# A hybrid AHP and K-means model for biopsychosocial surgical prioritization: validation in a high-complexity ENT unit

## TL;DR

This paper introduces a new model combining AHP and K-Means to prioritize surgery patients based on biopsychosocial factors, improving outcomes and efficiency in a high-complexity ENT unit.

## Contribution

Operationalizing an AHP-K-Means hybrid with ethical safeguards and real-world validation in a high-complexity ENT unit.

## Key findings

- The model reduced mean clinical risk by 27% and urgent hospitalizations by 41%.
- High-priority patients experienced over 12 days of accelerated access to surgery.
- The approach improved fairness, efficiency, and responsiveness in healthcare delivery.

## Abstract

•AHP and K-Means enable prioritizing surgical patients using biopsychosocial criteria.•Expert knowledge and clustering identify clinically relevant patient groups.•Structured prioritization boosts clinical outcomes and operational efficiency.•Model validation shows significant improvements in clinical risk, hospitalization rates, and waiting times for high-priority patients.

AHP and K-Means enable prioritizing surgical patients using biopsychosocial criteria.

Expert knowledge and clustering identify clinically relevant patient groups.

Structured prioritization boosts clinical outcomes and operational efficiency.

Model validation shows significant improvements in clinical risk, hospitalization rates, and waiting times for high-priority patients.

To address the critical challenge of efficiently and ethically managing surgical waiting lists in digital health systems by developing a decision support framework based on biopsychosocial prioritization.

The authors integrate the Analytic Hierarchy Process (AHP) with K-Means clustering to create a hybrid decision support model that prioritizes patients using multidimensional biopsychosocial variables. The model was applied in the otolaryngology (ENT) unit of a high-complexity public hospital in Chile. Expert-informed weightings guided the AHP process, while K-Means clustering enabled data-driven segmentation into clinically coherent patient groups.

The proposed methodology significantly outperformed traditional chronological scheduling approaches. Specifically, it achieved a 27 % reduction in mean clinical risk, a 41 % decrease in urgent hospitalizations, a 32 % reduction in urgent bed days, and more than 12-days of acceleration in access for high-priority patients.

While the AHP-clustering hybrid is established in prior literature, our contribution lies in operationalizing it with ethical safeguards and real-world validation within a high complexity ENT unit.

Our hybrid AHP and K-Means approach offers a transparent, scalable, and interpretable decision support tool for surgical prioritization. It aligns with the goals of digital health transformation by improving the fairness, efficiency, and responsiveness of healthcare delivery.