Abstract: Systems and methods for evaluating clinical comparative efficacy using real-world health data are provided herein. The method includes obtaining health trajectories for members of a healthcare system. The method also includes identifying index events in the health trajectories, and segmenting the health trajectories with index events into sub-trajectories such that each sub-trajectory ends at a different index event. The method also includes generating a digital fingerprint for each sub-trajectory by either (i) generating a raw data vector, or (ii) applying representation learning to generate an embedding vector. The method also includes identifying sub-trajectories that are similar to a patient sub-trajectory by either (i) performing a stratified search, or (ii) performing a nearest-neighbor search on the embedding vectors of the members. The method also includes ranking treatment strategies for the patient based on outcomes of the treatment strategies, according to the identified similar sub-trajectories.

Abstract: The present application describes a system and method for implementing a causal recommender for personalized disease treatment selection using machine learning. The method includes obtaining health trajectories for patients. Each health trajectory includes sub-trajectories. Each sub-trajectory includes a treatment event and ends at a respective index event. The method further includes stratifying the sub-trajectories for each patient to form stratified patient segments. Each segment corresponds to a separate and distinct health condition and includes the sub-trajectories for patients that have the health condition. For each segment, the method includes performing pairwise causal inference analysis on one or more treatments to estimate average treatment effect (ATE) values, and performing network meta-analysis on the ATE values, thereby ranking the one or more treatments.