A Statistical Model for the Prediction of Patient Non-Attendance in a Primary Care Clinic

Xiuli Qu, University of North Dakota
Ronald L. Rardin, Purdue University, Industrial Engineering

Document Type Article

Abstract

High patient non-attendance rates cause critical problems for many outpatient clinics in the United States. For historical patient attendance data in a primary care clinic, six categorical factors are analyzed in this paper: appointment type, session, patient attendance history, insurance, age group and weather. The main effects of session, insurance, age group and weather are statistically significant, and nine 2-factor interaction effects of the six factors are significant too. Furthermore, using logistic regression, a statistical model is built to predict the non- attendance rate of each classified group of appointments.