Target Trials and Structural Nested Models: Emulating RCTs using Observational Longitudinal Data
Conference
65th ISI World Statistics Congress
Format: IPS Abstract - WSC 2025
Session: IPS 736 - Causal Inference for Complex Data
Thursday 9 October 2 p.m. - 3:40 p.m. (Europe/Amsterdam)
Abstract
Target trials are RCTs one would like to conduct but cannot for ethical, financial, and/or logistical reasons. As a consequence, we must emulate such trials from observational data. A novel aspect of target trial methodology is that, for purposes of data analysis, each subject in the observational study is ‘enrolled’ in all target trials for which the subject is eligible, instead of a single trial. Subjects are typically censored if they deviate from the protocol, with inverse probability weighting to adjust for the censoring. If many subjects are censored, the estimates will be inefficient. In this talk I will propose more efficient estimators via the introduction of regime-specific structural nested target trial emulation models (SNTTEM). Given a regime, a SNTTEM imposes parametric models for all time-specific conditional treatment effects of the eligible subjects and leaves those for the ineligible unrestricted. We show that SNTTEMs can be fitted using g-estimation, a method that is less dependent on censoring than current estimation methods and hence generally more efficient.