Interpretable Machine Learning for Resource Allocation with Application to Ventilator Triage

TL;DR

This paper introduces an interpretable, data-driven model for resource allocation during crises, specifically creating simple decision tree-based triage guidelines for ventilator allocation in COVID-19, improving outcomes over existing protocols.

Contribution

The paper develops a novel algorithm for optimal tree policies in Markov Decision Processes, applying it to create transparent triage guidelines for ventilator allocation during pandemics.

Findings

Reduced excess deaths with the new guidelines

Identified limitations of existing triage protocols

Demonstrated the effectiveness of interpretable decision trees

Abstract

Rationing of healthcare resources is a challenging decision that policy makers and providers may be forced to make during a pandemic, natural disaster, or mass casualty event. Well-defined guidelines to triage scarce life-saving resources must be designed to promote transparency, trust, and consistency. To facilitate buy-in and use during high-stress situations, these guidelines need to be interpretable and operational. We propose a novel data-driven model to compute interpretable triage guidelines based on policies for Markov Decision Process that can be represented as simple sequences of decision trees ("tree policies"). In particular, we characterize the properties of optimal tree policies and present an algorithm based on dynamic programming recursions to compute good tree policies. We utilize this methodology to obtain simple, novel triage guidelines for ventilator allocations for COVID-19 patients, based on real patient data from Montefiore hospitals. We also compare the performance of our guidelines to the official New York State guidelines that were developed in 2015 (well before the COVID-19 pandemic). Our empirical study shows that the number of excess deaths associated with ventilator shortages could be reduced significantly using our policy. Our work highlights the limitations of the existing official triage guidelines, which need to be adapted specifically to COVID-19 before being successfully deployed.