Women are among the fastest growing population subgroups of Department of Veteran Affairs (VA) health care users. Approximately 6% of current VA users are women veterans and although the number is expected to double in the next 10 years, women are still underrepresented within the health care system.1 Compared with men, women are more likely to be diagnosed with a mental health condition, more likely to use mental health services, and on average have more mental health visits.2 The influx of women to the system has resulted in increased specialty mental health services designated for women and expansion of colocated mental health services in women’s primary care clinics.3 Organization of these designated women’s services varies widely, and limited empirical support or clinical consensus exists regarding the need for such services or how they should be implemented.4 VA policy recognizes gender-sensitive issues, such as choice of provider gender, as important components of mental health care.5 Some data suggest that women veterans’ concerns about comfort in the predominantly male VA environment limit health care utilization,6 but we know little about how such concerns may specifically impact access to mental health care. Gender-specific data are essential to inform the current efforts to expand mental health resources for women veterans and assure their access to and engagement with mental health services.
Patient-centered reconceptualizations of access to VA care are focusing on the set of dimensions that describe the fit between characteristics and preferences of the individual and qualities of available health care services.7 This model extends the quantification of access from observations of service utilization among those with perceived need for care, or unmet need, to examinations of how well these services meet an individual’s needs. An individual’s judgments regarding the extent to which services used meet their needs can therefore be conceptualized as perceptions of access to care. Unmet mental health care need is associated with younger age, being unmarried,8 nonwhite,9,10 Hispanic ethnicity,11 low income,8,11 and not having a usual source of care.12 Among women veterans, rural residence13 is also associated with unmet need. These factors have not yet been examined among mental health service utilizers as predictors of perceived access, or how well utilized services meet patient’s needs.
We propose that indicators of perceived access will have the greatest utility for defining patient-centered care when tailored to the experiences of specific populations, such as women veterans. Negative perceptions about the gender sensitivity of VA health care are associated with unmet need for care among women veterans.14 Minority race and ethnicity, which tend to be associated with unmet need, also predict stronger preferences for designated women’s mental health services among women veterans in VA health care.15 To date, few studies have examined the impact of perceptions of mental health care on mental health utilization.16 Among VA users, the impact of women’s experiences with access to female providers, designated women’s mental health clinics, or same-gender group treatments is largely anecdotal. These gender-sensitive dimensions would provide measureable indicators of perceived access and important points of intervention to promote access to care.
The current study is a national survey of women veteran VA primary care users designed to identify gender-sensitive dimensions of perceived access to mental health care. We used women’s subjective ratings of how well care met their needs as a global indicator of perceived access, that is, the match between patient preferences and mental health treatment experiences. We examine this construct in the context of perceived need for mental health care, and mental health care utilization within and outside VA. We also examine the extent to which veteran characteristics and gender-related experiences of VA mental health care are associated with subjective ratings that past-year mental health care met women’s needs. These care experiences examined included if a female provider was available as often as preferred; whether same-gender group treatments were available as often as preferred; whether gender-specific settings were available as often as preferred; and the frequency that women felt uncomfortable or out of place in VA mental health treatment settings because of gender. Furthermore, we examine overall ease of use as a proxy for logistical barriers to mental health care.17 We propose that factors that are significantly associated with perceptions that care meets women’s needs bear further investigation as key indicators of gender-sensitive access to care among women VA users.
Sample and Recruitment Procedures
The sample is drawn from the Women’s Overall Mental Health Assessment of Needs (WOMAN) Survey, a cross-sectional telephone survey conducted with a population-based stratified random sample of women VA primary care users within the 50 United States and the District of Columbia between October 1, 2010 and September 30, 2011. Sampling was stratified based on age and race/ethnicity and included an over-sample of nonwhite women and women aged 18–44. The sampling frame was identified from VA administrative data using the VA National Patient Care Database.18 Exclusion criteria were having an incomplete or international mailing address, and an active diagnosis of dementia or psychosis.
Study information, including a prepaid opt-out letter and a $2 cash incentive, was mailed to 15,328 women, and 7462 were able to be contacted by telephone to confirm eligibility. Eighty-four percent (n=6287) agreed to participate. Respondents were mailed a $20 check to thank them for their time. Surveys were conducted between June and September 2012 by trained survey interviewers. All study procedures were approved by the Stanford University Institutional Review Board. Participants did not significantly differ from the sampling frame in the likelihood of past-year mental health diagnosis or past-year VA mental health utilization.
The WOMAN Survey consists of 145 items designed to determine the prevalence of a wide range of mental health symptoms and conditions among women veterans in VA care, and to assess their mental health care needs, utilization, and preferences. The survey assessed patient characteristics (eg, demographics, family composition), presence of possible active mental health issues, and mental health services utilization and preferences. For the current study, a variable from the Planning Systems Support Group Enrollee File,19 a file of geographic information on VA patients, was matched to the survey to classify a veteran’s residence as urban, rural, or highly rural.
Perceived need for mental health care was defined according to prior research,11 with an item that asked, in the past year, “did you ever want (or need) help with personal, family, or emotional problems from a doctor or mental health professional, such as a psychologist, psychiatrist, counselor, or therapist?” Women who reported perceived need were queried for past-year self-reported mental health service utilization. Items were adapted from the National Survey on Drug Use and Health20 and included use of inpatient (6 items), outpatient (6 items), and pharmacotherapy (6 items) mental health services from a variety of VA and non-VA sources. Women who responded positively to any of these items were categorized as mental health users. Past-year VA mental health utilization was defined as use of a hospital stay overnight or longer in a psychiatric or residential unit of a VA medical center; receipt of outpatient therapy or counseling for emotional problems in a VA mental health clinic, primary care clinic, or women’s clinic; or receipt of prescription medication to treat a mental or emotional condition (exclusive of medication solely received during an inpatient stay) from a VA mental health clinic, primary care clinic, or women’s clinic. Self-report items for VA mental health care utilization demonstrated 90% agreement with VA administrative data, with a positive predictive value of 97.6%, suggesting good validity for these items.
Perceived access was measured using a global item, “Overall, how well did the mental health care you received in the past year meet your needs?” Responses were on a 5-point Likert scale ranging from “not at all” to “completely met your needs.” Unmet need was indicated by responses from “not at all” to “moderately,” and women who responded “completely” or “ very well” were considered to have had their mental health care needs met.
Mental health care experiences were assessed for VA mental health services. These items were rated in 5-point Likert scales and included ease of use; seeing a female mental health provider as often as wished; participating in women-only group treatments as often as wished; using a women-only clinic or setting as often as wished; and gender-related comfort, or how often a woman felt out of place, uncomfortable or uneasy in her mental health treatment setting because of her gender. Women who reported perceived need but did not use VA mental health care were asked to respond to similar statements describing their reasons for not using care (Table 3). Responses were scored on a 4-point Likert scale ranging from “strongly agree” to “strongly disagree.”
Data were weighted in analysis for design characteristics of the survey and to adjust for nonresponse and poststratification to represent the population of VA female primary care users. Objective indicators of access to mental health care were examined by calculating nested proportions of primary care users who reported past-year perceived need, past-year mental health utilization, and past-year VA mental health care use. Because large proportions of women who reported perceived need used mental health services, and large proportions of those women used VA care, we did not model the odds of objective access indicators (utilization among women with perceived need). Rather, we examined the odds of perceived access to VA care (ie, odds that mental health care needs were met) as a function of veteran characteristics. Veteran characteristics that were significantly associated (P<0.05) with perceived access were retained in models for the odds of perceived access as a function of VA mental health care experiences. Because veteran characteristics were associated with perceived access, we also examined whether gender-related mental health care experiences varied according to the same veteran characteristics by modeling the odds of positive experiences as a function of veteran characteristics. Finally, among the VA primary care users with perceived need but who did not use VA mental health services, we compared women who used non-VA mental health services to women who did not use any care on reasons for not using VA mental health care. All analyses were conducted using STATA version 10 (StataCorp, College Station, TX).
A total of 6287 women veteran VA primary care users participated in the study. As shown in Figure 1, half of all women in primary care reported perceived need for mental health care in the past year. Among these women, 84.3% received needed care, and nearly all of these women (90.9%) used VA for at least some of this mental health care. However, only about half (48.8%) of women who received mental health care at the VA in the past year reported that this mental health care met their needs completely or very well.
Table 1 displays individual characteristics of women VA mental health users (n=2466) and associations with perceived access, or met need. Over one third of VA mental health care users were reproductive aged, and nearly one third were parenting or caring for a child. A large proportion of VA mental health care users reported low household income and 37% were living in rural or highly rural areas. Thirteen percent reported no usual source of primary care. One quarter of women reported dual use of both VA and non-VA mental health care in the past year. Women who were younger, nonwhite, Hispanic, parenting, without a usual source of care, or dual users of VA and non-VA mental health services were significantly less likely to report that mental health care met their needs.
Table 2 presents VA mental health care experiences and associations with perceived access, or met need. Ease of use was the most robust indicator of perceived access, with high ratings associated with over 5 times greater odds that care met needs. Women who were able to see a female provider, participate in women-only groups, and receive care in women-only settings as often as desired, along with women who had no preferences for these gender-related services, were twice as likely to report met need when compared with women who were not able to access these gender-related services as often as desired. Gender-related comfort was also associated with over twice the odds of met need as compared with women who reported frequently feeling uncomfortable or out of place.
Because ease of use emerged as such a strong correlate of perceived access, we conducted post hoc analyses repeating models of the associations between gender-related health care experiences and perceived access, adjusting for ease of use. Gender-related experiences (female providers [adjusted odds ratio (AOR)=1.93; 95% confidence interval (CI), 1.53–2.44], women-only group treatments (AOR=1.75; 95% CI, 1.31–2.34), women-only treatment settings (AOR=1.77; 95% CI, 1.42–2.21), and gender-related comfort (AOR=1.71; 95% CI, 1.36–2.18) remained significantly associated with perceived access.
Table 3 displays associations of positive mental health care experiences to veteran characteristics. African American women were more likely to use women-only groups and women-only treatment settings as often as desired, but all nonwhite women reported lower gender-related comfort. A non-VA usual source of care was associated with lower ratings of access to female providers, women-only groups, and women-only settings, and VA-only mental health care was associated with higher ratings for using female providers as often as desired and for gender-related comfort.
Table 4 describes reasons for not using VA mental health care. Negative perceptions regarding ease of use and gender-related mental health experiences were significantly more common among women who sought mental health care only outside VA, as compared with women who perceived a need for mental health care but did not receive any care.
The WOMAN study is the first population-based study of the mental health care needs of women veterans using VA services. These results documented high levels of perceived need for mental health care, reported by half of all women using VA primary care services. Objective measures found very good access to mental health services: 84% of women with perceived need used mental health care in the past year, and nearly all of those women used VA for at least some of that care (Fig. 1). Access to mental health care was substantially higher than is observed in studies of the general US population, where approximately 20%–30% of individuals report a perceived need for mental health care in the past year, and only 50%–60% of individuals with perceived need receive mental health services.11,21,22 VA is a capitated system where many of the economic and insurance barriers that impact access in general population studies9,23 are not factors for VA patients. Routine universal screening for mental health conditions and integration of mental health into primary care have been demonstrated to increase use of mental health services among VA users with demonstrated need, especially women,24,25 and may underlie the high proportions of women who receive needed care.26
In contrast, measures of perceived access found that only half of women who used VA mental health care reported that these services met their needs very well or completely. Although women are engaging with mental health services, these services are not fully meeting women’s needs. Gender-related care experiences demonstrated strong associations with perceived access. Each of the gender-related care experiences was associated with a 2-fold increase in perceived access. These associations remained significant even after adjusting for ease of use, a proxy for logistical barriers to care. Furthermore, corresponding reasons for not using VA mental health services were cited among VA primary care patients who chose solely to use mental health services outside VA. The most frequently reported gender-related access issue was lack of access to designated women’s mental health treatment settings as often as desired, reported by nearly half of all women VA mental health users. Substantial proportions of women also reported lack of access to same-gender therapy groups. The frequency of the gender-sensitive care experiences observed in our study and their demonstrated impact on perceived access suggests that at a minimum, these issues should be discussed between patient and provider in the shared decision-making process of mental health referrals and treatment planning. In addition, attention to women’s perceptions and needs with respect to the care environment may help meet women’s needs. One in 5 women reported frequently feeling uncomfortable or out of place in their mental health treatment setting because of their gender. While these perceptions may be addressed by increasing access to designated settings for women, educational interventions are effective in improving staff and provider gender sensitivity in VA settings.27 Such interventions may also have potential to improve women’s comfort and engagement with care.
Younger women, nonwhite and Hispanic women, and women without a usual source of primary care were more likely to report that mental health services did not fully meet their needs. These characteristics are consistently associated with unmet need for mental health care in US population and health care samples,9,10,12 and unmet health care need among women veterans.14 Women VA users, on average, are younger and more racially and ethnically diverse as compared with men,28 and these characteristics may further exacerbate feelings of difference in treatment settings, as suggested by lower ratings of gender-related comfort. Women who were parents or caring for children also reported poorer perceived access to mental health services, although this did not appear to be a function of gender-related experiences among younger and nonwhite women. Parenting is associated with nonfinancial barriers to heath care,14,23 and is a relatively recent issue for VA mental health care with the influx of younger, current-era women veterans. Evaluation of pilot childcare programs and the dissemination of parenting resources and services throughout the system29,30 will further inform access in this population. Finally, women who were dual users of VA and non-VA mental health care reported poorer perceived access than VA-only users, and both dual users and women with a usual source of primary care outside VA reported poorer ratings for gender-related care experiences. Dual users may experience poorer care coordination as compared with VA-only health care users,31,32 especially when the usual source of care is outside VA, and the issue merits further investigation with respect to mental health.
Organizational factors will also play an important role in facilitating women’s perceived access to care. The recent increase in the population of women in VA has prompted increasing implementation of gender-specialized mental health treatment resources for women in VA, often integrated with women’s primary care clinics.4 However, implementation of these services poses a challenge. First, our data suggest that we should not, as a rule, assume women are best served by specialized women’s services. Substantial proportions of women reported no preferences for specialized women’s groups or settings, and these women reported similar perceived access compared with women with good access to such services. Women will be heterogeneous in the importance placed on gender-related preferences for mental health care, and our data suggest that it is the provision of services that correspond to these preferences that facilitates access to mental health care.
Second, in smaller or rural facilities, there may be too few women to support specialized services, despite veteran preferences for women’s clinics and groups. Research is needed to assess other delivery models and methods that could help deliver mental health care targeted to women’s needs. For example, can the availability of smaller, segregated waiting areas improve women’s comfort and meet needs for gender-specific settings? Similarly, would care environments that depict military-related images of women, as well as men, or include gender-targeted patient education materials address some of women’s needs for gender-sensitive care? If women-only groups are not feasible, would women feel better served in therapy groups that contained substantial proportions of women, rather than only 1 or 2 women? In addition, tele-mental health and eHealth resources may hold promise for virtual services targeted to women veterans.
Our results should be considered in light of several limitations. Despite our high participation rate and weighting methods, survey nonresponse could have affected the representativeness of our sample with respect to unmeasured variables or variables not available in VA administrative data. Our sample was limited to VA primary care users, in an effort to investigate gender-related access to VA mental health care distinct from barriers to VA utilization among women veterans that have been identified elsewhere.6,33 These results cannot be generalized to women veterans outside of the VA health care system, but do provide important information on how we might better meet the needs of those women currently using VA services.
Our results demonstrate very good objective access to mental health services among women veteran VA primary care users, and potential points of intervention for improving perceived access. We have identified several indicators of perceived access that provide a framework for examining mental health care for women VA users in the context of their gender-related experiences and preferences. The development of such targeted frameworks that define access to care for underrepresented health care users can play an important role in the continued evolution of patient-centered mental health care.
1. Friedman SA, Phibbs CS, Schmitt SK, et al.. New women veterans in the VHA: a longitudinal profile. Womens Health Issues. 2011;21supplS103–S111.
2. Frayne SM, Phibbs CS, Saecho F, et al.. Sociodemographics, Utilization, Costs of Care, and Health Profile. 2014.Washington, DC: Department of Veterans Affairs, Women’s Health Evaluation Initiative.
3. Oishi SM, Rose DE, Washington DL, et al.. National variations in VA mental health care for women veterans. Womens Health Issues. 2011;21supplS130–S137.
4. MacGregor C, Hamilton AB, Oishi SM, et al.. Description, development, and philosophies of mental health service delivery for female veterans in the VA: a qualitative study. Womens Health Issues. 2011;21supplS138–S144.
5. Veterans Health Administration. Uniform mental health services in VA Medical Centers and Clinics. In: Department of Veterans Affairs
. Washington, DC: VHA Handbook, 1160.01; 2008.
6. Washington DL, Kleimann S, Michelini AN, et al.. Women veterans’ perceptions and decision-making about Veterans Affairs health care. Mil Med. 2007;172:812–817.
7. Fortney JC, Burgess JF Jr, Bosworth HB, et al.. A re-conceptualization of access for 21st century healthcare. J Gen Intern Med. 2011;26suppl 2639–647.
8. Roll JM, Kennedy J, Tran M, et al.. Disparities in unmet need for mental health services in the United States, 1997-2010. Psychiatr Serv. 2013;64:80–82.
9. Stockdale SE, Tang L, Zhang L, et al.. The effects of health sector market factors and vulnerable group membership on access to alcohol, drug, and mental health care. Health Serv Res. 2007;423 pt 11020–1041.
10. Wang PS, Lane M, Olfson M, et al.. Twelve-month use of mental health services in the United States: results from the National Comorbidity Survey Replication. Arch Gen Psychiatry. 2005;62:629–640.
11. Kimerling R, Baumrind N. Access to specialty mental health services among women in California. Psychiatr Serv. 2005;56:729–734.
12. Jones E, Lebrun-Harris LA, Sripipatana A, et al.. Access to mental health services among patients at health centers and factors associated with unmet needs. J Health Care Poor Underserv. 2014;25:425–436.
13. Brooks E, Dailey N, Bair B, et al.. Rural women veterans demographic report: defining VA users’ health and health care access in rural areas. J Rural Health. 2014;30:146–152.
14. Washington DL, Bean-Mayberry B, Riopelle D, et al.. Access to care for women veterans: delayed healthcare and unmet need. J Gen Intern Med. 2011;26suppl 2655–661.
15. Kimerling R, Bastian L, Bean-Mayberry B, et al.. Patient centered mental health care for women veterans. Psychiatric Services. 2015;66:2.
16. Vogt D. Mental health-related beliefs as a barrier to service use for military personnel and veterans: a review. Psychiatr Serv. 2011;62:135–142.
17. Bauer MS, Williford WO, McBride L, et al.. Perceived barriers to health care access in a treated population. Int J Psychiatr Med. 2005;35:13–26.
18. VA Information Resource Center. VHA Medial SAS outpatient datasets and inpatient encounters dataset FY2009. VIReC Research User Guide. Hines, IL: US Department of Veterans Affairs, Health Services Research and Development Service, VA Information Resource Center; April, 2011. Available at: http://vaww.virec.research.va.gov/RUGs/MedSAS/RUG-MedSAS-OP-FY09-RA.pdf
19. Phibbs CS, Cowgill EH, Fan AY. Guide to the PSSG enrollee file. Guidebook
. Menlo Park, CA: VA Palo Alto, Health Economics Resource Center; February, 2013. Available at: http://www.herc.research.va.gov/files/BOOK_739.pdf
20. United States Department of Health and Human Services. Substance Abuse, Mental Health Services Administration. Center for Behavioral Health, Statistics Quality. National Survey on Drug Use and Health, 2010: 10-Year Substate R-DAS (2002 to 2011). ICPSR34947-v1. Ann Arbor, MI: Inter-university Consortium for Political and Social Research[distributor], 2013-12-17. Available at: http://doi.org/10.3886/ICPSR34947.v1
21. Grella CE, Cochran SD, Greenwell L, et al.. Effects of sexual orientation and gender on perceived need for treatment by persons with and without mental disorders. Psychiatr Serv. 2011;62:404–410.
22. Katz SJ, Kessler RC, Frank RG, et al.. The use of outpatient mental health services in the United States and Ontario: the impact of mental morbidity and perceived need for care. Am J Public Health. 1997;87:1136–1143.
23. Kullgren JT, McLaughlin CG, Mitra N, et al.. Nonfinancial barriers and access to care for US adults. Health Serv Res. 2012;471 pt 2462–485.
24. Bohnert KM, Pfeiffer PN, Szymanski BR, et al.. Continuation of care following an initial primary care visit with a mental health diagnosis: differences by receipt of VHA Primary Care-Mental Health Integration services. Gen Hosp Psychiatry. 2013;35:66–70.
25. Szymanski BR, Bohnert KM, Zivin K, et al.. Integrated care: treatment initiation following positive depression screens. J Gen Intern Med. 2012;28:346–352.
26. Washington DL, Davis TD, Der-Martirosian C, et al.. PTSD risk and mental health care engangement in a multi-war era community sample of women veterans. J Gen Intern Med. 2013;28:894–900.
27. Vogt DS, Barry AA, King LA. Toward gender-aware health care: evaluation of an intervention to enhance care for female patients in the VA setting. J Health Psychol. 2008;13:624–638.
28. Frayne SM, Phibbs CS, Friedman SA, et al.. Sociodemographic Characteristics and Use of VHA Care. 2010.Washington, DC: Department of Veterans Affairs, Women’s Health Evaluation Initiative.
29. Pemberton JR, Kramer TL, Borrego J Jr, et al.. Kids at the VA? A call for evidence-based parenting interventions for returning veterans. Psychol Serv. 2013;10:194–202.
30. Tsai J, David DH, Edens EL, et al.. Considering child care and parenting needs in Veterans Affairs mental health services. Eval Program Plann. 2013;39:19–22.
31. Bean-Mayberry B, Yano EM, Washington DL, et al.. Systematic review of women veterans’ health: update on successes and gaps. Womens Health Issues. 2011;21supplS84–S97.
32. Nayar P, Yu F, Apenteng B. Improving care for rural veterans: are high dual users different? J Rural Health. 2014;30:139–145.
33. Mengeling MM, Sadler AG, Torner J, et al.. Evolving comprehensive VA women’s health care: patient characteristics, needs, and preferences. Womens Health Issues. 2011;214SS120–S129.