A Systematic Approach to Surveillance and Detection of Hierarchical Healthcare Cost Drivers and Utilization Offsets

TL;DR

This paper presents a systematic, hierarchical approach utilizing advanced statistical algorithms to identify and analyze emerging healthcare cost drivers and treatment offsets, providing actionable insights for early intervention.

Contribution

It introduces a novel hierarchical search and SPC-based methodology for detecting high-impact healthcare cost drivers and offsets in large, noisy datasets.

Findings

Identified five categories of emerging cost drivers in healthcare data.

Provided actionable insights for early intervention and cost mitigation.

Demonstrated approach effectiveness on IBM Watson MarketScan data.

Abstract

There is strong interest among healthcare payers to identify emerging healthcare cost drivers to support early intervention. However, many challenges arise in analyzing large, high dimensional, and noisy healthcare data. In this paper, we propose a systematic approach that utilizes hierarchical search strategies and enhanced statistical process control (SPC) algorithms to surface high impact cost drivers. Our approach aims to provide interpretable, detailed, and actionable insights of detected change patterns attributing to multiple clinical factors. We also proposed an algorithm to identify comparable treatment offsets at the population level and quantify the cost impact on their utilization changes. To illustrate our approach, we apply it to the IBM Watson Health MarketScan Commercial Database and organized the detected emerging drivers into 5 categories for reporting. We also discuss some findings in this analysis and potential actions in mitigating the impact of the drivers.