to how decades of research have affected epidemiology and its practice. I address the many advances that resulted from epidemiologic research on smoking and health, such as demonstration of the utility of observational designs and important parameters (the odds ratio and the population attributable risk), guidelines for causal inference, and systematic review approaches. I also cover unintended and adverse consequences for the field, including the strategy of doubt creation and the recruitment of epidemiologists by the tobacco industry to serve its mission. The paradigm of evidence-based action for addressing noncommunicable diseases began with the need to address the epidemic of tobacco-caused disease, an imperative for action documented by epidemiologic research. [ABSTRACT FROM AUTHOR]

In the accompanying article (Am J Epidemiol. 2017;186(6):646-647), Dr. Timothy Lash makes a forceful case that the problems with reproducibility in science stem from our "culture" of null hypothesis significance testing. He notes that when attention is selectively given to statistically significant findings, the estimated effects will be systematically biased away from the null. Here I revisit the recent history of genetic epidemiology and argue for retaining statistical testing as an important part of the tool kit. Particularly when many factors are considered in an agnostic way, in what Lash calls "innovative" research, investigators need a selection strategy to identify which findings are most likely to be genuine, and hence worthy of further study. [ABSTRACT FROM AUTHOR]

Effect measure modification is often evaluated using parametric models. These models, although efficient when correctly specified, make strong parametric assumptions. While nonparametric models avoid important functional form assumptions, they often require larger samples to achieve a given accuracy. We conducted a simulation study to evaluate performance tradeoffs between correctly specified parametric and nonparametric models to detect effect modification of a binary exposure by both binary and continuous modifiers. We evaluated generalized linear models and doubly robust (DR) estimators, with and without sample splitting. Continuous modifiers were modeled with cubic splines, fractional polynomials, and nonparametric DR-learner. For binary modifiers, generalized linear models showed the greatest power to detect effect modification, ranging from 0.42 to 1.00 in the worst and best scenario, respectively. Augmented inverse probability weighting had the lowest power, with an increase of 23% when using sample splitting. For continuous modifiers, the DR-learner was comparable to flexible parametric models in capturing quadratic and nonlinear monotonic functions. However, for nonlinear, nonmonotonic functions, the DR-learner had lower integrated bias than splines and fractional polynomials, with values of 141.3, 251.7, and 209.0, respectively. Our findings suggest comparable performance between nonparametric and correctly specified parametric models in evaluating effect modification. [ABSTRACT FROM AUTHOR]

Observational epidemiologic studies typically face challenges of exposure measurement error and confounding. Consider an observational study of the association between a continuous exposure and an outcome, where the exposure variable of primary interest suffers from classical measurement error (i.e. the measured exposures are distributed around the true exposure with independent error). In the absence of exposure measurement error, it is widely recognized that one should control for confounders of the association of interest to obtain an unbiased estimate of the effect of that exposure on the outcome of interest. However, here we show that, in the presence of classical exposure measurement error, the net bias in an estimate of the association of interest may increase upon adjustment for confounders. We offer an analytical expression for calculating the change in net bias in an estimate of the association of interest upon adjustment for a confounder in the presence of classical exposure measurement error, and we illustrate this problem using simulations. [ABSTRACT FROM AUTHOR]

Quantitative bias analysis comprises the tools used to estimate the direction, magnitude, and uncertainty from systematic errors affecting epidemiologic research. Despite the availability of methods and tools, and guidance for good practices, few reports of epidemiologic research incorporate quantitative estimates of bias impacts. The lack of familiarity with bias analysis allows for the possibility of misuse, which is likely most often unintentional but could occasionally include intentional efforts to mislead. We identified 3 examples of suboptimal bias analysis, one for each common bias. For each, we describe the original research and its bias analysis, compare the bias analysis with good practices, and describe how the bias analysis and research findings might have been improved. We assert no motive to the suboptimal bias analysis by the original authors. Common shortcomings in the examples were lack of a clear bias model, computed example, and computing code; poor selection of the values assigned to the bias model's parameters; and little effort to understand the range of uncertainty associated with the bias. Until bias analysis becomes more common, community expectations for the presentation, explanation, and interpretation of bias analyses will remain unstable. Attention to good practices should improve quality, avoid errors, and discourage manipulation. [ABSTRACT FROM AUTHOR]

Machine learning is gaining prominence in the health sciences, where much of its use has focused on data-driven prediction. However, machine learning can also be embedded within causal analyses, potentially reducing biases arising from model misspecification. Using a question-and-answer format, we provide an introduction and orientation for epidemiologists interested in using machine learning but concerned about potential bias or loss of rigor due to use of "black box" models. We conclude with sample software code that may lower the barrier to entry to using these techniques. [ABSTRACT FROM AUTHOR]

In this study, we aimed to estimate the causal effect of normalized protein catabolic rate (nPCR) on mortality among end-stage renal disease (ESRD) patients in the presence of time-varying confounding affected by prior exposure using g-estimation. Information about 553 ESRD patients was retrospectively collected over an 8-year period (2011–2019) from hemodialysis facilities in Kerman, Iran. nPCR was dichotomized as <1.2 g/kg/day versus ≥1.2 g/kg/day. Then a standard time-varying accelerated failure time (AFT) Weibull model was built, and results were compared with those generated by g-estimation. After appropriate adjustment for time-varying confounders, weighted g-estimation yielded 78% shorter survival time (95% confidence interval (95% CI): −81, −73) among patients with a continuous nPCR <1.2 g/kg/day than among those who had nPCR ≥1.2 g/kg/day during follow-up, though it was 18% (95% CI: −57, 54) in the Weibull model. Moreover, hazard ratio estimates of 4.56 (95% CI: 3.69, 5.37) and 1.20 (95% CI: 0.66, 2.17) were obtained via weighted g-estimation and the Weibull model, respectively. G-estimation indicated that inadequate dietary protein intake characterized by nPCR increases all-cause mortality among ESRD patients, but the Weibull model provided an effect estimate that was substantially biased toward the null. [ABSTRACT FROM AUTHOR]

Suppose that an investigator wants to estimate an association between a continuous exposure variable and an outcome, adjusting for a set of confounders. If the exposure variable suffers classical measurement error, in which the measured exposures are distributed with independent error around the true exposure, then an estimate of the covariate-adjusted exposure-outcome association may be biased. We propose an approach to estimate a marginal exposure-outcome association in the setting of classical exposure measurement error using a disease score–based approach to standardization to the exposed sample. First, we show that the proposed marginal estimate of the exposure-outcome association will suffer less bias due to classical measurement error than the covariate-conditional estimate of association when the covariates are predictors of exposure. Second, we show that if an exposure validation study is available with which to assess exposure measurement error, then the proposed marginal estimate of the exposure-outcome association can be corrected for measurement error more efficiently than the covariate-conditional estimate of association. We illustrate both of these points using simulations and an empirical example using data from the Orinda Longitudinal Study of Myopia (California, 1989–2001). [ABSTRACT FROM AUTHOR]

The coronavirus disease 2019 (COVID-19) pandemic and the coming transition to a postpandemic world where COVID-19 will likely remain as an endemic disease present a host of challenges and opportunities in epidemiologic research. The scale and universality of this disruption to life and health provide unique opportunities to study phenomena and health challenges in all branches of epidemiology, from the obvious infectious disease and social consequences to less clear impacts on chronic disease and cancer. If we are to both take advantage of the largest natural experiment of our lifetimes and provide evidence to inform the numerous public health and clinical decisions being made every day, we must act quickly to ask critical questions and develop new methods for answering them. In doing so, we should build on each of our strengths and expertise and try to provide new insights rather than become yet another voice commenting on the same set of questions with limited evidence. [ABSTRACT FROM AUTHOR]

When estimating causal effects, careful handling of missing data is needed to avoid bias. Complete-case analysis is commonly used in epidemiologic analyses. Previous work has shown that covariate-stratified effect estimates from complete-case analysis are unbiased when missingness is independent of the outcome conditional on the exposure and covariates. Here, we assess the bias of complete-case analysis for adjusted marginal effects when confounding is present under various causal structures of missing data. We show that estimation of the marginal risk difference requires an unbiased estimate of the unconditional joint distribution of confounders and any other covariates required for conditional independence of missingness and outcome. The dependence of missing data on these covariates must be considered to obtain a valid estimate of the covariate distribution. If none of these covariates are effect-measure modifiers on the absolute scale, however, the marginal risk difference will equal the stratified risk differences and the complete-case analysis will be unbiased when the stratified effect estimates are unbiased. Estimation of unbiased marginal effects in complete-case analysis therefore requires close consideration of causal structure and effect-measure modification. [ABSTRACT FROM AUTHOR]

The literature on the relationship between adiposity and the onset of puberty is extensive, both in size and in the length of time this question has been alive in the biomedical literature. It is easy to wonder, then, whether there is anything new to be learned in this crowded field. In this issue of the Journal , Deardorff et al. (Am J Epidemiol. 2022;191(1):7–16) show that it is still possible to innovate in this area. The authors report on the relationship between body mass index at age 5 years and a variety of markers of the onset of puberty. Consistent with the current literature, they show no association between body mass index at age 5 and pubertal onset in boys. They also show an association between obesity at age 5 with earlier onset of secondary sexual characteristics in girls, making an important subject-matter contribution that addresses many of the concerns of the existing literature in assigning causality. At the same time, the authors make an important, although less evident contribution to the practice of epidemiology for studies where the assessment of pubertal status is of interest. [ABSTRACT FROM AUTHOR]

Over the past century, the field of epidemiology has evolved and adapted to changing public health needs. Challenges include newly emerging public health concerns across broad and diverse content areas, new methods, and vast data sources. We recognize the need to engage and educate the next generation of epidemiologists and prepare them to tackle these issues of the 21st century. In this commentary, we suggest a skeleton framework upon which departments of epidemiology should build their curriculum. We propose domains that include applied epidemiology, biological and social determinants of health, communication, creativity and ability to collaborate and lead, statistical methods, and study design. We believe all students should gain skills across these domains to tackle the challenges posed to us. The aim is to train smart thinkers, not technicians, to embrace challenges and move the expanding field of epidemiology forward. [ABSTRACT FROM AUTHOR]

There has been much debate about the relative emphasis of the field of epidemiology on causal inference. We believe this debate does short shrift to the breadth of the field. Epidemiologists answer myriad questions that are not causal and hypothesize about and investigate causal relationships without estimating causal effects. Descriptive studies face significant and often overlooked inferential and interpretational challenges; we briefly articulate some of them and argue that a more detailed treatment of biases that affect single-sample estimation problems would benefit all types of epidemiologic studies. Lumping all questions about causality creates ambiguity about the utility of different conceptual models and causal frameworks; 2 distinct types of causal questions include 1) hypothesis generation and theorization about causal structures and 2) hypothesis-driven causal effect estimation. The potential outcomes framework and causal graph theory help efficiently and reliably guide epidemiologic studies designed to estimate a causal effect to best leverage prior data, avoid cognitive fallacies, minimize biases, and understand heterogeneity in treatment effects. Appropriate matching of theoretical frameworks to research questions can increase the rigor of epidemiologic research and increase the utility of such research to improve public health. [ABSTRACT FROM AUTHOR]

With advances in natural language processing and machine learning, researchers are leveraging social media as a low-cost, low-burden method for measuring various psychosocial factors. However, it is unclear whether information derived from social media is generalizable to broader populations, especially middle-aged and older adults. Using data on women aged 53–70 years from Nurses' Health Study II (2017–2018; n  = 49,045), we assessed differences in sociodemographic characteristics, health conditions, behaviors, and psychosocial factors between regular and nonregular users of Facebook (Facebook, Inc., Menlo Park, California). We evaluated effect sizes with phi (φ) coefficients (categorical data) or Cohen's d (continuous data) and calculated odds ratios with 95% confidence intervals. While most comparisons between regular and nonregular users achieved statistical significance in this large sample, effect sizes were mostly "very small" (conventionally defined as φ or d <0.01) (e.g., optimism score: meanregular users = 19 vs. meannonregular users = 19 (d  = −0.03); physical activity: meanregular users = 24 metabolic equivalent of task (MET)-hours/week vs. meannonregular users = 24 MET-hours/week (d =  0.01)). Some factors had slightly larger differences for regular users versus nonregular users (e.g., depression: 28% vs. 23% (φ = 0.05); odds ratio = 1.27 (95% confidence interval: 1.22, 1.33); obesity: 34% vs. 26% (φ = 0.07); odds ratio = 1.42 (95% confidence interval: 1.36, 1.48)). Results suggest that regular Facebook users were similar to nonregular users across sociodemographic and psychosocial factors, with modestly worse health regarding obesity and depressive symptoms. In future research, investigators should evaluate other demographic groups. [ABSTRACT FROM AUTHOR]

Heterogeneous exposure associations (HEAs) can be defined as differences in the association of an exposure with an outcome among subgroups that differ by a set of characteristics. In this article, we intend to foster discussion of HEAs in the epidemiologic literature and present a variant of the random forest algorithm that can be used to identify HEAs. We demonstrate the use of this algorithm in the setting of the association between systolic blood pressure and death in older adults. The training set included pooled data from the baseline examination of the Cardiovascular Health Study (1989–1993), the Health, Aging, and Body Composition Study (1997–1998), and the Sacramento Area Latino Study on Aging (1998–1999). The test set included data from the National Health and Nutrition Examination Survey (1999–2002). The hazard ratios ranged from 1.25 (95% confidence interval: 1.13, 1.37) per 10-mm Hg increase in systolic blood pressure among men aged ≤67 years with diastolic blood pressure greater than 80 mm Hg to 1.00 (95% confidence interval: 0.96, 1.03) among women with creatinine concentration ≤0.7 mg/dL and a history of hypertension. HEAs have the potential to improve our understanding of disease mechanisms in diverse populations and guide the design of randomized controlled trials to control exposures in heterogeneous populations. [ABSTRACT FROM AUTHOR]

Matching methods are assumed to reduce the likelihood of a biased inference compared with ordinary least squares (OLS) regression. Using simulations, we compared inferences from propensity score matching, coarsened exact matching, and unmatched covariate-adjusted OLS regression to identify which methods, in which scenarios, produced unbiased inferences at the expected type I error rate of 5%. We simulated multiple data sets and systematically varied common support, discontinuities in the exposure and/or outcome, exposure prevalence, and analytical model misspecification. Matching inferences were often biased in comparison with OLS, particularly when common support was poor; when analysis models were correctly specified and common support was poor, the type I error rate was 1.6% for propensity score matching (statistically inefficient), 18.2% for coarsened exact matching (high), and 4.8% for OLS (expected). Our results suggest that when estimates from matching and OLS are similar (i.e. confidence intervals overlap), OLS inferences are unbiased more often than matching inferences; however, when estimates from matching and OLS are dissimilar (i.e. confidence intervals do not overlap), matching inferences are unbiased more often than OLS inferences. This empirical "rule of thumb" may help applied researchers identify situations in which OLS inferences may be unbiased as compared with matching inferences. [ABSTRACT FROM AUTHOR]

The potential for immortal time bias is pervasive in epidemiologic studies with left truncation or time-varying exposures. Unlike other biases in epidemiologic research (e.g., measurement bias, confounding due to unmeasured factors, and selection based on unmeasured predictors of the outcome), immortal time bias can and should be avoided by the correct assignment of person-time during follow up. However, even when handing person-time correctly, allowing late entry into a study or into an exposure group can open the door to more insidious sources of bias, some of which we explore here. Clear articulation of the study question, including the treatment plans of interest, can provide navigation around these sources of bias and elucidate the assumptions needed for inference given the available data. Here, we use simulated data to illustrate the assumptions required under various approaches to estimate the effect of a time-varying treatment and describe how these assumptions relate to the assumptions necessary to estimate single sample rates and risks in settings with censoring and truncation. [ABSTRACT FROM AUTHOR]

Mendelian randomization (MR) is used to answer a variety of epidemiologic questions. One stated advantage of MR is that it estimates a "lifetime effect" of exposure, though this term remains vaguely defined. Instrumental variable analysis, on which MR is based, has focused on estimating the effects of point or time-fixed exposures rather than "lifetime effects." Here we use an empirical example with data from the Rotterdam Study (Rotterdam, the Netherlands, 2009–2013) to demonstrate how confusion can arise when estimating "lifetime effects." We provide one possible definition of a lifetime effect: the average change in outcome measured at time t when the entire exposure trajectory from conception to time t is shifted by 1 unit. We show that MR only estimates this type of lifetime effect under specific conditions—for example, when the effect of the genetic variants used on exposure does not change over time. Lastly, we simulate the magnitude of bias that would result in realistic scenarios that use genetic variants with effects that change over time. We recommend that investigators in future MR studies carefully consider the effect of interest and how genetic variants whose effects change with time may impact the interpretability and validity of their results. [ABSTRACT FROM AUTHOR]

The popular inverse probability weighting method in causal inference is often hampered by extreme propensity scores, resulting in biased estimates and excessive variance. A common remedy is to trim patients with extreme scores (i.e. remove them from the weighted analysis). However, such methods are often sensitive to the choice of cutoff points and discard a large proportion of the sample. The implications for bias and the precision of the treatment effect estimate are unclear. These problems are mitigated by a newly developed method, the overlap weighting method. Overlap weights emphasize the target population with the most overlap in observed characteristics between treatments, by continuously down-weighting the units in the tails of the propensity score distribution. Here we use simulations to compare overlap weights to standard inverse probability weighting with trimming, in terms of bias, variance, and 95% confidence interval coverage. A range of propensity score distributions are considered, including settings with substantial nonoverlap and extreme values. To facilitate practical implementation, we further provide a consistent estimator for the standard error of the treatment effect estimated using overlap weighting. [ABSTRACT FROM AUTHOR]

Loss to follow-up is an endemic feature of time-to-event analyses that precludes observation of the event of interest. To our knowledge, in typical cohort studies with encounters occurring at regular or irregular intervals, there is no consensus on how to handle person-time between participants' last study encounter and the point at which they meet a definition of loss to follow-up. We demonstrate, using simulation and an example, that when the event of interest is captured outside of a study encounter (e.g., in a registry), person-time should be censored when the study-defined criterion for loss to follow-up is met (e.g., 1 year after last encounter), rather than at the last study encounter. Conversely, when the event of interest must be measured within the context of a study encounter (e.g., a biomarker value), person-time should be censored at the last study encounter. An inappropriate censoring scheme has the potential to result in substantial bias that may not be easily corrected. [ABSTRACT FROM AUTHOR]

Influenza viruses are associated with a substantial global burden of morbidity and mortality every year. Estimates of influenza-associated mortality often vary between studies due to differences in study settings, methods, and measurement of outcomes. We reviewed 103 published articles assessing population-based influenza-associated mortality through searches of PubMed and Embase, and we identified considerable variation in the statisticalmethods used across studies. Studies using regression models with an influenza activity proxy applied 4 approaches to estimate influenza-associated mortality. The estimates increased with age and ranged widely, from -0.3-1.3 and 0.6-8.3 respiratory deaths per 100, 000 population for children and adults, respectively, to 4-119 respiratory deaths per 100, 000 population for older adults. Meta-regression analysis identified that study design features were associated with the observed variation in estimates. The estimates increased with broader cause-of-death classification and were higher for older adults than for children. The multiplier methods tended to produce lower estimates, while Serfling-type models were associated with higher estimates than other methods. No "average" estimate of excess mortality could reliably be made due to the substantial variability of the estimates, partially attributable to methodological differences in the studies. Standardization of methodology in estimation of influenza-associated mortality would permit improved comparisons in the future. [ABSTRACT FROM AUTHOR]

One important assumption in case-control studies is that control selection should be independent of exposure. Nevertheless, it has been hypothesized that virus interference might lead to a correlation between receipt of influenza vaccination and increased risk of infection with other respiratory viruses. We investigated whether such a phenomenon might affect a study design commonly used to estimate influenza vaccine effectiveness (VE). We searched publications in MEDLINE, PubMed, and Web of Science. We identified 12 studies using the test-negative design (2011-2017) that reported VE estimates separately derived by 3 alternative control groups: 1) all patients testing negative for influenza (FLU), VEFLU-; 2) patients who tested positive for other/another respiratory virus (ORV), VEORV+; and 3) patients who tested negative for all viruses in the panel (PAN), VEPAN-. These included VE estimates from 7 countries for all age groups from 2003/2004 to 2013/2014. We observed no difference in vaccination coverage between the ORV-positive and PAN-negative control groups. A total of 63 VEFLU- estimates, 62 VEORV+ estimates, and 33 VEPAN- estimates were extracted. Pooled estimates of the difference in VE (ΔVE) were very similar between groups. Inmeta-regression, no association was found between the selection of control group and VE estimates. In conclusion, we did not find any differences in VE estimates based on the choice of control group. [ABSTRACT FROM AUTHOR]

In the last few years, stakeholders in the scientific community have raised alarms about a perceived lack of reproducibility of scientific results. In reaction, guidelines for journals have been promulgated and grant applicants have been asked to address the rigor and reproducibility of their proposed projects. Neither solution addresses a primary culprit, which is the culture of null hypothesis significance testing that dominates statistical analysis and inference. In an innovative research enterprise, selection of results for further evaluation based on null hypothesis significance testing is doomed to yield a low proportion of reproducible results and a high proportion of effects that are initially overestimated. In addition, the culture of null hypothesis significance testing discourages quantitative adjustments to account for systematic errors and quantitative incorporation of prior information. These strategies would otherwise improve reproducibility and have not been previously proposed in the widely cited literature on this topic. Without discarding the culture of null hypothesis significance testing and implementing these alternative methods for statistical analysis and inference, all other strategies for improving reproducibility will yield marginal gains at best. [ABSTRACT FROM AUTHOR]

Neighborhood conditions may influence a broad range of health indicators, including obesity, injury, and psychopathology. In particular, neighborhood physical disorder--a measure of urban deterioration--is thought to encourage crime and high-risk behaviors, leading to poor mental and physical health. In studies to assess neighborhood physical disorder, investigators typically rely on time-consuming and expensive in-person systematic neighborhood audits. We compared 2 audit-based measures of neighborhood physical disorder in the city of Detroit, Michigan: One used Google Street View imagery from 2009 and the other used an in-person survey conducted in 2008. Each measure used spatial interpolation to estimate disorder at unobserved locations. In total, the virtual audit required approximately 3% of the time required by the in-person audit. However, the final physical disorder measures were significantly positively correlated at census block centroids (r = 0.52), identified the same regions as highly disordered, and displayed comparable leave-one-out cross-validation accuracy. The measures resulted in very similar convergent validity characteristics (correlation coefficients within 0.03 of each other). The virtual audit-based physical disorder measure could substitute for the in-person one with little to no loss of precision. Virtual audits appear to be a viable and much less expensive alternative to in-person audits for assessing neighborhood conditions. [ABSTRACT FROM AUTHOR]

The advent of counterfactual-based mediation analysis has triggered enormous progress on how, and under what assumptions, one may disentangle path-specific effects upon combining arbitrary (possibly nonlinear) models for mediator and outcome. However, current developments have largely focused on single mediators because required identification assumptions prohibit simple extensions to settings with multiple mediators that may depend on one another. In this article, we propose a procedure for obtaining fine-grained decompositions that may still be recovered fromobserved data in such complex settings.We first show that existing analytical approaches target specific instances of a more general set of decompositions and may therefore fail to provide a comprehensive assessment of the processes that underpin cause-effect relationships between exposure and outcome. We then outline conditions for obtaining the remaining set of decompositions. Because the number of targeted decompositions increases rapidly with the number of mediators, we introduce natural effects models along with estimation methods that allow for flexible and parsimonious modeling. Our procedure can easily be implemented using off-the-shelf software and is illustrated using a reanalysis of theWorld Health Organization's Large Analysis and Review of European Housing and Health Status (WHO-LARES) study on the effect ofmold exposure onmental health (2002-2003). [ABSTRACT FROM AUTHOR]