"Variability in Data Presentation in the Medical Student Performance Evaluation (MSPE)" and "A Generative Case System That Reflects Our Patient Population" - September 29

12:00-12:30pm ET - Variability in Data Presentation in the Medical Student Performance Evaluation (MSPE)

Every year, medical schools across the country spend a significant amount of time composing the Medical Student Performance Evaluation (MSPE) for graduating students. The MSPE contains data on student performance across the first three years and at times, the early fourth year of medical education. It is one of many important items residency program directors use when selecting incoming interns. 

Over the past few years, prompted by the 2016 recommendations of the AAMC MSPE Task Force, the medical education community has witnessed a change in the content of the MSPE. As reported by Giang, the majority of US medical schools have adopted the task force guidelines. While this adoption is occurring, wide variability still exists in the actual content and display of data within the MSPE. 

To better understand the variability we have created a 23-item data abstraction tool to code key MSPE data components. We have already used this tool to evaluate MSPEs written for the 2019 graduates of over 95% of US medical schools. We are now in the process of continuing this analysis for the 2020 graduates. We will present the results of the coding process, focusing on the variations in data presentation in the following areas: 

1. Clerkship grades, including comparative performance;
2. Overall comparative performance, including summary adjectives; and
3. Competency-level performance data. 

12:30-1:00pm ET - A Generative Case System That Reflects Our Patient Population

After cataloging the patient demographics in our cardiac and respiratory pre-clerkship course, we recognized a bias towards male patients, and an absence of ethnic descriptors unless clinically relevant. Utilizing a patient case generation system in use at Cumming School of Medicine, we  further developed a mechanism that generated pulmonary function tests that were based on patient height, weight and gender. This allowed us to present students with a limitless supply of pulmonary function test results across a patient population that was reflective of Statscan data for the Canadian population. This intervention has generated 2316 patient cases and is open access.

This system, cards, in addition to the above material has now generated over 1 million cases. By using a system to manage these variables whenever they are not clinically relevant, we are able to tackle a long standing issue in our curriculum.

This webinar will discuss how we recognized the bias in our patient case content, and recognize how components of patient case content that can be freed of inherent gender and ethnicity bias. Attendees will be able to see a mechanism for generating patient case content that is based on population statistics. 

“MedBiquitous Community Connection” is a series of free one-hour webinars featuring health professions educators and technology innovators from around the globe. “MedBiquitous Community Connection” provides a forum for community experts to share how digital technologies are promoting improvement and better outcomes across the continuum of health professions education. A portion of each webinar will reserve time for a brief Q&A session, allowing attendees the opportunity to dialogue with guest speakers.