Talk to the Veterans Crisis Line now
U.S. flag
An official website of the United States government

VA Health Systems Research

Go to the VA ORD website
Go to the QUERI website

IIR 07-233 – HSR Study

IIR 07-233
Access Factors Affecting VA Enrollees' Use of Non-VA Medical Care
Alan N West, PhD
White River Junction VA Medical Center, White River Junction, VT
White River Junction, VT
Funding Period: April 2008 - December 2011
VA enrollees' use of VA versus private sector care depends on multiple factors such as specific medical needs, proximity to services, VA priority status, income, and insurance coverage. While their use of Medicare-funded services has been studied extensively, there are few sources of non-VA utilization data for VA enrollees of all ages and functional levels. To understand how they use hospital care, we acquired administrative discharge data for all hospitalizations, VA or non-VA, of VA enrollees in eight states, regardless of age, residence, or insurance coverage.

To assess: 1) For which major diagnostic or procedure categories do VA enrollees obtain more inpatient care in the VA or the private sector? 2) How do rural residence and distance to care interact with other patient characteristics to affect reliance on VA versus non-VA care? 3) Among patients who have multiple hospitalizations within a few years, how do patients who use both VA and non-VA hospitals (dual users) differ from single-system users?

From state health departments or hospital associations in New York, Pennsylvania, South Carolina, Tennessee, Florida, Louisiana, Iowa, and Arizona, we acquired administrative hospital discharge data for all non-VA admissions that resident VA enrollees underwent in recent years. We also acquired discharge data for any VA hospitalizations that enrollees in these states had. Reported here are analyses of 2004-2007 data for seven states (Tennessee excluded) whose residents could be matched to detect those who had admissions to both VA and non-VA hospitals, permitting comparisons of one-time inpatients, multiply-hospitalized patients who used only VA or non-VA hospitals, and dual users (total hospitalizations: VA=386,948; non-VA=1,406,960). For each of three age groups (18-44, 45-65, 65+), we compared VA and non-VA hospitalizations with respect to major diagnostic and procedure categories, co-morbidities (Elixhauser), gender, VA priority status, urban or rural residence (RUCA categories), and distance to care (ZIP-to-ZIP driving times from Googlemaps). We also compared high-population states (NY, PA, FL) to the less densely populated states.

Among individuals younger than 65 with any hospitalization(s), 39% had one or more VA admission(s) while 74% had one or more non-VA admission(s); for those 65 or older, comparable rates were 16% and 90%. Regardless of age, about 1/4 of VA patients and 2/5 of non-VA patients lived more than an hour from a VAMC; compared to VA patients, non-VA users were less likely to be in VA priority 5 and more likely to be in priorities 6-8, but rates of VA and non-VA hospitalization were similar for priorities 1-4. Among patients younger than 45, 17% of VA inpatients and 30% of non-VA inpatients were women.

By age range, the most common principal diagnostic categories, percent of admissions they represented, and odds of a VA admission for this diagnosis versus others (all p<.0001), were:

18 to 44 years old: Mental: 32%; OR=2.84; Circulatory: 11%; OR=0.63;
Digestive: 8%; OR=1.36;
Pregnancy/Childbirth/Puerperium: 7%; OR=0.01;
Injury/Poisoning: 6%; OR=0.77;
Endocrine/Nutritional/Metabolic/Immunity: 6%; OR=0.28

45 to 64 years old: Circulatory: 27%; OR=0.66; Mental: 17%; OR=2.38; Respiratory: 9%; OR=0.57;
Endocrine/Nutritional/Metabolic/Immunity: 8%; OR=0.35;
Digestive: 7%; OR=1.67; Injury/Poisoning: 6%; OR=1.13 (ns)

65 or older: Circulatory: 39%; OR=0.68; Respiratory: 12%; OR=0.87;
Endocrine/Nutritional/Metabolic/Immunity: 9%; OR=0.45;
Genitourinary: 8%; OR=0.83; Digestive: 5%; OR=2.40; Injury/Poisoning: 5%; OR=1.43

Inpatients were much more likely to receive diagnostic procedures (MRI, CT scan, ultrasound), blood transfusions, psychiatric and substance abuse treatment, and physical therapy, but much less likely to undergo cardiovascular or other surgical procedures, in VAMCs than non-VA hospitals. Nearly all obstetric procedures (almost 6,000 admissions) were in non-VA hospitals.

VA admission rates were similar for enrollees living in cities versus large, small, or isolated rural towns, whether their states were more or less heavily populated. But non-VA admission rates were much higher for enrollees, urban or rural, living in the higher population states than those in the "rural" states. Average distances to VA and non-VA hospitals were greater in rural states, and were associated with fewer psychiatric but not medical or surgical hospitalizations. In higher population states, VA and non-VA mental health admission rates were equivalent, but in the rural states enrollees relied much more on VA mental health care. Overall, patients were less likely to have multiple admissions if they lived farther from hospitals or in more rural settings.

In any age range, reliance on VA hospitals was greater for multiply-hospitalized patients than single-admission patients (p<.0001 for all tests): Among multiply-hospitalized patients younger than 65, 21% used only VAMCs, 28% were dual users, and 51% used only non-VA hospitals; for those 65 or older, comparable rates were 8%, 11%, and 81%. Single-system users averaged 3+ hospitalizations in four years, while dual users averaged 5, equally apportioned between VA and non-VA. Mean co-morbidities were lower for those who used only non-VA hospitals than those who used VAMCs at all. Dual users younger than 65 had the highest rates of psychiatric hospitalization, in both VA and non-VA hospitals.

VA enrollees rely heavily on non-VA hospitals, regardless of age or priority for VA care. They are more likely to use VAMCs for mental health or digestive system diagnoses, and non-VA hospitals for cardiovascular and other surgical care. They receive considerably more physical therapy and diagnostic procedures in VAMCs. Enrollees in less populous states must travel greater distances for hospital care, and they appear to have less access to non-VA hospitals. Yet greater distance to care seems to reduce the likelihood of multiple admissions but not admission in general, and the rate of psychiatric hospitalizations but not medical or surgical hospitalizations. Patients who are hospitalized multiple times within a few years rely more on VA care than single-admit patients, and dual users have more co-morbidities and higher rates of mental health admissions in both VA and non-VA hospitals. Non-VA admissions for obstetric care represent a substantial portion of hospitalizations for younger enrollees. VA might meet the healthcare needs of certain VA enrollees (women, psychiatric, and complex cases) by coordinating inpatient care options with non-VA hospitals.

External Links for this Project

Dimensions for VA

Dimensions for VA is a web-based tool available to VA staff that enables detailed searches of published research and research projects.

Learn more about Dimensions for VA.

VA staff not currently on the VA network can access Dimensions by registering for an account using their VA email address.
    Search Dimensions for this project


Journal Articles

  1. West AN, Lee PW. Associations between childbirth and women veterans' VA and non-VA Hospitalizations for major diagnostic categories. Military medicine. 2013 Jan 1; 178(11):1250-5. [view]
  2. West AN, Lee RE, Shambaugh-Miller MD, Bair BD, Mueller KJ, Lilly RS, Kaboli PJ, Hawthorne K. Defining "rural" for veterans' health care planning. The Journal of rural health : official journal of the American Rural Health Association and the National Rural Health Care Association. 2010 Jan 1; 26(4):301-9. [view]
  3. West AN, Weeks WB. Health care expenditures for urban and rural veterans in Veterans Health Administration care. Health services research. 2009 Oct 1; 44(5 Pt 1):1718-34. [view]
  4. West AN, Mackenzie TA. Time Trends In Expenditures For Rural Veterans' Healthcare. Journal of Rural Social Sciences. 2011 Oct 1; 26(3):181-200. [view]
HSR&D or QUERI Articles

  1. West AN. Health care access and quality for rural veterans. HSR&D Forum. 2010 Nov 1; 4. [view]
Conference Presentations

  1. West AN. Disparities in healthcare utilization among rural veterans who use VA care. Paper presented at: AcademyHealth Annual Research Meeting; 2008 Jun 9; Washington, DC. [view]
  2. West AN, Lee RE. Effect of distance on urban and rural VA enrollees' use of VA and non-VA hospitalizations in four states. Paper presented at: American Public Health Association Annual Meeting and Exposition; 2010 Nov 6; Denver, CO. [view]
  3. West AN. Medicare Part D increased prescription expenditures among rural residents, except VA patients. Poster session presented at: VA HSR&D National Meeting; 2011 Feb 1; Washington, DC. [view]
  4. West AN. VA enrollees use non-VA hospitals heavily, but less in rural states than urban states. Paper presented at: National Rural Health Association Annual Meeting; 2011 May 1; Austin, TX. [view]
  5. West AN. VA users also use non-VA care heavily, but rural veterans rely increasingly on the VA. Paper presented at: VA HSR&D National Meeting; 2009 Feb 12; Baltimore, MD. [view]
  6. Lee RE, West AN, Lee PW. Varied geographical increases in the rural veteran population. Paper presented at: American Public Health Association Annual Meeting and Exposition; 2010 Nov 6; Denver, CO. [view]

DRA: Aging, Older Veterans' Health and Care, Health Systems
DRE: none
Keywords: Access, Cost, Utilization patterns
MeSH Terms: none

Questions about the HSR website? Email the Web Team

Any health information on this website is strictly for informational purposes and is not intended as medical advice. It should not be used to diagnose or treat any condition.