Prolonged and frequent combat operations since 2001 have resulted in a substantial number of U.S. military personnel reporting psychological problems. One study found prevalence rates between 23% and 31% for posttraumatic stress disorder (PTSD) or depression (
1). Despite efforts to improve availability and decrease stigma related to receiving treatment, studies consistently find that only a fraction of those reporting mental health problems seek care (
2–
4).
Studies on the utilization of mental health help have focused on aspects such as stigma and barriers. Less research has focused on the potential association between types and sources of help and rates of utilization. Although research on help seeking among service members has largely focused on care from providers, such as mental health professionals and general medical doctors (
5), less attention has been given to nonprovider sources. These include chaplains, unit medics, and fellow soldiers. There is evidence to suggest that in the civilian population, individuals with mental health needs view nonprovider sources of help more positively than provider sources (
6). Recent trends indicate that among individuals in the general population with mental disorders, a larger proportion seek help from the clergy than from psychiatrists (
7).
Studies of military samples show similar preferences for help for mental health problems from nonproviders. Studies of service members in the United Kingdom indicate that nonprovider sources of help for mental health problems (for example, chaplains and social workers) and informal sources (for example, friends and family) were more widely used than mental health providers, such as doctors and psychologists (
8,
9). U.S. service members also report substantial openness to talking with clergy about emotional problems (
10).
Although stigma and barriers to treatment are frequently reported by soldiers with mental health problems (
2,
3,
11,
12), recent research has investigated the role of attitudes about mental health care in the decision to seek help. Negative attitudes, such as lack of trust in mental health professionals, have been shown to be more strongly associated with help seeking than stigma or organizational barriers (
13). Evidence within the civilian literature indicates that, although mental health practitioners are viewed as more competent to address mental health problems, nonproviders are viewed as more caring, stable, and professional (
14). Nonproviders may likely be used by individuals who may otherwise avoid or resist seeking help from providers because of their attitudes toward mental health care and providers. No research, to our knowledge, has explored the association between barriers to care and help seeking from nonproviders.
This survey-based study aimed to explore rates of help seeking from provider and nonprovider sources among persons reporting mental health problems in a military sample. We also examined differences in perceived stigma, negative attitudes, and organizational barriers between those seeking different sources of help and those not seeking help. In addition, we examined the degree to which these impediments to care were associated with help seeking from either providers or nonproviders.
Methods
Sample
Data for this study were collected under the Land Combat Study, a series of data collections that assessed the impact of combat on service members. A total of 3,380 active duty soldiers were surveyed cross-sectionally from four brigades at a large U.S. military installation. Data were collected between December 2008 and June 2009, six months after the service members’ return from a deployment. Approximately 5,666 soldiers were available to complete the survey, yielding an overall consent rate of 60%, similar to rates in other military population–based studies (
1,
2,
12). Reasons for not being available to complete the survey included work-related duties and being on leave or on temporary duty elsewhere. Soldiers who chose to complete the survey provided their consent under a protocol approved by the Walter Reed Army Institute of Research Institutional Review Board.
Measures
Mental health status was determined by screening for major depressive disorder, generalized anxiety disorder, and PTSD. Major depressive disorder was assessed with the nine-item Patient Health Questionnaire, a commonly used and well-validated tool for studies of soldiers (
1,
2). Soldiers reporting at least five of nine depression symptoms in the past month, including either “little interest or pleasure in doing things” or “feeling down, depressed, or hopeless” for more than half the days in the past month, were identified as screening positive for major depressive disorder (
15).
Generalized anxiety disorder was identified by using the GAD-7 (
16), a seven-item questionnaire measuring the frequency of a broad range of anxiety symptoms within the past month. Symptoms were assessed with a 4-point response scale, ranging from not at all, 0, to nearly every day, 3. The seven items were summed, which resulted in a composite score ranging from 0 to 21. A score of ≥10 was used to indicate a positive screen for anxiety.
PTSD was assessed by using the 17-item PTSD Checklist (
17), a well-validated measure that follows
DSM-IV guidelines (
18). Each item is reported on a 5-point scale, with a possible score range of 17 to 85. Soldiers reporting moderate or greater levels (a score of ≥3) of at least one intrusion symptom, three avoidance symptoms, and two hyperarousal symptoms in the past month and whose summed score was ≥50 met criteria for a positive screen.
Need for mental health help was determined by asking participants if they were “currently experiencing a stress, emotional, alcohol, or family problem.” Those responding “yes, moderate” or “yes, severe” were considered as needing mental health help. This is a commonly used item for measuring general behavioral health issues in military samples (
2,
12,
13).
Use of provider mental health care was assessed by asking soldiers whether they had received help in the past three months for a stress, emotional, alcohol, or family problem from a military or civilian mental health professional, general medical doctor, or Military OneSource (a Department of Defense–sponsored hotline that provides service members with licensed professionals and counselors). Help seeking from nonproviders was assessed by asking soldiers whether they had received help in the past three months from a fellow soldier, a medic in his or her unit, someone in the chain of command, or a chaplain or clergy.
Barriers to care were measured with 17 items related to soldiers’ concerns about receiving treatment (
2,
13). Each item had a 5-point response scale indicating degree of endorsement (1, strongly disagree, to 5, strongly agree). The items were grouped into three categories corresponding to the three factors identified by Kim and colleagues (
13): stigma, negative attitudes, and organizational barriers to care. Seven items were included in the stigma dimension of barriers to care; the items pertained to perceived adverse reactions from others in relation to the person’s receiving mental health care (for example, “It would be too embarrassing”). Six items constituted the negative attitudes dimension; these items assessed the person’s negative reaction to receiving mental health care (for example, “I don’t trust mental health professionals”). Four items made up the organizational barriers dimension; the items measure practical impediments to accessing care (for example, “Mental health services aren’t available”).
Analyses
Independent-samples t tests were conducted to examine mean differences in barriers to care between participants who used care only from providers, those who used care only from nonproviders, those who used no care, and those who used both types of care. Binomial logistic regression analyses were conducted to examine the association between barriers to care and sources of help used. The variables in each model were entered simultaneously. All analyses were conducted with SPSS software, version 20 (
19).
Results
Demographic Characteristics
The demographic characteristics of soldiers in the sample were similar to those of military samples in recent studies (
1,
2,
13) (
Table 1). Most soldiers were young (under 30), junior-enlisted men with a high school diploma. Approximately half were married. Race-ethnicity data were not collected.
Frequency of Use of Sources of Help
A total of 1,048 soldiers (31%) screened positive for a mental health problem (
Table 2). In this subsample, 74% reported not using any help, and 26% reported using some source of help, (provider, nonprovider, or both). In this subsample, 14% reported using provider help exclusively; 5% reported using nonprovider help exclusively; and 7% reported using both sources. Among those who received any help (N=275), 54% used provider help exclusively, 18% used nonprovider help exclusively, and 28% used both sources.
Table 2 presents rates of utilization for each source of help in the subsample that screened positive for any mental disorder (N=1,048). Among soldiers who used provider sources, military mental health professionals were the most frequently used type of care (15%), followed by Military OneSource (8%), civilian mental health professionals (6%), military general medical doctors (3%), and civilian general medical doctors (1%). Among those who used nonprovider sources of help, the most frequently used was a fellow soldier (7%), followed by chain of command (5%), chaplain or clergy (5%), and medic (2%).
Scores on Barriers to Care by Source of Help
Table 3 reports mean scores for three types of barriers to care—stigma, negative attitudes, and organizational barriers—by sources of help used. Soldiers who used military mental health professionals had significantly lower mean scores for all three barriers, compared with those who did not use any source of help. Soldiers who used civilian mental health professionals or Military OneSource and those who used a fellow soldier and chain of command had significantly lower scores for negative attitudes, compared with those who did not use any source of help.
Compared with soldiers who did not use any source of help, those who used help only from providers had significantly lower scores for all three barriers: stigma, negative attitudes, and organizational barriers. In addition, those who used help only from providers had significantly lower organizational barriers scores than those who used nonprovider help exclusively. Among soldiers who used nonprovider help exclusively, scores on the three barriers were not significantly different from scores among those who did not use any source of help.
Soldiers who used both sources of help had significantly lower scores for negative attitudes than those who did not use any source of help. Compared with soldiers who did not use any help, those who used any source of help (provider or nonprovider or both) had significantly lower scores for all three barriers: stigma, negative attitudes, and organizational barriers.
Effects of Barriers to Care on Source of Help Used
Table 4 presents results of the logistic regression analyses of the association of stigma, negative attitudes, and organizational barriers with source of help (provider exclusively, nonprovider exclusively, and any source). Three separate models were conducted—one for each category of help seeking (provider exclusively and nonprovider exclusively) and one predicting any source of help seeking (both provider and nonprovider). Each model controlled for sex and rank; the reference group was those who did not use any source of help. Organizational barriers were associated with use of care from nonproviders (Wald=5.69, df=1, p<.017). Each unit increase in organizational barriers was associated with approximately 65% higher odds of using help from nonproviders exclusively, compared with those who did not seek care from any source. Stigma and negative attitudes were not associated with use of nonprovider help.
Negative attitudes were associated with use of help from any source. Each unit increase in negative attitudes was associated with an approximately 34% decrease in use of any source of help (Wald=9.59, df=1, p<.01). Stigma and organizational barriers were not associated with help seeking of any kind within this model (that is, use of any type of help versus no help).
A fourth model compared soldiers who used help only from providers and those who used help only from nonproviders (
Table 4). Higher organizational barriers were associated with the use of nonprovider help, compared with the use of help from providers. Each unit increase in organizational barriers was associated with 47% lower odds of using help from providers (Wald=5.46, df=1, p<.05). No significant associations of stigma or negative attitudes with use of help from providers or nonproviders were noted.
Research has indicated that symptom severity is associated with use of mental health help (
20). To account for this effect, additional analyses were conducted that added depression, anxiety, and PTSD symptom severity to each of the four models (results not shown). The inclusion of symptom severity did not substantively alter the effects of barriers to care on type of help used in any of the four models shown in
Table 4.
Discussion
The proportion of soldiers reporting mental health problems in this study (31%) is similar to rates in recent postdeployment studies of mental health (
1,
12,
13) and indicates the propensity of the rate—and the risk—of behavioral health problems to increase over time following a deployment (
21). Military personnel who screen positive for a postdeployment mental health problem have been shown to have concerns about stigma and barriers to care (
2). After a deployment, soldiers may seek help from sources different from those used by nondeployed soldiers and may report different barriers to care. Further research is needed to address the potentially unique concerns of soldiers after deployment.
Research has shown that a notable number of soldiers with mental health problems turn to nonprovider sources of help. Recent studies support the effectiveness of nonprovider resources in recruiting soldiers into care and the willingness of soldiers to seek help from nonproviders, such as clergy and informal networks (
8–
10). Recently, military efforts to emphasize and incorporate these sources of help have been suggested, such as a “collaborative intervention” model that promotes the merging of expertise between medical officers and chaplains to treat psychological problems (
22). Other programs that incorporate peer support (
23–
25) or family support (
26) show promise to provide effective and favorable options for returning soldiers.
Barriers to care affect soldiers’ decisions to seek help (
13,
20). In this study, soldiers who used any source of help reported the three types of barriers—stigma, negative attitudes, and organizational barriers—to be lower than did soldiers who did not use help from any source. Those who used help exclusively from providers also had lower scores for the three types of barriers than did those who did not use help from any source. Soldiers who used some nonprovider sources of help had scores indicating lower negative attitudes. Military efforts to reduce barriers to care are not limited to provider help; efforts also extend to nonproviders.
Soldiers reporting organizational barriers had higher odds of using nonproviders, compared with using help from providers. This finding suggests that soldiers who perceive organizational barriers to care, such as not knowing where to get help or having difficulty scheduling an appointment, are more likely to seek help from nonproviders, perhaps to circumvent these organizational barriers. If this is the case, recent military efforts to embed behavioral health providers into military units are a step in the right direction. This approach gives soldiers an easily accessible mental health resource and directly addresses some of the organizational barriers that may prevent soldiers from seeking help. A recent assessment of such a program demonstrated its effectiveness in reducing symptoms of behavioral disorders (
27). However, accessibility and proximity to behavioral health services may be affected by whether soldiers live on or off the installation. Future studies should investigate if and how living on or off post affects perceptions of organizational barriers to mental health help.
The study had several limitations. First, the cross-sectional nature of the data did not enable us to draw causal connections between barriers to care and help seeking. Future research using longitudinal data could uncover the prospective effects of barriers on the use of these sources of help and whether nonprovider help serves as a “gateway” to provider help. Second, the survey did not include questions to measure the exact nature and effectiveness of the type of help used, which a longitudinal study could address.
Third, other sources of nonprovider help were not included. Future studies should examine the role of families, nonmilitary friends, or civilian mentors as sources of help. Other sources of provider help, such as marriage and family therapists and Army community services, should be examined. Fourth, although some sources of help are clearly military (for example, chain of command and fellow soldier), no delineation was made between military and civilian chaplains or clergy; such a delineation may have provided a more complete understanding of barriers to care.
Fifth, we were unable to separately analyze the effects of barriers on the use of different types of nonproviders because of the small sample sizes. For some types of nonproviders (fellow soldier and chain of command), results were similar in terms of stigma and negative attitudes. These types of nonproviders may be more easily accessible than chaplains or clergy or medics and thus may be primarily driving the effect of organizational barriers on use of nonprovider help. Future studies using adequate samples should investigate the effects of barriers to care on the use of different types of nonprovider help.
Sixth, because the survey items about types of help used covered the past three months and the mental health screen covered the past month, individuals who had sought help may no longer have screened positive for a mental health problem. Thus the sample may have disproportionately included those whose symptoms had not abated within two months, which may present a potential bias regarding attitudes about mental health help. Future research should account for those who have recovered from mental health problems and investigate the types of help they used.
Finally, use of self-reports in the surveys may present a potential bias for over- or underreporting of help used as well as self-assessed need for mental health help. For example, the self-assessment of having an alcohol problem could pose an issue of reporting bias. However, the confidentiality of the responses gives us confidence in the accuracy of much of the data.
Conclusions
Stigma, negative attitudes, and organizational barriers persist as significant factors that prevent soldiers from utilizing the help they need—from providers, nonproviders, or both. The results of this study expand our understanding of where soldiers seek help for their mental health concerns and suggest that more mental health training for all military personnel is warranted with a stronger focus on reducing organizational barriers to care by enhancing the skills of nonprovider resources and cultivating an awareness of these sources of help. Such training must also not overlook the large proportion of individuals in need of mental health care who do not seek help; this means facilitating the recognition of mental health problems among fellow soldiers and encouraging the use of nonprovider sources of help as a possible first step in addressing mental health concerns.
Acknowledgments
The authors thank the Land Combat Study team for their support with the data collection.