A strategy to identify event specific hospitalizations in large health claims database

TL;DR

This paper presents a new framework for accurately classifying hospitalizations due to specific events, such as COVID-19, in large health claims databases, addressing challenges of raw data disjointedness and lack of context.

Contribution

The paper introduces a novel classification framework for event-specific hospitalizations in claims data and validates it using COVID-19 hospitalization data from a large US claims database.

Findings

Claims-specific COVID-19 hospitalization rates align with CDC reports by age and sex.

The framework improves accuracy of event-specific hospitalization classification.

Demonstrates feasibility of large-scale disease outcome analysis using claims data.

Abstract

Health insurance claims data offer a unique opportunity to study disease distribution on a large scale. Challenges arise in the process of accurately analyzing these raw data. One important challenge to overcome is the accurate classification of study outcomes. For example, using claims data, there is no clear way of classifying hospitalizations due to a specific event. This is because of the inherent disjointedness and lack of context that typically come with raw claims data. In this paper, we propose a framework for classifying hospitalizations due to a specific event. We then test this framework in a health insurance claims database with approximately 4 million US adults who tested positive with COVID-19 between March and December 2020. Our claims specific COVID-19 related hospitalizations proportion is then compared to nationally reported rates from the Centers for Disease Control by age and sex.