Timezone: »

Identifying Disparities in Sepsis Treatment using Inverse Reinforcement Learning
Hyewon Jeong · Taylor Killian · Sanjat Kanjilal · Siddharth Nayak · Marzyeh Ghassemi

Sepsis is a severe reaction by the human body to infection and is associated with significant morbidity and mortality. Advances in the scale and granularity of electronic health record data offer the opportunity to apply reinforcement learning to understand clinician diagnostic and treatment policies for this complex condition, which can be used to understand the factors that drive disparities in sepsis care. The fundamental problem in using RL to model sepsis is that the reward function is unknown and involves tradeoffs between competing outcomes. In this work, we develop an inverse reinforcement learning (IRL) model to learn a reward function for patients being treated for sepsis, then leverage offline RL to map state-action pairs from retrospective data, thereby learning the expert policy. We will apply this approach to two large and independent datasets: part of MIMIC-IV data with sepsis patients admitted to ICU and the clinical data warehouse of the Mass General Brigham healthcare system which has detailed data from arrival in the emergency room until hospital discharge across 12 hospitals in the New England area from 2015 through the present. With learned policy, we will identify whether policies differ by gender and race/ethnicity subgroups, and finally, we will attempt to identify changes in recorded physician policies before and after the introduction of the national treatment guidelines. We hope this approach could help us understand the differential treatment policy across the subgroups of sepsis patients.

Author Information

Hyewon Jeong (MIT)
Taylor Killian (University of Toronto--Vector Institute Massachusetts Institute of Technology)
Sanjat Kanjilal (Harvard Medical School)

Dr. Kanjilal is an Instructor in the Department of Population Medicine at the Harvard Pilgrim Health Care Institute and the Associate Medical Director of Clinical Microbiology at the Brigham & Women’s Hospital (BWH). He is also an infectious diseases physician at the BWH and the course director of HST 040, Mechanisms of Microbial Pathogenesis, at Harvard Medical School. Dr. Kanjilal's research interests focus on harnessing observational and experimental data to improve the diagnosis and management of infectious diseases. Specific areas of work include the development of decision support tools built over machine learning models that assist healthcare providers in a variety of tasks such as antibiotic treatments and diagnostic testing strategies. His long term goal is to improve medical decision making by integrating accurate and interpretable artificial intelligence-supported meta-diagnostics into clinical workflows and to use these models on a broader scale to inform public health policy.

Siddharth Nayak (Massachusetts Institute of Technology)
Marzyeh Ghassemi (MIT)

More from the Same Authors