U.S. Female Veterans Who Do and Do Not Rely on VA Health Care: Needs and Barriers to Mental Health Treatment

Women constitute the fastest growing segment of the U.S. veteran population (1), and their health care needs have become a top priority for the U.S. Department of Veterans Affairs (VA) (2,3). In light of national concern about the needs of female veterans (3–5), the VA has focused on providing outreach and on developing specialized services for women (6–8). Recent national one-year estimates report that 20%−24% of female veterans use VA services (9,10). Among female veterans, those who do not use VA health care, compared with those who do, have been found to be older, more educated, more likely to be unmarried, have lower rates of insurance coverage, poorer general health and mental health, and greater incidence of military sexual trauma (11,12). Reasons cited for VA use include affordability, availability of women’s health clinics, quality of care, and convenience (13).

Further, studies have reported that compared with male VA service users, female VA users tend to be younger and more likely to be African American, unmarried, to have a VA service-connected disability, and to have musculoskeletal and skin disorders, depression, or adjustment disorders (14–18). Female VA patients, compared with male VA patients, have also been found to use outpatient VA general medical and mental health services more heavily and non-VA care more frequently (14–18).

Engaging female veterans in VA care has been a challenge, and data from the National Survey of Women Veterans found that predictors of attrition from VA care were better overall health, higher income, having other forms of health insurance, and less positive perceptions of quality of care and gender-specific features in the VA (19). However, few studies have examined gender-specific barriers to care and differences between those who use VA as their primary source of health care and those who do not (20). There is little extant research using nationally representative samples of female veterans who do not rely on VA care (8,21). This knowledge is critical to understanding the unique needs of female veterans served within and outside the VA at a population level, which is necessary for the design of large-scale, gender-specific outreach initiatives and for ensuring the effectiveness of service care delivery.

Although the VA operates the largest integrated health care network in the United States, a majority of veterans are not enrolled in VA care (22,23). The VA has been considered a health care “safety net” for veterans, serving a large number of patients who are uninsured, on low income, or otherwise vulnerable (22,24). Because the VA serves a predominantly male population, it is unclear to what extent this patient profile is also applicable to female veterans and what special needs may characterize this population.

Numerous studies have highlighted stigma and negative beliefs about mental health care among predominantly male samples of veterans (25,26), but these issues are not well understood among female veterans. Studies of male veterans have found that they often have been influenced by a military culture of masculinity and self-reliance that discourages help seeking for mental health problems (27,28). It is not clear to what extent female veterans perceive similar barriers to mental health care, particularly in the VA context where care is sought in an environment that serves mostly men.

Implementation of the Patient Protection and Affordable Care Act requires all veterans to obtain health insurance as part of the individual mandate, and enrollment in VA health care satisfies that requirement (23). Thus the role of VA health care for female veterans and potential barriers to mental health care in this population need to be better understood for policy, planning, and health services development.

Despite some progress in understanding the demographic profile, unique health care needs, and barriers to seeking mental health care among female users of VA health care (11–18), there has been only one known national study comparing utilization among female veterans who do and do not use VA health care. The study controlled for demographic and clinical characteristics but was conducted over a decade ago (11). Further, no known nationally representative studies have examined gender differences with regard to barriers to seeking care among veterans of all war eras.

In this study, we analyzed data from a nationally representative sample of U.S. veterans to evaluate differences between female veterans who use the VA as their main source of health care, female veterans who do not use VA as their main source of health care, and male veterans who use VA as their main source of health care. To keep our research objectives focused on female veterans, we chose to exclude male veterans who primarily use other health care. We had three objectives: examine the proportion of female veterans compared with male veterans who report that the VA is their main source of health care; identify differences in sociodemographic, military, and clinical characteristics and mental health service use between women using and not using VA care and men using VA care; and compare perceived barriers to mental health care between these groups. On the basis of previous work (9,10), we hypothesized that a similar proportion of female and male veterans would report the VA as their main source of health care. We also hypothesized that women using VA care would have worse general and mental health than women using other care and that female veterans would perceive less stigma about mental health care compared with their male counterparts.

Methods

Sample

Data were based on two combined nationally representative surveys of 3,157 and 1,484 (total N=4,641; mean±SD age=60.3±15.1) U.S. veterans. The surveys were conducted October–December 2011 and September–October 2013, respectively, as part of the National Health and Resilience in Veterans Study (NHRVS) (29,30). In this study, we examined data from 4,631 NHRVS participants who responded to a question asking whether VA was their main source of health care, and we conducted further analyses for the 1,202 (26%) veterans who responded, including women who used or did not use VA care and men who used VA care. Participants completed a 60-minute confidential Web-based survey. NHRVS samples were drawn from a research panel of more than 80,000 households maintained by GfK Knowledge Networks, Inc., a survey research firm that uses KnowledgePanel, a probability-based, online survey panel of a nationally representative sample of U.S. adults that covers approximately 98% of U.S. households. Veterans were identified through an initial screening question that asked about ever serving in the U.S. military, and only veterans were asked to complete the Web survey. Details regarding the KnowledgePanel sampling methodology may be found elsewhere (31). Households were provided with access to the Internet and computer hardware if needed. All participants provided informed consent, and the study was approved by the local institutional review board. To permit generalizability of study results to the entire population of U.S. veterans, poststratification weights were applied according to demographic distributions (specifically, age, gender, race-ethnicity, education, Census region, and metropolitan area) from the most contemporaneous Current Population Survey (32).

Assessments

To assess main source of health care, veterans were asked, “Is the VA your main source of health care?” Response options were yes or no. This questioning is similar to that used in other studies, which have asked about a “usual source of care” (33).

Several validated clinical measures were used to assess clinical characteristics. The Short Form–8 Health Survey (SF-8 [34]) is a widely used measure of health-related functioning that generates standardized physical and mental component summary scores (range 0–100). A score of 50 represents the average level of functioning in the general population, with each 10-point interval representing one standard deviation. Higher scores reflect better functioning.

Medical status was assessed with a list of 19 general medical conditions (including arthritis, cancer, diabetes, and heart disease), which were summed for a total score. Disability status was assessed with a list of five activities of daily living (ADLs) and a list of six instrumental activities of daily living (IADLs); endorsement of needing help in performing any of the ADLs or IADLs was coded as having a disability in the respective scale (35).

Lifetime major depression, alcohol use disorder, and drug use disorder were assessed with modules from the Mini-International Neuropsychiatric Interview (36). Lifetime posttraumatic stress disorder (PTSD) was assessed with the Posttraumatic Stress Disorder Checklist–Specific Stressor version (37). A positive screen for PTSD was operationalized as at least “moderate” symptoms on the required number of symptoms according to DSM-IV and DSM-5 criteria (38,39). (The two national surveys were based on different editions of DSM.)

Mental health service use was assessed with a question about lifetime service use (“Have you ever received mental health treatment [e.g., prescription medication or psychotherapy for a psychiatric or emotional problem?”]) and two questions about current service use (“Are you currently taking prescription medication for a psychiatric or emotional problem?” and “Are you currently receiving psychotherapy or counseling for a psychiatric or emotional problem?”), all of which had yes-no response options.

Perceived barriers to mental health care were assessed with the 11-item Perceived Stigma and Barriers to Care for Psychological Problems scale (40). Two additional items were added to include negative beliefs about mental health care, which is a known barrier to care (“I do not trust mental health professionals” and “Mental health care does not work”) (41). Response options to these items were yes and no, and item responses were summed for a total score (range 0–13), with higher scores reflecting greater stigma and barriers.

Data Analysis

First, descriptive statistics were used to examine the weighted prevalence of female and male veterans who reported VA as their main health care source, and a chi square test compared VA use by gender. Second, bivariate analyses were conducted to compare the two groups of women with each other and with males using VA care on the basis of sociodemographic, military, and clinical characteristics and mental health service use. Multivariable logistic regressions were then conducted on clinical characteristics, with adjustment for sociodemographic and military characteristics that differed between groups, and on mental health service use, with adjustment for sociodemographic, military, and clinical characteristics that differed between groups. Third, weighted prevalence of perceived barriers to care were calculated, and multivariable logistic regressions were conducted to compare women using and not using VA care and men using VA care on perceived barriers to mental health care, with controls for differences in sociodemographic, military, and clinical characteristics and for mental health service use. Adjusted odds ratios were calculated along with 99% confidence intervals. The significance level for all analyses was set at .01, and analyses were performed with SPSS version 20.0 (42).

Data from the two NHRVS surveys were combined for all analyses because there was no significant difference between the two groups of women in these data sets, with the exception of lifetime depression and two items on the Perceived Stigma and Barriers to Care for Psychological Problems scale. Further analysis revealed that these differences had no impact on the study findings.

Results

Among all veterans (N=4,631), 829 (weighted 19.9%) reported that the VA was their main source of health care. Among male veterans (N=4,153; weighted 90.3% of full sample), 724 (19.6%) reported that the VA was their main source of health care. Among female veterans (N=478; weighted 9.7% of full sample), 105 (weighted 23.4%) reported that the VA was their main source of health care. There was no gender difference in the prevalence of veterans who reported the VA as their main source of health care (χ²=3.69, df=1, p=.06).

Table 1 shows differences between the three groups (total N=1,202) on sociodemographic and military characteristics. Compared with women not using VA care, women using VA care were more likely to be nonwhite, to have low income, to not be employed, and to be combat veterans, and they were more likely to have served in Iraq or Afghanistan and to have a deployment-related injury (assessed among combat veterans only). Table 2 shows the clinical characteristics and health service use of these groups. With analyses controlling for differences in sociodemographic and military characteristics, female VA patients were more likely than the other women to have any ADL and IADL disability, to screen positive for lifetime PTSD, and to have lower SF-8 mental component summary scores. After controlling for clinical differences, we found no significant differences in mental health service use between the two groups of women.

Compared with men using VA care, women using VA care were younger; more likely to be nonwhite, single, and from the South; and less likely to be combat veterans. Controlling for these differences, we found that among veterans using VA care, women were more likely than men to screen positive for lifetime depression and PTSD but were less likely to screen positive for lifetime alcohol use or drug use disorder. There was no gender difference in mental health service use.

As shown in Table 3, the most frequently endorsed barriers to mental health care among women were “Getting treatment costs too much money” and “I would be seen as weak.” Among the men, the most frequently endorsed barriers were “Getting treatment costs too much money,” “I do not trust mental health professionals,” and “I would be seen as weak.” Controlling for differences in sociodemographic, military, clinical characteristics, and mental health service use, we found no significant difference on the total barriers score between the two groups of women and between the men and women using VA care. There was also no significant group difference in the frequency of any individual barrier, with the exception that women using VA care were less likely than the men to endorse “Mental health care does not work.”

Discussion and Conclusions

Using data from a nationally representative sample of female U.S. veterans, we found that 23.4% identified the VA as their main source of health care, and this prevalence did not significantly differ from the 19.6% of male veterans who identified the VA as their main source of health care. According to the 2009 National Survey of Women Veterans, 9%−35% of female veterans reported past-year VA health care use, but the survey did not make any comparisons with men and did not assess whether the VA was used as a primary source of health care (20). A more recent report found that 24% of female veterans and 28% of male veterans reported past-year VA health care use, but only 5% and 4%, respectively, reported using VA health care exclusively (9), which is slightly different from our assessment of the VA as a primary source of health care. These differences in assessment of VA service use are important for future research to explore because veterans may use different types of health care for different general medical and mental health conditions. As the VA continues its targeted outreach efforts for female veterans (6–8), it is reassuring that we found that a comparable proportion of female and male U.S. veterans relied on VA services. However, it is worth noting that a higher proportion of female veterans than male veterans have reported poverty and no health insurance coverage (9). Given these circumstances, similar levels of reliance on VA care by male and female veterans could still reflect underutilization of VA services by women in need of care.

The VA may serve as a health care safety net (11,24) for many female veterans. We found that women using VA care were more likely than women using other care to be unemployed, on low incomes, and from racial-ethnic minority groups. These sociodemographic differences are consistent with previous national comparisons (17). Providing further evidence of the VA as a safety net, we also found that women using VA care were more likely to have served in combat and to have had greater general medical and mental health needs, as indicated by higher rates of disability in ADLs and IADLs, higher rates of PTSD, and reduced mental health–related functioning.

Veterans perceived similar barriers to mental health care, regardless of gender or VA use. However, a caveat of this finding is that our assessment of perceived barriers was not comprehensive and may not have been sensitive to gender differences (for example, child care was not assessed). We found no significant differences in the frequency or total number of perceived barriers to mental health care between women using or not using VA care and between men and women using VA care. The most commonly perceived barrier to mental health care endorsed by about a quarter of veterans was “Treatment costs too much money,” which was also a concern reported in the National Survey of Women Veterans (20). This finding may be important for community providers to consider because most female veterans do not rely on VA services. Also, a considerable number of veterans, including women, are uninsured (23), indicating that low-cost mental health services are needed. The Affordable Care Act may represent a way to reduce cost barriers to care; perhaps new forms of coverage can be offered for the mental health needs of Americans (43), including veterans enrolled in VA health care (23). Regardless, providing affordable mental health care remains a problem, including in the veteran population.

Cost concerns were surprising among veterans who relied on VA health care because the VA provides low-cost mental health services; concerns may be rooted either in copays for psychotropic medications, which are generally lower than for other forms of health coverage, or in misperceptions by veterans that VA mental health services are costly. Further study is needed to elucidate this finding, and outreach efforts that incorporate education related to VA service costs may be warranted.

The second most commonly endorsed barrier to mental health care was “I would be seen as weak,” which suggests that female veterans are as likely as male veterans to have fears about how others may view them if they were to use mental health services. This finding further suggests that female veterans experience similar stigma-related barriers to mental health care, which have mostly been examined among male veterans thus far (25,26). Continued efforts are required in both the development and the implementation of stigma reduction strategies related to female veterans seeking treatment for mental health problems either through the VA or in community mental health systems. To this end, enlisting the support of veteran service organizations with large female memberships may be a fruitful avenue to pursue.

This study had several limitations of note. In the NHRVS, use of VA health care was assessed with only one broad question inquiring about “main source” of health care. Assessment of perceived barriers to mental health services did not specifically examine potential barriers in VA settings, including recent concerns about waiting lists (44). Additional studies are needed to confirm these findings, but results of this study underscore the importance of addressing barriers to mental health care for female veterans and support the continued development of specialized VA services that cater to their unique needs. As the female veteran population grows and the Affordable Care Act is implemented, the VA must be prepared to act as a safety net to care for all of those who have served.