==== Front
Innov Aging
Innov Aging
innovateage
Innovation in Aging
2399-5300 Oxford University Press US 

10.1093/geroni/igaa057.3350
igaa057.3350
Abstracts
Session 10280 (Late Breaking Poster)
Health and Social Services
AcademicSubjects/SOC02600
Using Medicare Data to Inform Intervention and Care Delivery for the Most Expensive Patients
Amodeo Sam 1 Kowalkowski Henrik 2 Brantley Halley 3 Bangerter Lauren 3 Jones Nicholas 4 Cook David 5 1 
UnitedHealth Group, Eagan, Minnesota, United States
2 
UnitedHealth Group, Minneapolis, Minnesota, United States
3 
UnitedHealth Group, Minnetonka, Minnesota, United States
4 
Brown University School of Public Health, Providence, Rhode Island, United States
5 
United Health Group, Edina, Minnesota, United States
2020 
16 12 2020 
16 12 2020 
4 Suppl 1 Program Abstracts from The GSA 2020 Annual Scientific Meeting “Turning 75: Why Age Matters”911 911
© The Author(s) 2020. Published by Oxford University Press on behalf of The Gerontological Society of America.2020This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted reuse, distribution, and reproduction in any medium, provided the original work is properly cited.Abstract
Older adults with high medical spend require tailored interventions and care delivery models to meet their complex needs. Segmenting high-spend patients is a promising approach to designing such interventions. In this study we explored patient spend across 4 years (2016-2019) using claims from 799,205 patients continuously enrolled in UnitedHealth Group Medicare Advantage (mean age=73.7; S.E.=0.01). Patients with healthcare spend in the top decile were segmented into three subgroups: catastrophic, persistent, and semi-persistent. Catastrophic patients had more acute events (acute myocardial infarction and hip/pelvic fracture) driving their cost. Persistent patients were younger (mean age=67.8; S.E.=0.06) and had significantly more medications. Semi-persistent patients were older (mean age=76.6; S.E.=0.04) and had significantly more chronic conditions and frailty, indicating their cost was driven by medical complexity. These subgroups displayed different temporal stability in their healthcare costs over time. Each year, 79-81% of the catastrophic group dropped out of the top decile. In contrast, nearly 72% of the persistent group remained in the top decile whereas only 37% of the semi-persistent group remained year over year. As the global population continues to age, it will be necessary to design interventions and care delivery models that address the complex needs of older adults in the high-spend patient population. Our study suggests that segmenting high-spend patients into potentially actionable subgroups is an important first step in achieving these goals.