Health Services Research & Development

Veterans Crisis Line Badge
Go to the ORD website
Go to the QUERI website
HSRD Conference Logo



2017 HSR&D/QUERI National Conference Abstract

Printable View

4092 — Differences in Receipt of Hospice for Veterans Enrolled in Medicare Managed Care versus Fee-for-Service Medicare

Lead/Presenter: Winifred Scott, Resource Center - HERC
All Authors: Gidwani RA (Health Economics Resource Center, VA Palo Alto Health Care System, Geriatrics & Extended Care Data Analysis Center, Stanford University) Scott WJ (Health Economics Resource Center, VA Palo Alto Health Care System, Geriatrics & Extended Care Data Analysis Center) Kinosian B (Geriatrics & Extended Care Data Analysis Center, Philadelphia VA Medical Center) Phibbs CS (Health Economics Resource Center, VA Palo Alto Health Care System, Stanford University) Intrator OK (Geriatrics & Extended Care Data Analysis Center, Canandaigua VA Medical Center, University of Rochester)

Objectives:
We study hospice utilization by older veterans to evaluate whether there is a difference in the likelihood of receiving hospice through Medicare managed care (Medicare Advantage, or MA) versus Medicare fee-for-service (Traditional Medicare, or TM) mechanisms. MA and TM have opposite financial incentives associated with care delivery, especially as it relates to hospice. Leveraging the availability of comorbidity data in VA datasets, we conducted analyses while risk-adjusting for differences in disease status across groups. Previous MA versus TM comparisons have not been able to risk-adjust.

Methods:
We conducted a retrospective analysis of VA- and Medicare-provided hospice care in the last six months of life using a cohort of male veteran decedents, age 66 or greater at death, enrolled in Medicare, and dying between fiscal years 2008 - 2013. Using a VA facility random-effects model and controlling for patient demographics and co-morbidities, we evaluated the effect of Medicare insurance type on the likelihood of hospice receipt and duration.

Results:
There were 1,515,441 veterans in our cohort. Seventy eight (78%) percent of decedents were enrolled in TM only; 19% were enrolled in MA only. There was a 25% increase in hospice receipt over time, climbing from 41% in 2008 to 50% of decedents in 2013. MA enrollees were significantly more likely to receive hospice than TM enrollees (aOR = 1.47, p < 0.001). However, MA enrollees had shorter stays in hospice (aIRR = 0.83, p < 0.001). There were significant disparities in hospice receipt related to race, marital status, and age. Minority veterans, unmarried veterans, and younger veterans were less likely to receive hospice. Results were robust to three sensitivity analyses.

Implications:
MA enrollees are significantly more likely to receive hospice than TM enrollees, but also have shorter hospice stays. Hospice utilization is increasing but there are continuing disparities in care.

Impacts:
Reasons for later enrollment of MA Veterans than TM Veterans into hospice and reasons for less frequent hospice use in TM than in MA plans should be explored to ensure Veterans are not receiving sub-optimal care. Risk adjustment for comorbidities is necessary to make unbiased comparisons about differences in care access among MA and TM beneficiaries.