1. Estimation of complier causal treatment effects under the additive hazards model with interval-censored data.
- Author
-
Ma, Yuqing, Wang, Peijie, Li, Shuwei, and Sun, Jianguo
- Subjects
- *
TREATMENT effectiveness , *MAXIMUM likelihood statistics , *HAZARDS , *CENSORING (Statistics) , *DATA modeling , *EARLY detection of cancer , *CONFOUNDING variables - Abstract
Estimation of causal treatment effects has attracted a great deal of interest in many areas including social, biological and health science, and for this, instrumental variable (IV) has become a commonly used tool in the presence of unmeasured confounding. In particular, many IV methods have been developed for right-censored time-to-event outcomes. In this paper, we consider a much more complicated situation where one faces interval-censored time-to-event outcomes, which are ubiquitously present in studies with, for example, intermittent follow-up but are challenging to handle in terms of both theory and computation. A sieve maximum likelihood estimation procedure is proposed for estimating complier causal treatment effects under the additive hazards model, and the resulting estimators are shown to be consistent and asymptotically normal. A simulation study is conducted to evaluate the finite sample performance of the proposed approach and suggests that it works well in practice. It is applied to a breast cancer screening study. [ABSTRACT FROM AUTHOR]
- Published
- 2024
- Full Text
- View/download PDF