Improving the Prediction of Total Surgical Procedure Time Using Linear Regression Modeling

Area of Science:

• Health Services Research
• Medical Informatics
• Operations Research

Background:

• Efficient operating room (OR) utilization is critical for healthcare efficiency.
• Accurate prediction of total procedure time (TPT) is essential for effective OR scheduling and resource allocation.
• Current methods for predicting TPT may lack the necessary precision for optimal OR management.

Purpose of the Study:

• To enhance the accuracy of total procedure time (TPT) predictions for surgical cases.
• To evaluate the effectiveness of linear regression models incorporating estimated surgeon-controlled time (eSCT) and other variables for TPT prediction.
• To compare the performance of developed linear regression models against a fixed ratio model and separate prediction of anesthesia-controlled time (ACT).

Main Methods:

• Utilized a Dutch benchmarking database of 79,983 surgeries from six academic hospitals (2012-2016).
• Developed and tested various linear regression models to predict TPT using eSCT, patient age, operation type, ASA classification, and anesthesia type.
• Compared model performance against a fixed ratio model (eSCT * 1.33) and models predicting ACT separately.

Main Results:

• The most accurate TPT prediction was achieved using a linear regression model incorporating eSCT, operation type, ASA classification, and anesthesia type.
• This optimized linear regression model demonstrated significantly superior performance compared to the fixed ratio model.
• The developed model also outperformed methods that predicted anesthesia-controlled time (ACT) separately.

Conclusions:

• Linear regression models integrating eSCT and key patient/procedural variables offer superior TPT prediction accuracy.
• Improved TPT prediction accuracy can enhance OR scheduling and sequencing, leading to increased OR utilization.
• Adoption of these advanced prediction models can yield substantial financial and productivity benefits in healthcare settings.