Mireille E. Schnitzer

Causal variable selection in time‐varying treatment settings is challenging due to evolving confounding effects. Existing methods mainly focus on time‐fixed exposures and are not directly applicable to time‐varying scenarios. We propose a novel two‐step procedure for variable selection when modeling the treatment probability at each time point. We first introduce a novel approach to longitudinal confounder selection using a Longitudinal Outcome Adaptive LASSO (LOAL) that will data‐adaptively select covariates with theoretical justification of variance reduction of the estimator of the causal effect. We then propose an adaptive fused LASSO that can collapse treatment model parameters over time points with the goal of simplifying the models in order to improve the efficiency of the estimator while minimizing model misspecification bias compared with naive pooled logistic regression models. Our simulation studies highlight the need for and usefulness of the proposed approach in practice. We implemented our method on data from the Nicotine Dependence in Teens study to estimate the effect of the timing of alcohol initiation during adolescence on depressive symptoms in early adulthood.

Understanding treatment effect heterogeneity is important for decision-making in medical and clinical practices, or handling various engineering and marketing challenges. When dealing with high-dimensional covariates or when the effect modifiers are not predefined and need to be discovered, data-adaptive selection approaches become essential. However, with data-driven model selection, the quantification of statistical uncertainty is complicated by post-selection inference due to difficulties in approximating the sampling distribution of the target estimator. Data-driven model selection tends to favor models with strong effect modifiers with an associated cost of inflated type I errors. Although several frameworks and methods for valid statistical inference have been proposed for ordinary least squares regression following data-driven model selection, fewer options exist for valid inference for effect modifier discovery in causal modeling contexts. In this article, we extend two different methods to develop valid inference for penalized G-estimation that investigates effect modification of proximal treatment effects within the structural nested mean model framework. We show the asymptotic validity of the proposed inferential methods. In our simulation study, the proposed methods effectively controlled the false coverage rates for the target parameters, while the naive inference based on the sandwich variance estimator resulted in false coverage rates higher than the nominal level. Our work is motivated by the study of hemodiafiltration for treating patients with end-stage renal disease at the Centre Hospitalier de l’Université de Montréal. We apply these methods to draw inference about the effect heterogeneity of dialysis facility on the repeated session-specific hemodiafiltration outcomes.

Wang et al. (2025) use statistics to argue that sex at birth is not a biological coin toss, by noticing that repeated patterns such as Male Male Male and Female Female Female occur in the Nurses Health Study more often than patterns like Male Female Male, Male Female Female, Female Male Female, or Female Male Male. This letter shows that this over-representation is likely due to a statistical artifact, arising from parent preferences for mixed-sex children. As noticed in Angrist and Evans (1998) and supported by the data in Wang et al. (2025), parents are more likely to have a third child if their first two children are of the same sex. We show mathematically and statistically that mixed-sex preferences lead to the over-representation of patterns like Male Male Male and Female Female Female. In fact, the patterns seen in the Nurses Health Study are perfectly consistent with sex at birth being a random coin toss.

Background Limited data exist on the comparative effectiveness and safety of first-line medications to treat chronic hypertension in pregnancy. We compared nifedipine versus labetalol initiation before 23 weeks’ gestation for risk of pregnancy loss (miscarriage or stillbirth) and preterm birth (Methods We identified pregnancies in Merative’s MarketScan commercial claims data with estimated last menstrual periods between July 1, 2016 and February 1, 2022. We included those with a new nifedipine or labetalol fill before 23 weeks’ gestation. We estimated risks, risk differences (RDs) per 100 pregnancies, and risk ratios (RRs) among nifedipine versus labetalol initiators using a weighted Aalen-Johansen estimator to account for confounding and informative censoring. We also conducted a post hoc analysis to estimate the comparative effect of nifedipine versus labetalol on risk of preeclampsia. Results We identified 987 nifedipine and 4,570 labetalol initiators during the study period. The weighted risk of pregnancy loss was 7.0% among nifedipine and 6.0% among labetalol initiators (RD=1.1 [-0.7, 2.8]; RR=1.18 [0.87, 1.59]). Weighted risks of preterm birth were 29.4% versus 22.2%, respectively (RD=7.2 [3.9, 10.5], RR=1.32 [1.13, 1.55]). In post hoc analyses, the weighted risks of preeclampsia were 38.6% versus 33.9%, respectively (RD=4.7 [0.7, 8.7], RR=1.14 [1.01, 1.28]). Findings were consistent across various sensitivity analyses. Conclusions Compared to labetalol initiators, nifedipine initiators experienced an increased risk of preterm birth but not pregnancy loss. The elevated risk of preterm birth may be explained by an increased risk of preeclampsia among nifedipine initiators, resulting in medically indicated preterm birth. What is new? Initiating nifedipine, compared to labetalol, before 23 weeks’ gestation as a first-line antihypertensive was associated with a 7.2 percentage point increased risk of preterm live birth (95% CI: 3.9, 10.5) and 4.7 percentage point increased risk of preeclampsia (95% CI: 0.7, 8.7). Nifedipine and labetalol were associated with a similar risk of pregnancy loss (risk difference: 1.1 per 100 pregnancies [95% CI: -0.7, 2.8]). These findings were robust across multiple sensitivity analyses. What are the clinical implications? Initiating labetalol, compared to nifedipine, prior to 23 weeks’ gestation for treatment of chronic hypertension in pregnancy may lead to lower risk of preeclampsia and preterm live birth in real-world settings. While high-quality data from clinical trials clearly support antihypertensive therapy for chronic hypertension in pregnancy, our findings illustrate that the first-line medications may have different comparative safety and effectiveness profiles. These data are important for informing shared decision-making between providers and patients managing chronic hypertension in pregnancy.

Prediction models are important in medical research, as such models enable health researchers to gain deeper insights into disease epidemiology and clinicians to identify patients at higher risk of adverse outcomes. One commonly employed approach to developing prediction models is variable selection through penalized regression. Integrating natural variable structures and predefined inclusion requirements into variable selection not only enhances model interpretability but can also potentially boost prediction accuracy. For example, the latent overlapping group Lasso can force the inclusion of the main terms in the resulting model if their interaction term is selected. However, when variable structures are complex, it is challenging to integrate such structures into the penalized regression. In this work, we first demonstrate how to convert variable structures and predefined variable inclusion requirements into “selection rules” (which represent rules for which or how variables can be included in the final prediction model) and present these rules mathematically. Then, we provide a structured approach for integrating complex rules into variable selection through the latent overlapping group Lasso so that the resulting prediction model follows the given selection rules. To illustrate our methodology, we applied these techniques to construct a coherent prediction model for major bleeding in hypertensive patients recently hospitalized for atrial fibrillation and subsequently prescribed oral anticoagulants. In this application, we account for a proxy of anticoagulant adherence and its interaction with dosage and the type of oral anticoagulants, in addition to drug‐drug interactions.

Targeted maximum likelihood estimation (TMLE) is an increasingly popular framework for the estimation of causal effects. It requires modeling both the exposure and outcome but is doubly robust in the sense that it is valid if at least one of these models is correctly specified. In addition, TMLE allows for flexible modeling of both the exposure and outcome with machine learning methods. This provides better control for measured confounders since the model specification automatically adapts to the data, instead of needing to be specified by the analyst a priori. Despite these methodological advantages, TMLE remains less popular than alternatives in part because of its less accessible theory and implementation. While some tutorials have been proposed, none address the case of a time‐to‐event outcome. This tutorial provides a detailed step‐by‐step explanation of the implementation of TMLE for estimating the effect of a point binary or multilevel exposure on a time‐to‐event outcome, modeled as counterfactual survival curves and causal hazard ratios. The tutorial also provides guidelines on how best to use TMLE in practice, including aspects related to study design, choice of covariates, controlling biases and use of machine learning. R‐code is provided to illustrate each step using simulated data ( https://github.com/detal9/SurvTMLE). To facilitate implementation, a general R function implementing TMLE with options to use machine learning is also provided. The method is illustrated in a real‐data analysis concerning the effectiveness of statins for the prevention of a first cardiovascular disease among older adults in Québec, Canada, between 2013 and 2018.

The Test-Negative Design (TND), which involves recruiting care-seeking individuals who meet predefined clinical case criteria, offers valid statistical inference for Vaccine Effectiveness (VE) using data collected through passive surveillance, making it cost-efficient and timely. Infectious disease epidemiology often involves interference, where the treatment and/or outcome of one individual can affect the outcomes of others, rendering standard causal estimands ill-defined; ignoring such interference can bias VE evaluation and lead to ineffective vaccination policies. This article addresses the estimation of causal estimands for VE in the presence of partial interference using TND samples. Partial interference means that the vaccination of units within the same group/cluster may influence the outcomes of other members of the cluster. We define the population direct, spillover, total, and overall effects using the geometric risk ratio, which are identifiable under TND sampling. We investigate various stochastic policies for vaccine allocation in a counterfactual scenario, and identify policy-relevant VE causal estimands. We propose inverse-probability weighted (IPW) estimators for estimating the policy-relevant VE causal estimands with partial interference under the TND, and explore the statistical properties of these estimators.

The test‐negative design (TND), which is routinely used for monitoring seasonal flu vaccine effectiveness (VE), has recently become integral to COVID‐19 vaccine surveillance, notably in Québec, Canada. Some studies have addressed the identifiability and estimation of causal parameters under the TND, but efficiency bounds for nonparametric estimators of the target parameter under the unconfoundedness assumption have not yet been investigated. Motivated by the goal of improving adjustment for measured confounders when estimating COVID‐19 VE among community‐dwelling people aged years in Québec, we propose a one‐step doubly robust and locally efficient estimator called TNDDR (TND doubly robust), which utilizes cross‐fitting (sample splitting) and can incorporate machine learning techniques to estimate the nuisance functions and thus improve control for measured confounders. We derive the efficient influence function (EIF) for the marginal expectation of the outcome under a vaccination intervention, explore the von Mises expansion, and establish the conditions for ‐consistency, asymptotic normality, and double robustness of TNDDR. The proposed estimator is supported by both theoretical and empirical justifications.

Effect modification occurs when the impact of the treatment on an outcome varies based on the levels of other covariates known as effect modifiers. Modeling these effect differences is important for etiological goals and for purposes of optimizing treatment. Structural nested mean models (SNMMs) are useful causal models for estimating the potentially heterogeneous effect of a time-varying exposure on the mean of an outcome in the presence of time-varying confounding. A data-adaptive selection approach is necessary if the effect modifiers are unknown a priori and need to be identified. Although variable selection techniques are available for estimating the conditional average treatment effects using marginal structural models or for developing optimal dynamic treatment regimens, all of these methods consider a single end-of-follow-up outcome. In the context of an SNMM for repeated outcomes, we propose a doubly robust penalized G-estimator for the causal effect of a time-varying exposure with a simultaneous selection of effect modifiers and prove the oracle property of our estimator. We conduct a simulation study for the evaluation of its performance in finite samples and verification of its double-robustness property. Our work is motivated by the study of hemodiafiltration for treating patients with end-stage renal disease at the Centre Hospitalier de l’Université de Montréal. We apply the proposed method to investigate the effect heterogeneity of dialysis facility on the repeated session-specific hemodiafiltration outcomes.