BACKGROUND/RATIONALE:
This project examined health outcomes for non-elderly, uninsured veterans and non-veterans. In doing so, this project's products filled a void in the research literature, which has largely ignored the unique characteristics and health care options of the uninsured veteran population. The large body of research on the uninsured non-elderly documents poorer outcomes for individuals lacking health insurance. While veterans are not exempt from the barriers to insurance (e.g., lack of employer offer, low income, high premiums) some are able to turn to VA-provided care as an alternative to other sources. In our prior work we have found that veterans who do receive care at the VA are about two times more likely than the general population and about four times more likely than other veterans to have no other source of coverage. Hence, this population is particularly vulnerable to the changes in VA eligibility and benefits. The overarching question that the proposed project addresses is: to what extent does the VA meet the need for high-quality health care for otherwise uninsured veterans?

OBJECTIVE(S):
This 1.5-year study will used multiple years of Medical Expenditure Panel Survey (MEPS) data to meet three objectives:

Objective 1: Investigate the characteristics of uninsured non-elderly veterans.
Objective 2: Develop and estimate multivariate models of insurance status and demand for VA care.
Objective 3: Develop and estimate multivariate models of outcomes as a function of being uninsured and/or using VA care.

METHODS:
This study used existing survey data and other publicly available data to achieve its objectives. We estimated statistical models using multiple years of MEPS data linked to characteristics of state Medicaid programs and other publicly available geographic data. The first objective focused on descriptive quantification of basic characteristics of the population under study: uninsured veterans and non-veterans. The second and third objectives developed multivariate statistical models that, together, related insurance status and VA use to health outcomes, controlling for observable and unobservable baseline characteristics.

FINDINGS/RESULTS:
Findings published to date pertain to all adult U.S. residents, not just veterans. Among individuals in this population, uninsurance rates increased from 2000 to 2005. Rates were particularly high for low-income individuals reporting serious health conditions (25%) or disabilities (15%). If they lived in the South, members of these groups had even higher rates (32% health conditions, 22% disabilities). In addition, such persons who did not belong to a federally mandated Medicaid eligibility category were about twice as likely to be uninsured overall and experienced more rapid increases over time. These regional and categorical differences reflect gaps in current policy that pose a challenge to any incremental health insurance reform.

IMPACT:
This project's products will enhance VA's understanding of the non-elderly veteran population, particularly the sub-population that does not use VA care. In addition, results will be relevant to the broader non-elderly population. As such, results will assist VA administrators plan for the future needs of veterans.

---

External Links for this Project

---

PUBLICATIONS:

Conference Presentations

1. Frakt AB, Pizer SD, Iezzoni L. Holes in the Safety Net: Trends in Uninsurance Rates for Individuals with Functional Impairment or Chronic Illness. Poster session presented at: AcademyHealth Annual Research Meeting; 2009 Jun 29; Chicago, IL. [view]
2. Frakt AB, Pizer SD, Iezzoni L. The Effects of Health Reform on Public and Private Insurance in the Long Run. Paper presented at: American Society of Health Economists Annual Conference; 2010 Jun 1; Ithaca, NY. [view]
3. Frakt AB, Pizer SD, Iezzoni L. The Effects of Health Reform on Public and Private Insurance in the Long Run. Paper presented at: Physicians for a National Health Program Annual Meeting; 2010 Nov 10; Birmingham, AL. [view]

DRA: Health Systems Science
DRE: none
Keywords: Chronic disease (other & unspecified), Outcomes, Utilization patterns
MeSH Terms: none