Patient No-Show Prediction: A Systematic Literature Review.

Area of Science:

• Healthcare Management
• Health Informatics
• Predictive Analytics

Background:

• Patient no-shows pose significant challenges for healthcare centers, leading to revenue loss and extended patient waiting lists.
• Effective patient scheduling systems often rely on accurate prediction of appointment attendance.
• Predicting patient no-shows accurately remains a complex and unresolved issue in healthcare.

Purpose of the Study:

• To conduct a systematic literature review on predicting patient no-shows.
• To establish the current state-of-the-art in patient no-show prediction methodologies.
• To identify challenges and future research directions in this field.

Main Methods:

• Systematic literature review following the PRISMA methodology.
• Analysis of 50 relevant articles published in the field of patient no-show prediction.
• Examination of prediction techniques, database sizes, and performance metrics.

Main Results:

• The majority of reviewed studies (82%) were published in the last decade, indicating recent research interest.
• Logistic regression was the most frequently employed technique for predicting patient no-shows.
• Only two studies exceeded the baseline show rate in accuracy, and one achieved an Area Under the Curve (AUC) above 0.9, underscoring the difficulty of the problem.

Conclusions:

• Accurate prediction of patient no-shows is a significant challenge in healthcare, despite advancements in data and techniques.
• The current state-of-the-art indicates a need for further research to develop more effective predictive models.
• Improving no-show prediction can help mitigate financial losses and optimize healthcare resource allocation.