Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events

Marginal structural models (MSMs) are recognized as useful methods for addressing the issue of time-varying confounding in longitudinal studies. In the analyses of longitudinal data with binary outcomes, using the generalized estimating equation (GEE) logistic regression model within the MSM framewo...

Full description

Saved in:
Bibliographic Details
Main Authors: Hiroyuki Shiiba, Hisashi Noma, Keisuke Kuwahara, Tohru Nakagawa, Tetsuya Mizoue
Format: Article
Language:English
Published: Taylor & Francis Group 2025-12-01
Series:Data Science in Science
Subjects:
Online Access:https://www.tandfonline.com/doi/10.1080/26941899.2025.2527144
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850087290611695616
author Hiroyuki Shiiba
Hisashi Noma
Keisuke Kuwahara
Tohru Nakagawa
Tetsuya Mizoue
author_facet Hiroyuki Shiiba
Hisashi Noma
Keisuke Kuwahara
Tohru Nakagawa
Tetsuya Mizoue
author_sort Hiroyuki Shiiba
collection DOAJ
description Marginal structural models (MSMs) are recognized as useful methods for addressing the issue of time-varying confounding in longitudinal studies. In the analyses of longitudinal data with binary outcomes, using the generalized estimating equation (GEE) logistic regression model within the MSM framework is a common approach to estimate the odds ratio. However, due to the interpretive issues with the odds ratio, recent statistical guidelines recommend the use of the risk ratio and the risk difference. Nevertheless, there are no applicable MSM methods using GEE that enable straightforward estimation for the causal risk ratio and risk difference. In this article, we provide two straightforward and effective methods for estimating the causal risk ratio and risk difference based on the MSM-GEE framework, using Poisson regression and normal linear regression models. We validated these methods through comprehensive simulation studies, confirming their unbiased estimation of the causal risk ratio and risk difference. Importantly, these methods remain effective even in situations where the MSM-GEE logistic regression model, which is the most widely used method for binary outcome data, yields biased estimates of the causal odds ratio. In addition, we applied the proposed methods to real-world longitudinal data and clearly demonstrated their practical effectiveness.
format Article
id doaj-art-14f468ec802648d8b37adb6acccf88d1
institution DOAJ
issn 2694-1899
language English
publishDate 2025-12-01
publisher Taylor & Francis Group
record_format Article
series Data Science in Science
spelling doaj-art-14f468ec802648d8b37adb6acccf88d12025-08-20T02:43:15ZengTaylor & Francis GroupData Science in Science2694-18992025-12-014110.1080/26941899.2025.2527144Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome EventsHiroyuki Shiiba0Hisashi Noma1Keisuke Kuwahara2Tohru Nakagawa3Tetsuya Mizoue4The Graduate Institute for Advanced Studies, The Graduate University for Advanced Studies (SOKENDAI), Tokyo, JapanDepartment of Interdisciplinary Statistical Mathematics, The Institute of Statistical Mathematics, Tokyo, JapanDepartment of Epidemiology and Prevention, Center for Clinical Sciences, National Center for Global Health and Medicine, Tokyo, JapanHitachi Health Care Center, Hitachi, Ltd, Hitachi, JapanDepartment of Epidemiology and Prevention, Center for Clinical Sciences, National Center for Global Health and Medicine, Tokyo, JapanMarginal structural models (MSMs) are recognized as useful methods for addressing the issue of time-varying confounding in longitudinal studies. In the analyses of longitudinal data with binary outcomes, using the generalized estimating equation (GEE) logistic regression model within the MSM framework is a common approach to estimate the odds ratio. However, due to the interpretive issues with the odds ratio, recent statistical guidelines recommend the use of the risk ratio and the risk difference. Nevertheless, there are no applicable MSM methods using GEE that enable straightforward estimation for the causal risk ratio and risk difference. In this article, we provide two straightforward and effective methods for estimating the causal risk ratio and risk difference based on the MSM-GEE framework, using Poisson regression and normal linear regression models. We validated these methods through comprehensive simulation studies, confirming their unbiased estimation of the causal risk ratio and risk difference. Importantly, these methods remain effective even in situations where the MSM-GEE logistic regression model, which is the most widely used method for binary outcome data, yields biased estimates of the causal odds ratio. In addition, we applied the proposed methods to real-world longitudinal data and clearly demonstrated their practical effectiveness.https://www.tandfonline.com/doi/10.1080/26941899.2025.2527144Longitudinal studytime-varying confoundermarginal structural modelrisk ratiorisk difference
spellingShingle Hiroyuki Shiiba
Hisashi Noma
Keisuke Kuwahara
Tohru Nakagawa
Tetsuya Mizoue
Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
Data Science in Science
Longitudinal study
time-varying confounder
marginal structural model
risk ratio
risk difference
title Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
title_full Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
title_fullStr Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
title_full_unstemmed Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
title_short Causal Risk Ratio and Causal Risk Difference in Longitudinal Studies With Frequent Outcome Events
title_sort causal risk ratio and causal risk difference in longitudinal studies with frequent outcome events
topic Longitudinal study
time-varying confounder
marginal structural model
risk ratio
risk difference
url https://www.tandfonline.com/doi/10.1080/26941899.2025.2527144
work_keys_str_mv AT hiroyukishiiba causalriskratioandcausalriskdifferenceinlongitudinalstudieswithfrequentoutcomeevents
AT hisashinoma causalriskratioandcausalriskdifferenceinlongitudinalstudieswithfrequentoutcomeevents
AT keisukekuwahara causalriskratioandcausalriskdifferenceinlongitudinalstudieswithfrequentoutcomeevents
AT tohrunakagawa causalriskratioandcausalriskdifferenceinlongitudinalstudieswithfrequentoutcomeevents
AT tetsuyamizoue causalriskratioandcausalriskdifferenceinlongitudinalstudieswithfrequentoutcomeevents