Title: Estimating individualized treatment rules from distributed data collection sites
Authors: Erica Moodie - McGill University (Canada) [presenting]
Abstract: Precision medicine is a rapidly expanding area of health research wherein patient-level information is used to inform care via individualized treatment rules (ITR). Identifying the ITR typically requires large data sources, and may necessitate multi-centre collaborations. This may raise concerns about data privacy, management of massive data, and situations where confounding or data collected may differ across sites. We will introduce ITRs and a straightforward, doubly-robust estimation method and discuss approaches to the above-mentioned challenges while producing unbiased estimates of rules that tailor treatment to individual characteristics. The preferred approach is illustrated via an analysis of the impact of antidepressant treatments on BMI.