The design and evaluation of a novel algorithm for automated preference card optimization.

Area of Science:

• Health Informatics
• Surgical Operations Management
• Healthcare Supply Chain Management

Background:

• Inaccurate surgical preference cards lead to increased costs, waste, and delays.
• Existing methods for preference card improvement are often manual, institution-specific, and lack reproducibility.
• An automated, informatics-based approach was developed to address these limitations.

Purpose of the Study:

• To develop and test a novel algorithm for automating the optimization of surgical preference cards.
• To evaluate the algorithm's effectiveness in reducing direct costs and improving accuracy compared to manual methods.
• To establish a reproducible, informatics-driven solution for surgical supply management.

Main Methods:

• An algorithm was developed to cross-reference procedure supplies with preference card quantities, using time-series regression to identify inaccuracies.
• Algorithm performance was assessed by comparing direct costs of preference cards revised using the algorithm versus those revised manually or not revised.
• A difference-in-differences (DID) multivariate fixed-effects model analyzed cost changes over pre- and post-intervention periods.

Main Results:

• The algorithm estimated the accuracy of over 469,000 surgeon-procedure-specific items.
• Nurses revised 309 preference cards across eight surgical services, impacting thousands of procedures.
• The intervention group experienced an 8.38% decrease in average direct supply costs per case, while the control group saw a 13.21% increase (P < .001).

Conclusions:

• The automated algorithm achieved cost savings comparable to manual optimization methods.
• This informatics-based approach offers a more readily reproducible solution for optimizing surgical preference cards.
• Automated preference card optimization presents a scalable strategy for enhancing efficiency and cost-effectiveness in surgical services.