Predicting patient no-shows using machine learning: A comprehensive review and future research agenda

Patient no-shows for scheduled medical appointments pose significant challenges to healthcare systems, resulting in wasted resources, increased costs, and disrupted continuity of care. This comprehensive review examines state-of-the-art machine learning (ML) approaches for predicting patient no-shows in outpatient settings, analyzing 52 publications from 2010 to 2025.

The study reveals significant advancements in the field, with Logistic Regression (LR) as the most commonly used model in 68% of the studies. Tree-based models, ensemble methods, and deep learning techniques have gained traction in recent years, reflecting the field’s evolution. The best-performing models achieved Area Under the Curve (AUC) scores between 0.75 and 0.95, with accuracy ranging from 52% to 99.44%. Methodologically, researchers addressed common challenges such as class imbalance using various sampling techniques and employed a wide range of feature selection methods to improve model efficiency. The review also highlighted the importance of considering temporal factors and the context-dependent nature of no-show behavior across different healthcare settings.

Using the ITPOSMO framework (Information, Technology, Processes, Objectives, Staffing, Management, and Other Resources), the study identified several gaps in current ML approaches. Key challenges include data quality and completeness, model interpretability, and integration with existing healthcare systems. Future research directions include improving data collection methods, incorporating organizational factors, ensuring ethical implementation, and developing standardized approaches for handling data imbalance. The review also suggests exploring new data sources, utilizing ML algorithms to analyze patient behavior patterns, and using transfer learning techniques to adapt models across different healthcare facilities.

By addressing these challenges, healthcare providers can leverage ML to improve resource allocation, enhance patient care quality, and advance predictive analytics in healthcare. This comprehensive review underscores the potential of ML in predicting no-shows while acknowledging the complexities and challenges in its practical implementation.