Model Developed for Predicting Missed Appointments in Primary Care

TUESDAY, July 29, 2025 (HealthDay News) — The gradient boost model achieves the best performance for predicting no-shows and late cancellations in primary care practices, according to a study published in the July/August issue of the Annals of Family Medicine.

Wen-Jan Tuan, M.P.H., from the Penn State College of Medicine in Hershey, and colleagues conducted a retrospective longitudinal study leveraging geolinked clinical, care utilization, socioeconomic, and climate data from 15 family medicine clinics in Pennsylvania from January 2019 to June 2023. Machine learning models were developed to predict appointment outcomes; contributing factors for no-shows or late cancellations were identified at the population and patient levels.

The analysis included 109,328 patients and 1,118,236 appointments, including 77,322 and 75,545 (6.9 and 6.8 percent) no-shows and late cancellations, respectively. The best performance was achieved with the gradient boost model, with an area under the receiver operating characteristic curve of 0.852 and 0.921 for predicting no-shows and late cancellations, respectively. There was no bias detected against patient characteristics. The most important predictor of missed appointments was schedule lead time (i.e., number of days from a patient”s appointment request to the appointment date).

“This analytical framework lays a foundation for health systems to assess individual risk of missed appointments and design personalized strategies to help patients adhere to primary care appointments,” the authors write.

Abstract/Full Text