A discrete-time survival model to handle interval-censored covariates, with applications to HIV cohort studies

TL;DR

This paper introduces a flexible discrete-time survival model that effectively handles interval-censored covariates, enabling more accurate analysis of HIV-related mortality and health outcomes in cohort studies.

Contribution

The paper develops a novel class of discrete-time parametric survival models that simultaneously address interval censoring of secondary events and primary outcomes.

Findings

Applied the model to HIV cohort data in South Africa.

Estimated the impact of HIV status on mortality.

Demonstrated applicability to various health outcome analyses.

Abstract

Methods are lacking to handle the problem of survival analysis in the presence of an interval-censored covariate, specifically the case in which the conditional hazard of the primary event of interest depends on the occurrence of a secondary event, the observation time of which is subject to interval censoring. We propose and study a flexible class of discrete-time parametric survival models that handle the censoring problem through simultaneous modeling of the interval-censored secondary event, the outcome, and the censoring mechanism. We apply this model to the research question that motivated the methodology, estimating the effect of HIV status on all-cause mortality in a prospective cohort study in South Africa. Our model has applicability for many open questions, including estimating the impact of policy decisions on population level HIV-related outcomes and determining causes of morbidity and mortality for which the HIV positive population may be at increased risk. Examples include determining how the large-scale transition from efavirenz-based to dolutegravir-based first-line ART impacted mortality for people living with HIV and determining whether HIV status is associated with increased risk of stroke, diabetes, hypertension, and other non-communicable diseases.