A0348
Title: Using multi-source data to estimate subgroup effects in an external or external target population
Authors: Guanbo Wang - Harvard University (United States) [presenting]
Abstract: One major challenge in estimating effect heterogeneity is that the sample size of the data used is typically not enough to capture how effects vary according to the effect modifiers precisely. Therefore, there is interest in synthesizing evidence across multi-source data (e.g., multi-centre trials, meta-analyses of randomized trials, pooled analyses of observational cohorts) to improve the precision of estimators of heterogeneous treatment efficacy. Furthermore, when combining information from multi-source data, the samples typically do not represent a common target population of substantive interest. This raises the question of combining information from multi-source data in an interpretable way in the context of some meaningful target population of interest while using evidence across multi-source data to improve efficiency. Methods are developed and evaluated for using multi-source data to estimate subgroup treatment effects in an external target population or the populations underlying the data sources. A doubly robust estimator is proposed that, under mild conditions, is non-parametrically efficient and allows for nuisance functions to be estimated using machine learning methods. The methods are illustrated in meta-analyses of randomized trials for schizophrenia and bipolar disorder.
