Prevalence of Mental Health Problems and Functional Impairment Among Active Component and National Guard Soldiers 3 and 12 Months Following Combat in Iraq

Context  A growing body of literature has demonstrated the association of combat in Iraq and Afghanistan with postdeployment mental health problems, particularly posttraumatic stress disorder (PTSD) and depression. However, studies have shown varying prevalence rates of these disorders based on different case definitions and have not assessed functional impairment, alcohol misuse, or aggressive behavior as comorbid factors occurring with PTSD and depression.

Objectives  To (1) examine the prevalence rates of depression and PTSD using several case definitions including functional impairment, (2) determine the comorbidity of alcohol misuse or aggressive behaviors with depression or PTSD, and (3) compare rates between Active Component and National Guard soldiers at the 3- and 12-month time points following their deployment to Iraq.

Design  Population-based, cross-sectional study.

Setting  United States Army posts and National Guard armories.

Participants  A total of 18 305 US Army soldiers from 4 Active Component and 2 National Guard infantry brigade combat teams.

Interventions  Between 2004 and 2007, anonymous mental health surveys were collected at 3 and 12 months following deployment.

Main Outcome Measures  Current PTSD, depression, functional impairment, alcohol misuse, and aggressive behavior.

Results  Prevalence rates for PTSD or depression with serious functional impairment ranged between 8.5% and 14.0%, with some impairment between 23.2% and 31.1%. Alcohol misuse or aggressive behavior comorbidity was present in approximately half of the cases. Rates remained stable for the Active Component soldiers but increased across all case definitions from the 3- to 12-month time point for National Guard soldiers.

Conclusions  The prevalence rates of PTSD and depression after returning from combat ranged from 9% to 31% depending on the level of functional impairment reported. The high comorbidity with alcohol misuse and aggression highlights the need for comprehensive postdeployment screening. Persistent or increased prevalence rates at 12 months compared with 3 months postdeployment illustrate the persistent effects of war zone service and provide important data to guide postdeployment care.

Two longitudinal studies of veterans of Operational Iraqi Freedom (OIF) and Operational Enduring Freedom (OEF) have shown that the incidence of posttraumatic stress disorder (PTSD) is 2 to 3 times higher among those exposed to combat compared with those who did not report significant combat exposure.^1,2 These studies add to growing cross-sectional evidence linking combat duty in Iraq and Afghanistan to the development of postdeployment health problems including PTSD, depression, anxiety, and symptoms attributed to mild traumatic brain injury.^3-7

Despite the strong association between combat and mental health problems, prevalence rates between studies have varied widely, most likely owing to differences in levels of combat exposure and to the case definitions used. Studies that use structured clinical interviews are costly and difficult to conduct with working infantry units, and most large studies of OIF or OEF combat veterans have relied on standardized self-report symptom scales, particularly the PTSD Checklist (PCL), which can be scored in several ways.⁸ To our knowledge, functional impairment measures have not been incorporated into any case definitions. The lack of consistent definitions and functional impairment measures has made it difficult to determine the true effects of combat service in Iraq and Afghanistan or accurately project mental health service needs.

This study examines the prevalence rates of depression and PTSD in a large group of infantry soldiers at the 3- and 12-month time points following their deployment using several different case definitions including a measure of functional impairment. The study also evaluates alcohol misuse and aggressive behaviors, 2 comorbid conditions commonly reported in veteran populations.^9-11

In addition, this study provides a unique look at differences between Active Component and National Guard soldiers. While Active Component soldiers have a federal mission to provide full-time military support for the defense of the nation, National Guard soldiers are part of the Reserve Component and primarily have a state mission to provide support to the community as citizen soldiers. During wartime, however, National Guard units can be federally mobilized to active duty to participate in direct combat operations in the same role as Active Component infantry brigade combat teams (BCTs); they have played a central role in OIF and OEF. Although approximately one-third of service members deployed to OIF/OEF have come from Reserve Component units, there is little research on the effect of OIF/OEF deployment by component.^4,12

Between 2004 and 2007, we collected 18 305 anonymous surveys from members of 4 Active Component and 2 National Guard BCTs at 3 and 12 months postdeployment. Of the 18 305 surveys received, 13 226 were from veterans of OIF and were used for analysis. Our study extends a previous cross-sectional study conducted 3 to 4 months postdeployment³ by increasing the sample size and adding a second cross-sectional evaluation of the same units at 12 months postdeployment. The 3-month time point was chosen based on research that found that assessing mental health 3 to 4 months postdeployment is optimal.¹³ The follow-up 12-month period was selected because it is the latest time that would ensure maximal participation without interfering with preparations for subsequent deployment. It is important to note that these units are not intended to be representative of all deploying military personnel but rather typical combat maneuver units of similar structure and function that were known to be at high risk owing to their mission to conduct ground combat operations. The 13 226 surveys used for analysis were obtained from 4933 soldiers from 4 Active Component BCTs at 3 to 4 months postdeployment; 4024 soldiers from these BCTs at 12 months postdeployment; 2684 soldiers from 2 National Guard BCTs at 3 to 4 months postdeployment; and 1585 soldiers from the same National Guard BCTs at the 12-month time point.

The data were collected under a protocol approved by the Walter Reed Army Institute of Research institutional review board using procedures previously described.³ Recruitment briefings were scheduled at the convenience of the units, close to their work location. From the 6 participating BCTs, a combined total of approximately 29 460 soldiers were determined to be present in the units at the 2 time points based on information provided by unit personnel staff. Thus, the overall response rate was 62% (18 305 of 29 460 soldiers). This number is consistent with other research that used large population-based surveys and similar procedures.^3,7,14,15 Most nonparticipation was related to not attending the recruitment briefings. At each data collection, research personnel conferred with the unit leadership and medical personnel to ensure that there were no systematic reasons that soldiers were not available to attend the briefings on the scheduled days. Reasons for unavailability included other duty or training obligations, medical appointments, illness, or unscheduled or personal leave.

In addition to determining the overall response rate in the units, the participation (consent) rate was calculated for soldiers who attended the recruitment briefing. Completion of any part of the paper-and-pencil survey was considered a response. The participation rate was 97% overall, ranging from 94% to 99% for all 12 data collections. Missing data on the surveys ranged from 2% to 10%; there was an average 2% of values missing for the combat exposure items, 2% for the PTSD items, 5% for the depression items, 7% for the aggression items, and 10% for the alcohol items. Finally, to assess the representativeness of our sample, we compared the demographics of soldiers from our sample to the demographic characteristics of all Active Component and National Guard OIF veterans with a combat occupational specialty. Demographics of OIF veterans were compiled from Post-Deployment Health Assessments from Army personnel obtained after their return from Iraq using the Defense Medical Surveillance System database (Table 1).

Demographic measures included component, age, sex, race/ethnic group, education level, rank, and marital status (Table 1). A combat events checklist measured whether each event had occurred at least once during deployment (Table 2).

We examined 7 case definitions of PTSD defined in Table 3. Posttraumatic stress disorder was measured using the 17-item PCL,⁸ well-validated in civilian and military primary care and mental health settings.^8,16,17 The PCL is usually scored by summing the answers to the 17 questions (range, 17-85); a stringent cutoff of 50 or higher has been most widely used in military populations. The 7 definitions ranged from broad (liberal) to strict (conservative), based on DSM-IV B, C, and D symptom criteria, a high symptom severity cutoff score (≥50), and/or endorsement of functional impairment.^8,16,18

We examined 3 case definitions of depression defined in Table 3. Depression was measured with the 9-item Patient Health Questionnaire (PHQ-9), a well-validated clinical scale for depression based on the DSM-IV criteria that is widely used in primary care and specialty mental health settings.^18-20

Functional impairment due to depression was measured using the single item from the PHQ-9 that asks, “If you checked off any of the problems above, how difficult have these problems made it for you to do your work, take care of things at home, or get along with other people?” This question has been shown to have acceptable correlations with standard measures of functional impairment and to be predictive of those with a depression diagnosis.¹⁹ To vary specificity and remain consistent with other studies in civilian and military populations,^3,20 responses of “somewhat difficult” were considered to indicate some impairment²⁰ and responses of “very difficult” or “extremely difficult” were considered to indicate serious impairment.²⁰ Because the PCL lacks any measure of functional impairment and the single functional impairment question has been used with both the PHQ-9 and PHQ generalized anxiety scale,²¹ the same question was added below the 17 PCL questions.

Soldiers were asked 2 yes/no screening questions about their alcohol use using a validated measure widely used in primary care,²² postdeployment screening,^6,15,23 and in a predeployment sample.³ The questions asked, in the past 4 weeks, “Have you felt you wanted or needed to cut down on your drinking?” and “Have you used alcohol more than you meant to?” A “yes” endorsement on either item was considered a positive screening for alcohol misuse.

Because anger and aggression are frequently reported among combat veterans,^10,11 we asked soldiers to indicate how often in the past month they “got angry with someone and kicked or smashed something, slammed the door, punched the wall, etc,” “threatened someone with physical violence,” or “got into a fight with someone and hit the person.” These questions were based on previous research that demonstrated a strong association between combat exposure and these items as well as other risk-taking behaviors.¹¹ Endorsement of any of the items 1 or more times in the past month was an indicator of aggressive behavior. When investigating the comorbidity of aggressive behavior with depression and PTSD, we used only the most specific item, “got into a fight with someone and hit the person,” as evidence of aggressive behavior. Although there are no published baseline comparisons, this item on fighting was used in the predeployment infantry sample included in Hoge and colleagues' study³; in that sample, 11.2% of soldiers endorsed this item.

Surveys were optically scanned using the Scan Tools software package (Pearson Assessments Inc, Bloomington, Minnesota). Protocol quality control procedures stipulated that 10% of the surveys were randomly chosen and checked for scanning errors in all fields. The overall scanning error rate was 0.24% of all fields, a 99.76% accuracy rate.

The SPSS software (version 15.0; SPSS Inc, Chicago, Illinois) was used for analyses. Analyses included simple frequency and descriptive statistics, χ² tests of association, and logistic regression analysis. Logistic regression was used to determine whether differences observed from the 3- and 12-month time points were significant in both the Active and National Guard groups while controlling for rank, marital status, and combat exposure, and to determine if changes in the prevalence rates from the 3- to 12-month time points differed between the Active Component and National Guard participants.

Table 1 displays the demographic characteristics for Active Component and National Guard soldiers at the 3- and 12-month time points following deployment. The demographic characteristics of personnel who took part in the study at the 3- and 12-month time points were similar to the reference population demographics of OIF combat veterans in both the Active Component and the National Guard. There were slight differences between the 3- and 12-month time points on rank, age, and marital status. Within a year of returning from deployment, it is common for soldiers to get promoted or married, so these were not unexpected. The demographic characteristics of the study groups were similar to those of the general reference population, except that officers were undersampled, resulting in lower rank and age distributions than in the reference groups. This is likely because officers were less available owing to work-related duties. Additionally, National Guard participants included a higher percentage of African American respondents than the reference group. This may be because the 2 National Guard BCTs are from states with a higher proportion of African American individuals than the reference group, which represented 50 states. Rates of combat exposure were similar to rates reported elsewhere.^3,24 Active Component and National Guard participants reported similar rates of exposure (Table 2).

Estimated PTSD prevalence based on DSM-IV symptom criteria ranged from 20.7% (Active Component, 3-month time point) to 30.5% (National Guard, 12-month time point). Most of these soldiers reported functional impairment at the “somewhat difficult” level, and a relatively high percentage reported serious impairment (“very difficult” or “extremely difficult” level). For example, PTSD rates based on DSM-IV criteria and serious functional impairment were 7.7% for the Active Component at 3 months and 8.9% at 12 months and 6.7% for the National Guard at 3 months and 12.4% at 12 months. More than 6% of the soldiers (and up to 11% of National Guard soldiers at 12 months) met the most stringent case criteria defined by the DSM-IV plus a high level of symptoms (PCL score, ≥50) and serious functional impairment.

For depression, estimated prevalence rates ranged from 11.5% (National Guard, 3-month time point) to 16.0% (Active Component, 3-month time point) using the PHQ definition alone. When using the PHQ definition plus serious functional impairment, rates were 8.3% and 8.5% for the Active Component (3- and 12-month time points) and 5.0% and 7.3%, respectively, for the National Guard. The estimated prevalence of either depression or PTSD based on the DSM-IV and using a high-specificity cutoff (PCL score, >50) ranged from 21.8% to 22.8% for Active Component (3 and 12 months) and 18.7% to 27.8% for National Guard soldiers, with most reporting some functional impairment. Prevalence rates for PTSD or depression with serious functional impairment ranged from 11.1% to 12.3% for Active Component and 8.5% to 14.0% for National Guard soldiers at 3 and 12 months, respectively.

Estimated prevalence rates of depression, PTSD, or the combination of PTSD or depression between the 3- and 12-month time points reveal a clear pattern across case definitions (Table 4).

(1) Symptoms of PTSD increased significantly in both groups but with much larger increases observed in National Guard participants. The Active Component group showed increased prevalence rates on 4 of the 7 PTSD diagnostic criteria: PTSD according to DSM-IV (broad); PTSD according to DSM-IV with some functional impairment; PTSD according to DSM-IV and a PCL score of 50 or higher with some functional impairment; and PTSD according to DSM-IV and a PCL score of 50 or higher with serious functional impairment. In the National Guard group there were significant increases from the 3- to the 12-month time point on all 7 case definitions of PTSD. (2) Estimated prevalence rates for depression symptoms increased significantly from the 3- to 12-month time points across all case definitions in the National Guard group only. (3) Estimated prevalence rates for PTSD or depression case definitions increased significantly from the 3- to 12-month time points for both the Active Component and National Guard groups on all 4 defined criteria. (4) The magnitude of the increase between 3 and 12 months was greater for the National Guard than for the Active Component group for several outcomes using logistic regression models that included a time-component interaction (Table 4). If the interaction term was significant (P < .05), the magnitude of the change from 3 to 12 months for the National Guard group was significantly greater than the magnitude of the change for the Active Component group. In summary, depression and/or PTSD symptoms increased significantly in the National Guard from the 3- to 12-month time points across all case definitions, whereas in the Active Component, depression symptoms remained stable and PTSD symptoms increased across several criteria.

We also examined rates of alcohol misuse and aggressive behaviors (Table 5). We found a significant increase from the 3- to 12-month time point for National Guard soldiers in aggressive behaviors (ie, threatening someone with physical violence and getting into a fight and hitting the person). The prevalence rates of depression or PTSD with accompanying alcohol misuse or aggressive behavior were 9.7% and 14.7% for the 3- and 12-month time points, respectively, in the National Guard group. Using the most stringent criteria for a combination of depression or PTSD, serious functional impairment, and either alcohol misuse or aggressive behavior, the rates were 4.2% and 8.0%, respectively, for the 3- and 12-month time points in the National Guard group. The National Guard sample had higher prevalence rates across all combined criteria at the 12-month time point.

There were no significant changes from 3 to 12 months in the Active Component group for alcohol misuse or aggressive behaviors. A component effect was found; the magnitude of the increase on all comorbid criteria from the 3- to 12-month time points was significantly different for the National Guard compared with Active Component soldiers.

The 3 objectives of the study, focused on delineating the broad mental health effects of combat deployment during the first year after return from Iraq, were to (1) examine the prevalence rates of depression and PTSD using several case definitions, with the addition of functional impairment, (2) examine the comorbidity of alcohol misuse or aggression behaviors with depression or PTSD, and (3) compare prevalence rates between Active Component and National Guard study groups.

Available data from deploying military samples indicate that the expected baseline (predeployment) rates of PTSD and depression are comparable with large population samples such as those found in the National Comorbidity Study.^1-3,25 The baseline PTSD prevalence ranges from 3% to 5% (defined using the DSM definition combined with a PCL score ≥50); depression rates are also in the same range (3%-5%). It has been estimated that 9.3% of soldiers have either PTSD, depression, or generalized anxiety symptoms before deployment.³

Using the least stringent definition, we observed PTSD rates across Active Component and National Guard study groups, study time points ranging from 20.7% to 30.5%, and depression rates ranging from 11.5% to 16.0%. Using the strictest definitions with high symptom rates and serious functional impairment, PTSD prevalence ranged from 5.6% to 11.3% and depression prevalence from 5.0% to 8.5%. Between 8.5% and 14.0% of all soldiers reported serious functional impairment due to either PTSD or depression symptoms.

This is the first study, to our knowledge, that looked at the prevalence in a population of veterans of OIF across different case definitions of PTSD with the addition of a functional impairment measure. The PTSD functional impairment question was modeled after the PHQ-9 measure for comparability and validity and showed that almost all soldiers who reported PTSD symptoms according to the DSM-IV also reported some functional impairment; roughly half reported serious impairment (at the “very difficult” or “extremely difficult” level).

The selection of both the DSM and the 50-point cutoffs as anchors for several of the PCL case definitions in this study is supported by a recent review of PCL studies that showed that a high cutoff is necessary to achieve the most accurate prevalence estimate in population research (as distinct from using the test in primary or specialty care settings).¹⁶ In this review, a cutoff of 48 to 50 produced the most accurate estimate of PTSD prevalence in a hypothetical population, with a true prevalence rate of PTSD of 15%. Lower cutoff values produced significant overestimates of prevalence because of the higher number of false positives and lower positive predictive value. Applying the DSM definition to the PCL has been shown to correspond to a somewhat lower PCL cutoff of 44 in a military sample.¹⁶ Our study expands on this knowledge by applying the different definitions with the addition of functional impairment.

These prevalence rates based on functional impairment are consistent with earlier estimates in OIF/OEF infantry populations based only on high symptom endorsement.³ For example, Hoge et al reported that 12.9% of Active Component soldiers met PTSD criteria 3 to 4 months postdeployment using a DSM definition combined with a PCL score of 50 or higher. In our study, in the Active Component sample at 3 months, 14.8% met the criteria for PTSD using the same criteria that Hoge used, 12.6% met criteria using the DSM plus high sympton severity (PCL score ≥50) and some functional impairment, and 6.3% met the most stringent definition, DSM plus high symptom severity (PCL score, ≥50) and serious functional impairment.

This study also showed that comorbid alcohol misuse or aggressive behavior was common across all case definitions. Around 50% of soldiers who screened positive for depression or PTSD, based on the strict definition, also met criteria for alcohol misuse or aggressive behavior (Table 5); aggressive behaviors showed increases in both Active Component and National Guard soldiers from the 3 and 12 months time points. The significant overlap between alcohol misuse, aggressive behavior, and mental disorders highlights the high rate of comorbidity in this population. Responses to the items that assessed aggressive behavior and alcohol misuse had slightly more missing data (7% and 10%, respectively) than other study measures, likely owing to the sensitive nature of endorsing alcohol misuse and aggressive behaviors in this occupational setting. However, we do not believe that the missing data for these items occurred at high enough rates to affect the results given the high valid response rate (more than 90%) and the consistency in the findings across different population subgroups and definitions. These findings indicate that it may be beneficial to screen for alcohol and aggressive behaviors when soldiers present for treatment of PTSD or depression.

Despite efforts to systematically assess soldiers following deployment, dispel stigma, encourage treatment, and improve access to care, the prevalence rates across the study's case definitions showed increases from the 3- to 12-month time points. These data make clear that, at 12 months following deployment, many combat soldiers have not psychologically recovered, which has immediate implications for current Department of Defense policy and troop rotations supporting OIF and OEF. The time between deployments (dwell time) has been 12 to 18 months for many Active Component combat units. While these data do not directly assess whether increasing dwell time between deployments would be associated with lower mental health disorder rates in returning veterans, the data indicate that, for many, 12 months appears to be insufficient time to recover. Because PTSD may develop or persist months after exposure to trauma, interventions are thought to be best provided as early as possible after returning home.⁶ Providing the time for treatment, intervention, and psychological recovery following deployment is particularly important because many Active Component BCTs have deployed 3 or 4 times to Iraq or Afghanistan, and many National Guard BCTs have also deployed on more than 1 combat tour in the past 8 years, with each extending greater than 12 months with predeployment training.

Increases in the prevalence rates of mental health problems from the 3- to 12-month time points postdeployment were significantly greater among National Guard soldiers. National Guard and Active Component soldiers reported similar rates of combat experiences and similar prevalence rates of mental health problems 3 months postdeployment. Therefore, the emergence of differences by 12 months likely does not have to do with differences in the health effects of combat but rather with other variables related to readjustment to civilian life or access to health care (as noted previously by Department of Defense researchers⁶). Because National Guard soldiers return to civilian status following their deployment, they do not have the same uninterrupted access to military medical care as Active Component soldiers. National Guard soldiers have access to free military medical coverage until 6 months following their deployment, after which they may purchase additional coverage or receive care at veterans affairs medical facilities.²⁶ However, these facilities are often not as conveniently located compared with medical facilities on the same posts as the Active Component units. Other potential differences between National Guard and Active Component soldiers include the time spent continuing to work with unit peers who may provide support for deployment-related problems and the stresses of reintegrating with civilian society and civilian employment. Another difference is that at 12 months postdeployment, Active Component soldiers are becoming collectively focused on the multiple tasks required to prepare a unit for another deployment including heavy training schedules (ie, field exercises). This high level of work engagement may result in less time to address personal problems or lingering mental health issues from a previous deployment.

The data reported here were collected using a cross-sectional design similar to previous studies with large intact military units.^3,7 Because the survey was anonymous, the data are not longitudinal. However, we are confident that the cross-sectional data were representative of soldiers in combat infantry units who returned from Iraq. Although some soldiers move to other duty locations or leave service shortly after returning home, most remain in the unit with which they deployed. The estimated turnover is 28% during the first year post-deployment (based on the surveys received from those who did not deploy), and it is likely that many soldiers in our study completed both surveys. The study groups also had comparable demographics with the Army combat occupational reference groups, and combat exposure levels were similar to those found in other studies.^3,24

There is potential selection bias in terms of soldiers who were available to take part in the survey. The participating units scheduled the recruitment and survey sessions to minimize adverse effects on the work requirements. However, some soldiers may not have been available owing to reasons such as temporary duty elsewhere, being on leave, or attending schools or training. Soldiers who were ill or injured, or who had been removed from the unit for administrative reasons such as drug abuse or misconduct, would also not have been available to take part in the survey. This potential bias is most likely to lead to underreporting of mental health problems in the study population compared with the larger population. Compensation-seeking bias is not likely, given that the surveys were anonymous and not linked in any way with health care or disability agency processes. Although officers typically have lower rates of mental health problems than enlisted soldiers, the undersampling of officers would have had a minimal effect on overall prevalence rates because officers account for less than 15% of all personnel in these combat units.

Additionally, although the sample only included 2 National Guard BCTs (compared with 4 Active Component BCTs), the findings of this study were consistent with another study indicating that soldiers from Reserve Component units appear to have greater increases in mental health problems after returning home than soldiers from active units.⁶ As a whole, our sample of National Guard soldiers was similar to the National Guard reference population on demographic variables other than race/ethnicity. Most importantly, National Guard soldiers were comparable with Active Component soldiers on combat exposure, a leading predictor of postdeployment mental health problems.

From an epidemiological perspective, the present study is an important contribution to the literature because it provides a comprehensive assessment of how different case definitions influence PTSD and depression prevalence in combat veterans. Using the same case definition reported by Hoge and colleagues³ across study groups and time points, 23% to 31% of soldiers described symptoms that met DSM criteria for PTSD or depression along with some functional impairment. Using an even stricter case definition including reporting serious functional impairment across study groups and time points, 9% to 14% still met criteria. Further studies are needed to better understand the nature and severity of the impairment along multiple dimensions to include work, family, and social relationships. Research is also needed to better quantify the effect of comorbid factors such as alcohol misuse, aggression, risk-taking behaviors, and physical symptoms to understand the full extent of the effects of war-related trauma.

We believe that these data have clear implications for the care of soldiers and their families. The findings of the study show that at 12 months following combat, the prevalence of mental health problems among veterans does not abate, and in many cases, increases. It is a virtual certainty that soldiers who remain in service will deploy again; this study shows that a sizable proportion (9%-14%) have depression or PTSD symptoms with serious functional impairment. Data collected from the US Army's Mental Health Advisory Teams has clearly demonstrated that multiple deployments are associated with a higher prevalence of mental health problems,²⁴ and the cumulative effects of combat deployments are worrisome. These data also have implications for individual soldiers and unit peers. If soldiers who are struggling with serious functional impairment as the result of a previous deployment are deployed again, there is potential that this could impair their performance in combat. This has implications for the safety of unit members and mission success. Further research is needed to understand the effects of self-perceived serious impairment and military occupational performance.

Our findings showed that National Guard soldiers' mental health problems increased dramatically from the 3- to 12-month time points. Therefore, it is imperative that members of the National Guard and other Reserve Component units have as ready access to care as Active Component members beyond the first few months of returning home. Functionally and culturally, National Guard soldiers are different from Active Component soldiers in that they return to civilian life after combat and have more restricted health care access.²⁶ It will continue to take a collective effort from Department of Defense, Veterans Affairs, community providers, and veteran organizations to help this generation's veterans readjust after service in Iraq and Afghanistan.

Correspondence: Jeffrey L. Thomas, PhD, Walter Reed Army Institute of Research, Department of Military Psychiatry, Walter Reed Army Institute of Research, 503 Robert Grant Ave, Silver Spring, MD 20910 (jeffrey.l.thomas@us.army.mil).

Submitted for Publication: March 24, 2009; final revision received November 3, 2009; accepted November 9, 2009.

Author Contributions: Dr Thomas had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Financial Disclosure: None reported.

Funding/Support: This study was supported by the Military Operational Medicine Research Area Directorate, US Army Medical Research and Materiel Command, Fort Detrick, Maryland.

Disclaimer: The material has been reviewed by the Walter Reed Army Institute of Research. There is no objection to its presentation and/or publication. The opinions or assertions contained herein are the private views of the authors, and are not to be construed as official, or as reflecting true views of the Department of the Army or the Department of Defense.

Additional Contributions: We thank the leaders and the soldiers of the units studied for their service to our nation and their participation in the study. We thank Paul Bliese, PhD, and Richard Herrell, PhD, for their guidance on the manuscript. We thank the Walter Reed Army Institute of Research Land Combat Study Team who collected the data: Wanda Cook, Allison Whitt, Tony Cox, MSW, Oscar Cabrera, PhD, Michael Wood, PhD, Dave Cotting, PhD, Tim Allison-Aipa, PsyD, Julie Clark, MA, Paul Kim, MA, Karen Eaton, MA, Matthew Baker, MA, Athena Kendall-Robbins, MA, Kyle Schaul, Megan Legenos, Nickolas Hamilton, Lloyd Shanklin, Nadia Kendall-Diaz, Duriel Randolph, Michael Brouillard, Lakisha Holley, and Akeiya Briscoe-Cureton.