Short-term Mortality Prediction for Elderly Patients Using Medicare Claims Data

TL;DR

This study demonstrates that machine learning classifiers, when combined with an improved feature set derived from clinical insights, significantly outperform traditional methods in predicting six-month mortality among elderly Medicare beneficiaries using claims data.

Contribution

The paper introduces a novel feature set for machine learning models that enhances mortality prediction accuracy in elderly patients using Medicare claims data.

Findings

Machine learning classifiers outperform traditional risk prediction methods.

Improved feature set enhances model performance.

Potential applications in end-of-life decision making and population health.

Abstract

Risk prediction is central to both clinical medicine and public health. While many machine learning models have been developed to predict mortality, they are rarely applied in the clinical literature, where classification tasks typically rely on logistic regression. One reason for this is that existing machine learning models often seek to optimize predictions by incorporating features that are not present in the databases readily available to providers and policy makers, limiting generalizability and implementation. Here we tested a number of machine learning classifiers for prediction of six-month mortality in a population of elderly Medicare beneficiaries, using an administrative claims database of the kind available to the majority of health care payers and providers. We show that machine learning classifiers substantially outperform current widely-used methods of risk prediction but only when used with an improved feature set incorporating insights from clinical medicine, developed for this study. Our work has applications to supporting patient and provider decision making at the end of life, as well as population health-oriented efforts to identify patients at high risk of poor outcomes.