The additive hazard estimator is consistent for continuous-time marginal structural models

DOI10.1007/S10985-019-09468-YzbMATH Open1436.62629arXiv1802.01946OpenAlexW2962695369WikidataQ91860601 ScholiaQ91860601MaRDI QIDQ83455FDOQ83455

Mats J. Stensrud, Pål C. Ryalen, Pål Christie Ryalen, Kjetil Røysland, Kjetil Røysland, Mats J. Stensrud

Publication date: 23 February 2019

Published in: Lifetime Data Analysis (Search for Journal in Brave)

Abstract: Marginal structural models (MSMs) allow for causal analysis of longitudinal data. The MSMs were originally developed as discrete time models. Recently, continuous-time MSMs were presented as a conceptually appealing alternative for survival analysis. In applied analyses, it is often assumed that the theoretical treatment weights are known, but these weights are usually unknown and must be estimated from the data. Here we provide a sufficient condition for a class of continuous-time MSMs to be consistent even when the weights are estimated, and we show how additive hazard models can be used to estimate such weights. Our results suggest that the continuous-time weights perform better than IPTW when the underlying treatment process is continuous. Furthermore, we may wish to transform effect estimates of hazards to other scales that are easier to interpret causally. We show that a general transformation strategy can be used on weighted cumulative hazard estimates to obtain a range of other parameters in survival analysis, and demonstrate how this strategy can be applied on data using our R packages ahw and transform.hazards.

Full work available at URL: https://arxiv.org/abs/1802.01946

zbMATH Keywords

additive hazard modelscausal inference in survival analysiscontinuous time marginal structural modelscontinuous time weights

Mathematics Subject Classification ID

Censored data models (62N01) Estimation in survival analysis and censored data (62N02) Applications of statistics to biology and medical sciences; meta analysis (62P10) Markov processes: hypothesis testing (62M02) Non-Markovian processes: estimation (62M09)

Cites Work

• Transforming cumulative hazard estimates
• Statistical models based on counting processes
• Revisiting g-estimation of the effect of a time-varying exposure subject to time-varying confounding
• Does Cox analysis of a randomized survival study yield a causal treatment effect?
• The Probability of Causation under a Stochastic Model for Individual Risk
• Multivariate point processes: predictable projection, Radon-Nikodym derivatives, representation of martingales
• Title not available (Why is that?)
• Title not available (Why is that?)
• Title not available (Why is that?)
• On null hypotheses in survival analysis
• A martingale approach to continuous-time marginal structural models

Cited In (4)

• ahw
• transform.hazards
• Causal inference with recurrent and competing events
• Simulating longitudinal data from marginal structural models using the additive hazard model

Uses Software

• R

   Recommendations

• The additive hazards model with time-varying regression coefficients 👍 👎
• Simulating longitudinal data from marginal structural models using the additive hazard model 👍 👎
• Robust estimation in the additive hazards model 👍 👎
• The additive hazards model with high-dimensional regressors 👍 👎
• A note on the asymptotic relative efficiency of estimators from the additive hazards model 👍 👎
• Adjusted empirical likelihood inference for additive hazards regression 👍 👎
• Some results on an additive hazards model 👍 👎
• Statistical inference for the additive hazards model under outcome-dependent sampling 👍 👎
• Additive hazard regression with auxiliary covariates 👍 👎
• Empirical Bayes estimation for additive hazards regression models 👍 👎

This page was built for publication: The additive hazard estimator is consistent for continuous-time marginal structural models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q83455)