Sufficient component cause simulations: an underutilized epidemiologic teaching tool

Simulation studies are a powerful and important tool in epidemiologic teaching, especially for understanding causal inference. Simulations using the sufficient component cause framework can provide students key insights about causal mechanisms and sources of bias, but are not commonly used. To make them more accessible, we aim to provide an introduction and tutorial on developing and using these simulations, including an overview of translation from directed acyclic graphs and potential outcomes to sufficient component causal models, and a summary of the simulation approach. Using the applied question of the impact of educational attainment on dementia, we offer simple simulation examples and accompanying code to illustrate sufficient component cause-based simulations for four common causal structures (causation, confounding, selection bias, and effect modification) often introduced early in epidemiologic training. We show how sufficient component cause-based simulations illuminate both the causal processes and the mechanisms through which bias occurs, which can help enhance student understanding of these causal structures and the distinctions between them. We conclude with a discussion of considerations for using sufficient component cause-based simulations as a teaching tool.

Effect Measure Modification by Covariates in Mediation: Extending Regression-based Causal Mediation Analysis

Existing methods for regression-based mediation analysis assume that the exposure-mediator effect, exposure-outcome effect, and mediator-outcome effect are constant across levels of the baseline characteristics of patients. However, investigators often have insight into how these underlying effects may be modified by baseline characteristics and are interested in how the resulting mediation effects, such as the natural direct effect (NDE), the natural indirect effect. (NIE), and the proportion mediated, are modified by these baseline characteristics. Motivated by an empirical example of anti-interleukin-1 therapy's benefit on incident anemia reduction and its mediation by an early change in an inflammatory biomarker, we extended the closed-form regression-based causal mediation analysis with effect measure modification (EMM). Using a simulated numerical example, we demonstrated that naive analysis without considering EMM can give biased estimates of NDE and NIE and visually illustrated how baseline characteristics affect the presence and magnitude of EMM of NDE and NIE. We then applied the extended method to the empirical example informed by pathophysiologic insights into potential EMM by age, diabetes, and baseline inflammation. We found that the proportion modified through the early post-treatment inflammatory biomarker was greater for younger, nondiabetic patients with lower baseline level of inflammation, suggesting differential usefulness of the early post-treatment inflammatory biomarker in monitoring patients depending on baseline characteristics. To facilitate the adoption of EMM considerations in causal mediation analysis by the wider clinical and epidemiologic research communities, we developed a free- and open-source R package, regmedint.

Safety of prenatal opioid analgesics: Do results differ between public health insurance beneficiary and population-based cohorts?

BACKGROUND

Pregnant patients with particular types of health insurance may have distinct demographic and medical characteristics that have a biologic effect on associations between opioid analgesics and congenital anomalies (CA).

METHODS

We followed 199,884 pregnant prescription beneficiaries in Ontario, Canada (1996-2018). Opioid analgesics dispensed in the first trimester and CA were identified from universal-access administrative health records. We estimated propensity score adjusted risk ratios (RR) between first trimester exposure and CA (any, major, minor, specific). RRs were compared to those published from an Ontario population-based cohort (N = 599,579, 2013-2018).

RESULTS

15,724 (7.9%) were exposed to first trimester opioid analgesics, mainly codeine (58.1%) or oxycodone (21.3%); CA prevalence in exposed was 3.1%. RRs in the beneficiary cohort appeared higher than the population-based cohort for any CA with hydromorphone (RR = 2.34, 95% CI: 1.65, 3.30) and oxycodone (RR = 1.73, 95% CI: 1.46, 2.05) and major CA with hydromorphone (RR = 2.74, 95% CI: 1.91, 3.94) and oxycodone (RR = 1.72, 95% CI: 1.42, 2.08). Other RRs that appeared higher in the beneficiary cohort included cardiovascular (codeine, oxycodone), gastrointestinal (oxycodone), musculoskeletal (any, hydromorphone, oxycodone), CNS (oxycodone), chromosomal (codeine), and neoplasm and tumor (oxycodone) anomalies. The beneficiary cohort had higher opioid doses, was younger, had lower socioeconomic status, and greater comorbidities.

CONCLUSIONS

Increased risks of CA after first trimester opioid analgesics were observed in low-income prescription beneficiaries, and some estimates were higher than a population-based cohort from the same setting. Biological differences associated with younger age, lower socioeconomic status and greater comorbidity may affect generalizability of results from pregnant low-income beneficiaries.

How Effect Measure Choice Influences Minimally Sufficient Adjustment Sets for External Validity

Epidemiologic researchers generalizing or transporting effect estimates from a study to a target must account for effect measure modifiers (EMMs) on the scale of interest. Little attention is paid to how the EMMs required may vary depending on the mathematical nuances of each effect measure, however. We defined two types of EMM: marginal EMM, where the effect on the scale of interest differs across levels of a variable; and conditional EMM, where the effect differs conditional on other variables associated with the outcome. These types define three classes of variables: Class 1 (conditional EMM), Class 2 (marginal, but not conditional, EMM), or Class 3 (neither marginal nor conditional EMM). Class 1 variables are necessary to achieve a valid estimate of the RD in a target, while a RR requires Class 1 and Class 2 and an OR requires Class 1, Class 2, and Class 3 (i.e., all variables associated with the outcome). This does not mean that fewer variables are required for an externally valid RD (because variables may not modify effects on all scale) but does suggest researchers should consider the scale of the effect measure when identifying EMM necessary for an externally valid treatment effect estimate.

Toward a Clearer Definition of Selection Bias When Estimating Causal Effects

Selection bias remains a subject of controversy. Existing definitions of selection bias are ambiguous. To improve communication and the conduct of epidemiologic research focused on estimating causal effects, we propose to unify the various existing definitions of selection bias in the literature by considering any bias away from the true causal effect in the referent population (the population before the selection process), due to selecting the sample from the referent population, as selection bias. Given this unified definition, selection bias can be further categorized into two broad types: type 1 selection bias owing to restricting to one or more level(s) of a collider (or a descendant of a collider) and type 2 selection bias owing to restricting to one or more level(s) of an effect measure modifier. To aid in explaining these two types-which can co-occur-we start by reviewing the concepts of the target population, the study sample, and the analytic sample. Then, we illustrate both types of selection bias using causal diagrams. In addition, we explore the differences between these two types of selection bias, and describe methods to minimize selection bias. Finally, we use an example of "M-bias" to demonstrate the advantage of classifying selection bias into these two types.

Performance Evaluation of Parametric and Nonparametric Methods When Assessing Effect Measure Modification

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.

Examining Optimism, Psychosocial Risks, and Cardiovascular Health Using Life's Simple 7 Metrics in the Multi-Ethnic Study of Atherosclerosis and the Jackson Heart Study

Background: Optimism has been shown to be positively associated with better cardiovascular health (CVH). However, there is a dearth of prospective studies showing the benefits of optimism on CVH, especially in the presence of adversities, i.e., psychosocial risks. This study examines the prospective relationship between optimism and CVH outcomes based on the Life's Simple 7 (LS7) metrics and whether multilevel psychosocial risks modify the aforementioned relationship. Methods: We examined self-reported optimism and CVH using harmonized data from two U.S. cohorts: Multi-Ethnic Study of Atherosclerosis (MESA) and Jackson Heart Study (JHS). Modified Poisson regression models were used to estimate the relationship between optimism and CVH using LS7 among MESA participants (N = 3,520) and to examine the relationship of interest based on four biological LS7 metrics (body mass index, blood pressure, cholesterol, and blood glucose) among JHS and MESA participants (N = 5,541). For all CVH outcomes, we assessed for effect measure modification by psychosocial risk. Results: Among MESA participants, the adjusted risk ratio (aRR) for ideal or intermediate CVH using LS7 comparing participants who reported high or medium optimism to those with the lowest level of optimism was 1.10 [95% Confidence Interval (CI): 1.04-1.16] and 1.05 (95% CI: 0.99-1.11), respectively. Among MESA and JHS participants, the corresponding aRRs for having all ideal or intermediate (vs. no poor) metrics based on the four biological LS7 metrics were 1.05 (0.98-1.12) and 1.04 (0.97-1.11), respectively. The corresponding aRRs for having lower cardiovascular risk (0-1 poor metrics) based on the four biological LS7 metrics were 1.01 (0.98-1.03) and 1.01 (0.98-1.03), respectively. There was some evidence of effect modification by neighborhood deprivation for the LS7 outcome and by chronic stress for the ideal or intermediate (no poor) metrics outcome based on the four biological LS7 metrics. Conclusion: Our findings suggest that greater optimism is positively associated with better CVH based on certain LS7 outcomes among a racially/ethnically diverse study population. This relationship may be effect measure modified by specific psychosocial risks. Optimism shows further promise as a potential area for intervention on CVH. However, additional prospective and intervention studies are needed.

Directed Acyclic Graphs, Effect Measure Modification, and Generalizability

Directed acyclic graphs (DAGs) have had a major impact on the field of epidemiology by providing straightforward graphical rules for determining when estimates are expected to lack causally interpretable internal validity. Much less attention has been paid, however, to what DAGs can tell researchers about effect measure modification and external validity. In this work, we describe 2 rules based on DAGs related to effect measure modification. Rule 1 states that if a variable, $P$, is conditionally independent of an outcome, $Y$, within levels of a treatment, $X$, then $P$ is not an effect measure modifier for the effect of $X$ on $Y$ on any scale. Rule 2 states that if $P$ is not conditionally independent of $Y$ within levels of $X$, and there are open causal paths from $X$ to $Y$ within levels of $P$, then $P$ is an effect measure modifier for the effect of $X$ on $Y$ on at least 1 scale (given no exact cancelation of associations). We then show how Rule 1 can be used to identify sufficient adjustment sets to generalize nested trials studying the effect of $X$ on $Y$ to the total source population or to those who did not participate in the trial.

Psychological Status as an Effect Modifier of the Association Between Sinonasal Instrument and Imaging Results

OBJECTIVE

To determine whether psychological status is an effect modifier of the previously observed low discriminatory capacity of Sinonasal Outcome Test-22 (SNOT-22) scores for Lund-Mackay computed tomography (CT) results.

STUDY DESIGN

Observational outcomes study.

SETTING

Tertiary care center.

SUBJECTS AND METHODS

We assessed patients presenting with chronic sinonasal complaints who underwent CT of the sinuses within 1 month of completing the SNOT-22 instrument. SNOT-22 overall and domain scores were calculated, as were Lund-Mackay CT scores. The discriminatory capacity of SNOT-22 scores for CT results was determined using the receiver-operator characteristic area under the curve (ROC-AUC). Patient-Reported Outcome Measurement Information System (PROMIS) mental health T-scores were assessed, and stratified analyses were used to test for effect modification by psychological status.

RESULTS

In stratified analyses, patients with better PROMIS mental health scores had SNOT-22 overall (ROC-AUC 0.96) and nasal domain scores (ROC-AUC 0.97-0.98) that were highly discriminatory for Lund-Mackay scores, while those with worse mental health scores did not (ROC-AUC 0.42-0.55, P < .007). Patients with better SNOT-22 psychological domain scores also had nasal scores that discriminated among CT results significantly better than those with worse psychological domain scores (ROC-AUC 0.65-0.69 and 0.34-0.35, respectively, P < .013).

CONCLUSIONS

Psychological status is an effect modifier of the relationship between SNOT-22 and Lund-Mackay scores. SNOT-22 scores were discriminatory for Lund-Mackay CT results in patients with better psychological status, while they were nondiscriminatory in those with worse psychological status. When assessing the relationship between subjective and objective measures of chronic rhinosinusitis, accounting for effect modification may have practical utility.

Heterogeneity in the Relationship between Disinfection By-Products in Drinking Water and Cancer: A Systematic Review

The epidemiological evidence demonstrating the effect of disinfection by-products (DBPs) from drinking water on colon and rectal cancers is well documented. However, no systematic assessment has been conducted to assess the potential effect measure modification (EMM) in the relationship between DBPs and cancer. The objective of this paper is to conduct a systematic literature review to determine the extent to which EMM has been assessed in the relationship between DBPs in drinking water in past epidemiological studies. Selected articles (n = 19) were reviewed, and effect estimates and covariates that could have been used in an EMM assessment were gathered. Approximately half of the studies assess EMM (n = 10), but the majority of studies only estimate it relative to sex subgroups (n = 6 for bladder cancer and n = 2 both for rectal and colon cancers). Although EMM is rarely assessed, several variables that could have a potential modification effect are routinely collected in these studies, such as socioeconomic status or age. The role of environmental exposures through drinking water can play an important role and contribute to cancer disparities. We encourage a systematic use of subgroup analysis to understand which populations or territories are more vulnerable to the health impacts of DBPs.

Using group data to treat individuals: understanding heterogeneous treatment effects in the age of precision medicine and patient-centred evidence

Although often conflated, determining the best treatment for an individual (the task of a doctor) is fundamentally different from determining the average effect of treatment in a population (the purpose of a trial). In this paper, we review concepts of heterogeneity of treatment effects (HTE) essential in providing the evidence base for precision medicine and patient-centred care, and explore some inherent limitations of using group data (e.g. from a randomized trial) to guide treatment decisions for individuals. We distinguish between person-level HTE (i.e. that individuals experience different effects from a treatment) and group-level HTE (i.e. that subgroups have different average treatment effects), and discuss the reference class problem, engendered by the large number of potentially informative subgroupings of a study population (each of which may lead to applying a different estimated effect to the same patient), and the scale dependence of group-level HTE. We also review the limitations of conventional 'one-variable-at-a-time' subgroup analyses and discuss the potential benefits of using more comprehensive subgrouping schemes that incorporate information on multiple variables, such as those based on predicted outcome risk. Understanding the conceptual underpinnings of HTE is critical for understanding how studies can be designed, analysed, and interpreted to better inform individualized clinical decisions.

Diabetes mitigates the role of memory complaint in predicting dementia risk: Results from the Prevention of Alzheimer's Disease with Vitamin E and Selenium Study

BACKGROUND

Subjective memory complaints (SMCs) are associated with increased risk of dementia in older adults, but the role of comorbidities in modifying this risk is unknown.

OBJECTIVES

To assess whether comorbidities modify estimated dementia risk based on SMCs.

DESIGN

The Prevention of Alzheimer's Disease with Vitamin E and Selenium Study (PREADVISE) was designed as an ancillary study to the Selenium and Vitamin E Cancer Prevention Trial (SELECT), a randomized, multi-center prostate cancer prevention trial with sites in the Unites States, Puerto Rico, and Canada. In 2009, PREADVISE and SELECT were changed into cohort studies.

SETTING

Secondary analysis of PREADVISE data.

PARTICIPANTS

PREADVISE recruited 7,540 non-demented male volunteers from participating SELECT sites from 2002 to 2009. SMCs, demographics, and comorbidities including hypertension, diabetes, coronary artery bypass graft (CABG), stroke, sleep apnea, and head injury were ascertained by participant interview.

MEASUREMENTS

Cox models were used to investigate whether baseline comorbidities modified hazard ratios (HR) for SMC-associated dementia risk using two methods: (1) we included one interaction term between SMC and a comorbidity in the model at a time, and (2) we included all two-way interactions between SMC and covariates of interest and reduced the model by "backward" selection. SMC was operationalized as any complaint vs. no complaint.

RESULTS

Baseline SMCs were common (23.6%). In the first analyses, with the exception of stroke, presence of self-reported comorbidities was associated with lower estimated HR for dementia based on SMC status (complaint vs. no complaint), but this difference was only significant for diabetes. In the second analysis, the two-way interactions between SMC and race as well as SMC and diabetes were significant. Here, black men without diabetes who reported SMC had the highest estimated dementia risk (HR=5.05, 95% CI 2.55-10.00), while non-black men with diabetes who reported SMC had the lowest estimated risk (HR=0.71, 95% CI 0.35-1.41).

CONCLUSIONS

SMCs were more common among men with comorbidities, but these complaints appeared to be less predictive of dementia risk than those originating from men without comorbidities, suggesting that medical conditions such as diabetes may explain SMCs that are unrelated to an underlying neurodegenerative process.

Independent and dependent contributions of advanced maternal and paternal ages to autism risk

Reports on autism and parental age have yielded conflicting results on whether mothers, fathers, or both, contribute to increased risk. We analyzed restricted strata of parental age in a 10-year California birth cohort to determine the independent or dependent effect from each parent. Autism cases from California Department of Developmental Services records were linked to State birth files (1990-1999). Only singleton births with complete data on parental age and education were included (n=4,947,935, cases=12,159). In multivariate logistic regression models, advancing maternal age increased risk for autism monotonically regardless of the paternal age. Compared with mothers 25-29 years of age, the adjusted odds ratio (aOR) for mothers 40+ years was 1.51 (95% CI: 1.35-1.70), or compared with mothers <25 years of age, aOR=1.77 (95% CI, 1.56-2.00). In contrast, autism risk was associated with advancing paternal age primarily among mothers <30: aOR=1.59 (95% CI, 1.37-1.85) comparing fathers 40+ vs. 25-29 years of age. However, among mothers >30, the aOR was 1.13 (95% CI, 1.01-1.27) for fathers 40+ vs. 25-29 years of age, almost identical to the aOR for fathers <25 years. Based on the first examination of heterogeneity in parental age effects, it appears that women's risk for delivering a child who develops autism increases throughout their reproductive years whereas father's age confers increased risk for autism when mothers are <30, but has little effect when mothers are past age 30. We also calculated that the recent trend towards delayed childbearing contributed approximately a 4.6% increase in autism diagnoses in California over the decade.