|
Abstract:
|
The Botswana Combination Prevention Project (BCPP) was a cluster-randomized HIV prevention trial whose follow-up period coincided with Botswana’s adoption of a universal test-and-treat strategy for HIV management. Of interest is whether, and to what extent, this change in policy (i) modified the observed preventative effects of the study intervention and (ii) was associated with a reduction in the incidence of HIV in Botswana. To address these questions, we propose a stratified proportional hazards model for clustered interval-censored data with time-dependent covariates, and develop a composite expectation maximization algorithm that facilitates estimation of this model without placing parametric assumptions on either the baseline hazard functions or the within-cluster dependence structure; we also present a robust profile composite likelihood variance estimator for inference on the regression parameters. We characterize the finite-sample performance and robustness of these estimators through a series of simulation studies, and conclude by applying the model to a re-analysis of the BCPP, with the implementation of universal test-and-treat now modeled as a time-dependent covariate.
|
