BMC2 Work on a Novel Risk Calculator for Nonhome Discharge After Lower Extremity Bypass Presented at SAVS 2024

A microphone is in the foreground with an audience out of focus in the background.
Elizabeth Horn

Brian T. Fry, MD, MS presented the rapid-fire talk, “A Novel Risk Calculator for Nonhome Discharge After Lower Extremity Bypass,” during the Southern Association for Vascular Surgery (SAVS) Annual Meeting on Friday, January 26th, 2024.

Nonhome discharge (NHD [e.g., to a skilled nursing or rehabilitation facility]) is associated with increased discharge delays, adverse events post-discharge, and decreased patient quality of life.

Dr. Fry and his co-authors used BMC2 data to analyze patients undergoing lower extremity bypass. They then used XGBoost machine learning modeling to assess predictive accuracy based on factors associated with NHD, including patient age, gender, race, comorbidities, smoking and ambulatory status, tissue loss, procedure class (elective, urgent, emergent), and length of stay.

The top five most important factors predictive of NHD were longer length of stay, older age, presence of an ambulatory deficit, absence of claudication, and requiring an emergent procedure.

The machine learning model can accurately predict NHD for greater than 86% of patients, and study results could inform care providers how to reduce the risks of NHD.

Co-authors are Jeremy Albright, PhD; Shukri H.A. Dualeh, MD; Nicolas J. Mouawad, MD, MPH, MBA; Andrew Kimball, MD; Jordan Knepper, MD, MSc; Eugene W. Laveroni, DO; Chandu Vemuri, MD; Peter K. Henke, MD.