NOVEMBER 3, 2023
Wesley Marrero, PhD, MA, MEng
Assistant Professor of Engineering
Thayer School of Engineering at Dartmouth
About the Presentation: The increasing availability of healthcare data has provided a great opportunity to develop data-driven models to guide health policy and medical practice. This talk aims to present new methods that use these data to make better healthcare decisions at a population and patient level.
I will first discuss my team’s work identifying predictors of premature discontinuation of opioid use disorder treatment and where and when resources should be allocated to improve access to treatment. This work is based on multiple datasets, including the National Survey on Drug Use and Health, the National Emergency Medical Services Information System, and the National Center for Health Statistics mortality data. Afterward, I will give an overview of our research modeling the supply, demand, and allocation of organs for transplantation using data from the Organ Procurement and Transplantation Network and the US Census Bureau. Subsequently, I will present our work developing an app to assist in the decision-making process for accepting livers offered to patients on the organ transplantation waiting list. I will then introduce our research predicting anxiety disorder, depression, and burnout in medical students using passive sensing and data from the Healthy Minds Study. Lastly, I will propose a modeling framework to consider physicians’ judgment and patients’ preferences in implementing treatment protocols. To illustrate how this method can be implemented in medical practice, my team and I found sets of similar-performing antihypertensive treatment choices for over 16 million adults in the US using data from the National Health and Nutrition Examination Survey.
This research has the potential to improve healthcare practice by giving flexible and achievable guidelines to policymakers and medical professionals based on patient and population-level data.
About the Presenter: Professor Marrero’s research interest lies in developing decision-support tools that consider the challenges associated with their implementation in practice, such as inequity, irrational behavior, lack of interpretability, and need for flexibility. To this end, he designs and applies techniques from operations research, statistics, and algorithmic fairness, with an emphasis on simulation and optimization. His current work focuses on measuring and mitigating the effects of bias and inequity through statistical and optimization methods. Through this research, Dr. Marrero has ongoing collaborations with Dartmouth Health, the Geisel School of Medicine, the Massachusetts General Hospital, the University of Michigan Medical School and School of Public Health, and the U.S. Department of Veterans Affairs.