Providence, RI, July 29, 2025 / prnewswire/ – Survey results from a New research Published in Family Medici's chroniclene We show that machine learning models accurately predict primary care no-shows and late cancellations, and provide insights to develop personalized strategies to improve booking adherence.
Research conducted by researchers Pennsylvania State UniversityIntegrated Clinical, Geosocial, and Environmental Data from 15 Family Medicine Clinics Personalizes No Shows, Cancellations, and Primary Care Visits and Over 109,000 Patients forecasts and Over 1 million appointments. The team leveraged multi-class machine learning models to predict reservation results by comparing four different approaches, including gradient boost, random forest, neural networks, and lassologistic regression.
The gradient boost model achieves the best performance when classifying appointments as no show or slow cancellation, with the area below the receiver operating characteristic curve score (AUROC, 0-to-1 measure of overall predicted quality, 1.0 is perfect, 0.5 has a chance) being 0.85 for no show and 0.92 for late cancellation. An unbiased check of the gradient boost model showed that the predicted results were not biased against gender or racial/ethnic patient characteristics. Lead time (the number of days between booking and visit) was the most important predictor of missed bookings. Lead times of 60 days or more were associated with a higher risk of missing appointments.
“Given the strong effect of lead times, clinics may prioritize short waiting times for high-risk patients,” the author wrote, “machine learning-based analyses help clinicians predict patient-specific needs, personalize outreach efforts, and actively promote appointment scheduling.”
connection Special Report In the same issue of Family Medicine Chronicles We propose five high-level considerations regarding data transformation required to make research like the above more common. These considerations include automating data collection, organizing fragmented data, identifying primary care-specific use cases, integrating AI and machine learning and integration into human workflows, and monitoring for unintended outcomes.
Effective and cohesive efforts require collaboration between industrial and academic AI and machine learning communities with primary care, increased funding from the private and public sectors, and upgrades to human and data infrastructure.
“These types of inter-functional collaboration are key to realizing the transformation of primary care data into critical resources, allowing us to unlock the true potential of artificial intelligence and machine learning in primary care,” the author writes.
Article cited:
Predicting missed plans in primary care: a personalized machine learning approach
Wen Jiang TuanDHA, MS, MPH; Yi Hwang YangMS; Bidal Abou al ArdatMD, Todd FelixMD; Qiushi ChenPhD
Data conversion to advance research and implementation of AI/ML in primary care
Timothy Tsai,do,mmci; Julie J. LeeMD, MPH; Robert PhillipsMD, MSPH, Stephen LynnMD
Family Medicine Chronicles Open Access is a peer-reviewed, indexed research journal that provides an interdisciplinary forum for new evidence-based information affecting the field of primary care. It was released on May 2003, Family Medicine Chronicles It is sponsored by six family medical institutions, including the American Academy of Family Medicine, the American Board of Family Medicine, the Association of Family Medicine Teachers, the Association of Family Medicine Schools, the Association of Family Medicine Residencies, and the North American Primary Care Research Group. Family Medicine Chronicles It is published online six times a year, and there is no public fee. Includes original research from the fields of clinical, biomedical, social and health services, as well as contributions on methodologies and theories, selected reviews, essays, and editing. The complete editorial content and interactive discussion groups for each published article are available for free access on the journal's website www.annfammed.org.
Sources of Family Medicine
