2005 HSR&D National Meeting Abstract

3017 — VA-Medicare Dual Beneficiaries with Stroke: Where Do They Get Care?

Author List:
Shen Y (East Orange VA Medical Center, Rutgers University)
Findley PA (East Orange VA Medical Center, Rutgers University)
Maney M (East Orange VA Medical Center)
Weerts S (Rutgers University)
Crystal S (Rutgers University)
Findley TW (East Orange VA Medical Center)

Objectives:
To investigate dual system use of veterans with stroke.

Methods:
This is a cross-sectional analysis using FY 1999 VA Diabets Epidemiology Cohort (DEpiC). By merging DepiC with FY 1999 Medicare denominator file, 1998-2000 Medicare claims data and VA inpatient/outpatient files, we identified patients with full year Medicare Fee For Service in FY 1999 admitted to the hospital for an incident stroke in FY 99 but did not have a stroke in FY 1998 using Reker-specific Method (N=6,699). We investigated the source of the first incident stroke hospitalization, whether they were admitted to the VA or to a community-based hospital reimbursed by Medicare. For the two groups, we compared age, race, gender, and marital status. We also compared their subsequent care in one year rolling forward from the day of discharge. We used the chi-square test for category variables and the t-test for continuous variables to test for statistical significance.

Results:
Seventy-six percent of the patients had their initial stroke care under Medicare and 24% had their initial care in the VA. Caucasian, married, female patients, and those older than 74 were more likely to receive initial care in Medicare. Patients admitted to the VA had an average inpatient initial length of stay of 29.8 days compared to 6.2 days for Medicare. Over the rolling year period, patients initiating care in the VA spent 47 days in hospitals compared to 33 days for Medicare. Medicare-first users were more likely to use dual systems compared to VA first users (71% vs. 48%). However, the Medicare- first group was more likely to use the other system for outpatient while VA-first group was more likely to use the other system for inpatient. After three months from the stroke incident, fewer VA first users needed any further care compared to those with initial care in Medicare. All the differences are statistically significant.

Implications:
Most patients received their incident stroke hospital care in Medicare, but a majority sought VA care afterward. The VA provided an intensive inpatient care for stroke patients.

Impacts:
Quality care cannot be ensured without coordinating care across systems.