Implementing prospective probabilistic scheduling of surgical cases to yield more predictable utilisations

Area of Science:

• Health Services Research
• Operations Research
• Surgical Management

Background:

• Traditional surgeon-led booking for operating theatres often struggles to balance maximizing utilization with minimizing over-utilization risks.
• Accurate scheduling is crucial for efficient healthcare delivery and managing patient waiting lists.

Purpose of the Study:

• To evaluate the impact of a prospective, probabilistic scheduling method (Infix Schedule tool) on operating theatre utilization and patient cancellations.
• To compare the accuracy of probabilistic scheduling predictions against surgeon-led scheduling and surgeon overrides.

Main Methods:

• Analysis of 2420 surgical lists across multiple specialties in an NHS Scotland hospital.
• Utilized the Infix Schedule tool, employing median and median absolute deviation (MAD) of historical case times to predict completion probabilities.
• Compared outcomes across three groups: prior surgeon-led scheduling, surgeon overrides of the tool, and unadjusted Infix-generated schedules.

Main Results:

• Infix Schedule predictions (77.8% utilized time) were significantly closer to actual realized utilization (78.8%) than surgeon predictions (87.7%).
• Surgeon overrides resulted in lower utilized time (76.5%) compared to Infix predictions (78.0%).
• Infix scheduling led to lower patient cancellation rates (8.51%) compared to surgeon-led scheduling (10.96%).

Conclusions:

• Probabilistic scheduling tools like Infix Schedule can enhance operating theatre efficiency and reduce patient cancellations.
• Accurate surgical coding is essential for the optimal performance of such scheduling tools.
• Iterative learning and minimizing human overrides are key to further refining efficiency and waiting list management.