Adaptive Sparsening and Smoothing of the Treatment Model for Longitudinal Causal Inference Using Outcome‐Adaptive LASSO and Marginal Fused LASSO
Statistics in Medicine · 2026
DOI: 10.1002/sim.70316 ↗At a glance
Journal article in Statistics in Medicine, 2026. Written by Mireille E Schnitzer and 7 co-authors. Institution: Faculté de pharmacie, Université de Montréal, Quebec, Canada. Major MEDLINE topics: Models, Statistical, Causality. 51 references. The full text is freely available (open access, CC BY 4.0). Funded by: NIH HHS, NIEHS NIH HHS, Fonds de recherche du Québec.
Abstract (original language)
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.
Subjects
MeSH terms (MEDLINE)
Author keywords
Authors and institutions
- Faculté de pharmacie, Université de Montréal, Quebec, Canada.
- Département de médecine sociale et préventive, Université de Montréal, Quebec, Canada.
- Département de médecine sociale et préventive, Université Laval, Quebec, Canada.
- Centre de recherche du CHU de Québec, Université Laval, Quebec, Canada.
- Department of Computer Science, Université de Montréal, Quebec, Canada.
- The Dartmouth Institute for Health Policy and Clinical Practice, Geisel School of Medicine at Dartmouth, Hanover, New Hampshire, USA.
ORCID profiles: Mireille E Schnitzer, Denis Talbot, Jennifer O'Loughlin
Funding
- NIH HHS
- NIEHS NIH HHS
- Fonds de recherche du Québec
- NIDA NIH HHS
- Natural Sciences and Engineering Research Council of Canada
- Canada Research Chairs
- NINDS NIH HHS
- National Institutes of Health
Cite
APA
Schnitzer, M. E., Talbot, D., Liu, Y., Berger, D., Wang, G., O'Loughlin, J., Sylvestre, M. P., & Ertefaie, A. (2026). Adaptive Sparsening and Smoothing of the Treatment Model for Longitudinal Causal Inference Using Outcome‐Adaptive LASSO and Marginal Fused LASSO. Statistics in Medicine. https://doi.org/10.1002/sim.70316
Vancouver
Schnitzer ME, Talbot D, Liu Y, Berger D, Wang G, O'Loughlin J, et al. Adaptive Sparsening and Smoothing of the Treatment Model for Longitudinal Causal Inference Using Outcome‐Adaptive LASSO and Marginal Fused LASSO. Statistics in Medicine. 2026. doi:10.1002/sim.70316
BibTeX
@article{Schnitzer2026,
author = {Schnitzer, Mireille E and Talbot, Denis and Liu, Yan and Berger, David and Wang, Guanbo and O'Loughlin, Jennifer and Sylvestre, Marie-Pierre and Ertefaie, Ashkan},
title = {Adaptive Sparsening and Smoothing of the Treatment Model for Longitudinal Causal Inference Using Outcome‐Adaptive LASSO and Marginal Fused LASSO},
journal = {Statistics in Medicine},
year = {2026},
doi = {10.1002/sim.70316}
}
EUVIMED