Daily surgery caseload prediction: towards improving operating theatre efficiency.

Area of Science:

• Healthcare Operations Research
• Health Informatics
• Predictive Analytics

Background:

• Operating theatre inefficiencies stem from dynamic demand and complex scheduling, leading to access block and treatment delays.
• Suboptimal theatre utilization impacts patient care and hospital resource management.
• Improved theatre management is crucial for efficient healthcare delivery.

Purpose of the Study:

• To develop and evaluate predictive models for operating theatre demand using historical surgical data.
• To provide decision support for enhanced theatre management and resource allocation.
• To improve the efficiency and effectiveness of hospital services.

Main Methods:

• Utilized historical operating theatre data from a major metropolitan hospital over four years.
• Developed statistical and machine learning models to predict daily surgery demand (elective and emergency).
• Employed time-series forecasting techniques, including rolling windows, calendar variables, and regression algorithms.

Main Results:

• Achieved approximately 90% accuracy in predicting overall daily hospital-level surgery counts.
• Demonstrated consistent predictive performance across a one-year forecast horizon.
• Identified that predictions for smaller subgroups (e.g., by medical specialty) were less predictable.

Conclusions:

• Operating theatre demand prediction is a valuable tool for effective theatre management.
• Accurate forecasting enables optimized service provision, leading to better patient health outcomes.
• The predictive approach supports both short-term operational decisions and long-term strategic planning.