RESEARCH FINDINGS #50:
Understanding Veterans' Healthcare Use and Experience, 2018–2019

Abstract

This report examines the prevalence of selected medical conditions, use of healthcare both inside and outside of the healthcare system of the Department of Veterans Affairs (VA), and experience with VA and non-VA healthcare providers among Veterans in the U.S. civilian non-institutionalized population. We use the Veteran Self-Administered Questionnaire (VSAQ) fielded in 2018 and 2019 as part of the Medical Expenditure Panel Survey (MEPS). In the first section of this report, we present sociodemographic characteristics, medical conditions, service use, and factors that affect primary care provider choice among all Veterans. In addition to estimates for all Veterans, we also present separate estimates for Veterans with and without VA coverage in the year they completed the VSAQ to identify differences and similarities in health status and reliance on the VA between Veterans with and without VA coverage. Reliance on the VA is defined as use of any healthcare services provided by the VA in the past 12 months. In the second section of this report, we present results on patient experience with VA and non-VA providers and care coordination between VA and non-VA providers.

Introduction

Examining Veterans' use of healthcare and patient experience both inside and outside of the healthcare system of the Department of Veterans Affairs (VA) provides helpful information for predicting Veterans' future demand for VA healthcare as the demographics and needs of the Veteran population change. Predicting such demand is important for ensuring that adequate resources and capabilities will be available to meet the unique and varying healthcare needs of Veterans. This report uses pooled data from the Veteran Self-Administered Questionnaire (VSAQ) fielded in 2018 and 2019 as part of the Medical Expenditure Panel Survey (MEPS). Since the VSAQ contains information on healthcare use and patient experience both inside and outside of the VA health system, it is a unique data source for analyzing Veterans' healthcare needs.

Veterans' eligibility for VA healthcare use is based on many factors, including active duty military service, type of military discharge, service-connected disabilities, medical conditions incurred while in the service, locations and dates of military service, and deployments. Veterans' reliance on VA healthcare versus outside care is also based on multiple factors, such as whether Veterans are eligible to receive free VA care, whether they have private or public health insurance coverage, the generosity of their insurance coverage relative to the cost of receiving VA care, their proximity to a VA healthcare facility, wait times at the nearest VA healthcare facility, and perceived quality of care at VA and non-VA healthcare facilities.

In the first section of this report, we present descriptive statistics on sociodemographic characteristics, medical conditions, service use, and factors that affect primary care provider (PCP) choice among Veterans. In addition to estimates for all Veterans, we present separate estimates for Veterans with and without VA coverage to identify differences and similarities in sociodemographics, health status, and reliance on VA healthcare. Since Veteran women are younger than Veteran men and differ from them in terms of healthcare use and medical conditions, we also present sociodemographic characteristics separately for men and women Veterans. In addition, to assess unique characteristics of Veterans, Veteran men are compared with non-Veteran men in terms of sociodemographics and the prevalence of priority medical conditions collected in the non-VSAQ sections of the MEPS questionnaire. The rest of the estimates in this report, such as service use, factors that affect primary care choice, and patient experience, are not broken down by gender or Veteran status.

In the second section of this report, we present results on patient experience with VA and non-VA providers and care coordination among VA and non-VA providers. The patient experience measures were collected for the VA PCPs or Patient Aligned Care Teams (PACTs), VA specialists, and non-VA providers who Veterans saw most often. A PACT includes a PCP, nurse care manager, clinical associate, and administrative clerk.

These estimates represent average annual estimates for the 2018–2019 period. Only differences that are statistically significant at the 0.10 level or better are described in the text unless otherwise noted.

Findings

Sociodemographic characteristics of Veterans by VA coverage (table 1a)

The average annual population of non-institutionalized Veterans was 16.4 million during the 2-year period 2018–2019 (table 1a), as estimated by MEPS¹. Among all Veterans, 62.4 percent (10.2 million) did not have VA coverage and 37.6 percent (6.2 million) had VA coverage (table 1a). There was no significant difference between Veterans with and without VA coverage in terms of sex and region. Veterans with VA coverage were significantly more likely to be in the youngest age group; 10.5 percent of Veterans with VA coverage were 18-34 years old, compared with 6.9 percent of Veterans without VA coverage. The majority of Non-Hispanic White Veterans did not have VA coverage; 35.5 percent had VA coverage and 64.5 percent did not have VA coverage. In contrast, Non-Hispanic Black Veterans were more likely than not to have VA coverage; 52.1 percent had VA coverage and 47.9 percent did not have VA coverage (results not shown). Those with VA coverage were less likely to have private health insurance and more likely to have public health insurance than Veterans without VA coverage. Those with higher income, with higher education, living in the South Atlantic region, and living in metropolitan areas were less likely to have VA coverage. Veterans with worse physical and mental health were more likely to have VA coverage. The differences in sociodemographic characteristics were similar for male Veterans with VA coverage and male Veterans without VA coverage.

Sociodemographic characteristics of males by Veteran status (table 1b)

Veteran men account for 8.6 percent of the male population age 18 and above (table 1b). Veteran men are older than non-Veteran men; 55.2 percent of Veteran men are 65 years or older, compared with 16.8 percent of non-Veteran men. Compared with the general male population, Veteran men are more likely to be non-Hispanic White and less likely to be non-Hispanic Asian and Hispanic. Among the elderly, Veteran men are more likely to have Medicare and private coverage than non-Veteran men. Veteran men are more likely to have some college education and less likely to have incomes below the Federal poverty level. Veteran men are less likely to live in the Northeast and in metropolitan areas and more likely to live in the South than the civilian male population. Veteran men are also more likely to report fair/poor physical and mental health than non-Veteran men.

Sociodemographic characteristics of all Veterans by gender (table 1c)

Veteran women account for 8.9 percent of the Veteran population (table 1c). They are younger than Veteran men; only 26.5 percent of women Veterans are aged 65 or older, compared with 53.6 percent of Veteran men. Compared with Veteran men, Veteran women are less likely to be non-Hispanic White and more likely to be non-Hispanic Black. Veteran women are more likely to have private health insurance, have some college education, live in the South Atlantic region, and have better self-reported physical health than Veteran men. There was no significant difference between male and female Veterans in terms of poverty level, region, urbanicity, and self-reported mental health.

Prevalence of selected medical conditions common among Veterans by VA coverage (table 2a)

Fifteen conditions were selected based on the recommendations of an expert panel, including conditions that have a higher prevalence among Veterans than in the non-Veteran population. For each condition, the VSAQ asked if the Veterans had ever been told by a doctor or another healthcare provider that they have had the condition. The 15 conditions include four general conditions, seven musculoskeletal conditions, and four mental health conditions. One musculoskeletal and three mental health conditions were removed from the analysis because their low prevalence made estimation problematic. "Any mental health diagnosis" includes schizophrenia, bipolar disorder, depression, and other mood disorder. Eleven conditions—COPD, hearing loss, back pain, gout, neck pain, osteoarthritis, alcohol abuse, any mental health condition, depression, other mood disorders, and post-traumatic stress disorder (PTSD)—were more prevalent among Veterans with VA coverage than Veterans without VA coverage (table 2a).

Prevalence of selected medical condition common among males by Veteran status (table 2b)

Due to potential differences in the prevalence of health conditions by gender, in this section we focus on the prevalence of selected medical conditions among Veteran and non-Veteran men. The sample size was too small for an accurate comparison of Veteran and non-Veteran women. This section presents priority conditions collected in the non-VSAQ sections of the MEPS questionnaire. The conditions are selected because of their high prevalence and/or cost. For each condition, MEPS respondents were asked if they had ever been told by a doctor or another heath care provider that they have had the condition. The prevalence of the following conditions were higher among Veteran men than non-Veteran men: coronary heart disease, angina, heart attack, other heart disease, any heart disease, stroke, arthritis, cancer, chronic bronchitis, diabetes, emphysema, high blood pressure, high cholesterol, and joint pain. Similarly, the prevalence of the following limitations were higher among Veteran men than non-Veteran men: limitations with instrumental activities of daily living (IADL); limitations with activities of daily living (ADL); cognitive limitations; limitations with work, housework, or school; any limitations; serious difficulty hearing; and serious difficulty seeing with glasses (table 2b). All of these conditions, with the exception of limitations with IADL and ADL, were statistically more prevalent in non-elderly Veteran men than non-elderly non-Veteran men. In elderly men, the following conditions were more prevalent in Veterans than non-Veterans: coronary heart disease, angina, heart attack, other heart disease, any heart disease; stroke; cancer; diabetes; emphysema; limitations with IADL; limitations with work, housework, or school; any imitations; and serious difficulty hearing (results not shown).

Number and percentage of Veterans with only VA, only non-VA, and both VA and non-VA healthcare use by VA coverage (table 3)

Among all Veterans, 83.0 percent had at least one healthcare visit, 10.2 percent had only VA care visits, 55.0 percent had only non-VA care visits, and 17.8 percent had both VA and non-VA care visits in the past 12 months (table 3). Probability of any use was higher among elderly Veterans (87.7%) than non-elderly Veterans (77.6%) (results not shown). Veterans with VA coverage were more likely to have any healthcare visits than those without VA coverage (88.6 percent and 79.6 percent, respectively). Among those with VA coverage, 24.7 percent had VA care only within the calendar year, 26.1 percent had non-VA care only, and 37.8 percent had both VA and non-VA care. Among those without VA coverage, 1.5 percent had VA care only within the calendar year, 72.4 percent had non-VA care only, and 5.8 percent had both VA and non-VA care. VA coverage was reported for the calendar year 2018 or 2019, whereas the VSAQ collected data on care in the "past 12 months," which may account for VA use among Veterans without VA coverage.

Use of selected health services common among Veterans (table 4)

The VSAQ collected data on the use of seven selected health services and whether they were received from the VA or outside of the VA since discharge from military service. These services are caregiver support, group counseling for mental healthcare, individual mental healthcare, assistive mobility device, prosthesis, rehabilitation services, and prescription medications. For all seven types of services, Veterans with VA coverage were more likely to have any care and more likely to have VA services only. Among Veterans with VA coverage, 20.8 percent had any individual mental healthcare and 17.1 percent had VA care only (table 4). Among Veterans without VA coverage, 8.5 percent had any individual mental healthcare and 5.7 percent had non-VA care only. Among Veterans with VA coverage, 69.2 percent had any prescription medications and 51.9 percent had VA prescription medications only. Among Veterans without VA coverage, 46.2 percent had any prescription medications and 39.3 percent had non-VA prescription medications only. For these seven services, most Veterans either used VA care only or non-VA care only. Among Veterans with VA coverage, the majority used only VA care for these services. Prescription medications had the highest rate of both VA and non-VA care use among Veterans. Among Veterans with VA coverage, 6.6 percent obtained prescription medications both from the VA and from non-VA sources.

Importance of factors for choosing a primary care provider (table 5)

As shown in table 5, among Veterans with VA coverage, the top three major factors in choosing a PCP were the reputation of the personal doctor (ranked as major by 44.9 percent), whether the doctor understands the special needs of Veterans (41.1 percent), and whether the doctor is in their health plan's network (39.6 percent). Among Veterans without VA coverage, the top three major factors were whether the doctor is in their health plan's network (52.3 percent), the reputation of the personal doctor (50.8 percent), and the location of the doctor's practice (39.6 percent). Whether the doctor understands the special needs of Veterans was significantly more important for Veterans with VA coverage than those without VA coverage; 41.1 percent and 18.8 percent ranked it as a major factor, respectively.

Patient experience with VA primary care provider or Patient Aligned Care Team (table 6)

Patient experience with providers is presented in tables 6, 7, and 8. The questions concern healthcare received within the past 12 months. Among all Veterans visiting a VA PCP/PACT in the past 12 months, 67.6 percent reported that their VA PCP/PACT provider "always" knew about their past health problems and treatments, 41.1 percent reported that their VA PCP/PACT provider "always" knew about tests or results from other VA providers, and 70.2 percent reported that their VA PCP/PACT provider "always" kept health information complete and up to date (table 6). Among all Veterans visiting a VA PCP/PACT who needed their medical records, 54.9 percent reported that they "always" got their medical records from their VA PCP/PACT provider as soon as they needed them. Among all Veterans visiting a VA PCP/PACT who needed a referral, 44.6 percent reported that they "always" got a referral to see non-VA providers from their VA PCP/PACT provider as soon as they needed it. On all five measures, elderly Veterans were significantly more likely to rate their patient experience higher than non-elderly Veterans (results not shown).

Patient experience with non-VA providers (table 7)

Among all Veterans visiting a non-VA provider, 50.6 percent reported that their non-VA provider "always" knew about their past health problems and treatments and 62.2 percent reported that their non-VA provider "always" kept health information complete and up to date (table 7). Among all Veterans visiting a non-VA provider who needed their medical records, 41.2 percent reported that they "always" got their medical records from their non-VA provider as soon as they needed it. On all three measures, elderly Veterans were significantly more likely to rate their patient experience higher than non-elderly Veterans (results not shown).

Patient experience with VA specialists (table 8)

Among all Veterans with care from a VA specialist, 90.1 percent reported that the VA specialist had enough information about their medical history (table 8). This satisfaction rate was significantly higher among elderly Veterans than non-elderly Veterans (94.1 percent and 86.2 percent, respectively).

Number and percent of Veterans with health care use from multiple types of providers (table 9)

The VSAQ collected data on patients who visited three types of providers, VA PCP/PACT, VA specialist, and non-VA providers. Patient experience questions regarding care coordination between VA and non-VA providers are relevant only to Veterans who received care from VA and non-VA providers. Similarly, obtaining medical records and getting referrals to non-VA providers are only relevant for a subset of Veterans. Therefore, table 9 presents the number and percentage of Veterans with healthcare visits by type of provider, with multiple provider types, who asked for their medical records, and who needed referrals to non-VA providers from a VA PCP/PACT. Among all Veterans, 23.5 percent had at least one visit with a VA PCP, 14.5 percent had at least one visit with a VA specialist, and 72.8 percent had at least one visit with a non-VA provider. Among all Veterans, 14.8 percent had visits with both a VA PCP and a non-VA provider, and 9.8 percent had visits with both a VA specialist and a non-VA provider. Among all Veterans, 20.8 percent had healthcare from at least two types of providers. Among all Veterans, 5.1 percent asked for his/her medical records from a VA PCP/PACT, and 17.1 percent asked for his/her medical records from a non-VA provider. Among all Veterans, 7.6 percent needed referral from a VA PCP/PACT to see a non-VA provider.

Care coordination between VA and non-VA providers (table 10)

Among Veterans with care from both a VA PCP/PACT and a non-VA provider, 63.2 percent said their VA PCP/PACT was aware of their healthcare services outside the VA. Among Veterans with care from both a VA specialist and a non-VA provider, 40.8 percent said their VA specialist was aware of healthcare services they received outside the VA. Among Veterans with care from both a VA provider and a non-VA provider, 64.8 percent said their non-VA provider was aware of healthcare services they received at the VA.

Care coordination problems between VA and non-VA providers (table 11)

Among Veterans who received care from both a VA PCP/PACT and a non-VA provider but whose VA PCP/PACT was not aware of the services from the non-VA provider, 83.0 percent reported that the lack of awareness was not a problem. Among Veterans who received care from both a VA specialist and a non-VA provider but whose VA specialist was not aware of the services from the non-VA provider, 89.6 percent reported that the lack of awareness was not a problem. Among Veterans who received care from both a non-VA provider and a VA provider but whose non-VA provider was not aware of the services received at the VA, 84.2 percent reported that the lack of awareness was not a problem.

Data Source

The estimates in the research findings are based on data from the MEPS 2018 and 2019 Full Year Consolidated Data Files (HC-209, HC-216).

These files are available at https://meps.ahrq.gov/mepsweb/data_stats/download_data_files.jsp . The MSA variable is available through the AHRQ Data Center.

The VSAQ was fielded during Panel 22 Round 5, Panel 23 Round 3, and Panel 24 Round 1. The VSAQ data for Panels 22 and 23 were included in the 2018 FY file with weight VSAQW18F, and the data for Panels 23 and 24 were included in the 2019 FY file with weight VSAQW19F. The data for these three panels were combined to produce more accurate estimates of VSAQ variables. This was done by pooling 2018 FY and 2019 FY files and dividing each year's VSAQ weight by 2 to derive a VSAQ weight for the pooled file, as is done in the usual 2-year pooling of MEPS FY files. However, for variance estimation, the common variance structure was obtained from the pooled linkage public use file HC-036.

Definitions

Veterans
Veterans are defined as people who reported as having been honorably discharged from active duty in the Armed Forces.

Health insurance status
Individuals ages 18-64 were classified in the following three insurance categories based on household responses to health insurance status questions:

• Any private health insurance: Individuals who, at any time during the year, had insurance that provided coverage for hospital and physician care (other than Medicare, Medicaid, or other public hospital/physician coverage) were classified as having private insurance. Coverage by TRICARE (Armed Forces-related coverage) was treated as private health insurance coverage. Insurance that provided coverage for a single service only, such as dental or vision coverage, was not included.
• Public coverage only: Individuals were considered to have public coverage only if they met both of the following criteria: (1) they were not covered by private insurance at any time during the year and (2) they were covered by any of the following public programs at any point during the year: Medicare, Medicaid, or other public hospital/physician coverage.
• Uninsured: The uninsured were defined as people not covered by private hospital/physician insurance, Medicare, TRICARE, Medicaid, or other public hospital/physician programs at any time during the entire year or period of eligibility for the survey.

Individuals ages 65 and over were classified in the following three insurance categories based on household responses to health insurance status questions:

• Medicare only (65+): Individuals who are 65 years old or more and are covered by Medicare only.
• Medicare and private (65+): Individuals who are 65 years old or more and are covered by Medicare and private insurance or TRICARE/CHAMPVA.
• Medicare and other public only (65+): Individuals who are 65 years old or more and are covered by Medicare and Medicaid/SCHIP or another public program.

Race/Ethnicity
Classification by race/ethnicity is based on information reported for each family member. First, respondents were asked if the person's main national origin or ancestry was Puerto Rican; Cuban; Mexican, Mexican-American, or Chicano; other Latin American; or other Spanish. All people whose main national origin or ancestry was reported as being in one of these Hispanic groups, regardless of racial background, were classified as Hispanic. All other people were classified according to their reported race. For this analysis, the following classification by race/ethnicity was used: Hispanic (any race), Black non-Hispanic only, White/other.

About the Medical Expenditure Panel Survey-Household Component

Footnotes

1. By design, the MEPS estimates of Veterans are less than the American Community Survey estimates. The latter survey gets its information on Veterans from a household respondent, who is asked to identify Veterans in the household. In the MEPS, a respondent for each household initially identifies Veterans in the household, then each identified Veteran is asked to self-identify as a Veteran. The second requirement reduces the estimated number of Veterans, all other things being equal, by about 5.5 percent.

Suggested Citation