Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 1.
Published in final edited form as: J Trauma Stress. 2020 Sep 23;34(2):394–404. doi: 10.1002/jts.22585

A Latent Class Analysis of Mental Health Beliefs Related to Military Sexual Trauma

Christine K Hahn 1, Jessica Turchik 2, Rachel Kimerling 2
PMCID: PMC7985046  NIHMSID: NIHMS1632740  PMID: 32969098

Abstract

Military veterans with histories of military sexual trauma (MST) are at risk for several negative mental health outcomes and report perceived barriers treatment engagement. To inform interventions to promote gender-sensitive access to MST-related care, we conducted an exploratory, multiple-group latent class analysis of negative beliefs about MST-related care. Participants were U.S. veterans (N = 1,185) who screened positive for MST within the last 2 months and reported a perceived need for MST-related treatment. Associations between class membership, mental health screenings, logistical barriers, difficulty accessing care, and unmet need for MST-related care were also examined. Results indicated a four-class solution, with classes categorized as (a) low barrier, with few negative beliefs; (b) high barrier, with pervasive negative beliefs; (c) stigma-related beliefs; and (d) negative perceptions of care (NPC). Men were significantly less likely than women to fall into the low barrier class (27.9% vs. 34.5%). Relative to participants in the low barrier class, individuals in all other classes reported more scheduling-, ps < .001; transportation-, p < .001 to p = .014; and work-related barriers, p < .001 to p = .031. Participants in the NPC class reported the most difficulty with access, p < .001, and those in the NPC and high barrier classes were more likely to report unmet needs compared to other classes, ps < .001. Brief cognitive and behavioral interventions, delivered in primary care settings and via telehealth, tailored to address veterans’ negative mental health beliefs may increase the utilization of mental health treatment related to MST.

A Latent Class Analysis of Mental Health Beliefs Related to Military Sexual Trauma

Despite a disproportionately high risk for posttraumatic stress disorder (PTSD) following sexual assault relative to other traumatic events (Kilpatrick et al., 2013), only 29%–37% of sexually assaulted men and women ever receive mental health services (Kirkner et al., 2018; Light & Monk-Turner, 2009). In addition to logistical issues, such as transportation, and attitudinal barriers to mental health treatment, such as a belief that treatment will not work (Andrade et al., 2014), there are also unique negative beliefs related to sexual trauma. Negative mental health beliefs related to sexual trauma are hypothesized to stem from the pervasive social stigma as well as an internalized prevailing cultural schema that normalizes harassment and assault or attribute responsibility to victims, often referred to as “rape myths” (Edwards et al., 2011). Perhaps the most pervasive beliefs relevant to military sexual trauma relate stigma, gender-related concerns, and secondary victimization, which involves others’ behaviors that increase the impact of the traumatic event on health outcomes, such as blaming the victim (Donne et al., 2018; Turchik et al., 2013, 2014). Research into modifiable negative beliefs is critical to inform interventions that can increase the utilization of sexual trauma–related mental health care.

The U.S. Veterans Health Administration (VHA) is in many ways an ideal in which setting to investigate mental health beliefs related to sexual trauma in the United States, as a sizable population of veterans has experienced sexual trauma, and there is widespread availability of specialized services for sexual trauma across the system. Military sexual trauma (MST), defined as sexual assault and sexual harassment experienced during a veteran’s military service (Title 38 U.S. Code 1720D), is a pervasive issue across all branches of the armed forces (Morral et al., 2015). Among recent-era Veterans, 41.5% of women and 4.0% of men have reported MST (Barth et al., 2016), which is a risk factor for a wide range of health conditions, including PTSD and depressive disorders (Kimerling et al., 2007). Provisions made by the VHA aimed at facilitating access to MST-related mental health care are essential to connecting veterans to care.

The VHA has implemented extensive measures to increase access to MST-related care, including system-wide provider training, patient education and outreach, and national clinical performance monitoring. Moreover, the VHA has implemented universal MST screening to facilitate disclosure and reduce economic barriers, whereby all care for health conditions designated by the provider as being MST-related are exempt from co-pays. A recent report noted that among VHA users, 28.3% of women (n = 117,991) and 1.5% of men (n = 71,851) reported a history of MST when screened, and 62.8% of female veterans and 50.6% of male veterans who screened positive for MST attended at least one MST-related mental health visit (Military Sexual Trauma Support Team, 2018). These access rates are higher than those reported among the general population (Kirkner et al., 2018; Light & Monk-Turner, 2009), suggesting that VHA efforts have likely facilitated access to services among VHA users. Veterans who screen positive for MST but still do not receive care may have potentially modifiable negative beliefs about MST-related health care.

Stigma-related beliefs are frequently reported by survivors of sexual assault and include self-stigma, or negative beliefs about having a condition or seeking help (e.g., being embarrassed and ashamed), as well as the anticipation that others will have negative reactions, such as viewing the individual as weak (Andersen & Blais, 2019; Holland et al., 2016; Turchik et al., 2013). Veterans have also described a fear that others would find out about their MST as a reason for not seeking treatment (Turchik et al., 2013). Specific stigma-related beliefs have demonstrated unique associations with disclosure of MST (Andersen & Blais, 2019), suggesting that endorsement of various stigma-related beliefs may differ across veterans.

Military personnel and individuals diagnosed with PTSD have reported beliefs about mental health treatment providers and the health care system to be perceived barriers to treatment (Ouimette et al., 2011; Smith et al., 2020; Vogt et al., 2014). These types of beliefs include concerns related to privacy and fears that providers may have a negative reaction or lack skills or sensitivity. In addition to apprehension related to different types of stigma, active duty women who have experienced MST have described fears regarding secondary victimization, such as professionals blaming or not believing them, as barriers to seeking MST -related health care (Burns et al., 2014). Concerns about providers and the quality of services within the VHA may be common because of military experiences that foster a sense of betrayal due to the failure of the system to protect individuals during their service. Beliefs related to self-reliance, such as believing that it is best to handle one’s problem on their own; avoidance of seeking treatment; and difficulties with experiencing emotion have also been identified as important factors that contribute to treatment-seeking among veterans, military personnel, and individuals with PTSD (Smith et al., 2020; Williston et al., 2019). Beliefs related to self-reliance have also been shown to be associated with less treatment-seeking among active duty military personnel who have experienced MST (Zinzow et al., 2015). These beliefs might be reinforced by a military culture that encourages toughness and stoicism.

Given that a smaller proportion of men than women who have experienced MST have received related health care services (Turchik et al., 2012), there may also be gender-specific barriers to seeking MST treatment. Yet, few studies have examined barriers to receiving MST-related health care among men (Wilson, 2018). The findings from a qualitative study of a sample of male veterans who were enrolled in the VHA but had not received any MST-related mental health care demonstrated that men endorsed barriers including anticipated stigma, self-stigma (e.g., shame), concerns about others finding out, and fear that providers would not believe them (Turchik et al., 2013). Several additional self-reliance–related barriers that are consistent with traditional gender-role norms, such as the belief that men should be tough, were also described. Men identified concerns that providers would question their sexual orientation, and most men indicated that a provider’s gender might be perceived as a barrier to seeking services (Turchik et al., 2013). In a similar qualitative study, female veterans reported stigma-related beliefs, concerns about emotions, and gender-related concerns, such as being uncomfortable in a male-dominated environment and the prospect of seeking services from a male provider, as barriers to MST-related health care (Turchik et al., 2014). This preliminary research suggests that differences exist between male and female veterans with regard to perceived barriers specific to rape myths and gender roles.

Many various barriers to receiving health care following sexual trauma have been described in the current literature. However, barriers that are specific to receiving MST-related mental health care among VHA users have been understudied, particularly among men (Wilson, 2018). A person-centered approach to understanding negative beliefs about MST-related mental health care is an important step for informing tailored interventions that can facilitate the utilization of mental health services. The primary goal of the current study was to identify classes of negative beliefs about MST-related mental health care among a national sample of male and female veterans who screened positive for MST within the VHA. We hypothesized there would be distinct patterns of negative mental health beliefs reflecting qualitatively different types of concerns, such as beliefs related to stigma, secondary victimization, and gender-related concerns. We also hypothesized there would be two classes, one with a high probability of endorsing negative beliefs and another with a low probability. The second study aim was to examine if underlying typologies of negative beliefs about MST-related mental health care differed by gender. The final aim was to describe typologies of negative beliefs by examining associations with mental health symptoms, logistical barriers, difficulty with access, and unmet need for MST-related care.

Method

Participants and Procedure

This study was part of the Understanding Veterans’ Opinions and Attitudes about VA Health Care Survey (Garneau-Fournier et al., 2017) and was designed to identify factors associated with access to MST-related mental health care. Participants were randomly sampled from among U.S. veteran VHA users within 3 months of a positive screen for MST. Sampling occurred in waves between August 2013 and March 2014. Inclusion required a valid mailing address in the medical record and exclusion criteria were current diagnoses indicating cognitive impairment (i.e., dementia, brain injury), legal blindness, or an indication of a conservator or legally authorized representative. A total of 11,042 veterans met the inclusion criteria, and 8,409 eligible veterans were randomly selected for participation. Participants could opt-in to the study by returning a mailed envelope or contacting the researchers via telephone. Of 2,682 veterans who opted-in to the study (31.9% of eligible veterans), 2,220 returned a completed survey and were compensated $20 for their participation (26.4% of eligible veterans). Based on administrative data, responders were significantly more likely than nonresponders to be older (M age = 50.09 years vs. M age = 43.00 years), male (38.1% vs. 35.7%), non-Hispanic (85.4% vs. 82.9%), and White (61.6% vs. 56.9%). The Stanford University School of Medicine approved this study. We limited the sample in the current study to veterans with a perceived need for care. As with other traumatic events, many veterans are resilient to MST experiences; thus, our results would be most actionable in promoting access to MST-related care if they focused on veterans who wanted or needed these services. The present study used a subsample of 1,185 veterans who responded positively to the following survey item: “In the past year, did you ever want or need help with any emotional or mental health concerns related to any military sexual trauma experience?” Women veterans, χ2(2, N = 2,182) = 25.61, p < .001, who identified as white, χ2(2, N = 2,036) = 45.50, p < .001; and younger veterans (M age = 49.17 vs. 54.60 years, SD = 0.48), t(2,180) = −8.84, p = .001, were more likely to report perceived need. This item followed behaviorally specific items related to MST experiences.

Measures

Demographic Characteristics

Participants filled out a questionnaire to ascertain information related to demographic characteristics and military background.

Military Sexual Trauma–Related Care Information

Immediately following items that queried perceived need, as previously described, care experiences were assessed with two additional items, adapted from prior survey research (Kimerling et al., 2015). Perceived access to MST-related care was measured with the item “How difficult was it for you to get the VA health care related to military sexual trauma that you wanted in the past year?” Responses were given a 5-point Likert scale, with options ranging from 1 (very difficult) to 5 (very easy), with ratings of 2 or below indicating difficulty with access. Appropriateness of care was measured with the item “Overall, how well did the VA health care you received for military sexual trauma meet your needs in the past year?” Responses ranged from 1 (not at all) to 5 (completely). Ratings of 2 or below indicated unmet needs.

Perceived Barriers to Care

To ensure a wide range of negative beliefs about MST-related mental health care were assessed, we included 20 items from the Perceived Barriers to Care (Brown et al., 2011), Perceived Stigma and Barriers to Care for Psychological Problems (Pietrzak et al., 2009), Endorsed and Anticipated Stigma Inventory (EASI; Vogt et al., 2014), and items created by Fikretoglu and colleagues (2008). To assess gender-specific beliefs, we created four items based on qualitative research of veterans with histories of MST (e.g., “The large number of men at VA make me uncomfortable;” “It is embarrassing to talk to some of the same sex about MST;” “It is embarrassing to talk to the opposite sex about MST;” “Other people would make judgments about my sexuality;” Turchik et al., 2013, 2014). Veterans were prompted to “Imagine you are seeking mental healthcare from VA related to military sexual trauma. Rate how much you agree or disagree with each statement;” an example item is “It is embarrassing to receive services for MST.” The 20 items were rated used a 5-point Likert scale ranging from 1 (strongly disagree) to 5 (strongly agree). Participants were coded as endorsing a belief if they reported a 4 or above, which is consistent with previous research that used this cutoff to measure endorsement of mental health beliefs (Gorman et al., 2016; Valenstein et al., 2014). In addition, four items were selected from the instruments to assess logistical barriers (e.g., childcare, work, transportation, scheduling difficulties); the items were used as correlates of the latent classes.

PTSD Screening

The Primary Care PTSD screen (PC-PTSD; Prins et al., 2003) is a four-item screen used to detect probable PTSD. Participants were coded positive for PTSD if they reported experiencing a potentially traumatic event and endorsed a minimum of three out of four items assessing past-month hyperarousal, avoidance, reexperiencing, and numbing symptoms. A score of three was demonstrated to be an optimally efficient cutoff among a sample of veterans (sensitivity = 78.0, specificity = 87.0; Prins et al., 2003). In the current sample, the reliability for the total PTSD scale from the PC-PTSD was excellent, Cronbach’s α = .96.

Depression Screening

The Patient Health Questionnaire–2 (PHQ-2; Kroenke et al., 2003) was used to assess depressive symptoms over the past two weeks. Participants were asked to rate items using a scale of 0 (not at all) to 3 (nearly every day); those who endorsed the two items with a score of 2 or higher were coded as having a positive depression screen. A cutoff score of 2 has demonstrated adequate sensitivity (82.1) and specificity (80.4) for detecting depressive disorders (Kroenke et al., 2003). In the current sample, the reliability of the total PHQ-2 score was high, Cronbach’s α = .84.

Data Analysis

Multiple-group latent class analysis (LCA) was conducted using maximum likelihood estimation with robust standard errors for two to six classes. Analyses were conducted using the LCA Stata Plugin (Version 1.2) in STATA 14 (StataCorp, 2017). The LCA method classifies individuals into mutually exclusive groups based on patterns of responses to discrete observed variables, termed “indicators” (Dean & Raftery, 2010; Lanza et al., 2015). We used LCA to determine the nature and number of mutually exclusive groups of veterans based on responses to negative beliefs about MST-related mental health care. Dichotomous indicators were used to achieve model identification by minimizing the number of response patterns (Schinka & Velicer, 2012) and to facilitate interpretation of results. Multiple-group LCA allowed us to determine if patterns of item-response probabilities differed between men and women and compare latent class prevalence across men and women (Schinka & Velicer, 2012). Composite indicators composed of related items protect against violations of local independence assumptions as well as lack of convergence of the LCA models due to sparse pattern combinations (Dean & Raftery, 2010; Swanson et al., 2012). We assessed clustering among items using hierarchical item clustering via the iclust package in R (Revelle, 1979), a method similar to exploratory factor analysis but more effective in identifying the hierarchical relations among highly correlated variables (Bacon, 2001). Composite indicators were created for seven indicators from the 24 negative beliefs (see Table 1).

Table 1.

Frequency of Latent Class Indicators

Indicator Women Men
n % n %
Avoidance: It is better to avoid talking about or seeking help for mental health problems. 126 16.2 79 21.4
Fear of affect: I would not be able to control my emotions. 508 65.3 207 55.5
Self-reliance: Tough people can handle their problems on their own; It is better to cover up how I feel. 418 53.6 223 59.6
Internal stigma: I would feel ashamed to talk to my provider about MST; It is embarrassing to receive services for MST. 400 51.3 242 64.9
Public stigma: Other people would find out I was seeking mental health care; Other people would find out I experienced MST. 380 48.7 222 59.5
Anticipated stigma: Other people would think I was weak; Other people might think I caused the event to happen. 487 62.4 247 66.2
Evidence: I would not be believed without evidence. 354 45.6 208 55.9
Staff sensitivity: My provider would blame me for the events; The provider would not keep it confidential. 172 22.1 113 30.4
Provider reaction: My provider would react negatively; My VA provider would not believe me; My provider would not want to hear about my problems. 336 43.1 170 45.6
Care: No one cares about my experiences. 245 31.5 152 40.8
Staff skill: It is better to seek services outside of the VA for MST; I would not receive good mental health services for MST; Staff is not well trained to handle problems related to MST. 306 39.3 154 41.3
Opposite-sex concerns: It is embarrassing to talk to a person of the opposite sex about MST. 482 62.0 98 26.3
Same-sex concerns: It is embarrassing to talk to a person of the same sex about MST. 131 16.8 203 54.6
Sexuality concerns: Other people would make judgments about my sexuality. 213 27.4 220 59.1
Discomfort with men: The large number of men at the VA makes me uncomfortable. 322 41.4 179 48.1
Open in a new tab

Note. MST = military sexual trauma; VA = Veterans Affairs.

We did not constrain groups for measurement invariance to observe class differences by gender. The LCA procedure in Stata employs the expectation-maximization (EM) algorithm to produce maximum likelihood estimates of all model parameters. Fit statistics, entropy, parsimony, and interpretability of classes were examined to identify the number of classes that best represented the observed negative beliefs (Schinka & Velicer, 2012). The sample-size adjusted Bayesian information criteria (aBIC) was used to compare classes, as it has demonstrated superior performance compared to other fit indices (Swanson et al., 2012). Next, the four-class model was run again with the inclusion of demographic variables (i.e., age, race, and service during U.S. conflicts in Afghanistan and Iraq) as covariates in the model to test associations with class membership and to condition class assignment on demographic characteristics. Participants were allocated to a class based on the posterior probabilities of this LCA model to reduce bias due to classification uncertainty (Bray et al., 2015). To inform our interpretation of the classes, we calculated a series of chi-square tests to explore variation across classes with regard to the proportions of veterans who reported clinically meaningful mental health symptoms (i.e., screening results for PTSD and depression), barriers to care (i.e., childcare, scheduling, transportation, and work-related barriers), and care experiences (i.e., perceived access to care and unmet need for care). For all statistically significant chi-square test results, we examined pairwise comparisons of each class with each of the other classes, using Wald tests to identify statistically significant differences. Missing data ranged from 0 to 11 data points across all outcomes; we employed maximum likelihood estimation to allow for missingness on these cases in the LCA analysis (Schinka & Velicer, 2012). Cases with missing data were excluded in the bivariate analyses.

Results

There were 796 (67.2%) women and 389 (32.8) men included in the sample. Participants primarily identified as White (59.7% women, 65.1% men), followed by Black (29.9% women, 23.4% men). Regarding other races and ethnicities, 10.4% of women and 11.5% of men identified as American Indian/Alaska Native or Native Hawaiian/other Pacific Island or chose “other” for the racial/ethnic background item. With respect to service era, 28.1% of women and 10.5% of men served during U.S. conflicts in Afghanistan and Iraq. The frequencies of MST-related mental health beliefs that comprised the LCA indicators are described in Table 1. In the LCA, 1,153 participants were included due to missing data on covariates (n = 780 women; n = 373 men). The four-class LCA was selected as the best-fitting model (see Table 2). With 1,000 random starts, the log-likelihood was replicated in the majority of solutions, suggesting we reached the optimal solution rather than a local solution (Lanza et al., 2007). Median posterior probabilities of class assignment ranged from .91 (interquartile range [IQR]: .74–.98) to .98 (IQR: .87–.99), with all average class posterior probabilities higher than .80, suggesting good reliability for class assignment. A class characterized as “low barrier” included veterans with a relatively low probability of endorsing all negative beliefs about MST-related mental health care, and a “high barrier” class included veterans with a high probability of endorsing all indicators (Figures 1 and 2). Although the low and high barrier classes generally had low and high probabilities of endorsing barriers, respectively, there were still differences in the pattern of perceived barriers in these classes for men and women. For instance, in the low-barrier class, the indicator with the highest probability for women was fear of affect, whereas anticipated stigma had the highest probability in the high barrier class (see Figure 1). After controlling for covariates, there were a higher proportion of women compared to men in the low barrier class (27.9% of men, 34.5% of women) F(1, 1,152) = 5.25 p = .022, but gender differences were not significant in the high barrier class (25.7% of men, 22.4% of women) F(1, 1,152) = 1.48, p = .224. Two additional classes emerged: One characterized by stigma and one by the perception of care. There were no significant differences in the proportion of men and women assigned to the stigma class (36.2% of men, 33.1% of women) F(1, 1,152) = 1.07, p = .301; or perception of care class (10.2% of men, 10.0% of women), F(1, 1,152) = .01, p = .921. In the class characterized by stigma, there were various forms of stigma with high prevalence probabilities, in addition to self-reliance and avoidance indicators, for both men and women. Further, men had a high probability of endorsing the same-sex provider concern indicator, whereas women had a high probability of endorsing the opposite-sex provider concern indicator. In addition, sexuality concerns were elevated among men. In the perception of care class, indicators related to staff sensitivity and skills were elevated for men and women; however, there were gender differences. For instance, women also endorsed indicators related to stigma, whereas the highest prevalence probabilities for men were indicators related to staff skill and not being believed without evidence.

Table 2.

Goodness of Fit Criteria

Number of Classes df Gⁿ2 aBIC Entropy
2 65, 473 5,277.98 5,517.18 .86
3 65, 441 4,902.48 5,265.14 .77
4 65, 409 4,733.54 5,219.66 .78
5 65, 377 4,643.96 5,253.54 .76
6 65, 345 4,510.21 5,846.74 .75
Open in a new tab

Note. BIC = Bayesian information criteria.

Figure 1.

Figure 1

Open in a new tab

Four-Class Model for Women

Figure 2.

Figure 2

Open in a new tab

Four-Class Model for Men

Classes differed regarding demographic characteristics. Omnibus tests indicated differences across classes in the likelihood of reporting race as White versus non-White, p = .041; Black versus non-Black, p = .006; and age 65 years or older versus under age 65, p = .020. Specifically, female participants in the perception of care class were significantly less likely to be White, odds ratio (OR) = 0.284, 95% CI [0.119, 0.673]; and Black, OR = 0.398, 95% CI [0.250, 0.634], than members of the low barrier class. In addition, men who were 65 years of age or older were significantly less likely to be in the high barrier class than members of the low barrier class, OR = 0.649, 95% CI [0.488, 0.863].

Table 3 presents the proportions of veterans within each class who reported mental health symptoms, barriers to care, and negative care experiences. All chi-square tests indicated significant variation across classes, ps < .001. Pairwise comparisons of classes revealed differences in the associations of class membership with symptoms, barriers to care, and care experiences that supported the proposed typologies of beliefs. Despite the large proportion of the sample with a positive PTSD screen, the proportions were significantly lower within the low barrier class as compared to all other classes: F(1, 1,139) = 29.10, p < .001 for low barrier versus high barrier; F(1, 1,139) = 7.71, p = .006, for low barrier versus perceptions of care; and F(1, 1,139) = 6.02, p = .014 for low barrier versus stigma. The proportions of participants with positive PTSD screens were also higher in the high barrier class compared to the stigma class, F(1, 1,139) = 9.88, p = .002. These patterns were similar for depression, whereby the proportion of positive depression screenings was highest within the high barrier class compared to all other classes: F(1, 1,147) = 55.00, p < .001 for high barrier versus low barrier; F(1, 1,147) = 29.86, p < .001 for high barrier versus stigma; and F(1, 1,147) = 22.43, p < .001 for high barrier versus perceptions of care. Further, the proportion of positive screenings was significantly lower within the low barrier class compared to the stigma class, F(1, 1,147) = 4.33, p = .038.

Table 3.

Class-Specific Proportions of Veterans who Reported Symptoms, Barriers to Care, and Care Experiences

Variable Low (n = 373) Stigma (n = 393) Care (n = 116) High (n = 271) Total (n = 1,153) χ2(df, N)
% SE % SE % SE % SE %
Symptoms
 PTSDa 78.5 0.02 85.4 0.02 88.7 0.03 92.9 0.02 85.3 26.69(3, N = 1,140)***
 Depressionb 50.4 0.03 57.9 0.02 52.2 0.05 77.4 0.03 59.5 51.64(3, N = 1,148)***
Barriers to care
 Childcare 4.3 0.01 8.7 0.01 15.0 0.03 12.0 0.02 8.68 18.39(3, N = 1,140)***
 Schedule 22.1 0.02 37.0 0.02 61.2 0.05 56.1 0.03 39.1 102.37(3, N = 1,150)***
 Transportation 7.5 0.01 14.8 0.02 20.7 0.04 24.4 0.03 15.3 37.22(3, N = 1,152)***
 Work 19.2 0.02 28.1 0.02 36.5 0.05 26.5 0.03 25.7 16.51(3, N = 1,145)***
Care experiences
 Accessc 46.8 0.03 60.1 0.02 88.9 0.03 77.5 0.03 62.7 96.09(3, N = 1,102)***
 Unmet needd 54.0 0.03 67.7 0.03 88.6 0.03 85.2 0.02 69.3 77.69(3, N = 951)***
Open in a new tab

Note. PTSD = posttraumatic stress disorder; VA = Veterans Affairs.

a

Screned positive PTSD screen based on the Primary Care Screen for PTSD.

b

Screened positive for depression based on the Patient Health Questionnaire–2.

c

Measured with the item “How difficult was it for you to get the VA health care related to military sexual trauma that you wanted in the past year?”

d

Measured with the item “Overall, how well did the VA health care you received for military sexual trauma meet your needs in the past year?” Participants (n = 198) who did not report any Veterans Health Administration MST-related mental health care at the time of the survey responded “n/a” to the item.

*

p < .001.

As expected, participants in the low barrier class were significantly less likely than participants in all other classes to endorse barriers to care, including (a) childcare: low barrier vs. stigma, F(1, 1,139) = 6.05, p = .014; low barrier vs. perceptions of care, F(1, 1,139) = 9.17, p = .003; low barrier vs. high barrier, F(1, 1,139) = 11.61, p < .001; (b) scheduling: low barrier vs. stigma, F(1, 1,149) = 20.88, p < .001; low barrier vs. perceptions of care, F(1, 1,149) = 60.44, p < .001; low barrier vs. high barrier, F(1, 1,149) = 83.85, p < .001; (c) transportation: low barrier vs. stigma, F(1, 1,151) = 10.28, p = .014; low barrier vs. perceptions of care, F(1, 1,151) = 10.73, p < .001; low barrier vs. high barrier, F(1, 1,151) = 32.55, p < .001; and (d) work: low barrier vs. stigma, F(1, 1,144) = 8.40, p = .004; low barrier vs. perceptions of care, F(1, 1,144) = 12.24, p < .001; low barrier vs. high barrier F(1, 1,144) = 4.64, p = .031. Scheduling difficulties were significantly more frequent among participants in both the high barrier class, F(1, 1,149) = 24.18, p < .001; and the perceptions of care class, F(1, 1,149) = 22.04, p < .001, compared to those in the stigma class. Transportation barriers were more frequently endorsed among participants in the high barrier class compared to those in the stigma class, F(1, 1,151) = 9.18, p = .003.

In contrast to patterns of class differences across symptoms and barriers to care, the highest proportions of access difficulties were endorsed among veterans in the perceptions of care class compared to participants in all other classes: perceptions of care vs. low barrier, F(1, 1,101) = 109.47, p < .001; perceptions of care vs. stigma, F(1, 1,101) = 53.05, p < .001;, perceptions of care vs. high barrier, F(1, 1,101) = 8.18, p = .004. The proportions of participants who endorsed unmet need for care were commensurate in the high barrier and perceptions of care classes and significantly higher than the proportions of unmet need among participants in the (a) low barrier class: high barrier vs. low barrier, F(1, 950) = 72.27, p < .001; perceptions of care vs. low barrier, F(1, 950) = 61.67, p < .001; and (b) stigma class: perceptions of care vs. stigma, F(1, 950) = 23.73, p < .001; high barrier vs. stigma, F(1, 950) = 24.49, p < .001.

Discussion

The adoption of a person-centered approach allowed us to classify individuals in a national sample of male and female veterans who reported MST and a perceived need for care into mutually exclusive subgroups based on typologies of negative beliefs about MST-related mental health care. Most veterans comprised subgroups characterized by reports of significant but potentially modifiable negative beliefs about MST-related mental health care. Because men were less likely than women to be in the low barrier subgroup, which was characterized by few negative beliefs, it is likely that traditional male gender role expectations contribute to increased concerns about seeking treatment following MST among men (Turchik et al., 2013). Consistent with previous research, the current results suggest a subset of veterans are primarily concerned about seeking MST treatment because of potential stigma, including self- and anticipated stigma, and feelings of embarrassment about seeking services from providers of a certain gender (Andrade et al., 2013). The use of a person-centered approach allowed us to identify that individuals in the stigma group had a low-level endorsement of beliefs related to provider skills and services, unlike the subset of veterans in the perceptions of care class, whose most frequently endorsed beliefs were related to secondary victimization.

After controlling for demographics, the associations among most likely class membership, mental health symptoms, logistical barriers provided further support for the four-class solution. Consistent with the present results, previous researchers have reported higher levels of negative mental health beliefs to be associated with more logistical barriers (Andrade et al., 2014) and symptoms of PTSD and depression (Cadigan et al., 2019; Ouimette et al., 2011). When veterans experience logistical barriers and/or mental health symptoms, this may further maintain negative mental health-related beliefs. Conversely, negative mental health-related beliefs may increase the likelihood of experiencing logistical barriers and distress. For instance, having thoughts such as “I would not be able to control my emotions” may lead to distress and make problem-solving to overcome logistical barriers more difficult. Veterans who experience negative mental health beliefs may also be more likely to perceive more difficulty in accessing care as well as more unmet care needs, or negative experiences with the VHA may contribute to negative beliefs.

Although there are various potential negative beliefs about MST-related mental health care, it is important to consider the combination of veterans’ specific beliefs. The typologies of negative beliefs about MST-related mental health care identified in the current study described groups of veterans who may benefit from different approaches to enhancing access to care. Services geared at promoting access to MST-related mental health care should be delivered in settings where veterans who lack access are already located, such as primary care settings, and via telehealth (Zinzow et al., 2012). Brief, one-session cognitive behavioral interventions aimed at facilitating mental health treatment and delivered in primary care settings or via telephone to veterans who are not yet receiving VHA services have been shown to decrease PTSD symptoms and increase completion of mental health treatment sessions among veterans who have screened positive for PTSD (Gallegos et al., 2015; Possemato et al., 2017; Stecker et al., 2014). To further guide the development of brief interventions aimed at increasing access to MST-related mental health care, more research is needed to define typologies of negative mental health beliefs and test associations between typologies and access to care. Tailoring brief interventions to the unique needs of different subgroups of veterans may be more effective than delivering a cognitive behavioral intervention that uniformly addresses all negative mental health beliefs.

Brief cognitive behavioral interventions that address modifiable beliefs are especially needed for individuals who endorse high levels of barriers to care. Brief cognitive behavioral approaches may also work exceptionally well for veterans with more concerns regarding stigma, as beliefs related to internal stigma and fear of affect may be reframed with psychoeducation and cognitive techniques (Stecker et al., 2014). There may be unique clinical implications for brief interventions among men and women with concerns similar to those characterized by the current perception of care subgroup. For example, interventions for men should focus on reframing male rape myths and increasing acknowledgment that men can experience MST and have related mental health problems that warrant services (Hoyt et al., 2012). Female veterans may benefit from not only receiving cognitive behavioral interventions focused on reframing stigma-related beliefs but also being connected to different VHA service options, such as receiving treatment in women’s health clinics or via telemedicine, which are effective ways to increase access to evidence-based, trauma-focused treatment for PTSD-related MST (Zinzow et al., 2012). One-quarter of veterans in the present study were grouped into the low barrier class, indicating that a significant portion of veterans does not need brief interventions to reduce negative treatment-related beliefs. It may be beneficial to gather information from individuals who are classified as having low levels of negative beliefs about the types of MST-related resources they have accessed. It is also important to recognize that negative mental health beliefs related to MST stem from rape myths and cultural schema that promote stigma and sexual assault (Edwards et al., 2011). Community-based approaches that attempt to create healthier beliefs about mental health and debunk rape myths are also essential.

The present results should be interpreted within the context of the study’s limitations. To conduct the LCA with the current sample size, we dichotomized indicators and reduced items to 15 indicators. Although these decisions were based on previous research, this approach precludes exploration of the variation in the severity of treatment beliefs within each class. It is possible that relevant mental health beliefs were not or adequately captured by the dichotomous indicators. The LCA focused on negative beliefs about MST-related mental health care. To further describe classes of perceived barriers, researchers should also focus on other factors, including institutional barriers (e.g., limited hours) to treatment. As responders were more likely to identify as non-Hispanic and White compared to other racial and ethnic groups, there may be additional classes of perceived barriers among racial and ethnic minority veterans that were not captured in the current results. Future research is needed to test the associations between class association and treatment utilization.

In the current study, we observed several noteworthy patterns of negative beliefs about MST-related mental health care, including unique differences in patterns of barriers for men and women. It is essential to further research the typologies of negative mental health beliefs and their associations with veterans’ receipt of MST mental health care. Addressing the different typologies of negative beliefs about MST-related mental health care may be a promising approach to facilitate access.

Footnotes

Open Practices Statement

Neither of the studies reported in this article was formally preregistered. Neither the data nor the materials have been made available on a permanent third-party archive; requests for the data or materials can be sent via email to the lead author at hahnc@musc.edu

Author Note

The institution of origin for this manuscript was Veterans Affairs (VA) Palo Alto Health Care System, National Center for PTSD. This material is based upon work supported or supported in part by the Department of Veterans Affairs, Veterans Health Administration, Office of Research and Development (CDA 12–273), and the National Center for PTSD. Manuscript preparation was partially supported by a grant from the National Institute of Mental Health (T32MH018869) and the National Institute on Drug Abuse (2U54DA016511–16; PIs: Mcrae-Clark, Brady).

The views expressed in this article are those of the authors and do not necessarily reflect the position or policy of the Department of Veterans Affairs or the United States government.

References

  1. Andresen FJ, & Blais RK (2019). Higher self-stigma is related to lower likelihood to disclosing military sexual trauma during screening in female veterans. Psychological Trauma: Theory, Research, Practice, and Policy, 11(4), 372–378. 10.1037/tra0000406 [DOI] [PubMed] [Google Scholar]
  2. Andrade LH, Alonso J, Mneimneh Z, Wells JE, Al-Hamzawi A, Borges G, Bromet E, Bruffaerts R, de Girolamo G, de Graff R, Florescu S, Gureje O, Hinkov HR, Hu C, Huang Y, Hwang I, Jin R, Karam EG, Kovess-Masfety V, … Kessler RC (2014). Barriers to mental health treatment: Results from the WHO World Mental Health surveys. Psychological Medicine, 44, 1303–1317. 10.1017/S0033291713001943 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Bacon DR (2001). An evaluation of cluster analytic approaches to initial model specification. Structural Equation Modeling: A Multidisciplinary Journal, 8(3), 397–429. https://doi.org/0.1207/S15328007SEM0803_4 [Google Scholar]
  4. Barth SK, Kimerling RE, Pavao J, McCutcheon SJ, Batten SV, Dursa E, Peterson MR, & Scneiderman AI (2016). Military sexual trauma among recent veterans. American Journal of Preventative Medicine, 50, 77–86. 10.1016/j.amepre.2015.06.012 [DOI] [PubMed] [Google Scholar]
  5. Brown MC, Creel AH, Engel CC, Herrell RK, & Hoge CW (2011). Factors associated with interest in receiving help for mental health problems in combat veterans returning from deployment to Iraq. Journal of Nervous and Mental Disease, 199, 797–801. 10.1097/NMD.0b013e31822fc9bf [DOI] [PubMed] [Google Scholar]
  6. Burns B, Grindlay K, Holt K, Manski R, & Grossman D. (2014). Military sexual trauma among U.S. service women during deployment: A qualitative study. American Journal of Public Health, 104, 345–349. 10.2105/AJPH.2013.301576 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cadigan JM, Lee CM, & Larimer ME (2019). Young adult mental health: A prospective examination of service utilization, perceived unmet service needs, attitudes, and attitudes to services use. Prevention Science, 20(3), 366–376. 10.1007/s11121-018-0875-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Title 38 U.S. Code § Sec. 1720D. (2006). Counseling and treatment for sexual trauma.
  9. Dean N & Raftery AE (2010). Latent class analysis variable selection. Annals of the Institute of Statistical Mathematics, 62(1), 11–35. 10.1007/s10463-009-0258-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Donne MD, DeLuca J, Pleskach P, Bromson C, Mosley MP, Perez ET, Mathews SG, Stephenson R, & Frye V. (2018). Barriers to and facilitators of help-seeking behavior among men who experience sexual violence. American Journal of Men’s Health, 12(2), 189–201. 10.1177/1557988317740665 [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Edwards KM, Turchik JA, Dardis CM, Reynolds N, & Gidycz CA (2011). Rape myths: History, individual and institutional-level presence, and implications for change. Sex Roles, 65(11–12), 761–773. 10.1007/s11199-011-9943-2 [DOI] [Google Scholar]
  12. Fikretoglu D, Guay S, Pedlar D, & Brunet A. (2008). Twelve-month use of mental health services in a nationally representative, active military sample. Medical Care, 46(2), 217–223. 10.1097/mlr.0b013e31815b979a [DOI] [PubMed] [Google Scholar]
  13. Gallegos AM, Streltzov NA, & Stecker T. (2016). Improving treatment engagement for returning Operation Enduring Freedom and Operation Iraqi Freedom veterans with posttraumatic stress disorder, depression, and suicidal ideation. Journal of Nervous and Mental Disease, 204(5), 339–343. 10.1097/NMD.0000000000000489 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Garneau-Fournier J, Habarth J, & Turchik JA (2017). Factors associated with sexual dysfunction symptoms among veterans who have experienced military sexual trauma. International Journal of Sexual Health, 30(1), 28–41. 10.1080/19317611.2017.1404541 [DOI] [Google Scholar]
  15. Gorman LA, Sripada RK, Ganocyz D, Walters HM, Bohnert KM, Dalack GW, & Valenstein M. (2016). Determinates of national guard mental health service utilization in VA versus non-VA Settings. Health Services Research, 51(5), 1814–1837. 10.1111/1475-6773.12446 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Holland KJ, Rabelo VC, & Cortina LM (2016). Collateral damage: Military sexual trauma and help-seeking barriers. Psychology of Violence, 6(2), 253–261. 10.1037/a0039467 [DOI] [Google Scholar]
  17. Hoyt T, Rielage JK, & Williams LF (2012). Military sexual trauma in men: Exploring treatment principles. Traumatology, 18, 29–40. 10.1177/1534765611430724 [DOI] [Google Scholar]
  18. Kilpatrick DG, Resnick HS, Milanak ME, Miller MW, Keyes KM, & Friedman MJ (2013). National estimates of exposure to traumatic events and PTSD prevalence using DSM-IV and DSM-5 criteria. Journal of Traumatic Stress, 26, 537–547. 10.1002/jts.21848 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kimerling R, Gima K, Smith MW, Street A, & Frayne S. (2007). The Veterans Health Administration and military sexual trauma. American Journal of Public Health, 97(12), 2160–2166. 10.2105/AJPH.2006.092999 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Kimerling R Pavao J Greene L, Karpenko J, Rodriguez A, Saweikis M, & Washington DL (2015). Access to mental health care among women veterans: Is VA meeting women’s needs? Medical Care, 53, S97–S104. 10.1097/MLR.0000000000000272 [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Kirkner A, Relyea M, & Ullman SE (2018). PTSD and problem drinking in relation to seeking mental health and substance use treatment among sexual assault survivors. Traumatology, 24(1), 1–7. 10.1037/trm0000126 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kroenke K, Spitzer RL, & Williams JB (2003). The Patient Health Questionnaire–2: Validity of a two-item depression screener. Medical Care, 41, 1284–1292. 10.1097/01.MLR.0000093487.78664.3C [DOI] [PubMed] [Google Scholar]
  23. Lanza ST, Collins LM, Lemmon DR, & Schafter JL (2007). PROC LCA: A SAS procedure for latent class analysis. Structural Equation Modeling, 14(4), 671–694. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Lanza ST, Dziak JJ, Huang L, Wagner AT, & Collins LM (2015). LCA Stata plugin users’ guide (Version 1.2). The Methodology Center, Penn State University. https://methodology.psu.edu [Google Scholar]
  25. Light D, & Monk-Turner E. (2009). Circumstances surrounding male sexual assault and rape: Findings from the National Violence Against Women Survey. Journal of Interpersonal Violence, 24(11), 1849–1858. 10.1177/0886260508325488 [DOI] [PubMed] [Google Scholar]
  26. Military Sexual Trauma Support Team. (2018). Summary of military sexual trauma-related outpatient care: Fiscal year 2017. Department of Veterans Affairs, Office of Mental Health Services. [Google Scholar]
  27. Morral AR, Gore KL, & Schell TL (2015). Sexual assault and sexual harassment in the U.S. military. Volume 2. Estimates for Department of Defense service members from the 2014 RAND Military Workplace Study. Rand National Defense Research Institute. https://www.rand.org/pubs/research_reports/RR870z2-1.html [Google Scholar]
  28. Ouimette P, Vogt D, Wade M, Tirone V, Greenbaum MA, Kimerling R, Laffaye C, Fitt JE, & Rosen CS (2011). Perceived barriers to care among Veterans Health Administration patients with posttraumatic stress disorder. Psychological Services, 8(3), 212–223. 10.1037/a0024360 [DOI] [Google Scholar]
  29. Pietrzak RH, Johnson DC, Goldstein MB, Malley JC, & Southwick SM (2009). Perceived stigma and barriers to mental health care utilization among OEF-OIF veterans. Psychiatric Services, 60(8), 1118–1112. 10.1176/ps.2009.60.8.1118 [DOI] [PubMed] [Google Scholar]
  30. Prins A, Ouimette P, Kimerling R, Cameron RP, Hugelshofer DS, Shaw-Hegwer J, Thraikill A, Gusman FD, & Sheikh JI (2003). The Primary Care PTSD Screen (PC-PTSD): Development and operating characteristics. Primary Care Psychiatry, 9(1), 9–14. 10.1185/135525703125002360 [DOI] [Google Scholar]
  31. Possemato K, Johnson EM, Wray LO, Webster B, & Stecker T. (2018). The implementation and testing of a referral management system to address barriers to treatment-seeking among primary care veterans with PTSD. Psychological Services, 15(4), 457–469. 10.1037/ser0000150 [DOI] [PubMed] [Google Scholar]
  32. Revelle W. (1979). Hierarchical cluster analysis and the internal structure of tests. Multivariate Behavioral Research, 14, 57–74. http://personality-project.org/revelle/publications/iclust.pdf See also more extensive documentation at http://personality-project.org/r/r.ICLUST.html [DOI] [PubMed] [Google Scholar]
  33. Schinka JA, & Velicer WF (2012). An introduction to latent class and latent transition analysis. In Weiner IB (Ed.), Handbook for psychology (Volume 2): Research methods in psychology (pp. 691–715). John Wiley & Sons. [Google Scholar]
  34. Smith JR, Workneh A, & Yaya S. (2020). Barriers and facilitators to help-seeking for individuals with posttraumatic stress disorder: A systematic review. Journal of Traumatic Stress, 33(2), 137–150. 10.1002/jts.22456 [DOI] [PubMed] [Google Scholar]
  35. StataCorp. (2017). Stata statistical software: Release 15. StataCorp LLC [Google Scholar]
  36. Stecker T, McHugo G, Xie H, Whyman K, & Jones M. (2014). RCT of a brief phone-based CBT intervention to improve PTSD treatment utilization by returning service members. Psychiatric services (Washington, D.C.), 65(10), 1232–1237. 10.1176/appi.ps.201300433 [DOI] [PMC free article] [PubMed] [Google Scholar]
  37. Swanson SA, Lindenberg K, Bauer S, & Crosby RD (2012). A Monte Carlo investigation of factors influencing latent class analysis: An application to eating disorder research. International Journal of Eating Disorders, 45(5), 677–684. 10.1002/eat.20958 [DOI] [PubMed] [Google Scholar]
  38. Turchik JA, Bucossi MM, & Kimerling R. (2014). Perceived barriers to care and gender preferences among veteran women who experienced military sexual trauma: A qualitative analysis. Military Behavioral Health, 2(2), 180–188. 10.1080/21635781.2014.892410 [DOI] [PubMed] [Google Scholar]
  39. Turchik JA, & Edwards KM (2012). Myths about male rape: A literature review. Psychology of Men & Masculinity, 13(2), 211–226. 10.1037/a0023207 [DOI] [Google Scholar]
  40. Turchik JA, McLean C, Rafie S, Hoyt T, Rosen CS, & Kimerling R. (2013). Perceived barriers to care and provider gender preferences among veteran men who have experienced military sexual trauma: A qualitative analysis. Psychological Services, 10(2), 213–222. 10.1037/a0029959 [DOI] [PubMed] [Google Scholar]
  41. Turchik JA, Pavao J, Hyun J, Mark H, & Kimerling R. (2012). Utilization and intensity of outpatient care related to military sexual trauma for veterans from Afghanistan and Iraq. Journal of Behavioral Health Services & Research, 39(3), 220–233. 10.1007/s11414-012-9272-4 [DOI] [PubMed] [Google Scholar]
  42. Valenstein M, Gorman L, Blow AJ, Ganoczy D, Walters H, Kees M, Pfeiffer PN, Kim HM, Lagrou R, MacDermid Wadsworth S, Rauch SAM, & Dalack GW (2014). Reported barriers to mental health care in three samples of U. S. Army National Guard Soldiers at Three Time Points. Journal of Traumatic Stress, 27(4), 406–414. 10.1002/jts.21942 [DOI] [PubMed] [Google Scholar]
  43. Vogt D, Leone BAL, Wang JM, Sayer NA, & Pineles SL (2014). Endorsed and Anticipated Stigma Inventory (EASI): A tool for assessing beliefs about mental illness and mental health treatment among military personnel and veterans. Psychological Services, 11(1), 105–113. 10.1037/a0032780 [DOI] [PubMed] [Google Scholar]
  44. Williston SK, Roemer L, & Vogt DS (2019). Cultural and service factors related to mental health beliefs among post-9/11 veterans. International Journal of Social Psychiatry, 65(4), 313–321. 10.1177/0020764019842327 [DOI] [PubMed] [Google Scholar]
  45. Wilson LC (2018). The prevalence of military sexual trauma: A meta-analysis. Trauma, Violence, & Abuse, 19(5), 584–597. 10.1177/1524838016683459 [DOI] [PubMed] [Google Scholar]
  46. Zinzow HM, Britt TW, McFadden AC, Burnette CM, & Gillispie S. (2012). f Clinical Psychology Review, 32(8), 741–753. 10.1016/j.cpr.2012.09.002 [DOI] [PubMed] [Google Scholar]
  47. Zinzow HM, Britt TW, Pury CLS, Jennings K, Cheung JH, & Raymond MA (2015). Barriers and facilitators of mental health treatment-seeking in U.S. active duty soldiers with sexual assault histories. Journal of Traumatic Stress, 28(4), 289–297. 10.1002/jts.22026 [DOI] [PubMed] [Google Scholar]

RESOURCES