Data-driven identification of outpatient-suitable procedures: a machine learning approach

TL;DR

This paper uses machine learning to identify which medical procedures can be safely done as outpatient care, helping improve healthcare efficiency.

Contribution

A novel machine learning model with explainable AI methods is introduced to classify procedures suitable for outpatient care.

Findings

The model achieves 92% accuracy and >95% AUC in classifying outpatient-suitable procedures.

Explainable AI reveals that time from procedure to discharge is a key factor for outpatient suitability.

The model's scores can be applied to hospital data across countries to support policy decisions.

Abstract

Policymakers worldwide are encouraging a shift from inpatient to outpatient care to improve the efficiency of health systems. One of the first steps in such efforts typically involves allowing providers, often hospitals, to perform a designated list of procedures on an outpatient basis. However, determining which procedures are suitable for the hospital outpatient setting remains challenging and has traditionally relied on expert judgment and established practices. Our study advances this approach by employing supervised machine learning techniques to identify patterns in physician and expert decisions. We present a comprehensive classification of hospital procedures as either inpatient- or outpatient-suitable and use some methods of explainable AI methods to identify the main factors influencing these assessments. Our model achieves high accuracy (92%) and a robust area under the receiver operating characteristic curve (> 95%), assigning outpatient suitability scores to the entire German procedure catalog. These scores can easily be used with aggregate hospital data from different countries. To validate our approach, we applied the model to a surgical procedures shortlist from the OECD. Our findings provide decision-makers with a data-driven framework for developing targeted strategies to incentivize the provision of outpatient care.

The online version contains supplementary material available at 10.1007/s10729-026-09758-6.

We developed a machine learning classification model to enable the data-driven identification of medical procedures that may be suitable for outpatient settings.Our model assigns outpatient suitability scores, using data from 115 million hospital cases.Using some methods of Explainable AI, the model shows the extent to which different variables influence the outpatient suitability scores of individual procedures, supporting informed decision-making by stakeholders.Our study suggests that when evaluating the suitability of a medical procedure for outpatient provision, decision makers should prioritize the time from procedure to discharge for specific patient groups rather than focusing solely on the overall length of stay for all patients.The model equips decision makers with tools to develop targeted strategies that incentivize hospitals to optimize their service delivery.

We developed a machine learning classification model to enable the data-driven identification of medical procedures that may be suitable for outpatient settings.

Our model assigns outpatient suitability scores, using data from 115 million hospital cases.

Using some methods of Explainable AI, the model shows the extent to which different variables influence the outpatient suitability scores of individual procedures, supporting informed decision-making by stakeholders.

Our study suggests that when evaluating the suitability of a medical procedure for outpatient provision, decision makers should prioritize the time from procedure to discharge for specific patient groups rather than focusing solely on the overall length of stay for all patients.

The model equips decision makers with tools to develop targeted strategies that incentivize hospitals to optimize their service delivery.

The online version contains supplementary material available at 10.1007/s10729-026-09758-6.

Figures