ISE Disccusion: Disparities in Healthcare with Michele Samorani, Ph.D.

Details

Following Dr. Samorani’s seminar, we invite the ISE students and faculty to join via zoom or in person for a lunch discussion on “Disparities in Healthcare” with Michele Samorani, Ph.D.

Refreshments will be provided in Weil Hall Room 406.

Join virtually via Zoom: https://ufl.zoom.us/j/97583979403

Preceding Seminar information:
3/31/23 at 10:40 AM

Virtual Seminar
Zoom: https://ufl.zoom.us/j/97583979403

Dr. Michele Samorani

Title: “Studying and Fixing Algorithmic Racial Disparity in Medical Appointment Scheduling”

Abstract: To enhance clinic efficiency, cutting-edge scheduling algorithms utilize machine learning to forecast individual no-show probabilities for patients when booking medical appointments. Empirical evidence suggests that no-shows are correlated with patient socioeconomic status, and therefore, their race. This presentation will demonstrate how the integration of machine learning and optimization inadvertently creates racial disparities in outpatient appointment scheduling.

Using the real-world data from a large specialty clinic where black patients have higher no-show probabilities than non-black patients, we analytically study racial disparity in outpatient appointment scheduling. We discover that the conventional objective function used in practice generates racial disparities in two primary ways.

Firstly, prioritizing efficiency results in assigning patients with the highest risk of no-show to undesirable slots (overbooked or immediately following an overbooked slot), while assigning patients with the lowest risk of no-show to priority slots (preceding the day’s first overbooked slot). We propose various solutions and find that the best results are obtained through a race-aware objective function, which balances efficiency and racial fairness. A simpler heuristic method that strikes a good balance between disparity and efficiency is to simply eliminate priority slots.
Secondly, maximizing efficiency creates the perverse incentive of scheduling the patients with the highest risk of no-show farther into the future when compared to low-risk patients.

BIO:
Michele Samorani is an Associate Professor in Information Systems and Analytics at the Leavey School of Business at Santa Clara University.

He earned his undergraduate and Master’s degree in computer science from the University of Bologna and his Ph.D. in operations and information management from the University of Colorado at Boulder.

Michele’s research combines machine learning and optimization techniques to build decision support systems that improve companies’ business processes and information flow. His areas of research include machine learning and racial disparity in medical scheduling, relational data mining, text mining, and metaheuristic optimization.
His research has been published in outlets in Information Systems, Operations Management, and Operations Research, such as MIS Quarterly, Manufacturing & Service Operations Management, Production and Operations Management, INFORMS Journal on Computing, Decision Support Systems, and the European Journal of Operational Research.

Michele’s work on racial disparity in outpatient appointment scheduling has won the 2021 INFORMS Minority Issues Forum Paper Competition, has been featured by top-tier media, and was included in a United Nations report.

Please contact the ISE admin staff with any questions or information needed for the seminar: administration@ise.ufl.edu