Physician Characteristics Associated With Advance Care Planning After a Machine Learning-Based Nudge

Prognostic machine learning models can help identify patients in need of advance care planning (ACP) conversations. However, additional work is needed to understand which factors guide physicians’ comfort with and prioritization of ACP conversations, even after such nudges. In this exploratory secondary analysis of a cluster randomized trial, we examine the relationship between characteristics of internal medicine physicians (eg, training background and practice patterns) and likelihood of ACP conversation following notification by a mortality risk prediction machine learning model.