Predicting Health Access during Critical Public Health Crises: An Analysis of Medical Office Responses to the COVID-19 Pandemic

Abstract

Objectives This study explores how ambulatory medical practices adapted their policies in response to the global COVID-19 crisis. Practice and provider characteristics were used to build an artificial intelligence model that predicts future medical practice closures during critical events. Methods We surveyed 261 outpatient medical practices and collected information on clinician age, gender, the protective measures implemented, closure status, and utilization of telemedicine services. These data were used to build an artificial intelligence model that predicts practice closure in future critical public health events. Results Responses varied by specialty, location, and physician characteristics. These factors predicted closure status in 85.45% of test cases. Conclusions This paper demonstrates that practice characteristics can assist in predicting medical practice responses to public health events.