Exploring Factors Associated With Missed Dental Appointments: A Machine Learning Analysis of Electronic Dental Records.

Abstract

This study aimed to employ machine learning techniques to explore the factors that could be associated with missed dental appointments. This cross-sectional study analyzed a total of 14,066 electronic dental records. Dental appointment adherence was categorized as attended or missed. Descriptive statistics and machine learning techniques, including conditional inference regression trees (CTree) and random forests (RFs), were employed for the analyses. About 31% of dental appointments were missed. Among the study population, appointments scheduled on Monday of the first week in the school year had the highest percentage of missed appointments, reaching up to 60%. Similarly, appointments scheduled on weeks 9, 10, 15-19, on Mondays, and with female dental students had slightly above 40% of missed appointments. The random forest analysis identified the week of the dental appointment, age, clinical day, and dental education level of students as the most influential variables in predicting dental appointment adherence. The most significant factors associated with a higher proportion of missed dental appointments were scheduled during specific weeks, on Mondays, with younger patients (<50 years), and with female dental students. Therefore, identifying these factors may assist healthcare providers and dental institutions in planning strategies to improve appointment attendance.