Segmenting high-cost Medicare patients into potentially actionable cohorts

Area of Science:

• Healthcare economics
• Health services research
• Population health management

Background:

• Healthcare providers face increasing responsibility for healthcare expenditures.
• A small proportion of patients accounts for a significant concentration of healthcare spending.
• Simple patient segmentation strategies can aid in cost-control interventions.

Purpose of the Study:

• To assess the utility of basic demographics and comorbidities in segmenting Medicare beneficiaries.
• To identify clinically meaningful subgroups with distinct healthcare spending profiles.
• To inform targeted cost-reduction strategies for providers.

Main Methods:

• Utilized Medicare fee-for-service claims data from 2011 and 2012.
• Grouped beneficiaries into six cohorts based on demographics and comorbidities: under-65 disabled/ESRD, frail elderly, major complex chronic, minor complex chronic, simple chronic, and relatively healthy.
• Identified the highest 10% of spenders as "high-cost" beneficiaries.

Main Results:

• Identified 611,245 high-cost beneficiaries.
• Frail elderly (46.2%) and under-65 disabled/ESRD (14.3%) cohorts had the highest proportion of high-cost patients.
• High-cost frail elderly spending was driven by inpatient and post-acute care; under-65 disabled/ESRD spending was higher in Part D costs.

Conclusions:

• Basic demographic and comorbidity data effectively segment Medicare beneficiaries into clinically relevant subgroups.
• These subgroups exhibit significantly different healthcare spending patterns.
• Interventions focused on frail or disabled Medicare beneficiaries show high potential value for cost reduction initiatives.