# Bayesian semi-parametric G-computation for causal inference in a cohort study with MNAR dropout and death

## TL;DR

This paper introduces a Bayesian semi-parametric G-computation method using Bayesian additive regression trees to estimate causal effects in longitudinal studies with MNAR dropout and death, addressing limitations of traditional parametric approaches.

## Contribution

It develops a flexible Bayesian approach with sensitivity parameters for causal inference under MNAR, applicable to complex longitudinal observational data.

## Key findings

- Successfully applied to a cohort study on cognition and aging.
- Provided estimates of widowhood effect on memory accounting for MNAR dropout.
- Compared favorably to standard methods in simulation and real data analysis.

## Abstract

Causal inference with observational longitudinal data and time-varying exposures is often complicated by time-dependent confounding and attrition. The G-computation formula is one approach for estimating a causal effect in this setting. The parametric modeling approach typically used in practice relies on strong modeling assumptions for valid inference, and moreover depends on an assumption of missing at random, which is not appropriate when the missingness is missing not at random (MNAR) or due to death. In this work we develop a flexible Bayesian semi-parametric G-computation approach for assessing the causal effect on the subpopulation that would survive irrespective of exposure, in a setting with MNAR dropout. The approach is to specify models for the observed data using Bayesian additive regression trees, and then use assumptions with embedded sensitivity parameters to identify and estimate the causal effect. The proposed approach is motivated by a longitudinal cohort study on cognition, health, and aging, and we apply our approach to study the effect of becoming a widow on memory. We also compare our approach to several standard methods.