Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 May 1.
Published in final edited form as: Couns Psychol. 2022 Jan 27;50(4):506–535. doi: 10.1177/00110000221074019

Postdeployment Treatment Gap: Symptoms and Treatment Utilization Among Returning National Guard Soldiers

Simon B Goldberg 1, Anthony W P Flynn 2, Maleeha Abbas 3, Megan E Schultz 4, Michele Hiserodt 5, Kathryn A Thomas 6, Kasey Kallio 7, Mary F Wyman 8
PMCID: PMC9354595  NIHMSID: NIHMS1662562  PMID: 35936651

Abstract

This study evaluated rates of psychiatric symptoms and mental health treatment utilization among National Guard service members during the post-deployment period. National Guard service members (n=311) completed surveys assessing demographics, beliefs about mental health treatment, emotion regulation strategies, and psychiatric symptoms. Mental health treatment utilization was assessed at 6-month follow-up. Post-deployment, 41.2% of the sample had psychiatric symptoms above the clinical cut-off for at least one symptom measure. This proportion increased at follow-up (53.5%). Alcohol use disorder (AUD) symptoms showed the largest increase (d=0.66), although symptoms of depression and posttraumatic stress disorder (PTSD) also showed small magnitude increases. Among those with elevated symptoms post-deployment (n=128), only 27.8% received mental health treatment at follow-up. Severity of depression, anxiety, and PTSD were higher among those who utilized treatment. The post-deployment period is a vulnerable one. Continued efforts to understand and address barriers to treatment for this population are warranted.

Keywords: military veterans, post-deployment, mental health treatment, treatment utilization

High rates of psychiatric symptoms and low rates of mental health treatment utilization among United States military service members involved in Operation Enduring Freedom (OEF) and Operation Iraqi Freedom (OIF) have been documented for over a decade (Hoge et al., 2004; Hoge, Auchterlonie, & Milliken, 2006). In particular, military service members and veterans commonly endorse symptoms of depression, anxiety, posttraumatic stress disorder [PTSD], and alcohol use disorder ([AUD]; Seal et al., 2007; Seal et al., 2011). However, substantial variability has been reported in these estimates. Although the post-deployment period has been identified as a time of vulnerability (Sayer et al., 2010), rates of post-deployment mental health concerns among OEF/OIF service members and veterans have ranged widely. One of the earliest studies was conducted by Milliken et al. (2007) based on a large cohort of returning OEF/OIF soldiers (n = 88,235). They reported that 20.3% of active duty military (i.e., fulltime soldiers) and 42.4% of reservists (e.g., National Guard soldiers) showed elevated psychiatric symptoms reflecting a need for mental health care during the post-deployment period. A strength of this study was the large, population-based sample, although the assessments used were very limited in scope (e.g., two-item depression measure, four-item PTSD measure). Thomas et al. (2010) also assessed a large sample of returning service members (n = 18,305), with estimated rates of depression and PTSD ranging from 18.7% to 33.2%. In contrast, Interian et al. (2012) found that only 11% of returning National Guard soldiers scored positive for PTSD symptoms.

Estimates of post-deployment mental health treatment utilization have also varied considerably. Hoge et al. (2004) found that among returning Army and Marine soldiers who met criteria for major depression, generalized anxiety, or PTSD, between 13% and 21% received help from a mental health professional in the past year. These low rates of utilization differ markedly from Elbogen et al. (2013) who found that 69% of randomly sampled OEF/OIF veterans received psychiatric treatment (i.e., medications or psychotherapy) in the past year.

There are a variety of methodological features that may account for the variability in estimates of mental health concerns and treatment utilization. These include how cases are defined (e.g., symptom cut-offs used, measures used) and military subpopulations sampled (e.g., active component versus reservist), among other factors (e.g., sample demographics; Kok, Herrell, Thomas, & Hoge, 2012; Thomas et al., 2010). Further, it is entirely plausible that estimates are accurately dynamic, and may change in response to a variety of cohort-related variables such as amount and type of combat exposure (Fulton et al., 2015). This highlights the need for ongoing research examining symptoms and treatment utilization that may provide more population-specific (e.g., focused on a particular subpopulation of service members) and cohort-specific (e.g., more recently returning soldiers) estimates that can inform policy and outreach efforts targeted towards a specific group of service members.

Studies investigating psychiatric symptoms and treatment utilization have often used cross-sectional designs (Elbogen et al., 2013; Interian et al., 2012; Kulesza et al., 2015). While cross-sectional designs may facilitate collecting larger sample sizes, longitudinal designs are ideal for studying the development of symptoms post-deployment and investigating longitudinal predictors of treatment utilization. In particular, longitudinal designs allow assessment of changes in symptoms over time (e.g., immediately post-deployment and in the subsequent post-deployment period). A richer understanding of the trajectory of symptom change during the post-deployment period can directly inform intervention and outreach efforts (e.g., highlighting the need for outreach 6 months following deployment if symptoms increase during this period). Further, longitudinal models allow evaluation of factors that predict treatment utilization during the post-deployment period. Such analyses have the potential to clarify factors that may be associated with health disparities (e.g., demographic variables such as gender or race/ethnicity that differentially predict treatment utilization; Wang et al., 2005). These models can also enrich our understanding of what may motivate individuals to utilize treatment (e.g., the presence of elevated symptoms), which can likewise inform efforts to reduce barriers to care (e.g., through screening in primary care settings; Cornwell, Brockmann, Lasky, Mach, & McCarthy, 2018).

The variability in the specific estimates aside, the pairing of high rates of psychiatric symptoms and potentially low rates of treatment utilization represents a treatment gap (Kohn, Saxena, Levav, & Saraceno, 2004), given the availability of evidence-based treatments to address mental health concerns common among returning military service members (e.g., depression, PTSD, AUD; Elbogen et al., 2013; Thomas et al., 2010). This gap is particularly notable within military and veteran populations due to the availability of mental health treatment through military and Veterans Health Administration (VHA) providers coupled with outreach efforts by VHA and other veteran service organizations (National Academies of Sciences, Engineering, and Medicine, 2018; Straits-Tröster et al., 2011).

It is particularly vital that research clarify factors that predict mental health treatment utilization to facilitate targeting utilization efforts and reducing the treatment gap among returning OEF/OIF service members and veterans. Prior research has highlighted a variety of factors that predict post-deployment treatment utilization in these populations, including screening positive for mental health problems (Kehle et al., 2010), psychiatric symptom severity (Meis et al., 2010), readjustment stressors (Interian et al., 2012), race/ethnicity (Zinzow et al., 2015), stigma and career-related worries associated with receiving treatment (Stecker et al., 2010), and beliefs about treatment (Fox et al., 2015; Kehle et al., 2010; Pietrzak et al., 2009). However, as with prevalence and treatment utilization estimates, the precise characteristics that do or do not predict utilization have varied across studies. For example, symptom severity has proven an inconsistent predictor of treatment utilization. Kulesza et al. (2015) found that anxiety and PTSD symptoms positively correlated with treatment utilization in a cross-sectional sample of young adult veterans. Interian et al. (2012) found that depression symptom severity but not PTSD symptom severity positively correlated with likelihood of post-deployment mental health treatment in a cross-sectional sample of National Guard soldiers. Zinzow et al. (2015) found that depression, PTSD, and alcohol problems did not predict utilization in a cross-sectional sample of active duty soldiers with sexual assault histories. Beliefs about mental health treatment is also an inconsistent predictor of treatment utilization. Key beliefs assessed in prior research focus on the likelihood of benefit (e.g., “mental health treatment generally does not work”; Vogt et al., 2014). While some cross-sectional studies have found a positive association between positive beliefs and treatment utilization (Kehle et al., 2010) or a negative association between negative beliefs and treatment utilization (Pietrzak et al., 2009), other studies have failed to find a relationship (e.g., for female OEF/OIF veterans; Fox et al., 2015). These inconsistencies highlight the need for additional research clarifying predictors of utilization. In particular, longitudinal studies using robust symptom measures assessing disorders most relevant for veterans, that are focused on a particular military subpopulation (e.g., National Guard service members) may be particularly valuable for understanding the post-deployment experience of a specific group of veterans.

A further limitation of the existing literature has been the omission of constructs drawn from basic psychological science and social psychology that may prove to be important predictors of utilization. Emotion regulation strategies may be particularly relevant, given that these strategies could influence how an individual responds to and processes adverse experiences that could occur during combat. Gross and John (2003) identify two primary emotion regulation strategies including cognitive reappraisal (i.e., viewing an emotionally evocative situation in a way that influences its impact) and expressive suppression (i.e., inhibiting emotional expression). These constructs have direct relevance to military populations. Cognitive reappraisal strategies are foundational to many efficacious psychotherapies offered to veterans (e.g., cognitive therapy) and endorsement of emotional suppression values (e.g., emotional “toughness”) has been linked to elevated PTSD and depression symptoms among OEF/OIF veterans (Jakupcak, Blais, Grossbard, Garcia, & Okiishi, 2014). Use of these strategies have important psychological, physiological, and interpersonal consequences in the general population (Gross & John, 2003) and have been shown to predict amygdala activation in combat-exposed veterans (Fitzgerald et al., 2017). The link with amygdala activation is consequential given the central role of the amygdala in the neurobiological underpinnings of PTSD symptoms (e.g., hyperresponsivity during the processing of both trauma-related and trauma-unrelated affective information; Shin, Rauch, & Pitman, 2006). It may be that the use of certain strategies (e.g., emotional suppression) leads to a decreased willingness to utilize mental health care. To our knowledge, the potential link between emotion regulation strategies and treatment utilization has not been previously explored but may represent a psychological tendency that promotes or inhibits treatment receipt.

National Guard service members represent a military subpopulation who may face unique challenges as reserve component members returning from deployment (e.g., feeling isolated from the military community, balancing demands of civilian employment and military service, financial hardship; Gorman et al., 2011; Riviere et al., 2011). In 2010, there were 362,015 Army National Guard service members, 26% of whom were racial/ethnic minorities and 14% of whom were female (U.S. Army, 2020). In recent estimates, National Guard service members showed the highest rates of death by suicide across all branches of the military (DoDSER, 2016). Past research suggests that National Guard service members may be at increased risk for mental health problems relative to their active component peers post-deployment (Milliken et al., 2007; Thomas et al., 2010).

The current study thus sought to add to the literature on post-deployment mental health concerns and mental health treatment utilization among National Guard service members. Using a longitudinal design, we investigated changes in symptoms across several of the most common psychiatric conditions within military and veteran populations (i.e., depression, anxiety, AUD, PTSD; Seal et al., 2007, 2011) assessed immediately post-deployment and 6 months later. In addition to assessing factors shown to predict treatment utilization in prior studies (i.e., demographic variables, psychiatric symptoms, treatment beliefs), we also assessed service members’ use of emotion regulation strategies.

Method

Study procedures were approved by Institutional Review Boards at University of Wisconsin - Madison.

Participants and Procedure

Data were collected as part of a larger study of mental health and social functioning among Wisconsin Army National Guard service members who were deployed to combat theaters during 2008–2010. The original study included three assessment points: before deployment (pre-deployment), immediately after return from deployment (post-deployment), and 6 months after deployment (follow-up). At each time point, participants were recruited for voluntary survey participation during pre- and post-mobilization training events held at the Army installation. Participants could complete the survey anonymously or could provide identifiers that would allow their data to be linked across time points. Due to differences in matched sample sizes across time points and based on previous literature identifying the post-deployment period as one of potential vulnerability as well as opportunity for intervention or prevention, the present analysis examines post-deployment and follow-up data exclusively. All participants who had available symptom measures at post-deployment and treatment utilization data at follow-up were included.

Among the primary analytic sample (n = 311), the average age was 29.00 (SD = 8.08). The majority identified as male (82.9%) with the remainder (17.1%) identifying as female (non-binary gender identities were unfortunately not offered as a response option). In terms of race/ ethnicity, the majority of the sample (88.6%) identified as non-Hispanic White, 4.9% as Hispanic, 3.2% as non-Hispanic Black, 1.9% as American Indian or Alaskan Native, and 1.3% as Asian or Pacific Islander. The sample had an average of 13.55 years (SD = 1.73) of formal education. Participants had completed on average 1.55 deployments (SD = 0.67, range = 1 to 4).

At each time point, participants were provided with a verbal description of study procedures, given information regarding confidentiality, and informed they could withdraw from the study at any time without penalty. As noted, anonymous survey participation was offered. Response rates for survey completion were only available for the follow-up assessment and was 62%. No compensation was provided for study participation at post-deployment. At follow-up, participants were eligible to be included in a raffle draw for three cash prizes of $100.

Measures

Demographics.

Demographic data were assessed immediately post-deployment. Participants reported their age, self-identified (binary) gender, race/ethnicity (non-Hispanic White or racial/ethnic minority), and years of education.

Treatment utilization.

Consistent with prior studies in military samples (e.g., Elbogen et al., 2013), we conceptualized mental health treatment as including the use of psychiatric medications or mental health counseling. Two items assessed treatment utilization at follow-up. The first asked “Have you been in counseling or psychotherapy since your most recent deployment?” The second assessed pharmacotherapy, asking participants to indicate if they had taken any medications for mental health (e.g., depression, anxiety) since returning from deployment. For descriptive purposes, rates of utilization are reported separately for these two forms of mental health treatment. However, given our interest in utilization of any form of mental health treatment and in order to reduce the number of statistical tests conducted, responses from both items were combined into a single variable indicating whether any mental health treatment was received for use in longitudinal analyses. This variable was coded dichotomously (i.e., “1” if either psychotherapy or medications was utilized and “0” if neither was utilized).

Psychiatric symptom measures.

Four widely used and validated questionnaires were used to assess psychiatric symptoms within four disorder domains both immediately post-deployment and at follow-up. Established clinical cut-off scores were used to determine presence of symptom elevations.

Depression.

The Beck Depression Inventory (BDI) – II (Beck, Steer, & Brown, 1996) is a 21-item measure assessing symptoms of depression. Participants indicated the degree to which they experienced various symptoms during the past two weeks on a 0 to 3 scale. Item anchors vary across BDI-II items, with 0 reflecting an absence of symptoms (e.g., “I do not feel sad”) and 3 reflecting severe symptoms (e.g., “I am so sad or unhappy that I can’t stand it”). A higher score reflects greater depressive symptomatology. The BDI-II has shown high internal consistency and factorial validity (Dozois, Dobson, & Ahnberg, 1998), including specifically in veteran populations (Palmer et al., 2014). A total score was computed by summing across all items (scale range = 0 to 63). Internal consistency was high in the current sample (α = .96). A score of ≥13 has been recommended as a clinical cut-off optimizing sensitivity and specificity (Dozois et al., 1998) and was used in the present study to denote presence of clinically elevated symptoms.

Anxiety.

The Generalized Anxiety Disorders – 7 (GAD-7; Spitzer, Kroenke, Williams, & Lowe, 2006) is a 7-item measure assessing anxiety symptoms. Participants indicated the degree to which they experienced various symptoms during the past two weeks (e.g., “trouble relaxing”) from 0 (Not at all) to 3 (Nearly every day). The GAD-7 has shown high internal consistency as well as criterion, construct, factorial, and procedural validity (Spitzer et al., 2006). The GAD-7 has been widely used and shown to possess high internal consistency reliability and sensitivity to distress in veteran samples specifically (Rudd, Goulding, & Bryan, 2011). A total score was computed by summing across all items (scale range = 0 to 21). Internal consistency was high in the current sample (α = .91). A score of ≥5 has been recommended as a clinical cut-off (Spitzer et al., 2006).

Alcohol use.

The Alcohol Use Disorders Identification Test (AUDIT; Babor et al., 2001) is a 10-item measure assessing alcohol use and effects. Example items include “how often do you have a drink containing alcohol” and “have you or someone else been injured as a result of your drinking.” Items are scored from 0 to 4, with 0 reflecting the absence of alcohol-related symptoms (e.g., never having a drink, no on being injured) and 4 reflecting elevated symptoms (e.g., drinking 4 or more times a week, someone being injured during the last year). The AUDIT has been widely used to detect problematic alcohol use and has shown concurrent, construct, and discriminant validity (Bohn, Babor, & Kranzler, 1995) and validated specifically in veteran samples (Bradley et al., 2003). Participants were asked to indicate the responses that best fit their drinking during the time they were consuming the most alcohol, either during the most recent deployment (post-deployment) or since returning from the most recent deployment (follow-up). A total score was computed following the published scoring recommendations (scale range = 0 to 40; Babor et al., 2001), with higher scores indicating higher likelihood of problematic alcohol use. Internal consistency was acceptable in the current sample (α = .80). A score of ≥8 has been recommended as a clinical cut-off (Babor et al., 2001).

Posttraumatic stress.

The PTSD Checklist (PCL; Weathers, Litz, Herman, Huska, & Keane, 1993) is a 17-item measure assessing symptoms of PTSD. The PCL-Military (PCL-M; Weathers et al., 1993), which has slightly altered wording for use with a military population, was used in the current study. Participants indicated the degree to which they have been bothered by various symptoms (e.g., “repeated, disturbing memories, thoughts, or images of a stressful experience from the past”) during the past month from 1 (not at all) to 5 (extremely). The PCL-M has shown high internal consistency and predictive validity in National Guard samples specifically (Arbisi et al., 2012). A total score was computed by summing across all items (scale range = 17 to 85), with higher scores indicating higher PTSD symptoms. Internal consistency was high in the current sample (α = .94). A score of ≥50 has been recommended as a clinical cut-off for military samples.

Total number of clinical elevations.

The total number of clinical elevations was computed by summing across elevations within the four symptom domains (scale range = 0 to 4).

Mental health help-seeking beliefs.

Participants’ beliefs about mental health help-seeking were assessed at post-deployment via three dichotomously scored items (yes/no) created for the current survey, but similar in length and content to previous studies examining attitudes about mental health care (e.g., Fischer & Turner, 1970; Mojtabai, 2007). Items were: “I believe that those dealing with mental health issues can be helped by seeking treatment,” “If I personally were dealing with mental health issues, I would seek treatment,” and “I believe that mental health issues are best dealt with on your own, without professional help” (reverse scored). For the current study, we summed item scores (total possible points 0 to 3). Internal consistency reliability was slightly below the recommended range (α = .68).

Emotion regulation strategies.

The Emotion Regulation Questionnaire (ERQ; Gross & John, 2003) was administered immediately post-deployment to assess participants’ tendency to use emotion regulation strategies of reappraisal and suppression. Items were rated on a Likert-type scale ranging from 1 (strongly disagree) to 7 (strongly agree). Six items assessed reappraisal (e.g., “I control my emotions by changing the way I think about the situation I’m in”) and four items assessed suppression (e.g., “I control my emotions by not expressing them”). The ERQ has shown convergent and discriminant validity as well as correlations with measures of well-being and interpersonal functioning (Gross & John, 2003). The ERQ has shown desirable psychometric properties specifically in military veteran samples (Boden et al., 2013). In addition to demonstrating acceptable internal consistency reliability, ERQ scores have been shown to change in response to PTSD treatment, with decreases in emotional suppression predicting decreases in PTSD symptoms post-treatment (Boden et al., 2013). Separate subscale scores were computed by computing the mean across the reappraisal or suppression items (scale range = 1 to 7). Internal consistency reliabilities for each subscale were acceptable (αs = .84, .72, for reappraisal and suppression, respectively).

Analysis Procedures

Statistical analyses were conducted using the R programming language (R Core Team, 2018) and IBM SPSS Version 26 (IBM, 2019). Due to missingness in some variables above that which is typically considered inconsequential (e.g., 5%; see Supplemental Materials Table 1), we created 10 datasets using multiple imputation. The fully conditional specification was implemented in SPSS which is capable of handling both continuous and categorical variables (Enders, 2010). Unlike listwise deletion, the use of multiple imputation avoids loss of statistical power due to missingness and is robust to both missing completely at random (MCAR) and missing at random (MAR) assumptions (Enders, 2010). All statistical tests (i.e., p-values) were based on pooled estimates. For values that SPSS does not automatically pool (e.g., standard deviation for descriptive statistics), we took a simple average of analyses based on the 10 imputed datasets.

McNemar’s test for paired dichotomous data and paired t-tests were used to assess changes in psychiatric symptoms from post-deployment to follow-up. Cohen’s (1988) d was used as a standardized effect size reflecting symptom change and was calculated using methods common in meta-analysis (e.g., matched groups d; Borenstein, Hedges, Higgins, & Rothstein, 2009). We interpreted the magnitude of d based on Cohen’s (1988) guidelines (i.e., ds = 0.20, 0.50, 0.80 reflect small, medium, and large effects, respectively). A correlation matrix was constructed showing the relationships between post-deployment variables with one another and treatment utilization at follow-up. A multivariate logistic regression model was used to examine treatment utilization at follow-up from all post-deployment variables in order to determine which, if any, predictors appeared most robust. Models were conducted on both the full sample and restricted to participants who met any cut-offs for elevated psychopathology at post-deployment. We calculated Nagelkerke’s R2 as a measure of variance explained and report odds ratios for interpretation of model coefficients. Based on the full sample (n = 311), we had 80% power to detect post-deployment to follow-up changes of d ≥ 0.16 and differences between groups (i.e., treatment utilizers vs. non-utilizers) of d ≥ 0.32. Based on the sample with clinically elevated symptoms (n = 128), we had 80% power to detect post-deployment to follow-up changes of d ≥ 0.25 and differences between groups (i.e., treatment utilizers vs. non-utilizers) of d ≥ 0.50. Power calculations were conducted using the ‘pwr.t.test’ function of type “paired” or “two.sample” in the ‘pwr’ package in R (Champley, 2018).

Results

Psychiatric Symptoms Immediately Post-deployment and at 6-month Follow-up

Psychiatric symptoms assessed at post-deployment and at follow-up are reported in Table 1. At post-deployment, 41.2% of the sample had psychiatric symptoms above the clinical cut-off for at least one of the four symptom measures. Participants reported a mean of 0.66 (SD = 0.95) clinical elevations. Anxiety symptoms were the most commonly elevated (26.7%) and PTSD symptoms were the least commonly elevated (8.7%).

Table 1.

Psychiatric symptoms immediately post-deployment and at 6-month follow-up

Variable Post-deployment Follow-up
Elevated Mean SD Med Mode Min Max Elevated Mean SD Med Mode Min Max d p
Total # elevations 0.66 0.95 0.00 0.00 0 4 1.04 1.25 1.00 0.00 0 4 0.36 .019*
Depression 15.9% 5.89 8.36 2.09 0.00 0 48 21.0% 7.43 9.33 4.00 0.00 0 54 0.17 .001**
Anxiety 26.7% 3.50 4.27 2.00 0.00 0 20 27.1% 3.94 4.68 2.82 0.00 0 21 0.10 .098
Alcohol use 12.9% 3.42 4.81 3.42 1.00 0 32 33.6% 6.96 5.88 6.96 5.00 0 34 0.66 < .001***
PTSD 8.7% 26.85 11.87 22.00 17.00 17 80 8.6% 28.26 12.64 23.10 17.00 17 82 0.12 .020*
Open in a new tab

Note: Post-deployment = immediately post-deployment; Follow-up = 6 months post-deployment; Elevated = symptoms above clinical cut-off; SD = standard deviation; Med = median; Min = minimum; Max = maximum; d = Cohen’s (1988) standardized mean difference computed as follow-up minus post-deployment (i.e., a positive d reflects increased symptoms) with pooled standard deviation (Cooper et al., 2009); p = p-value for paired t-test assessing change in symptoms from post-deployment to follow-up; Total # elevations = total number of symptom measures with clinical elevations (scale range = 0 to 4); Depression = Beck Depression Inventory – II (BDI-II; scale range = 0 to 63); Anxiety = Generalized Anxiety Disorder – 7 (GAD-7; scale range = 0 to 21); Alcohol use = Alcohol Use Disorders Identification Test (AUDIT; scale range = 0 to 40); PTSD = PTSD Checklist – Military (PCL-M; scale range = 17 to 85).

*

p < .05

**

p < .01

***

p < .001

At follow-up, 53.5% of the sample endorsed clinically significant psychiatric symptoms in at least one of the four symptom domains. Participants reported a mean of 1.04 (SD = 1.25) clinical elevations. AUD symptoms were most commonly elevated (33.6%) and PTSD symptoms were the least commonly elevated (8.6%).

There was a significant increase at follow-up in the proportion reporting elevations above the clinical cut-off on at least one of the four symptom measures (McNemar’s χ2[1] = 14.35, p < .001). Participants also increased in the number of clinical elevations (d = 0.36, p = .019).1 The largest increase in symptom total score was for AUD symptoms (d = 0.66, p < .001), although very small magnitude increases were also observed for depressive symptoms (d = 0.17, p = . 001) and PTSD symptoms (d = 0.12, p = .020). No changes in anxiety symptoms were noted (d = 0.10, p = .098).

Treatment Utilization at Follow-up

A minority of the full sample (19.2%) reported receiving mental health treatment since return from deployment. Among those reporting clinically elevated symptoms immediately post-deployment (n = 128), 27.8% reported receiving mental health treatment within 6 months post-deployment. Among those reporting clinically elevated symptoms at follow-up (n = 153), 31.6% reported receiving mental health treatment within 6 months post-deployment. Utilization rates at follow-up were higher for psychotherapy than for medications within the full sample (16.1% vs. 11.5%, McNemar’s χ2 [1] = 4.92, p = .030), among those with clinically elevated symptoms at follow-up (26.7% vs. 17.2%, McNemar’s χ2 [1] = 7.14, p = .011), but not among those with clinically elevated symptoms post-deployment (24.3% vs. 17.2%, p = .074).

A correlation matrix of associations between post-deployment variables and treatment utilization is displayed in Table 2. As can be seen, several post-deployment variables showed bivariate associations with the likelihood of treatment utilization at follow-up.2 Among demographic variables, non-Hispanic White race/ethnicity was associated with a lower likelihood of treatment utilization (r = −.13, p = .022). No other demographic variables (age, gender, education) were correlated with treatment seeking. Total number of clinical elevations (r = .26, p < .001) as well as severity of depression (r = .24, p < .001), anxiety (r = .24, p < .001), and PTSD (r = .35, p < .001) symptoms were positively correlated with treatment utilization. Mental health help seeking beliefs and emotion regulation strategies were not correlated with treatment utilization.

Table 2.

Correlation matrix of post-deployment variables and treatment utilization at follow-up

1 2 3 4 5 6 7 8 9 10 11 12 13
1. Tx Utilization 1
2. Age .07 1
3. Male .03 .20** 1
4. Non-Hispanic White −.13* .06 .05 1
5. Education −.03 .25** −.06 .10 1
6. Depression .24** .03 .04 −.04 −.01 1
7. Anxiety .24** .01 −.02 −.03 .02 .64** 1
8. Alcohol use .06 −.06 .08 .02 −.05 .13* .06 1
9. PTSD .35** .07 .12* −.16* −.03 .66** .72** .12* 1
10. Total # elevations .26** .02 .08 −.03 −.05 .72** .75** .34** .75** 1
11. Health beliefs .04 .03 −.01 −.02 .17 −.24* −.03 −.13 −.06 −.15 1
12. Reappraisal .02 .11 −.04 −.05 .04 −.16 −.09 −.23 −.11 −.17 −.01 1
13. Suppression −.04 .13 .09 .01 −.03 .15 −.02 −.16 .00 .06 −.09 .06 1
Open in a new tab

Note: Correlation matrix based on pooled results using 10 multiply imputed datasets. For correlations between continuous variables, values represent Pearson’s r. For correlations between a continuous variable and a dichotomous variable, values represent point biserial correlation. For correlations between two dichotomous variables, value represents phi. Education = years of education; Depression = Beck Depression Inventory – II (BDI-II; scale range = 0 to 63); Anxiety = Generalized Anxiety Disorder – 7 (GAD-7; scale range = 0 to 21); Alcohol use = Alcohol Use Disorders Identification Test (AUDIT; scale range = 0 to 40); PTSD = PTSD Checklist – Military (PCL-M; scale range = 17 to 85); Health beliefs = mental health help seeking beliefs (scale range = 0 to 3); Reappraisal and Suppression = subscales of Emotion Regulation Questionnaire (ERQ; scale range = 1 to 7).

*

p < .05

**

p < .01

***

p < .001

We evaluated the assumptions of logistic regression which include a dichotomous outcome, a linear relationship between predictor variables and the logit of the outcome, and the absence of multicollinearity when including multiple predictors (Cohen et al., 2003). Our outcome was dichotomous (i.e., treatment utilization). Scatterplots of significant continuous predictors of treatment utilization and the logit of treatment utilization appeared generally linear (Supplemental Materials Figure 1). As SPSS does not allow estimation of the variance inflation factor (VIF) for logistic regression, this was estimated using the original data set in R. In multivariate models with all predictors, VIF values suggested multicollinearity was not a substantial concern in either the full or clinical samples (VIFs < 3.5; Cohen et al., 2003).

A multivariate model that included demographics, psychiatric symptoms, mental health help seeking beliefs, and emotion regulation strategies explained 19.8% of variance in treatment utilization at follow-up (Nagelkerke R2 = 0.198; Table 3). However, only PTSD symptoms emerged as a statistically significant predictor, with the odds ratio indicating that a one unit increase in PTSD symptoms was associated with a 6% increase in likelihood of treatment utilization at follow-up (odds ratio = 1.06, p = .003). To further characterize the magnitude of this effect, a model was estimated with PTSD symptoms excluded. Excluding PTSD was associated with a 0.048 decrease in R2. A subsequent model examined the same full set of predictors but was restricted to participants with clinically elevated symptoms at post-deployment (n = 128; Table 4). This model explained 22.3% of variance in treatment utilization at follow-up (Nagelkerke R2 = 0.223). However, none of the predictors were statistically significant, although PTSD symptoms showed a trend-level association with treatment utilization (odds ratio = 1.05, p = .061). To allow comparison with results from the full sample, as before a model was re-estimated with PTSD symptoms excluded. Excluding PTSD was associated with a 0.039 decrease in R2.

Table 3.

Multivariate logistic regression model predicting treatment utilization at follow-up from post-deployment variables in the full sample (n = 311)

Predictor B SE Wald df p Odd Ratio
Intercept −2.76 2.24 3.17 1 .223 0.06
Age 0.02 0.02 1.14 1 .350 1.02
Male −0.22 0.46 0.3 1 .627 0.8
Non-Hispanic White −0.74 0.45 2.85 1 .104 0.48
Education −0.05 0.1 0.27 1 .629 0.95
Depression 0.02 0.03 0.93 1 .411 1.02
Anxiety −0.03 0.06 0.36 1 .630 0.97
Alcohol use 0.02 0.04 0.59 1 .596 1.02
PTSD 0.06 0.02 10.26 1 .003** 1.06
Reappraisal 0.03 0.26 2.12 1 .912 1.03
Suppression −0.12 0.2 2.04 1 .552 0.89
Health beliefs 0.66 0.85 1.66 1 .442 1.94
Open in a new tab

Note: Nagelkerke R2 = 0.198. Pooled results based on 10 multiply imputed datasets. Education = years of education; Depression = Beck Depression Inventory – II (BDI-II; scale range = 0 to 63); Anxiety = Generalized Anxiety Disorder – 7 (GAD-7; scale range = 0 to 21); Alcohol use = Alcohol Use Disorders Identification Test (AUDIT; scale range = 0 to 40); PTSD = PTSD Checklist – Military (PCL-M; scale range = 17 to 85); Reappraisal and Suppression = subscales of Emotion Regulation Questionnaire (ERQ; scale range = 1 to 7); Health beliefs = mental health help seeking beliefs (scale range = 0 to 3).

*

p < .05

**

p < .01

***

p < .001

Table 4.

Multivariate logistic regression model predicting treatment utilization at follow-up from post-deployment variables in the sample with clinically elevated symptoms at post-treatment (n = 128)

Predictor B SE Wald df p Odd Ratio
Intercept −3.97 2.86 2.81 1 .167 0.02
Age 0.04 0.03 2.27 1 .167 1.05
Male −1.05 0.69 2.66 1 .128 0.35
Non-Hispanic White −0.93 0.67 2.06 1 .165 0.40
Education 0.01 0.13 0.02 1 .957 1.01
Depression 0.04 0.03 2.11 1 .239 1.04
Anxiety −0.02 0.07 0.17 1 .793 0.98
Alcohol use 0.02 0.05 0.35 1 .699 1.02
PTSD 0.04 0.02 3.98 1 .061 1.05
Reappraisal 0.02 0.27 0.86 1 .932 1.02
Suppression 0.00 0.26 0.87 1 .999 1.00
Health beliefs 1.28 1.1 2.27 1 .248 3.58
Open in a new tab

Note: Nagelkerke R2 = 0.223. Pooled results based on 10 multiply imputed datasets. Education = years of education; Depression = Beck Depression Inventory – II (BDI-II; scale range = 0 to 63); Anxiety = Generalized Anxiety Disorder – 7 (GAD-7; scale range = 0 to 21); Alcohol use = Alcohol Use Disorders Identification Test (AUDIT; scale range = 0 to 40); PTSD = PTSD Checklist – Military (PCL-M; scale range = 17 to 85); Reappraisal and Suppression = subscales of Emotion Regulation Questionnaire (ERQ; scale range = 1 to 7); Health beliefs = mental health help seeking beliefs (scale range = 0 to 3).

*

p < .05

**

p < .01

***

p < .001

Discussion

The current study examined psychiatric symptoms and mental health treatment utilization among National Guard service members immediately post-deployment and 6 months following return from deployment. Immediately post-deployment, a large proportion of the sample (41.2%) reported symptoms meeting the clinical cut-off for at least one of four symptom categories (depression, anxiety, AUD, PTSD). At follow-up, the majority of the sample (53.5%) reported symptoms meeting the clinical cut-off for at least one of four symptom categories. These rates are significant, particularly given a potential of under-reporting of symptoms (i.e., due to social desirability bias) and the possibility that individuals with severe psychopathology may not have been present to complete the surveys. The largest increase in symptoms were observed for AUD (d = 0.66), although very small but statistically significant increases in depression (d = 0.17) and PTSD (d = 0.12) were also observed. Consistent with prior studies (Milliken et al., 2007; Riviere et al., 2011; Thomas et al., 2010), this pattern of findings demonstrates the high rates of psychiatric symptoms immediately upon return from deployment among service members and highlights 6 months post-deployment also as a vulnerable period for this population. The notable increase in AUD symptoms endorsed during the period of heaviest use after return from deployment is particularly striking, with 33.6% of the sample reporting clinically elevated symptoms at follow-up. Several prior studies have highlighted elevations in AUD symptoms among National Guard service members in particular (Blow et al., 2013; Thomas et al., 2010). It cannot be ruled out that the increase in AUD rates reflect reduced access to alcohol during deployment in certain regions, and subsequently increased access in the home community post-deployment.

Despite high rates of psychiatric symptoms, utilization with mental health treatment was relatively modest during the post-deployment period. Only 27.8% of the 128 participants with clinically elevated symptoms immediately post-deployment engaged with mental health treatment within 6 months following their return home. While many service members do receive treatment, the majority of those with elevated symptoms do not. As an important point of comparison, a recent nationally representative survey of men in the general population found that 41.0% of those experiencing anxiety and depression symptoms received treatment in the previous year (i.e., took medication for anxiety or depression, spoke with a mental health professional; Blumberg, Clarke, & Blackwell, 2015). Thus, the treatment gap for National Guard service members appears to be larger than men generally. This is particularly notable, as veterans have mental health resources available (e.g., Veterans Health Administration [VHA] or other military-related health care options), making it is less likely that financial concerns played a role in low rates of utilization. Indeed, per the Defense Authorization Act of 2008, the vast majority of National Guard service members who serve on active duty are entitled to 5-years of VHA benefits (Veterans Health Administration, 2011). Nonetheless, National Guard service members may face unique barriers to accessing treatment after a combat deployment that are greater than those faced in the general population, such as increased difficulty taking time away from civilian employment post-deployment (Stecker et al., 2010). Logistical barriers may be present for Wisconsin National Guard service members living in rural areas (Stecker et al., 2013).

Aside from psychiatric symptom severity,3 only race/ethnicity predicted later treatment utilization, and this was seen only in the full sample (i.e., not when restricted to those with clinically elevated symptoms, potentially due to decreased statistical power in the subsample).4 The general lack of non-symptom predictors of treatment utilization supports universal efforts to screen for mental health problems and facilitate treatment access for National Guard service members. The fact that race/ethnicity predicted treatment utilization in the full sample is intriguing, particularly given that the direction of this effect suggested that racial/ethnic minority National Guard service members may be more likely than their non-Hispanic White peers to utilize treatment. This finding is in stark contrast to disparities in health care utilization observed in the general population (e.g., lower rates of mental health care among racial/ethnic minorities; Wang et al., 2005). However, this finding is in keeping with several recent, large-scale analyses that suggest racial/ethnic disparities in mental health treatment may be reduced or even reversed among veterans, with racial/ethnic minority veterans engaging with mental health treatment at similar or higher rates than non-Hispanic White veterans (Bensley et al., 2017; Glass et al., 2010; Goldberg et al., 2020).5

The current results contrast prior cross-sectional studies demonstrating a link between beliefs about treatment and treatment utilization (Fox et al., 2015; Stecker et al., 2013). While it is possible that beliefs about treatment are truly not associated with treatment utilization in National Guard service members when examined longitudinally, it is perhaps more likely that our assessment missed key aspects of these beliefs. In particular, the current study did not directly assess mental health stigma (e.g., perceived public stigma related to mental health treatment seeking, “My peers might treat me differently”; Kulesza et al., 2015) which has been more widely shown to predict treatment utilization in both veteran and civilian populations (Kulesza et al., 2015; Vogel, Wade, & Hackler, 2007; Vogt, 2011). In contrast, the lack of association between cognitive reappraisal and emotional suppression strategies with all study variables, including treatment utilization, may imply they are not in fact particularly important constructs for understanding patterns of distress and treatment utilization in this population. Other sources of variance (e.g., combat experience, social support) may be more important and worth investigating further in future work.

Limitations

Several important limitations are worth noting. As the current sample was restricted to Wisconsin Army National Guard service members, findings may not generalize to other military or veteran samples. For example, the high rates of AUD symptom elevations in the current sample (33.6% at follow-up) are considerably higher than past-year probable AUD estimates in the general veteran population (14.8%; Fuehrlein et al., 2016). This may be due to differences between recently returned National Guard and other veterans, but may also be related to the culture of heavy alcohol use in Wisconsin (Wisconsin Department of Health Services, 2018) or our assessment of symptoms during the period of heaviest use.6 Other aspects of the results (e.g., psychiatric symptoms, treatment utilization) may be influenced by practices within the Wisconsin National Guard (e.g., offering of reintegration programming; Gunter-Hunt et al., 2013). On the other hand, it is likely that some aspects of the Wisconsin National Guard experience (e.g., challenges with reintegration into family life; Lapp et al., 2010) are not unique to Wisconsin or the National Guard but rather are common across service members (Bowling & Sherman, 2008). Low racial/ethnic and gender diversity in the current sample also limits generalizability. The reversed racial/ethnic disparity in treatment utilization observed in the full sample should be interpreted cautiously as a result and replicated in a future sample with a larger number of racial/ethnic minority participants. The unknown response rate for the post-deployment assessment further compounds questions of generalizability and is not possible to evaluate. It is also possible that individuals with severe symptoms were not present at the reintegration events, leading to under-estimates of psychiatric morbidity. It may be that psychiatric symptoms were under-reported immediately post-deployment, given the military context in which they were being collected (although the high rates of symptom endorsement argue against this possibility).

Internal consistency reliability was below the recommended cut-off for the mental health help seeking measure which may have reduced the accuracy of ratings and attenuated associated with treatment utilization. While examination of the average inter-item correlations (rs = .32 to .50) suggests the items showed a recommended degree of internal consistency based on this alternative metric (rs = .10 to .50; Clark & Watson, 1995), it still would have been preferable to include a longer and previously validated mental health help seeking measure (e.g., Vogt et al., 2014). In addition, key predictors of treatment utilization assessed in previous studies (e.g., mental health stigma; Kulesza et al., 2015) were not included.

Implications for Practice, Advocacy, Education/Training, and Research

Practice.

At a basic level, our findings highlight the post-deployment period as a period of elevated risk for National Guard service members in terms of mental health symptomatology. Because early intervention may reduce the persistence of mental illness (Goldberg et al., 2019; Wang et al., 2005), the initial months after the return home may be a critical time for intervention. The high rates of psychiatric symptoms, which increased at follow-up, highlights potential difficulties associated with reintegration following a combat experience (Sayer et al., 2010) and supports continued attention to primary and secondary prevention efforts. These might include family-focused interventions (Sherman & Larsen, 2018) and other approaches aimed at supporting the tasks of reintegration (e.g., creating shared meaning, redefining roles, managing strong emotions; Bowling & Sherman, 2008).

Advocacy.

Given the treatment gap observed, ongoing efforts to engage returning service members in treatment are warranted. Both the Department of Defense and the VHA have implemented relevant programs and procedures in recent years, including post-discharge outreach efforts (Straits-Tröster et al., 2011), peer support programs (Hebert et al., 2008), universal post-deployment mental health screening (Hoge et al., 2006), integration of mental health services within primary care (Cornwell et al., 2018), and public awareness campaigns (e.g., Veterans’ Crisis Line; Knox et al., 2012). As reported by Goldberg et al. (2019), the reduced delay to mental health treatment among post-9/11 veterans relative to pre-9/11 veterans may be due to some of these factors. Efforts to increase public awareness and decrease stigma of PTSD may be part of why associations appeared particularly robust for PTSD symptoms, at least in the full sample. Rigorous epidemiological and ideally experimental studies evaluating specific efforts can help clarify the most effective methods for increasing treatment utilization.

Research.

It is important to continue to identify barriers to treatment amenable to intervention (Mojtabai et al., 2011). Such barriers may vary across military subpopulation, demographics, and a variety of other factors, necessitating research conducted with sensitivity to individual- and group-level differences. Our results suggest that beliefs about mental health treatment utilization are not a primary barrier to access for National Guard service members, at least as assessed in the current sample. However, future studies in this population should clarify the role of stigma as a predictor of treatment utilization, along with other candidate barriers identified previously (e.g., logistical barriers, emotional readiness, beliefs about treatment efficacy; Stecker et al., 2010; Stecker et al., 2013). It would be valuable to assess other military-specific factors that may relate to willingness to seek treatment such as endorsement of military beliefs (e.g., ideal of mental toughness; Zinzow et al., 2013), aspects of veteran identity (e.g., warrior ethos; Hack, DeForge, & Lucksted, 2017), and post-deployment social support (Polusny et al., 2011). Mobile health approaches (e.g., delivery through telehealth, smartphone applications; Knox et al., 2012; Sherman & Larsen, 2018) may be particularly relevant for addressing distance from care facilities as a logistical barrier among National Guard service members living in rural areas.

It is certain the experience of military service and post-deployment varies based on service members’ identity variables. Our results indicated, for example, that racial/ethnic minority status was associated with increased treatment utilization. As found in previous work (Glass et al., 2010; Goldberg et al., 2020), military veterans may be a population who do not demonstrate racial/ethnic disparities in mental health treatment engagement found in the general population. Future studies should examine what proximal factors are responsible for this reduced disparity (e.g., access to military health care coverage, universal screening; Bailey et al., 2017; Hoge et al., 2014), to inform broader efforts to reduce health disparities in the general population. It is crucial that future work closely examine the experiences of racial/ethnic minority National Guard service members and other military veterans to more fully characterize their military service and post-deployment experiences. Such efforts could build upon prior work showing that experiences of racism during military service account for variance in PTSD symptoms in Asian American Vietnamese veterans (Loo et al., 2001). Work in this area can be used to guide intervention and prevention strategies designed at creating a more inclusive and equitable military experience.

Education and Training.

Ultimately, insights gained from ongoing research in this area can be integrated into formal education and training opportunities in order to increase treatment utilization among military veterans. Clinicians and trainees can become better attuned to factors that may promote or inhibit veterans’ likelihood of engaging in treatment, as well as involved in efforts to reduce barriers that may be modifiable (e.g., stigma reduction, novel treatment delivery efforts).

In summary, the current study adds to the literature as one of the few to use a longitudinal design to examine post-deployment psychiatric symptoms and treatment utilization among National Guard service members, who may be particularly vulnerable during this period (Thomas et al., 2010; Riviere et al., 2011). Results highlight post-deployment as a period of increased risk to mental health. This is especially concerning in light of the large number (72.2%) of National Guard service members in our sample with elevated symptoms post-deployment who did not receive care at follow-up. Clarifying treatment barriers specific to this group and continuing efforts to engage returning service members in mental health treatment continue to be critical activities to reducing the psychiatric risks associated with military deployment.

Supplementary Material

1

Significance of the Scholarship to the Public:

The post-deployment period is a vulnerable one for National Guard service members. The combination of high rates of clinically elevated symptoms, increasing symptoms during the post-deployment period, and low of utilization of mental health treatment in this population highlights opportunities for advocacy, prevention, and treatment delivery innovation.


Acknowledgments

The authors report no conflicts of interest. M.F.W. was supported in part by fellowship funding from the U.S. Department of Veterans Affairs. S.B.G. was supported by the University of Wisconsin-Madison, Office of the Vice Chancellor for Research and Graduate Education with funding from the Wisconsin Alumni Research Foundation. Research reported in this publication was also supported by the National Center For Complementary & Integrative Health of the National Institutes of Health under Award Number K23AT010879 to S.B.G. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health. The authors wish to acknowledge the time and effort put forth by the service member respondents and study team members at the Zablocki VA Medical Center, Milwaukee, and W.S. Middleton Memorial Veterans Hospital, Madison, WI, and the resources provided by those institutions. The authors acknowledge that the data set used for this study belongs to the VA. The authors are grateful to Scott A. Baldwin for providing statistical consultation. This content does not represent the views of the U.S. Department of Veterans Affairs or the United States Government.

Author Biographies

Simon B. Goldberg, PhD, is an assistant professor in the Department of Counseling Psychology and affiliate faculty at the Center for Healthy Minds at the University of Wisconsin – Madison. His research focuses on psychotherapy, with a particular emphasis on meditation-based interventions and ways of increasing access to mental health care. He is currently completing an NIH-funded K23 award focused on the delivery of meditation training via mobile health technology.

Anthony W. P. Flynn, BA, is a doctoral student in counseling psychology at the University of Wisconsin-Madison. Prior to coming to Madison, Anthony worked as a study coordinator at the VA Desert Pacific Mental Illness Research and Education Center. His research interests include health and mental health disparities among gender and sexual minority populations and US military veterans, psychotherapy process and outcomes, and psychosocial HIV prevention efforts.

Maleeha Abbas, PhD, received her PhD in counseling psychology from the University of Wisconsin – Madison. She completed her internship and fellowship at the VA Puget Sound Health Care System. She is a specialist in the evidence-based treatment of PTSD, anxiety and related problems in adults at the Evidence Based Treatment Centers of Seattle.

Megan E Schultz, BA, is a doctoral student in clinical psychology at the University of Washington-Seattle. Her research focuses on the impact of stress on the etiology of substance use.

Michele Hiserodt, MA completed her undergraduate studies at the University of Wisconsin-Madison and received her Master’s from Boston University. She is currently the research coordinator at the Translational Research Program in the Center for Anxiety and Related Disorders at Boston University.

Kathryn A. Thomas, M.Ed., is a doctoral candidate in Counseling Psychology at University of Wisconsin-Madison. She is currently completing her predoctoral internship at the Yale School of Medicine in the Division of Law and Psychiatry. Her research interests include suicide prevention, the role of meaning making in recovery from PTSD, and the psychological impact of wrongful convictions.

Kasey C, Kallio, MSN, received his Masters of Science in Nursing from Marquette University. Most recently, Kasey has specialized in mental health nursing through the Rogers Behavioral Health system in West Allis, Wisconsin. His research interests include schizophrenia treatments and therapies, as well as trauma and addiction.

Mary F. Wyman, PhD trained as a clinical psychologist at Indiana University and the University of California-San Francisco. She is a VA Career Development Award recipient based in the Geriatric Research, Education, and Clinical Center at the Madison, WI Veterans Hospital, and is a Clinical Adjunct Associate Professor and funded investigator in the Department of Psychiatry at the University of Wisconsin.

Footnotes

1

Note that p-values are derived from statistical tests and do not reflect a test of the significance of the Cohen’s d effect size itself. This and subsequent p-values in this paragraph are based on paired t-tests which were used for continuous outcome measures, as described in the Methods section.

2

A reviewer raised the question of whether emotion regulation strategies and mental health help-seeking beliefs moderate the association between post-deployment symptoms and treatment utilization at follow-up. We conducted these exploratory analyses and found that reappraisal, suppression, and help-seeking beliefs did not moderate this association in the full sample or the subsample with clinical elevations (ps > .050).

3

Although our evaluation of the variance inflation factor (VIF) suggests multicollinearity did not unduly bias results (i.e., below serious multicollinearity; Cohen et al., 2003), there were large correlations between PTSD, depression, and anxiety symptoms at post-deployment (r ≥ .64). Thus, the change in statistical significance of depression and anxiety as predictors of treatment utilization when controlling for PTSD should be interpreted cautiously, as some degree of variance inflation is likely.

4

PTSD symptoms were similarly reduced to non-significance as a predictor of treatment utilization in the sample with clinically elevated symptoms. As the effect size was similar across both the full and clinical samples, this change in statistical significance may also be due to reduced statistical power.

5

In response to a reviewer’s suggestion, we examined whether higher treatment utilization among racial/ ethnic minority National Guard service members was related to higher distress at post-deployment (i.e., whether differences in treatment utilization were mediated by distress). Examination of the correlation matrix (Table 2) indicates that non-Hispanic White race/ethnicity was in fact negatively correlated with PTSD symptoms. We then conducted a logistic regression model predicting treatment utilization from both race/ethnicity and PTSD symptoms. Results indicated that PTSD symptoms remained a significant predictor of treatment utilization (odds ratio = 1.06, p < .001) while non-Hispanic White race/ethnicity did not (odds ratio = 0.50, p = .098). This pattern is consistent with the theoretical possibility that higher post-deployment distress mediates the link between race/ethnicity and treatment utilization at follow-up.

6

Data from the Centers for Disease Control and Prevention (n. d.) indicates that Wisconsin has the second highest rate of binge drinking among the fifty states (24.4%).

Contributor Information

Simon B. Goldberg, University of Wisconsin-Madison

Anthony W. P. Flynn, University of Wisconsin-Madison

Maleeha Abbas, VA Puget Sound Health Care System, Seattle Division.

Megan E. Schultz, William S. Middleton Memorial Veterans Hospital

Michele Hiserodt, William S. Middleton Memorial Veterans Hospital.

Kathryn A. Thomas, University of Wisconsin-Madison

Kasey Kallio, William S. Middleton Memorial Veterans Hospital.

Mary F. Wyman, William S. Middleton Memorial Veterans Hospital

References

  1. Arbisi PA, Kaler ME, Kehle-Forbes SM, Erbes CR, Polusny MA, & Thuras P (2012). The predictive validity of the PTSD Checklist in a nonclinical sample of combat-exposed National Guard troops. Psychological Assessment, 24(4), 1034–1040. 10.1037/a0028014 [DOI] [PubMed] [Google Scholar]
  2. Babor TF, Higgins-Biddle JC, Saunders JB, & Monteiro MG (2001). The Alcohol Use Disorders Identification Test: Guidelines for use in primary health care. (Second Ed.) Geneva: World Health Organization. [Google Scholar]
  3. Bailey ZD, Krieger N, Agénor M, Graves J, Linos N, & Bassett MT (2017). Structural racism and health inequities in the USA: Evidence and interventions. The Lancet, 389(10077), 1453–1463. 10.1016/S0140-6736(17)30569-X [DOI] [PubMed] [Google Scholar]
  4. Beck AT, Steer RA, & Brown GK (1996). Beck Depression Inventory manual (2nd ed.). San Antonio, TX: Psychological Corporation. [Google Scholar]
  5. Blow AJ, Gorman L, Ganoczy D, Kees M, Kashy DA, Valenstein M, Marcus SM, Fitzgerald HE, & Chermack S (2013). (2013). Hazardous drinking and family functioning in National Guard veterans and spouses post-deployment. Journal of Family Psychology, 27(2), 303–313. 10.1037/a0031881 [DOI] [PubMed] [Google Scholar]
  6. Bohn MJ, Babor TF, & Kranzler HR (1995). The Alcohol Use Disorders Identification Test (AUDIT): validation of a screening instrument for use in medical settings. Journal of Studies on Alcohol, 56(4), 423–432. [DOI] [PubMed] [Google Scholar]
  7. Boden MT, Westermann S, McRae K, Kuo J, Alvarez J, Kulkarni MR, Gross JJ, & Bonn-Miller MO (2013). Emotion regulation and posttraumatic stress disorder: A prospective investigation. Journal of Social and Clinical Psychology, 32(3), 296–314. [Google Scholar]
  8. Borenstein M, Hedges LV, Higgins JPT, & Rothstein HR (2009). Introduction to meta-analysis. New York: Wiley. [Google Scholar]
  9. Bowling UB, & Sherman MD (2008). Welcoming them home: Supporting service members and their families in navigating the tasks of reintegration. Professional Psychology: Research and Practice, 39(4), 451–458. 10.1037/0735-7028.39.4.451 [DOI] [Google Scholar]
  10. Blumberg SJ, Clarke TC, & Blackwell DL (2015). Racial and ethnic disparities in men’s use of mental health treatments. Hyattsville, MD: National Center for Health Statistics. [PubMed] [Google Scholar]
  11. Bradley KA, Bush KR, Epler AJ, Dobie DJ, Davis TM, Sporleder JL, Maynard C, Burman ML, & Kivlahan DR (2003). Two brief alcohol-screening tests From the Alcohol Use Disorders Identification Test (AUDIT): validation in a female Veterans Affairs patient population. Archives of Internal Medicine, 163(7), 821–829. [DOI] [PubMed] [Google Scholar]
  12. Centers for Disease Control and Prevention (n.d.). Data on excessive drinking. Retrieved on April 19, 2020 from: https://www.cdc.gov/alcohol/data-stats.htm
  13. Champely S (2018). pwr: Basic Functions for Power Analysis. R package version 1.2–2. https://CRAN.R-project.org/package=pwr [Google Scholar]
  14. Clark LA, Watson D (1995). Constructing validity: Basic issues in objective scale development. Psychological Assessment, 7(3), 309–319. [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Cohen J (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale, NJ: Erlbaum. [Google Scholar]
  16. Cohen J, Cohen P, West SG, & Aiken LS (2003). Applied multiple regression / correlation analysis for the behavioral sciences (3rd ed.). Mahwah, New Jersey: Lawrence Erlbaum Associates, Inc. [Google Scholar]
  17. Cornwell BL, Brockmann LM, Lasky EC, Mach J, & McCarthy JF (2018). Primary Care-Mental Health Integration in the Veterans Affairs Health System: Program Characteristics and Performance. Psychiatric Services, 69(6), 696–702. 10.1176/appi.ps.201700213 [DOI] [PubMed] [Google Scholar]
  18. Department of Defense Suicide Event Report (2016). Department of Defense Suicide Event Report Calendar Year 2015 Annual Report. Retrieved from: http://t2health.dcoe.mil/sites/default/files/DoDSER_2015_Annual_Report.pdf
  19. Dozois DJ, Dobson KS, & Ahnberg JL (1998). A psychometric evaluation of the Beck Depression Inventory–II. Psychological Assessment, 10(2), 83–89. [Google Scholar]
  20. Elbogen EB, Wagner HR, Johnson SC, Kinneer P, Kang H, Vasterling JJ, … & Beckham JC (2013). Are Iraq and Afghanistan veterans using mental health services? New data from a national random-sample survey. Psychiatric Services, 64(2), 134–141. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Enders CK (2010). Applied missing data analysis. New York: Guilford press. [Google Scholar]
  22. Fischer EH, & Turner JI (1970). Orientations to seeking professional help: Development and research utility of an attitude scale. Journal of Consulting and Clinical Psychology, 35(1), 79–90. [DOI] [PubMed] [Google Scholar]
  23. Fitzgerald JM, MacNamara A, Kennedy AE, Rabinak CA, Rauch SA, Liberzon I, & Phan KL (2017). Individual differences in cognitive reappraisal use and emotion regulatory brain function in combat-exposed veterans with and without PTSD. Depression and Anxiety, 34(1), 79–88. 10.1002/da.22551 [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. Fox AB, Meyer EC, & Vogt D (2015). Attitudes about the VA health-care setting, mental illness, and mental health treatment and their relationship with VA mental health service use among female and male OEF/OIF veterans. Psychological Services, 12(1), 49–58. 10.1037/a0038269 [DOI] [PubMed] [Google Scholar]
  25. Fuehrlein BS, Mota N, Arias AJ, Trevisan LA, Kachadourian LK, Krystal JH, … & Pietrzak RH (2016). The burden of alcohol use disorders in US military veterans: results from the National Health and Resilience in Veterans Study. Addiction, 111(10), 1786–1794. https://doi/10.1111/add.13423 [DOI] [PubMed] [Google Scholar]
  26. Fulton JJ, Calhoun PS, Wagner HR, Schry AR, Hair LP, Feeling N, … & Beckham JC (2015). The prevalence of posttraumatic stress disorder in Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF) veterans: A meta-analysis. Journal of Anxiety Disorders, 31, 98–107. doi: 10.1016/j.janxdis.2015.02.003 [DOI] [PubMed] [Google Scholar]
  27. Glass JE, Perron BE, Ilgen MA, Chermack ST, Ratliff S, & Zivin K (2010). Prevalence and correlates of specialty substance use disorder treatment for Department of Veterans Affairs Healthcare System patients with high alcohol consumption. Drug and Alcohol Dependence, 112(1–2), 150–155. 10.1016/j.drugalcdep.2010.06.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  28. Goldberg SB, Fortney JC, Chen JA, Young BA, Lehavot K, & Simpson TL (2020). Military service and military health care coverage are associated with reduced racial disparities in time to mental health treatment initiation. Administration and Policy in Mental Health and Mental Health Services Research. 10.1007/s10488-020-01017-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Goldberg SB, Simpson TL, Lehavot K Katon JG, Chen JA,…& Fortney JC (2019). Mental health treatment delay: A comparison among civilians and veterans of different service eras. Psychiatric Services, 70(5), 358–366. 10.1176/appi.ps.201800444 [DOI] [PMC free article] [PubMed] [Google Scholar]
  30. Gorman LA, Blow AJ, Ames BD, & Reed PL (2011). National Guard families after combat: Mental health, use of mental health services, and perceived treatment barriers. Psychiatric Services, 62(1), 28–34. [DOI] [PubMed] [Google Scholar]
  31. Gross JJ, & John OP (2003). Individual differences in two emotion regulation processes: Implications for affect, relationships, and well-being. Journal of Personality and Social Psychology, 85(2), 348–362. 10.1037/0022-3514.85.2.348 [DOI] [PubMed] [Google Scholar]
  32. Gunter-Hunt G, Feldman J, Gendron J, Bonney A, & Unger J (2013). Outreach to women veterans of Iraq and Afghanistan: A VA and National Guard collaboration. Federal Practitioner, 30(2), 25–31. [Google Scholar]
  33. Hack SM, DeForge BR, & Lucksted A (2017). Variation in veteran identity as a factor in veteran-targeted interventions. Psychiatric Services, 68, 727–729. 10.1176/appi.ps.201600237 [DOI] [PubMed] [Google Scholar]
  34. Hebert M, Rosenheck R, Drebing C, Young AS, & Armstrong M (2008). Integrating peer support initiatives in a large healthcare organization. Psychological Services, 5(3), 216. [Google Scholar]
  35. Hoge CW, Auchterlonie JL, & Milliken CS (2006). Mental health problems, use of mental health services, and attrition from military service after returning from deployment to Iraq or Afghanistan. JAMA, 295(9), 1023–1032. [DOI] [PubMed] [Google Scholar]
  36. Hoge CW, Castro CA, Messer SC, McGurk D, Cotting DI, & Koffman RL (2004). Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. New England Journal of Medicine, 351(1), 13–22. [DOI] [PubMed] [Google Scholar]
  37. Hoge CW, Grossman SH, Auchterlonie JL, Riviere LA, Milliken CS, & Wilk JE (2014). PTSD treatment for soldiers after combat deployment: Low utilization of mental health care and reasons for dropout. Psychiatric Services, 65(8), 997–1004. [DOI] [PubMed] [Google Scholar]
  38. IBM Corp. (2019). IBM SPSS Statistics for Mac, Version 26.0. Armonk, NY: IBM Corp. [Google Scholar]
  39. Interian A, Kline A, Callahan L, & Losonczy M (2012). Readjustment stressors and early mental health treatment seeking by returning National Guard soldiers with PTSD. Psychiatric Services, 63(9), 855–861. [DOI] [PubMed] [Google Scholar]
  40. Jakupcak M, Blais RK, Grossbard J, Garcia H, & Okiishi J (2014). “Toughness” in association with mental health symptoms among Iraq and Afghanistan war veterans seeking Veterans Affairs health care. Psychology of Men & Masculinity, 15(1), 100. [Google Scholar]
  41. Kehle SM, Polusny MA, Murdoch M, Erbes CR, Arbisi PA, Thuras P, & Meis LA (2010). Early mental health treatment-seeking among US National Guard soldiers deployed to Iraq. Journal of Traumatic Stress, 23(1), 33–40. 10.1002/jts.20480 [DOI] [PubMed] [Google Scholar]
  42. Kohn R, Saxena S, Levav I, & Saraceno B (2004). The treatment gap in mental health care. Bulletin of the World Health Organization, 82, 858–866. [PMC free article] [PubMed] [Google Scholar]
  43. Kok BC, Herrell RK, Thomas JL, & Hoge CW (2012). Posttraumatic stress disorder associated with combat service in Iraq or Afghanistan: Reconciling prevalence differences between studies. The Journal of Nervous and Mental Disease, 200(5), 444–450. 10.1097/NMD.0b013e3182532312 [DOI] [PubMed] [Google Scholar]
  44. Knox KL, Kemp J, McKeon R, & Katz IR (2012). Implementation and early utilization of a suicide hotline for veterans. American Journal of Public Health, 102(S1), S29–S32. 10.2105/AJPH.2011.300301 [DOI] [PMC free article] [PubMed] [Google Scholar]
  45. Kulesza M, Pedersen ER, Corrigan PW, & Marshall GN (2015). Help-seeking stigma and mental health treatment seeking among young adult veterans. Military Behavioral Health, 3(4), 230–239. 10.1080/21635781.2015.1055866 [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Lapp CA, Taft LB, Tollefson T, Hoepner A, Moore K, & Divyak K (2010). Stress and coping on the home front: Guard and reserve spouses searching for a new normal. Journal of Family Nursing, 16(1), 45–67. 10.1177/1074840709357347 [DOI] [PubMed] [Google Scholar]
  47. Loo CM, Fairbank JA, Scurfield RM, Ruch LO, King DW, Adams LJ, & Chemtob CM (2001). Measuring exposure to racism: Development and validation of a Race-Related Stressor Scale (RRSS) for Asian American Vietnam veterans. Psychological Assessment, 13(4), 503–520. [PubMed] [Google Scholar]
  48. Meis LA, Barry RA, Kehle SM, Erbes CR, & Polusny MA (2010). Relationship adjustment, PTSD symptoms, and treatment utilization among coupled National Guard soldiers deployed to Iraq. Journal of Family Psychology, 24(5), 560–567. 10.1037/a0020925 [DOI] [PubMed] [Google Scholar]
  49. Milliken CS, Auchterlonie JL, & Hoge CW (2007). Longitudinal assessment of mental health problems among active and reserve component soldiers returning from the Iraq war. JAMA, 298(18), 2141–2148. [DOI] [PubMed] [Google Scholar]
  50. Mojtabai R (2007). Americans’ attitudes toward mental health treatment seeking: 1990–2003. Psychiatric Services, 58(5), 642–651. [DOI] [PubMed] [Google Scholar]
  51. Mojtabai R, Olfson M, Sampson NA, Jin R, Druss B, Wang PS, … & Kessler RC (2011). Barriers to mental health treatment: Results from the National Comorbidity Survey Replication. Psychological Medicine, 41(8), 1751–1761. 10.1017/S0033291710002291 [DOI] [PMC free article] [PubMed] [Google Scholar]
  52. National Academies of Sciences, Engineering, and Medicine. (2018). Evaluation of the Department of Veterans Affairs Mental Health Services. Washington, DC: The National Academies Press. 10.17226/24915 [DOI] [PubMed] [Google Scholar]
  53. Palmer GA, Happe MC, Paxson JM, Jurek BK, Graca JJ, & Olson SA (2014). Psychometric properties of the Beck Depression Inventory-II for OEF/OIF veterans in a polytrauma sample. Military Medicine, 179(8), 879–884. 10.7205/MILMED-D-14-00048 [DOI] [PubMed] [Google Scholar]
  54. 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), 1118–1122. [DOI] [PubMed] [Google Scholar]
  55. Polusny MA, Erbes CR, Murdoch M, Arbisi PA, Thuras P, & Rath MB (2011). Prospective risk factors for new-onset post-traumatic stress disorder in National Guard soldiers deployed to Iraq. Psychological Medicine, 41(4), 687–698. 10.1017/S0033291710002047 [DOI] [PubMed] [Google Scholar]
  56. R Core Team (2018). R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. URL https://www.R-project.org/ [Google Scholar]
  57. Riviere LA, Kendall-Robbins A, McGurk D, Castro CA, & Hoge CW (2011). Coming home may hurt: risk factors for mental ill health in US reservists after deployment in Iraq. The British Journal of Psychiatry, 198, 136–142. 10.1192/bjp.bp.110.084863 [DOI] [PubMed] [Google Scholar]
  58. Rudd MD, Goulding J, & Bryan CJ (2011). Student veterans: A national survey exploring psychological symptoms and suicide risk. Professional Psychology: Research and Practice, 42(5), 354–360. 10.1037/a0025164 [DOI] [Google Scholar]
  59. Sayer NA, Noorbaloochi S, Frazier P, Carlson K, Gravely A, & Murdoch M (2010). Reintegration problems and treatment interests among Iraq and Afghanistan combat veterans receiving VA medical care. Psychiatric Services, 61(6), 589–597. [DOI] [PubMed] [Google Scholar]
  60. Seal KH, Bertenthal D, Miner CR, Sen S, & Marmar C (2007). Bringing the war back home: Mental health disorders among 103 788 US veterans returning from Iraq and Afghanistan seen at Department of Veterans Affairs Facilities. Archives of Internal Medicine, 167(5), 476–482. [DOI] [PubMed] [Google Scholar]
  61. Seal KH, Cohen G, Waldrop A, Cohen BE, Maguen S, & Ren L (2011). Substance use disorders in Iraq and Afghanistan veterans in VA healthcare, 2001–2010: Implications for screening, diagnosis and treatment. Drug and Alcohol Dependence, 116(1–3), 93–101. 10.1016/j.drugalcdep.2010.11.027 [DOI] [PubMed] [Google Scholar]
  62. Sherman MD, & Larsen JL (2018). Family-focused interventions and resources for veterans and their families. Psychological Services, 15(2), 146–153. 10.1037/ser0000174 [DOI] [PubMed] [Google Scholar]
  63. Shin LM, Rauch SL, & Pitman RK (2006). Amygdala, medial prefrontal cortex, and hippocampal function in PTSD. Annals of the New York Academy of Sciences, 1071(1), 67–79. 10.1196/annals.1364.007 [DOI] [PubMed] [Google Scholar]
  64. Spitzer RL, Kroenke K, Williams JBW, & Lowe B (2006). A brief measure for assessing generalized anxiety disorder: The GAD-7. Archives of Internal Medicine, 166, 1092–1097. [DOI] [PubMed] [Google Scholar]
  65. Stecker T, Fortney J, Hamilton F, Sherbourne CD, & Ajzen I (2010). Engagement in mental health treatment among veterans returning from Iraq. Patient Preference and Adherence, 4, 45–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
  66. Stecker T, Shiner B, Watts BV, Jones M, & Conner KR (2013). Treatment-seeking barriers for veterans of the Iraq and Afghanistan conflicts who screen positive for PTSD. Psychiatric Services, 64(3), 280–283. [DOI] [PubMed] [Google Scholar]
  67. Straits-Tröster KA, Brancu M, Goodale B, Pacelli S, Wilmer C, Simmons EM, & Kudler H (2011). Developing community capacity to treat post-deployment mental health problems: A public health initiative. Psychological Trauma: Theory, Research, Practice, and Policy, 3(3), 283–291. 10.1037/a0024645 [DOI] [Google Scholar]
  68. Thomas JL, Wilk JE, Riviere LA, McGurk D, Castro CA, & Hoge CW (2010). Prevalence of mental health problems and functional impairment among active component and National Guard soldiers 3 and 12 months following combat in Iraq. Archives of General Psychiatry, 67(6), 614–623. [DOI] [PubMed] [Google Scholar]
  69. U.S. Army. (2010). Demographics. Retrieved on April 23, 2020 from: https://www.army.mil/article/219140/demographics
  70. Veterans Health Administration. (2011). IB 10–438 Combat Veteran Eligibility. Washington DC: Veterans Health Administration. Accessed July 5th, 2019 from: https://www.va.gov/healthbenefits/assets/documents/publications/IB-10-438_Combat_Veteran_Eligibility.pdf [Google Scholar]
  71. Vogel DL, Wade NG, & Hackler AH (2007). Perceived public stigma and the willingness to seek counseling: The mediating roles of self-stigma and attitudes toward counseling. Journal of Counseling Psychology, 54(1), 40–50. 10.1037/0022-0167.54.1.40 [DOI] [Google Scholar]
  72. Vogt D (2011). Mental health-related beliefs as a barrier to service use for military personnel and veterans: a review. Psychiatric Services, 62(2), 135–142. [DOI] [PubMed] [Google Scholar]
  73. Vogt D, Di Leone BA, Wang JM, Sayer NA, Pineles SL, & Litz BT (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]
  74. Wang PS, Berglund P, Olfson M, Pincus HA, Wells KB,. & Kessler RC (2005). Failure and delay in initial treatment contact after first onset of mental disorders in the National Comorbidity Survey Replication. Archives of General Psychiatry, 62, 603–613. [DOI] [PubMed] [Google Scholar]
  75. Weathers F, Litz B, Herman D, Huska J, & Keane T (October, 1993). The PTSD Checklist (PCL): Reliability, validity, and diagnostic utility. Paper presented at the Annual Convention of the International Society for Traumatic Stress Studies, San Antonio, TX. [Google Scholar]
  76. Wisconsin Department of Health Services. (2018). 2017 Wisconsin mental health and substance use needs assessment. Madison, WI: Division of Care and Treatment Services. [Google Scholar]
  77. Zinzow HM, Britt TW, Pury CL, Jennings K, Cheung JH, & Raymond MA (2015). Barriers and facilitators of mental health treatment-seeking in US active duty soldiers with sexual assault histories. Journal of Traumatic Stress, 28(4), 289–297. 10.1002/jts.22026 [DOI] [PubMed] [Google Scholar]
  78. Zinzow HM, Britt TW, Pury CL, Raymond MA, McFadden AC, & Burnette CM (2013). Barriers and facilitators of mental health treatment seeking among active-duty army personnel. Military Psychology, 25(5), 514–535. 10.1037/mil0000015
 [DOI] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1662562-supplement-1.pdf (124.9KB, pdf)

RESOURCES