|Home | About | Journals | Submit | Contact Us | Français|
Using a sample of student service members/veterans, the current study aimed to examine the prevalence of psychiatric diagnoses and suicide-related outcomes and the association of hazardous duty with mental health.
Data are from the Fall 2011 National College Health Assessment (n=27,774).
Logistic regression was used to examine (1) the association of student service member/veteran status with mental health outcomes and (2) the association of hazardous duty with mental health outcomes among student service members/veterans (n=706).
Student service members/veterans had higher odds of self-harm than students without military experience. Among student service members/veterans, hazardous duty was positively associated (OR=2.00, 95% CI: 1.30–3.07) with having a psychiatric diagnosis but negatively associated (OR=0.41, 95% CI: 0.20–0.85) with suicidal ideation.
Self-harm may be a unique phenomenon among service members/veterans. Suicide prevention with this population should include information about self-harm, and future research should explore whether suicidal intent underlies self-harm.
Over 500,000 U.S. military veterans returning from post-9/11 service have utilized educational benefits through the Department of Veterans Affairs (VA).1 Recent evidence suggests that college-attending veterans experience high prevalence of symptoms of psychological problems, including depression, anxiety, and posttraumatic stress disorder (PTSD).2 Furthermore, veterans and service members have garnered particular attention as a possible population with mental health disparities, especially regarding suicide risk.3–6 Although the precise etiology of suicidal risk disparities among military experienced remains unknown, several factors that have been identified, such as relationship dissolution7 and posttraumatic stress disorder8 as well as hazardous duty-related consequences, such as traumatic brain injury9 and combat exposure.10–12 In spite of an increase in research about mental health and suicide risk among service members and veterans, little is known about the demographic and mental health characteristics of service members/veterans who are also currently enrolled as students in postsecondary education.
Mental health concerns and self-directed violence are prevalent among young adults, including college-attending populations.13–17 For instance, in an examination of data from the National Epidemiologic Study of Alcohol and Related Conditions, Blanco and colleagues noted that overall, college-attending respondents had similar prevalence of past 12-month psychiatric diagnoses when compared with their contemporaries who were not attending college. Additionally, Zivin et al. found approximately 6.4% of a sample of college students reported past-year suicidal ideation,14 compared with a general U.S. adult population prevalence of 3.7%.18 Research also has shown higher prevalence of self-reported suicide attempts. Data from over 80,000 students who participated in the Spring 2008 American College Health Association (ACHA) survey showed that 1.3% reported a suicide attempt in the past school year.19 The general U.S. adult population prevalence of past 12-months suicide attempt was approximately 0.5%.18 Moreover, very high prevalence of deliberate self-harm has been noted in young adult and college populations.17,20,21
Despite a large body of research about mental health and self-directed among college populations, very little is known about the subset of college populations who are service members/veterans. Barry et al. (2012) noted that binge drinking was positively associated with depression, anxiety, and posttraumatic stress among student service members/veterans but there was no association between binge drinking and these mental health problems among students without military service.22 In a different analysis, Barry and colleagues found that student service members/veterans who experienced trauma related to combat were more likely to report posttraumatic stress than either their civilian peers or their military-experienced peers without exposure to combat.23 In one of the only studies that currently focuses on student veterans and suicidal thoughts and behaviors, Rudd, Goulding, and Bryan (2011) noted elevated prevalence of symptoms of depression and anxiety in a national sample of student veterans.2 In particular, the authors noted that the prevalence of suicidal thoughts and behaviors was higher in their sample of student veterans than estimates from a large national sample of college students. However, the sample of student veterans in the Rudd et al. study was a convenience-based sample without a direct student non-veteran comparison group. Further study is needed to examine these findings within more robust sampling methodology.
Several subpopulations of veterans have been identified with elevated risk for suicide-related outcomes when compared with their non-military experienced peers, such as women veterans,24 veterans utilizing Veterans Health Administration (VHA) services,4 and sexual minority veterans.25 It is unclear whether student service members/veterans may also comprise a subpopulation of veterans with high prevalence of suicidal thoughts and behaviors. Student service members/veterans may have experiences that are associated with elevated prevalence of mental health problems (e.g., PTSD, suicidal ideation) that could result from consequences of hazardous duty, including traumatic brain injury and combat exposure.11,26 For example, Bryan and colleagues noted a stronger link between feelings of guilt and suicidal ideation among veterans who experienced combat.10 Similarly, Maguen et al. found an association between killing in combat and suicidal ideation among a representative sample of Vietnam veterans; an association that persisted even after adjusting for known correlates of suicidal ideation (e.g., PTSD, depression, substance use).11
While there is a flurry of research about suicidal ideation, suicide attempt, and suicide mortality among service members/veterans, there are very few studies that have examined the prevalence of deliberate self-harm behaviors among military/veteran populations in general. While deliberate self-harm is associated with suicide,27,28 it is conceptualized as a distinct phenomenon in that a person harms him/herself without the intent to die.29 For instance, in a an analysis of deliberate self-harm among VHA veterans, Kleespies et al. noted that one-quarter of patients with self-harm reported no intent to die. Among a clinical sample of over 500 veteran men who met criteria for PTSD, Sacks and colleagues found over half of the respondents indicated self-mutilating behavior in the last two weeks, and those who reported self-harm had significantly worse mental health profiles, including depression and more severe PTSD.30 It is unclear whether there are differences among deliberate self-harm among nonclinical samples of service members/veterans compared with those without military service experience. Since self-harm behaviors are prevalent among young adults,17,20,21 college attending samples may constitute a particularly useful population in which to investigate differences in self-harm behaviors by military experience.
The present study builds on previous research about mental health outcomes among student service members/veterans by examining prevalence of psychiatric diagnoses, past 12-months self-harm, and past 12-months suicidal ideation and suicide attempt among a large national sample of students with and without a history of military service. Moreover, the study seeks to explore if student service members/veterans with and without self-reported hazardous duty experience differ in mental health and suicide risk. Based on previous studies, the following hypotheses were proposed: (1) student service members/veterans will have significantly higher prevalence and odds of past 12-months psychiatric diagnoses, self-harm, suicidal ideation or attempt than their non-military experienced peers; and (2) among student service members/veterans, those with self-reported hazardous duty experience will report significantly higher prevalence and odds of past 12-months psychiatric diagnoses, self-harm, and suicidal ideation or attempt than their peers who do not self-report experiencing hazardous duty.
Data are from the Fall 2011 National College Health Assessment (NCHA), which is created and managed by the American College Health Association (ACHA). Postsecondary educational institutions purchase the survey to administer to enrolled students, and data from institutions that use a census or random sampling design of their students are concatenated by ACHA to create national datasets by semester.31 The Fall 2011 NCHA dataset includes 27,774 respondents from 44 participating institutions. Most institutions (95.4%) used web-based survey administration, and the mean response rate among the participating schools was 24%.32 Further details of the sample are available from the American College Health Association.32
In Fall 2011, ACHA introduced a new question to the NCHA-II survey that assessed history of military service [i.e., “Are you currently or have been a member of the United States Armed Services (Active Duty, Reserve, or National Guard)?”] Response options were no; yes and I have deployed to an area of hazardous duty; yes and I have not deployed to an area of hazardous duty. Student service member/veteran status was defined as those who reported yes (regardless of hazardous duty experience). Additionally, the student service members/veterans were divided into two groups based on their affirmative response to the aforementioned military service question: student service members/veterans with and without hazardous duty exposure.
Standard demographic variables included: age (in years), sex (male vs. female), and enrollment status (full-time vs. part-time). Race/ethnicity was coded as white, non-Hispanic; African American/black, non-Hispanic; Asian, non-Hispanic; multiple racial groups, non-Hispanic; other racial groups, non-Hispanic; and Hispanic. Additionally, since sexual orientation has been associated with elevated prevalence of psychiatric conditions, suicidal ideation, and suicide attempt,33 we also examined self-report sexual identity. Specifically, respondents were asked “What is your sexual orientation?” Response options were heterosexual, gay/lesbian, bisexual, unsure. Due to the low prevalence of veterans who identified as sexual minority, sexual orientation was dichotomized into heterosexual vs. gay/lesbian, bisexual, and unsure.
Prior research indicates that stress, alcohol use, and mental health service utilization are associated with suicidal risk,34–36 so we adjusted our analyses using variables within the NCHA survey that represent these factors. Past 12-month stress level (i.e., “Within the last 12 months, how would you rate the overall level of stress you have experienced? no stress; less than average stress; average stress; more than average stress; tremendous stress”) was coded 0 through 4, with higher values indicating higher stress levels. Alcohol use was defined by use in the past 30-days (i.e., “Within the last 30 days, on how many days did you use alcohol (beer, wine, liquor)? never used; have used, but not in the last 30 days; 1–2 days; 3–5 days; 6–9 days; 10–19 days; 20–29 days; used daily”). Those who indicated “never used” were recoded into the “no use in the last 30 days” category, resulting in a 7-item response category in which higher numbers indicated higher levels of alcohol use. Finally, mental health service utilization was defined by respondents’ indication of whether they had ever sought mental health services from four different types of providers: counselor, therapist, psychologist; psychiatrist; other medical provider (e.g., physician, nurse practitioner); or minister, priest, rabbi, other clergy.
Three items were used to assess self-directed violence. Respondents were asked three separate questions of have you ever 1) “intentionally cut, burned, bruised, or otherwise injured yourself,” 2) “seriously considered suicide,” or 3) “attempted suicide.” Response options included no, never; no, not in the last 12 months; yes in the last 2 weeks; yes in the last 30 days; yes in the last 12 months. For each of the three items, the response options were dichotomized into past 12-months (yes in the last 2 weeks; yes in the last 30 days; yes in the last 12 months) vs. not in the past 12-months (no, never; no, not in the last 12 months).
The outcome of psychiatric diagnosis was created from an item that asked respondents if they had been diagnosed or treated by a professional in the last 12 months for any of 15 listed conditions: anorexia, anxiety, attention deficit/hyperactivity disorder, bipolar disorder, bulimia, depression, insomnia, other sleep disorder, obsessive compulsive disorder, panic attacks, phobia, schizophrenia, substance abuse or addiction (alcohol or other drugs), other addiction (e.g., gambling, internet, sexual), and other mental health condition. Response options included no; yes, diagnosed but not treated; yes, treated with medication; yes treated with psychotherapy; yes treated with medication and psychotherapy; yes other treatment. A dichotomous variable of past 12-months psychiatric diagnosis was created, based on respondents who indicated a diagnosis or treatment for one or more of any of the listed conditions versus respondents who indicated no diagnoses/treatment.
Chi-square tests of independence were used to examine differences in categorical demographic characteristics, mental health outcomes, and service utilization variables between (1) student service members/veterans vs. students without military experience and (2) between student service members/veterans with hazardous duty experience vs. student service members/veterans without hazardous duty experience. Differences in age were assessed with t-tests, and Mann-Whitney tests were used to examine differences in stress and alcohol use.
Using the entire sample, four multiple logistic regression models were used to examine the association of student member/veteran status with mental health outcomes of psychiatric diagnosis, self-harm, suicidal ideation, and suicide attempt. Specifically, these models assessed the association of student service member/veteran status with each of the four mental health outcomes while adjusting for demographic information and covariates related to suicide risk.
Using only the student service members/veterans within the sample, a second set of multiple logistic regression models was used to examine the association of exposure to hazardous duty with each of the mental health outcomes. Because of the rarity of suicide attempt among the student service member/veteran group, the model examining suicide attempt created estimates that were unstable [data not shown]. Consequently, the second set of models only included psychiatric diagnosis, self-harm, and suicidal ideation.
For all multivariable models, race/ethnicity was recoded into a two-category variable of white and racial/ethnic minority due to small samples of specific racial/ethnic minority student service members/veterans. Furthermore, to create a more parsimonious model and preserve degrees of freedom to accommodate the small sample of student service members/veterans, the four individual variables of ever seeking mental health care from different types of providers were recoded into a single dichotomous variable of ever received care from any of these sources versus never received care from any of these sources. Variance inflation factors were used to assess for collinearity in multivariable models and showed no problematic correlations among any of the included variables. All analyses were conducted using Stata/SE Ver. 12. This study was approved by the [university name masked for peer review] Institutional Review Board.
Overall, 2.6% of the sample (n=706) self-identified as a service member/veteran. When compared with students without military service, student service members/veterans were significantly older, had a higher proportion of males, were less racially/ethnically diverse, and were more likely to attend school part-time (see Table 1). Student service members/veterans and students without military service did not differ in terms of sexual orientation. Among student service members/veterans, those with hazardous duty experience were older, had a higher proportion of males, and were less racially/ethnically diverse than student service members/veterans without hazardous duty experience.
Student service members/veterans and students without military service did not differ in crude prevalence of stress or prevalence of alcohol use (see Table 2). A significantly larger proportion of student service members/veterans reported a psychiatric diagnosis than students without military service (27.0% and 20.4%, respectively, p<.05), but the groups did not differ in terms of crude prevalence of self-harm, suicidal ideation, or suicide attempt. Significantly more student service members/veterans than students without military service reported seeking mental health services from all four different types of service providers, but the groups did not differ in their use of college/university mental health services. Among student service members/veterans, those with hazardous duty reported more alcohol use and were more likely to have a psychiatric diagnosis than student service members/veterans without hazardous duty experience (31.9% vs. 22.2%, respectively, p<.05).
After adjusting for covariates and demographic characteristics (see Table 3), student service members/veterans did not differ from students without military service in odds of psychiatric diagnosis (OR=1.17, 95% CI: 0.95–1.46), suicidal ideation (OR=1.34, 95% CI: 0.95–1.89) or suicide attempt (OR=1.35, 95% CI: 0.58–3.16). Student service members/veterans did, however, have significantly increased odds of self-harm when compared with students without military service (OR=1.83, 95% CI: 1.22–2.76). Models in Table 3 show multiple logistic regression models among only student service members/veterans in order to examine the association of hazardous duty with the outcomes. Student service members/veterans with hazardous duty had twice the odds of a psychiatric diagnosis than their student service member/veteran peers without hazardous duty, but they had a statistically significant 59% decrease in odds of past 12-months suicidal ideation (see Table 4).
The present findings are among the first to examine differences in correlates and indicators of suicidal thoughts and behaviors among a large national sample of students by service member/veteran status. It is also the first study to examine outcomes among student service member/veteran based on self-reported experiences of hazardous duty. The results supported only the hypotheses that (1) student service members/veterans would experience significantly higher odds of self-harm than their peers without military service experience and (2) student service members/veterans with hazardous duty experience would have higher odds of a psychiatric diagnosis than their student service member/veteran peers without hazardous duty experience.
In this sample, there was no difference among student service member/veteran and students without military service in odds of suicidal ideation or suicide attempt. The results seem to differ from the only other study to examine mental health and suicide risk among student veterans. For example, using the revised 4-item Suicide Behavior Questionnaire, Rudd and colleagues found that 10.4% of student veterans in their sample reported past-year suicidal ideation either often or very often; an estimate that likely would be higher had it included the other response categories of rarely and sometimes. In contrast, the prevalence of past 12-months suicidal ideation (yes/no) was 6.7% among student service members/veterans from the sample in the present analysis, which did not significantly differ from students without military service. It is important to note that direct comparison between these two studies is limited due to differences in survey items and sampling design. Further research is needed to replicate and compare the findings of these respective studies in order to more accurately assess the epidemiology of suicidal thoughts and behaviors among student service members/veterans.
Interestingly, hazardous duty had a negative association with suicidal ideation. Specifically, student service members/veterans with hazardous duty were significantly less likely to report suicidal ideation than their student service member/veteran peers without hazardous duty. This finding is particularly profound since the association persisted after adjusting for lifetime and past 12-months psychiatric service utilization. That is, even though student service members/veterans with hazardous duty had greater odds of having psychiatric care in the past 12 months, this characteristic did not fully explain the lower odds for suicidal ideation. It is possible that student service members/veterans with hazardous duty experience may under-report suicidal ideation for several reasons, such as stigma and unwanted potential hospitalization.37 Further research with additional samples of student service member/veteran is needed to corroborate this finding and explore explanations for it. For instance, studies with comparison groups of non-college attending service members/veterans would help to clarify if college-attending service members/veterans may represent a self-selected resilient group, since they met requirements for and are actively pursuing higher education.
After adjusting for other demographic characteristics and psychiatric diagnoses, student service member/veteran status was significantly associated with higher odds of self-harm. Relative to information about suicide risk, much less is known regarding self-harm among service persons and veterans. In one of the first studies of self-harm among veterans, Kleespies and colleagues examined interviews with VHA-utilizing veterans following self-harm requiring medical attention. They noted that over half of the sample of veterans with self-harm also had a history of attempted suicide,38 which aligned with previous research documenting co-occurrence of self-reported self-harm behavior and suicide attempts.39,40 Thus, it is possible that some reports of self-harm in this sample actually may have included suicide attempts.41 Unfortunately, the survey did not assess suicidal intent involved in an incident of deliberate self-harm. Further studies with veterans are needed to examine self-harm and suicidal intent involved in self-harm.
This is among the first studies to examine the use of mental health services in a population of student service members/veterans. In particular, a significantly greater proportion of student service members/veterans than students without military service reported seeking counseling from clergy or other faith-based providers. Previous studies suggest that some veterans may be more willing to seek help from a pastoral care provider as opposed to using specialty mental health services42–44. Additionally, considering the relatively young age of the sample population, the use of pastoral care services by college-aged veterans may be a practice carried over from time spent in active duty. Military chaplains serve as front-line providers to service members in need, effectively acting as gatekeepers for mental health professionals.45 Further research on pastoral care for mental health among student service members/veterans may offer unique insight to their help-seeking behaviors, such as preferences for mental health care, access to care, and potential partnerships between universities and faith-based organizations.
The results here must be viewed in light of several limitations. First, the data are cross-sectional, which preclude examination of causality. Second, misclassification bias may have resulted in defining student service members/veterans since history of military service was self-reported rather than from official records. Moreover, characteristics of military service were not available (e.g., active duty and ever served were combined in the survey question and period of service was not ascertained). Third, the term hazardous duty was not defined in the survey, and thus cannot be interpreted as a measure of more specific experiences, such as combat exposure and combat-related trauma. For example, hazardous duty as defined by the Department of Defense for payment purposes includes experiences such as exposure toxic chemicals, demolition of explosives, and hostile fire.46 Consequently, in the data used for this study it is unclear which type(s) of hazardous duty the respondents experienced or whether respondents were self-selected or assigned hazardous duty. Fourth, the measure of self-harm was a single item and did not assess the intent behind the self-harm (i.e., non-suicidal vs. suicidal). Finally, there may be selection bias among this student service members/veterans sample since this group was currently enrolled in postsecondary education, which may indicate the sample was healthier than those unable or unwilling to enroll in higher education.
The era that began access to higher education for veterans with Franklin D. Roosevelt’s signing of the Servicemen’s Readjustment Act (G.I. Bill) seems worlds away.47 Just as warfare has changed dramatically in the seven decades since the G.I. Bill, the landscape of education has transformed. The culmination of those changes intersect with the over 830,000 veterans who used their VA education benefits in 2011, the vast majority of whom are veterans of wars in Afghanistan and Iraq. As college campuses begin initiatives to reach out to their student service members/veterans,48 there is a clear need to focus attention on the mental health needs of the women and men working to integrate the roles of students into their identities as service members/veterans.
Similar to previous findings with college populations,14,16,19 the crude prevalence of suicidal ideation and suicide attempt in this college sample, regardless of student service member/veteran status, were higher than general population18 and reiterate suicide risk as a public health imperative for this population. In particular, while student service members/veterans in this sample did not have higher odds of suicidal ideation or suicide attempt than students without military service, they had significantly higher odds of reporting self-harming behavior compared to students without military service. Health care providers who work with student service members/veterans may need to better understand and assess the phenomenon of self-harm as distinct from attempted suicide40 yet highly predictive of suicide.27 Epidemiologic and clinical questions about suicide risk among student service members/veterans should include the use of appropriate questions that can assess whether suicidal intent is involved in self-harm behaviors.
The authors thank the American College Health Association for use of the Fall 2011 National College Health Assessment data.
This work was partially supported by the VISN2 Center of Excellence for Suicide Prevention, a postdoctoral fellowship to JRB in an institutional National Research Service Award from the National Institute of Mental Health (5T32MH020061), and a postdoctoral fellowship to JRB through the Department of Veterans Affairs Office of Academic Affiliations and the Center for Health Equity Research and Promotion at the VA Pittsburgh Healthcare System. The opinions expressed in this work are the authors’ and do not reflect those of the funders, institutions, the Department of Veterans Affairs, or the U.S. Government.