Related Experiment Videos
Refining the predictive variables in the "Surgical Risk Preoperative Assessment System" (SURPAS): a descriptive
William G Henderson1,2,3, Michael R Bronsert1,2, Karl E Hammermeister1,2,4
11Surgical Outcomes and Applied Research program, Department of Surgery, University of Colorado School of Medicine, Aurora, CO USA.
Patient Safety in Surgery
|August 28, 2019
Summary
The Surgical Risk Preoperative Assessment System (SURPAS) was refined to improve accuracy and usability. Enhancements include updated risk variables and automated documentation for easier implementation in clinical practice.
Area of Science:
- Medical Informatics
- Surgical Risk Assessment
- Health Services Research
Background:
- The Surgical Risk Preoperative Assessment System (SURPAS) provides preoperative risk predictions for patients.
- Surgeon and patient feedback identified areas for improving SURPAS usability and usefulness.
- Eight key issues were systematically evaluated to enhance the SURPAS tool.
Purpose of the Study:
- To refine the SURPAS tool by addressing identified usability and usefulness concerns.
- To improve the accuracy and clinical applicability of preoperative risk prediction models.
- To enhance the implementation of SURPAS through improved documentation and user interface.
Main Methods:
- Logistic regression analysis of the ACS NSQIP PUF (2005-2015) was used for model refinement.
- Preoperative sepsis variable was replaced with a procedure-related risk variable.
- Automated electronic health record documentation and patient handouts were developed.
Main Results:
- Replacing the sepsis variable improved model discrimination and calibration.
- Adding a variable for multiple concurrent procedures did not significantly enhance model accuracy.
- Revised models predicted eleven adverse outcomes, including unplanned readmissions; automated documentation and visual risk displays were implemented.
Conclusions:
- Refinements successfully improved SURPAS model accuracy and reduced manual data entry.
- Graphical risk displays and automated EHR documentation facilitate easier SURPAS implementation.
- While accuracy improved, a variable for complex operations did not enhance model performance.