Mental disorders constitute the highest burden of disability worldwide, according to the Global Burden of Disease study (1). U.S. military personnel experience a significant mental disorder burden. Among active-duty military personnel hospitalized during the 1990s, the most common category of discharge diagnosis was mental disorders (2). In the conflicts in Afghanistan (Operation Enduring Freedom) and Iraq (Operation Iraqi Freedom), depression, substance abuse, and posttraumatic stress disorder (PTSD) and other anxiety disorders are common mental health problems (3).

A substantial percentage of persons with mental health problems, however, do not receive care for those problems (4–6). This is also true in the U.S. Army. Among U.S. active-duty soldiers returning from combat deployment, 44% reported at least one type of mental health problem at 12 months postdeployment (7). Of those, 13% reported using mental health services. Similarly, among veterans who served in Operation Iraqi Freedom who screened positive for a mental health problem, only 18% were referred to mental health services. Of those, 56% received mental health care (3). In a study of United Kingdom military personnel with a self-perceived mental health problem, 31% sought help from a general medical doctor, mental health professional, or other professional (for example, a medical officer, counselor, psychologist, psychiatric nurse, or chaplain) (8). Of these, 19% received this help from general medical or mental health professionals. In the U.S. Army, soldiers can access mental health services directly or by receiving a referral. Of note, the percentage of soldiers perceiving difficulty in scheduling a mental health appointment varied from 17% to 45% (7,9).

In the general population of the United States, a number of demographic factors may influence the use of mental health services. Women seek mental health services in both general and specialty care settings (that is, treatment that is provided by mental health care specialists, including psychiatrists, psychologists, psychiatric nurses, or clinical social workers), whereas men are more likely to seek mental health care in specialty care settings only (10). After adjusting for income, analyses have shown that African Americans who are not classified as poor are less likely than whites to receive specialty care (11). Compared with persons who are not married, those who are married use more general medical care for mental health care but less often utilize specialized mental health care (12).

Although persons in the general population with impaired functioning are generally believed to be more likely to seek services, this may not hold in the military setting. Military personnel screening positive for a mental disorder are more likely to identify concerns of being stigmatized and other barriers to mental health services (9,13). Whether soldiers with higher levels of impaired functioning are receiving mental health services is unclear.

Understanding the patterns of mental health service use among Army soldiers can facilitate mental health service planning and identify gaps in care. Using a representative sample of the U.S. Army, this study examined types of mental health services sought by Army soldiers and the sociodemographic characteristics and levels of functioning associated with obtaining mental health care.

Participants and procedures

Using the 2008 Department of Defense (DoD) Survey of Health Related Behaviors Among Active Duty Military Personnel (HRB) (14), we examined mental health service use in the U.S. Army. The DoD HRB study was a cross-service anonymous and voluntary self-report survey completed from May through July 2008. It sampled 45,800 active-duty service members in the Army, Navy, Air Force, Marines, and Coast Guard and obtained 28,546 responses. Among Army personnel, 10,400 were sampled, with 5,927 responses representing 508,088 soldiers. A two-stage sampling design was used. First, a methodology of stratified probability proportional to size by service and region was applied. Active-duty personnel were then stratified by pay grade and gender and randomly selected with replacement at the participating installations. Officers and women were oversampled to account for low numbers in these groups. The survey data were weighted to represent the active-duty population; sampling and nonresponse differences were accounted for in the weighting. The Uniformed Services University of the Health Sciences Institutional Review Board approved this study.

In the U.S. Army sample, nearly half of the participants were ≤25 years of age (46%, N=2,619) (all reported Ns are respondent Ns unless otherwise indicated). A majority were male (87%, N=4,320) and currently married (56%, N=3,204). Nearly two-thirds were white, non-Hispanic (63%, N=3,270); 20% (N=1,226) were African American, non-Hispanic; 11% (N=902) were Hispanic; and 7% (N=529) were classified as other. Approximately two-thirds of the participants had some college (67%, N=4,055), and a majority were in the enlisted ranks (83%, N=5,927); warrant officers were included in officer ranks.

Measures

We examined the use of any mental health service, three categories of mental health service use, use of the highest level of mental health services, and the number of services used. Mental health service included both mental health and substance abuse treatment services.

Visits with a general medical doctor.

We defined a visit with a general medical doctor as counseling or therapy for mental health or substance abuse from a general medical doctor in the past 12 months, alone or in combination with other mental health services.

Visits with a mental health professional.

We defined a mental health visit as counseling or therapy for mental health or substance abuse from a mental health professional (including a psychologist, psychiatrist, clinical social worker, or other mental health counselor) in the past 12 months, alone or in combination with other mental health services.

Visits for prescribed medication.

Medication visits involved visits with a doctor or other health professional in the past 12 months that resulted in a prescription for medication to treat depression, anxiety, or sleeping problems. This category included all respondents who answered affirmatively to being prescribed such medication alone or in combination with other mental health services.

Highest level of services.

This category includes Army personnel who used both mental health professional and prescribed medication services. Respondents may or may not have also seen a general medical doctor for services.

Any mental health services.

The category “any mental health services” covers use of any of the three previously defined categories of mental health services.

Impaired functioning days.

Current day-to-day function over the previous 30 days was assessed with a modified version of the activities limitation item of the Health-Related Quality of Life–4 instrument (15). Respondents selected one of seven categories ranging from “never” to “almost every day” that most accurately captured the number of days within the past month in which poor mental health hindered their performance of usual activities, such as work or recreation. Scores were dichotomized, with high scores defined as those that exceeded one week of impairment.

Statistical analysis

The relationship of impaired functioning and sociodemographic characteristics with the number and types of services utilized was initially explored by using prevalence rates across the U.S. Army and among those in the U.S. Army using mental health services. Potential associations between the mental health services used and both impaired functioning and sociodemographic factors (gender, race-ethnicity, age, education, marital status, and enlisted status) were evaluated with bivariate (unadjusted) and multivariate (adjusted) logistic regressions. In this and all subsequent multivariate models, all aforementioned sociodemographic factors and impaired functioning were included in the models. Further, all interactions between impaired functioning and each of the sociodemographic factors were explored individually and in the full multivariate models. Nonsignificant interactions were subsequently dropped from the final models.

In order to examine comparisons between different types of mental health services and the relationship to impaired functioning and sociodemographic factors, we used multinomial logistic regression analyses. Similarly, to examine the number of mental health services used and the relationship to impaired functioning and sociodemographic factors, we used Poisson regression analyses. In these regression analyses, interactions were similarly explored, and nonsignificant interactions were removed from final models as described above.

Estimates of the odds ratio (OR), relative risk ratio (RRR), incidence rate ratio (IRR), 95% confidence intervals, and p values are reported. All reported analyses were conducted while we accounted for weighted data and the complex survey design. Multinomial regression analyses compared the mental health service under consideration with the remaining mental health service options. Statistical analyses were conducted with Stata, release 11.0.1.

Mental health service utilization in the U.S. Army

Twenty-one percent (weighted N=97,293) of Army soldiers used at least one mental health service in the preceding 12 months (see Figure 1 and Table 1). Among the three mental health service options, 11% of soldiers (weighted N=50,274) used one service (52% among those using mental health services), 6% (weighted N=28,241) used two services (29% among those using mental health services), and 4% (weighted N=18,778) received all three services (19% among those using mental health services). Further, 7% (weighted N=31,945) received the highest level of care, or 33% among those using services. In addition, approximately 11% (weighted N=48,832) were prescribed medications (50% among those using mental health services) alone or in combination with counseling services from a mental health professional or a general medical doctor.

Predictors of types of mental health service utilization

Use of any mental health service.

Women were 1.33 times more likely than men to use a mental health service (25% versus 20%; Table 2). Enlisted soldiers were 1.91 times more likely than officers to report using any mental health service (23% versus 13%). Persons reporting more than one week per month of impaired performance of their usual activities (6% of the U.S. Army population) were 8.03 times more likely than soldiers with lower levels of impairment to use a mental health service (65% versus 19%). Mental health service use was not related to race or marital status.

In the multivariate logistic regression model, women were significantly more likely than men to use mental health services (OR=1.45). Similarly, enlisted soldiers were 1.87 times more likely than officers to use mental health services. Those who were married (OR=1.27) and those with more than one week per month of impaired function (OR=7.82) were more likely to use mental health services. The interactions between impaired function and each of the sociodemographic factors were examined and found to be nonsignificant in these and all subsequent multivariate regression models. These nonsignificant interactions were dropped from the final models.

Highest level of mental health services.

Nonwhites were less likely (RRR=.68) than whites to use the highest level of mental health services compared with other mental health services (6% versus 7%; Table 3). Enlisted personnel were 1.87 times more likely than officers to report using the highest level of mental health services (8% versus 3%). Persons reporting more than one week per month of impaired functioning were 3.19 times more likely to use the highest level of mental health services (36% versus 5%). Using the highest level of mental health services was not related to gender or marital status.

In the multivariate multinomial regression model, again, nonwhites were significantly less likely than whites to use the highest level of mental health services compared with other mental health services (RRR=.63). Enlisted soldiers were more likely than officers to use the highest level of mental health services (RRR=1.82). Those who reported more than one week per month of impaired function (RRR=3.18) were more likely than others to use the highest level of mental health services.

Number of services.

Women were 1.29 times more likely than men to use a greater number of mental health services (13% versus 10%) (Table 4). Enlisted personnel were 2.03 times more likely than officers to use a greater number of mental health services (11% versus 4%). Persons with more than one week per month of impaired function were 4.60 times more likely to use a greater number of mental health services (48% versus 8%). Using a greater number of mental health services was not related to race or marital status.

In the multivariate Poisson regression model, women remained significantly more likely than men to use a greater number of mental health services (IRR=1.39). Those who were married were more likely than nonmarried persons to use a greater number of mental health services (IRR=1.19). Enlisted personnel were more likely than officers to use a greater number of mental health services (IRR=1.93), and persons with more than one week per month of impaired function remained more likely than those with fewer days of impairment to use a greater number of mental health services (IRR=4.40).

In the general U.S. population, approximately 20% receive mental health treatment in a 12-month period (4). The results of this study indicate that among active-duty U.S. Army soldiers, approximately 21% used mental health services in the previous 12 months. These rates are quite similar, although the types of mental health services used are not always comparable between military and civilian populations. Approximately 7% of soldiers used the highest levels of specialty care (mental health professional and prescribed medication). That represents 33% of those using any mental health services. Further, 15% saw a mental health professional (69% of those using mental health services) and 10% saw a general medical doctor (49% of those using mental health services), compared with approximately 9% using each type of health service in the U.S. general population (16).Although the overall mental health service use rate and the rate of obtaining mental health services from a general medical doctor were quite similar between the U.S. Army and the general population, the rate of receiving services from a mental health professional was approximately 1.5 times higher in the U.S. Army. This difference may in part be related to the ability to obtain such services directly in the Army without cost to the service member.

In the prior year, approximately 11% of soldiers were prescribed medications, representing 50% of those using mental health services. A wide variety of medical professionals may have prescribed the medications, including surgeons and obstetricians/gynecologists, among others. The prescribed medication category includes medications used for sleep, which are widely prescribed in primary care. A recent analysis of prescription data for insured adults indicates that more than 20% of American adults took at least one drug for a mental or behavioral health condition in 2010 (17). Although methods for assessing psychotropic medication use and the average age across the populations differed, these data suggest that this rate is nearly twice that of Army soldiers and does not include sleep medications.

Mental health service use varied by the degree of impaired function and demographic characteristics. Soldiers reporting more than one week per month of impaired function were more likely to have received mental health services. This was true for all service types (visits with a mental health professional or general medical doctor or visits for prescription medications). Nearly 65% of soldiers who reported more than one week of impaired function used a mental health service compared with 19% of those reporting less impaired function. Importantly, those reporting more than one week per month of impaired functioning were also three to four times more likely to use the highest level of services and to use more services. The fact that self-reported impaired functioning was significantly associated with mental health service use suggests that those in the greatest need are obtaining care. However, because impairment was self-reported in this study, it is an indicator of self-perceived need rather than a direct, objective measure of impaired function. Also, approximately 35% of those with high levels of impaired functioning did not receive care. Issues of stigma and barriers to care should be further studied, and outreach to this specific population may be required (9,18).

In this study, women, married persons, and enlisted personnel were more likely to use a mental health service and receive prescriptions for depression, anxiety, or sleeping problems. Those who were white and enlisted were more likely to use the highest level of services, whereas women, married persons, and enlisted personnel used a greater number of services. Women are often more likely to obtain services (16,19,20). This may reflect that women are more willing to identify their own mental health symptoms or are less likely to associate stigma with receiving mental health services (19). Although results regarding married persons have been mixed (21–23), married persons may be more likely to obtain services when prompted to do so from an affected spouse (19). Persons with higher service ranks are generally less likely to seek mental health services (5). This phenomenon may reflect a fear of increased scrutiny or stigma or a greater commitment to self-reliance (9,18).

Compared with individuals of other races-ethnicities, whites in the general and military populations appear to be more likely to receive any psychiatric medications and to be prescribed a higher number of psychiatric medications. Further, whites in the general population have also been shown to obtain higher levels of mental health specialty care and are somewhat less likely than those of other races-ethnicities to obtain mental health services in a primary care setting (21,24,25). Attitudes regarding mental health treatment, stigma, and relationships with primary care physicians have been posed as potential reasons for this differential usage (24,26).

The findings from this study must be interpreted in terms of several methodological considerations. Because this was a cross-sectional study, further research using longitudinal designs is recommended to better determine the course of mental health service use and its relation to demographic factors and functional impairment. Because this was a self-report survey, recall bias is a possibility and could result in overreporting or underreporting of mental health service use and functional impairment. Fortunately, those administering the survey used procedures that had been shown to promote honesty on self-report validity studies, such as ensuring respondents’ anonymity, having command leadership leave the room during the survey, and explaining the purpose of the survey (27). Further, the survey questions were somewhat limited. Additional information regarding the specifics of mental health services used, such as the duration of service and whether the presenting concern was resolved, was not available. Further, we were not able to separate the types of prescribed medications. Finally, although this study addressed the question of need via functional impairment, the use of validated measures to assess psychiatric disorders should be evaluated in future research.

Understanding the relationship of these demographic factors to access to care, diagnosis, and receipt of appropriate care are important for determining necessary changes in health care services. Even among those who receive care for a mental health problem, ensuring appropriate care is important to health care delivery. Further research is needed to understand the relationship of these findings to service access, diagnosis, and receipt of appropriate care.

The opinions expressed in this article are those of the authors and therefore do not necessarily reflect the views of the Department of Defense, the Uniformed Services University of the Health Sciences, or the Center for the Study of Traumatic Stress.

Dr. Kessler has been a consultant for Analysis Group, AstraZeneca, Bristol-Myers Squibb, Cerner-Galt Associates, Eli Lilly and Company, GlaxoSmithKline, HealthCore, Health Dialog, Hoffman-LaRoche, Integrated Benefits Institute, John Snow, Inc., Kaiser Permanente, Matria, Mensante, Merck, Ortho-McNeil Janssen Scientific Affairs, Pfizer, Primary Care Network, Research Triangle Institute, Sanofi-Aventis Groupe, Shire U.S., SRA International, Takeda Global Research and Development, Transcept Pharmaceuticals, and Wyeth-Ayerst. Dr. Kessler has served on advisory boards for Appliance Computing II, Eli Lilly and Company, Johnson and Johnson, Mindsite, Ortho-McNeil Janssen Scientific Affairs, Plus One Health Management, and Wyeth-Ayerst. He has received research support for his epidemiological studies from Analysis Group, Bristol-Myers Squibb, Eli Lilly and Company, EPI-Q, GlaxoSmithKline, Johnson and Johnson Pharmaceuticals, Ortho-McNeil Janssen Scientific Affairs, Pfizer, Sanofi-Aventis Groupe, Shire U.S., and Walgreens Co. He also owns 25% share in DataStat. Dr. Stein is coeditor-in-chief of UpToDate and a consultant for Care Management Technologies. The other authors report no competing interests.