Volume 53 | Number S3 | December 2018

Abstract List

Paul L. Hebert Ph.D., Adam S. Batten B.A., Eric Gunnink M.S., Ashok Reddy M.D., M.Sc., Edwin S. Wong Ph.D., Stephan D. Fihn, Chuan‐Fen Liu M.P.H., Ph.D.


To estimate the effect of Medicare use on the receipt of outpatient services from 2001 through 2015 for a cohort of Veterans Administration () users who became age‐eligible for Medicare in 1998–2000.

Data Sources/Study Setting

administrative data linked with Medicare claims for veterans who participated in the 1999 Large Health Survey of Enrolled Veterans.

Study Design

We coded each veteran as ‐reliant or Medicare‐reliant based on the number of visits in each system and compared the health and social risk factors between ‐reliant and Medicare‐reliant veterans. We used bivariate probit and instrumental variables models to estimate the association between a veteran's reliance on Medicare and the receipt of outpatient procedures in Medicare and the .

Principal Findings

Veterans who chose to rely on the ( = 4,317) had substantially worse social and health risk factors than Medicare‐reliant veterans ( = 2,567). Medicare reliance was associated with greater use of outpatient services for 24 of the 28 types of services considered. Instrumental variable estimates found significant effects of Medicare reliance on receipt of advanced imaging and cardiovascular testing.


Expanded access to fee‐for‐service care in the community may be expensive, while the will likely continue to care for the most vulnerable veterans.