Authors alone are responsible for opinions expressed in the contribution and for its clearance through their federal health agency, if required.

Transcription

1 ORIGINAL ARTICLES Authors alone are responsible for opinions expressed in the contribution and for its clearance through their federal health agency, if required. MILITARY MEDICINE, 177, 4:366, 2012 Postdeployment Mental Health Screening: An Application of the Soldier Adaptation Model S. Cory Harmon, PhD; CPT Timothy V. Hoyt, MS USA; Michael D. Jones, PhD; Joseph R. Etherage, PsyD, ABPP; John C. Okiishi, PhD ABSTRACT The Global War on Terrorism and its corresponding frequent and long deployments have resulted in an increase in mental health concerns among active duty troops. To mitigate these impacts, the Department of Defense has implemented postdeployment screening initiatives designed to identify symptomatic soldiers and refer them for mental health care. Although the primary purpose of these screenings is to identify and provide assistance to individuals, macrolevel reporting of screening results for groups can assist Commanders, who are charged with ensuring the wellbeing of their soldiers, to make unit-level interventions. This study assesses the utility of a metatheory of occupational stress, the Soldier Adaptation Model, in organizing feedback information provided to Army Commanders on their units postdeployment screening results. The results of a combat brigade of 2319 soldiers who completed postdeployment screening following return from Iraq were analyzed using Structural Equation Modeling to assess the Soldier Adaptation Model s use for macrolevel reporting. Results indicate the Soldier Adaptation Model did not strengthen the macrolevel reporting; however, alcohol use and reckless driving were found to mediate the relationship between combat exposure and numerous mental health symptoms and disorders (e.g., post-traumatic stress disorder, anger, depression, anxiety, etc.). Research and practice implications are discussed. INTRODUCTION Over 2 million U.S. service members have deployed to Afghanistan or Iraq in the past decade with over a third of these soldiers deploying more than once. 1 Soldiers returning from deployments have increased mental health symptoms, with greater combat exposure associated with more reported symptoms. 2 Increased mental health problems significantly impact soldier functioning in multiple ways, including job success, family stability, physical health, and social relationships. Higher levels of mental health symptomatology are associated with increases in health care utilization, physical health symptoms, and absenteeism, even after controlling for injuries. 3 Mental health problems create a significant impact on military readiness. Over the past 10 years, mental disorders accounted for more service member hospitalizations than any other diagnostic category, and the overall incidence rates of mental disorders in the Army were higher during this time period than Madigan Healthcare System, Joint Base Lewis/McChord, MCHJ-CP-BH, Building 9040,9040A Fitzsimmons Avenue, Tacoma, WA any other military branch. 4 The effects of increased combat deployments on the physical and emotional well-being of service members are apparent, and efforts are underway to identify at-risk soldiers and route them to appropriate care. Postdeployment Screening The Department of Defense (DoD) elected to use postdeployment screening to identify military members who may need mental health services. 5 Postdeployment screening became mandatory in 1997, 6 with the creation of the Postdeployment Health Assessment (PDHA) occurring soon after. The PDHA is a two part process consisting of mental and physical health screening questions completed proximal to return from a deployment and a face-to-face encounter with a medical provider (e.g., not a mental health provider) who reviews the results and makes referrals. Initial reviews of PDHA data suggested immediate postdeployment screening resulted in significant identification errors. 7 A majority of soldiers who report initial post-traumatic stress disorder (PTSD) symptoms later indicate that the symptoms remit within a few months of returning home and do not 366 MILITARY MEDICINE, Vol. 177, April 2012

2 develop PTSD. Conversely, an important cohort of soldiers does not report PTSD symptoms until 3 to 6 months after deployment. 8,9 These findings raised questions regarding the timing of screenings 10 and led to the creation of the Post Deployment Health Re-Assessment (PDHRA). PDHRA is a screening initiated DoD-wide in 2005 that takes place 90 to 180 days after return from theatre. Like PDHA, this screening includes a physical health component while maintaining a focus on mental health symptoms. The PDHRA includes a process for providing Commanders a monthly unit update. These updates, however, only report soldier participation levels and do not address screening outcomes. 11 In conjunction with the advent of PDHRA, researchers and clinicians at Joint Base Lewis/McChord (JBLM) developed an enhanced comprehensive screening program (see Ref. 12), which expanded to Schofield Barracks in In this screening, soldiers are queried about mental and physical health concerns, meet with a credentialed mental health provider (e.g., psychologist, social worker) for a 15-, 30-, or 60-minute appointment depending on the concerns identified in the questionnaires, are educated about healthy postdeployment behaviors, and receive mental and physical health referrals, as needed. This represents an enhanced PDHRA, as the program developed at JBLM asks soldiers additional questions beyond those contained in the standard PDHRA and includes one-on-one, face-to-face contact with a mental health provider for 100% of soldiers, which is not part of the standard PDHRA process. Further, as part of the initiative to deploy JBLM s enhanced PDHRA to another installation, revised aggregate reports (e.g., Commander s Reports) were created for Schofield Barracks Commanders. Feedback to Commanders Although it has been suggested screenings should include feedback to Commanders on the mental health of their force to facilitate resource allocation decision making, 13 there is FIGURE 1. Commander s Report sample page. MILITARY MEDICINE, Vol. 177, April

3 no systemic method for providing this information to Commanders who are responsible for the soldiers they lead. 14 PDHRA-based Commander s Reports were developed at JBLM in order to facilitate the dissemination of feedback that can assist Commanders in making unit-level interventions. These reports provide detailed, aggregated information to Company Commanders on how the screening results of soldiers in their command (approximately 140 individuals) compare to the screening results of all soldiers who have completed the enhanced PDHRA screening (approximately 26,000 soldiers). See Figure 1 for a sample page of a Commander s Report detailing results, research, and recommendations for a unit s leadership climate. Nineteen health-related constructs are included in the report. A green/amber/red color coding system compares each Company to the population of soldiers who have taken the survey. Green indicates the Company s soldiers reported symptom levels that are less than or equivalent to the soldier population; amber and red indicate Company soldiers reported successively worse symptom levels than the overall population of soldiers. In addition to comparison of Company data to benchmarks, Commander s Reports include specific research-based recommendations for improving unit and soldier functioning for health-related behaviors classified as amber or red. The feedback includes a Research section that summarizes relevant research on the construct of interest. For example, for the Anxiety construct, Commanders are provided information about the relationship between performance and arousal and how high arousal can be optimal in combat yet may be harmful after deployment as soldiers may be more likely to miss important details and misinterpret situations when hyperaroused. The Recommendations section provides examples of specific actions Commanders can take. For example, in units with above average Anger levels, it is suggested Commanders consider encouraging soldiers to seek an assessment for PTSD and to encourage angry soldiers to seek support. Soldier Adaptation Model Information in the Commander s Reports is summarized according to the Soldier Adaptation Model (SAM). 15 The SAM was developed at Walter Reed Army Institute of Research as a metatheory to organize occupational stress research. The SAM is comprised of three domains: Stressors, Moderators, and Strains. Stressors consist of events or circumstances that place stress on soldiers physical and mental fitness (e.g., being overburdened at work). Moderators are attitudes and circumstances that increase or decrease the impact of stressors (e.g., job involvement, leadership climate), and Strains consist of potential outcomes from exposure to different stressors and moderators (e.g., job satisfaction, commitment to organization). The Commander s Report is organized according to this model, with Stressors, Moderators, and Strains classified overall as green/ amber/red based on the summary of effect size comparisons. The purpose of this study is to assess one method of providing unit-level postdeployment health screening information to Commanders. Specifically, this study assesses whether the SAM is an appropriate model for organizing postdeployment mental and physical health information. Further, it is unknown whether organization of common mental and physical health symptoms into a relational structure will be helpful in communicating important health information to Commanders. The current study tests the appropriateness of this rationally derived model by comparing it to real-world data collected during PDHRA screening. METHODS Participants Participants were 2,413 U.S. Army combat brigade soldiers who completed screening at Schofield Barracks, Hawaii, following return from deployment in Iraq. The median time since return from deployment was 4 months. The average number of TABLE I Summary of Participant Demographics Variable % n Men (%) ,226 Ethnicity (%) White ,527 Hispanic Black Pacific Islander Asian Native American/Alaskan Native Other Marital Status (%) Married ,370 Single Divorced/Separated Education (%) Some High School High School or Equivalent Some College Year Degree Year Degree Graduate or Professional Degree Rank (%) E-1 to E ,048 E-5 to E ,064 W-1 to W-5 and O-1 to O MOS (%) Infantry ,018 Combat Support Artillery Armor Military Intelligence Medical Signal Corps Military Police/JAG Engineers Human Resources N = 2319; Some categories may not sum to 100% because of skipped items, participant endorsement of more than 1 race/ethnicity, and rounding error. 368 MILITARY MEDICINE, Vol. 177, April 2012

4 deployments within the past 5 years was 1.89 (SD = 1.206). The modal number of deployments was one. The screening software permits soldiers to skip questions; however, missing data were minimal. 94 soldiers were excluded because of one or more missing items, yielding a total sample of 2,319. Of the 2,319, a total of 108 soldiers had partially complete demographic data. Data from these participants were retained as missing data was not directly related to constructs of interest. The average age of participants was 26.9 years (SD = 5.49, range: 19 54). Participant demographic characteristics are summarized in Table I. Procedure As described earlier, all returning service members are screened 90 to 180 days following return from deployment. Participant responses are reviewed by medical staff who discuss the information with the soldier, provide psychoeducation about postdeployment adjustment, and make referrals for care, as indicated (see Ref. 12). The Madigan Healthcare System Institutional Review Board approved use of de-identified screening data for analysis. Measures The health constructs were classified into Stressors, Moderators, and Strains by a team of psychologists based on clinical judgment. The measures were classified as follows: Stressors Four categories of hypothesized stressors were assessed. Combat Experience was assessed with four dichotomous questions: (1) During combat operations, did you kill others in combat (or have reason to believe others were killed as a result of your actions)? (2) During combat operations did you see the bodies of dead soldiers or civilians? (3) During combat operations did you personally witness anyone being killed? and (4) During combat operations did you become wounded or injured?. Yes responses were summed to create a Combat Experience total score, with range of 0 to 4. Relationship Health was assessed by one question: Since return from your deployment, have you had serious conflicts with your spouse, family members, close friends, or at work that continue to cause you worry or concern? (rated on a three-point scale of No, Unsure, and Yes ). Physical Health was assessed by one question: During the past 4 weeks, how difficult have physical health problems (illness or injury) made it for you to do your work or other regular daily activities? (rated on a 4-point scale ranging from Not difficult at all to Extremely difficult ). Finally, Legal and Financial Problems was assessed by three dichotomous questions: (1) Are you having legal problems? (2) Have you had a check returned in the last 30 days? and (3) Are you behind in your mortgage, rent, or loans?. Yes responses were summed to create a total score with range of 0 to 3. Moderators The first hypothesized moderators, Leadership Effectiveness and Unit/Military Morale, were assessed by 12 items from the Deployment Risk and Resilience Inventory (DRRI). 16 The DRRI scales have demonstrated good reliability and validity in a number of military samples. 17,18 Responses were summed to create a total score for Leadership Effectiveness, with range of 0 to 16, and a total score for Unit/Military Morale, with range of 0 to 32. Coefficient alpha for these scales was and 0.915, respectively. Sexual Assault was assessed by two dichotomous questions: (1) While you were deployed, did you experience any unwanted sexual attention, like verbal remarks, touching, or pressure for sexual favors? (2) While you were deployed, did anyone use force, threat of force, or coerce you to have sex against your will?. Yes responses were summed to create a total score, ranging from 0 to 2. The final moderator, Sleep, was assessed by one question: How many hours of sleep do you get per night? (rated on a 6-point scale ranging from 4 hours or less to 9 or more hours). Strains Five strains were assessed. Anger was evaluated using the Dimensions of Anger Reactions scale (DAR), 19 a 7-item, selfreport measure with strong psychometric properties. 20 Coefficient alpha for the DAR in the current sample was Anxiety and Depression were measured using scales from the Patient Health Questionnaire (PHQ). 21 The PHQ Anxiety scale is a 7-item gated measure; respondents who deny feeling nervous in the prior 4 weeks are not asked the six remaining questions. In this study, all soldiers answering Several days or More than half the days to the question Over the last 4 weeks, how often have you felt nervous, anxious, on edge, or worrying a lot about different things? were presented with the full PHQ Anxiety scale, and a final score was calculated by summing the items using the following values (0 = Not at all, 1 = Several days, and 2 = More than half the days). The PHQ scales have been validated in a number of veteran samples, showing good psychometric properties. 22,23 Coefficient alpha for the PHQ Anxiety and PHQ-9 Depression scales in the current sample was and 0.870, respectively. PTSD Symptoms were measured using the primary care-ptsd screen (PC-PTSD), 24 a four-item measure of PTSD symptoms shown to be psychometrically strong in military samples. A positive score on the PC-PTSD indicates PTSD may be present; further assessment is needed to determine whether the soldier has the disorder. Coefficient alpha for the PC-PTSD screen in the current sample was Alcohol Abuse was measured by the Alcohol Use Disorder Identification test (AUDIT), 25 a10-item screening measure developed by the World Health Organization to assess excessive drinking. The AUDIT was gated in this study; soldiers denying all alcohol use were not asked the remaining nine questions. The AUDIT has strong reliability and validity 26 and has been used widely in military veteran MILITARY MEDICINE, Vol. 177, April

5 samples. 27,28 Coefficient alpha for the AUDIT in the current sample was Three additional strains were measured using specific items. Suicidal Ideation was measured by one item from the PHQ 21 : Over the last 2 weeks, how often have you had thoughts that you would be better off dead, or of hurting yourself in some way? (rated on a four-point scale from Not at all to Nearly every day ). Driving Problems were measured by two questions assessing problematic driving behaviors: (1) How many times in the past month did you drive 15 mph or more over the speed limit? (rated on a 5-point scale ranging from None to 4 or more ), and (2) How many times in the last month did you drive or ride when the driver had perhaps too much to drink? (rated on an 11-point scale ranging from 0 to 10 or more). Finally, Hearing Protection was measured by one question: What percent of the time do you wear your hearing protection when you are around loud noise in your work environment? (rated on a 3-point scale from Less than 75% to 100% ). Data Analytic Strategy Structural Equation Modeling was used to evaluate the suitability of the SAM for explaining the relationships among constructs. The data were divided into two random halves in order to test and confirm theoretical models of the postdeployment adjustment process. Half of the sample was utilized in order to test two initial theoretical models. The second half of the sample was utilized in order to crossvalidate the better fitting model. Models were constructed using AMOS RESULTS Stressors On average, soldiers reported experiencing 0.91 (SD = 1.07) of the Combat Experience events. The most frequently reported event, seeing dead bodies, was endorsed by 50.6% of the sample (n = 1174). Additionally, 23.2% (n = 537) indicated witnessing someone being killed. Approximately 1 in 10 soldiers (11.9%; n = 276) reported killing others or believing their actions were responsible for another s death. Finally, 5.4% (n = 126) indicated they were wounded or injured. Relationship Health findings were consistent with Milliken et al 8 who found 14% of soldiers returning from Iraq reported interpersonal conflict at PDHRA screening. A total of 16.8% (n = 389) of respondents in this sample endorsed relationship conflicts. Most soldiers (65.8%; n = 1526) denied functional impairment stemming from Physical Health problems in the 4 weeks before screening. However, 29.2% (n = 677) reported physical health problems were Somewhat difficult, 3.8% (n = 89) reported Very difficult, and 1.2% (n = 27) indicated physical health problems were making functioning Extremely difficult. Generally, most respondents denied Legal/Financial Problems ; 3.4% (n = 79) reported legal problems, 2.2% (n = 51) reported falling behind on financial obligations, and 1% (n = 24) reported a recently returned check. Moderators On average, soldiers reported Leadership Effectiveness and Unit/Military Morale scores that are lower than published reports of non active duty samples. On average, the DRRI score was (SD = 11.43). In contrast, earlier studies yielded average scores of (SD = 10.08) and 41.3 (SE = 0.7) for National Guard/Reserve soldiers and OEF/ OIF veterans, respectively. 18,30 Regarding Sexual Assault, unwanted sexual attention was endorsed by 9.7% (n = 9) of females and 0.4% (n = 9) of males. Additionally, one male and one female reported forced sexual intercourse during deployment. Soldiers reported a median of 6 hours of Sleep per night, with 15.1% (n = 351) indicating 4 or fewer hours and 9% (n = 208) reporting 8 hours. Strains Soldiers reported an average DAR ( Anger ) score of (SD = 11.48); nearly a quarter of the sample (28.7%; n = 666) denied all anger symptoms. It is unknown how this compares to other population screens utilizing DAR, but is significantly lower than the average intake score of 34.7 (SD = 12.2) for Australian veterans diagnosed with PTSD. 20 Nearly half of the soldiers (49.3%; n = 1144) denied feeling anxious or nervous and were not presented with the remaining PHQ Anxiety questions. With the inclusion of these soldiers, the mean PHQ Anxiety score was 3.41 (SD = 4.06). The average PHQ-9 Depression score was (SD = 4.48), which is similar to the mean for nondepressed patients reported in the validation sample. 31 Soldiers reported fewer PTSD Symptoms on the PTSD screener than the Milliken et al 8 sample. A total of 9.5% (n = 221) scored 2 or more, and 5.0% (n = 116) scored three or more, compared with 16.7% and 9.1%, respectively. Further, 84.6% of soldiers denied experiencing any of the four PC-PTSD Screen symptoms. With regard to Alcohol Use, 17.4% (n = 403) scored at or above the AUDIT cut score of 8. This is consistent with results from a multisite study of male veterans that indicated 18.6% of respondents met or exceeded the cutoff. 32 A total of 2.1% (n = 50) endorsed experiencing Suicidal Ideation in the prior 2 weeks. A prior study 8 found that when soldiers were asked this question by a primary care provider during a PDHRA screening interview, 0.6% of soldiers endorsed suicidal ideation. Driving Problems in the form of drinking and driving were endorsed by 4.7% (n = 110) of the soldiers. A total of 14.7% (n = 340) reported driving 15 mph or more over the speed limit four or more times in the previous month. Soldiers report some failure to wear appropriate hearing protection at work; 24.5% (n =569)ofthesoldiers reported utilizing appropriate hearing protection less than 75% of the time. 370 MILITARY MEDICINE, Vol. 177, April 2012

6 FIGURE 2. Initial model. Testing the SAM Statistical comparisons of the split-half datasets were completed to test for equivalence of data and to ensure randomization was accomplished. The two datasets were equivalent in age t (2319) = 0.37, ns; gender, x 2 (1, n = 2319) = 0.010, ns; pay grade x 2 (17, n = 2298) = , ns; marital status x 2 (3, n = 2301) = 0.457, ns; education, x 2 (5, n = 2299) = 4.947, ns; and race x 2 (1, n = 2311) = , ns. The initial hypothesized model organized the postdeployment survey measures into Stressors, Moderators, and Strains to test the fit of the Commander s Report Feedback proposed by the SAM (see Figure 2). Fit statistics for Model 1 indicated a poor fit, x 2 (117, n = 1160) = , p < 001, CFI = 0.720, GFI = 0.877, TLI = 0.675, RMSEA = Modification indexes were consulted to determine variables and regression paths contributing to poor fit in Model 1. Three items with restricted variance were removed. These included: sexual assault, which was denied by 99% of the sample; drinking and driving, which was denied by 95.3%; and suicidal thoughts, which were denied by 97.8%. Because of high correlation (r = 0.768) and typical use of the DRRI scale, Leadership Effectiveness and Unit Morale were combined into a single variable, which showed greater correlation with other strains. Hearing Protection was excluded because of poor loading and less conceptual relevance to a model of postdeployment adaptation to stressors. The inclusion of hearing protection in the screening highlights a potential issue in the development of postdeployment assessment programs. Although hearing loss is a chronic problem among soldiers and results in significant disability rates, 33 hearing protection behavior may not directly relate to combat experience. Regarding stressors, follow-up exploratory analyses using item-level correlations suggested combat exposure was the predominant stressor. Thus, combat experience was retained as the sole stressor, with each associated item contributing unique variance. The remaining stressor variables (physical health, relationship, and financial/legal problems) showed stronger relationships to other strain variables and were realigned. When reviewing the model modification indices, two proposed strains (alcohol use and reckless driving) showed a weak relationship to other identified strains and stronger relationships with combat experiences. These variables were realigned as mediators and conceptually relabeled risky behavior. These modifications resulted in a new model (see Fig. 3) that identifies direct and indirect effects of combat experiences on strains. In addition to a direct effect on mental health outcomes, combat exposure seems to also have a relationship with risky behavior (as measured by alcohol use and driving behavior), which in turn has its own impact on mental health outcomes. Fit statistics for Model 2 were adequate, x 2 (87, n = 1160) = , p < 0.001, CFI = 0.949, GFI = 0.964, TLI = 0.938, RMSEA = Cross validation similarly indicated acceptable fit of the model, x 2 (87, n = 1159) = , p <0.001,CFI=0.948,GFI=0.963,TLI=0.938, RMSEA = DISCUSSION The SAM is an attractive model to describe how individual and unit-level protective factors can influence the relationship MILITARY MEDICINE, Vol. 177, April

7 FIGURE 3. Final model. between soldier stressors and resulting symptoms. It is well suited to the military because it reduces the range of postdeployment reactions into components that are separately addressable by military leadership or medical providers. Unfortunately, when using postdeployment screening instruments, this initial evaluation failed to adequately explain the observed relationships between variables in a combat brigade of postdeployment soldiers. The observed relationships do not support a model by which unit support and sleep mediate the relationship between stressors and strains. Following modification, the final model shows combat exposure has an effect on postdeployment mental health symptoms as well as impacting alcohol use and reckless driving, which also have an impact on mental health symptoms. These relationships held up in a cross-validation sample. Organizing aggregate results of postdeployment screening for Commanders using the SAM is not indicated. Clinical/Practice Implications These results indicate risky behavior mediates the relationship between combat exposure and mental health outcomes like PTSD. Clinically, this indicates additional time might be well-spent treating risky behaviors as a primary concern. Providers should consider focusing assessments and interventions on specific risky behaviors instead of PTSD in general. This is in concert with Jakupcak et al. 34 who found PTSD treatment had little direct effect on anger symptoms even though anger had been previously considered to be an effect of PTSD. Local military leaders can have confidence that an increased focus on treating risky behaviors will pay dividends in reducing other mental health concerns. Methodological Limitations Methodological limitations likely impacted the results of this analysis. The questionnaire items were part of a standard (although enhanced) screening event designed for clinical, rather than research, purposes. As a result, many items were chosen for brevity and clinical utility and not to maximize research potential. Further, some constructs were measured with one-item scales, and others were doubled-barreled. This led to measurement models with potentially constrained variability and may have made it difficult to tease out meaningful relationships as reduced item and scale variability can artificially reduce correlations with external variables. Additionally, some data were highly skewed because of low base rates. Greater numbers of items evaluating each content area would likely yield better variability for hypothesis testing. Finally, as the measure is entirely self-report, it is possible that soldiers may have de-emphasized or overemphasized symptoms. As the enhanced PDHRA did not include a measure of impression management, it is not possible to directly assess this. Future Directions Although the study sample was adequate for this evaluation (i.e., a postdeployment combat brigade with a large number of soldiers across a wide range of supportive and combat roles), the nature of standard postdeployment screening constrained the study methodology. Attempts to replicate these results with 372 MILITARY MEDICINE, Vol. 177, April 2012

8 similar samples using better instruments may be useful. If results remain consistent, it may suggest policy makers might better balance resources by treating risky behaviors as separate concerns worthy of primary attention and interventions. Future studies could both address the clinical implications of this study and evaluate alternative methods for effectively presenting actionable aggregate postdeployment mental health results to Commanders. Additional assessment of the impact of the feedback information on the actions of Commanders is also warranted. REFERENCES 1. Tan M: Two million troops have deployed since 9/11. Marine Corps Times, December 18, Available at com/news/2009/12/military_deployments_121809w/; accessed June 6, Hoge C, Castro C, Messner S, McGurk D, Cotting D, Koffman R: Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. N Engl J Med 2004; 351(1): Hoge C, Terhakopian A, Castro C, Messer S, Engel C: Association of posttraumatic stress disorder with somatic symptoms, health care visits, and absenteeism among Iraq war veterans. Am J Psychiatry; 164(1): AFHSC: Selected mental health disorders among active component members. U.S. Armed Forces, Available at viewmsmr?file=2010/v17_n11.pdf; accessed September 14, Department of Defense: Department of Defense Instruction : Deployment health. Washington, DC, Department of Defense, Available at accessed September 14, Department of Defense: Department of Defense Instruction : Implementation and application of joint medical surveillance for deployments. Washington, DC, Department of Defense, Available at background_inception/dodi_6490.3_aug97.pdf; accessed September 14, Gray M, Bolton E, Litz B: A longitudinal analysis of PTSD symptom course: delayed-onset PTSD in Somalia peacekeepers. J Consult Clin Psychol 2004; 72(5): Milliken C, Auchterlonie J, Hoge C: Longitudinal assessment of mental health problems among active and reserve component soldiers returning from the Iraq war. JAMA 2007; 298(18): Southwick S, Morgan C, Darnell A, et al: Trauma-related symptoms in veterans of Operation Desert Storm: a 2-year follow-up. Am J Psychiatry 1995; 152: Bliese P, Wright K, Adler A, Thomas J, Hoge C: Timing of postcombat mental health assessments. Psychol Services 2007; 4(3): Appenzeller G, Warner C, Grieger T: Postdeployment health reassessment: a sustainable method for brigade combat teams. Mil Med; 172(10): Gahm G, Swanson R, Lucenko B, Reger M: History and implementation of the Fort Lewis Soldier Wellness Assessment Program (SWAP). Mil Med; 174: Wright K, Thomas J, Adler A, Ness J, Hoge C, Castro C: Psychological screening procedures for deploying U.S. Forces. Mil Med 2005; 170(7): Department of the Army: Army Regulation : Army Command Policy. Washington, DC, Department of Defense, Available at accessed December 9, Bliese P, Castro C: The Soldier Adaptation Model (SAM): applications to peacekeeping research. In: The Psychology of the Peacekeeper: Lessons From the Field, pp Edited by T. Britt and A. Adler Westport, CT, Praeger Press, King L, King D, Vogt D, Knight J, Samper R: Deployment Risk and Resilience Inventory: a collection of measures for studying deploymentrelated experiences of military personnel and veterans. Mil Psychol 2006; 18(2): Vogt D, Proctor S, King D, King L, Vasterling J: Validation of scales from the Deployment Risk and Resilience Inventory in a sample of Operation Iraqi Freedom veterans. Assessment 2008; 15: Renshaw K: Deployment experiences and postdeployment PTSD symptoms in National Guard/Reserve service members serving in Operations Enduring Freedom and Iraqi Freedom. J Trauma Stress 2010; 23(6): Novaco R: Dimensions of Anger Reactions. Irvine, CA, University of California, Forbes D, Hawthorne G, Elliott P, et al: A concise measure of anger in combat-related posttraumatic stress disorder. J Trauma Stress 2004; 17(3): Spitzer R, Kroenke K, Williams J: Validation and utility of a self-report version of PRIME-MD. JAMA 1999; 282(18): Maguen S, Lucenko B, Reger M, et al: The impact of reported direct and indirect killing on mental health symptoms in Iraq War veterans. J Trauma Stress 2010; 23(1): Reger M, Gahm G, Swanson R, Duma S: Association between number of deployments to Iraq and mental health screening outcomes in US Army soldiers. J Clin Psychiatry 2009; 70: Prins A, Ouimette P, Kimmerling R, et al: The primary care PTSD screen (PC-PTSD): development and operating characteristics. Primary Care Psychiatry 2004; 9: Babor T, Higgins-Biddle J, Saunders J, Monteiro M: The Alcohol Use Disorders Identification Test: Guidelines for use in primary health care, Available at a.pdf; accessed Decemer 7, Fiellin D, Carrington R, O Connor P: Outpatient management of patients with alcohol problems. Ann Intern Med 2000; 133(10): Bradley K, Maynard C, Kivlahan D, McDonell M, Fihn S: The relationship between alcohol screening questionnaires and mortality among male veteran outpatients. J Stud Alcohol 2001; 62(6): Hawkins E, Lapham G, Kivlahan D, Bradley K: Recognition and management of alcohol misuse in OEF/OIF and other veterans in the VA: a cross-sectional study. Drug Alcohol Depend 2010; 109(1 3): Arbuckle J: (2009). AmosÔ 18 user s guide. Chicago: SPSS, Inc. 30. Pietrzak R, Goldstein M, Malley J, Rivers A, Southwick S: Structure of posttraumatic stress disorder symptoms and psychosocial functioning in Veterans of Operations Enduring Freedom and Iraqi Freedom. Psychiatry Res 2010; 178(2): Kroenke K, Spitzer R, & Williams J: The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001; 16: Bradley K, Bush K, McDonell M, Malone T, Fihn S: Screening for problem drinking: comparison of CAGE and AUDIT. J. Gen Intern Med 1998; 13(6): Humes L, Joellenbeck L, Durch J: (2005). Noise and Military Service: Implications for Hearing Loss and Tinnitus. Washington, DC, Institute of Medicine, National Academy of Sciences. 34. Jakupcak M, Conybare D, Phelps L, et al: Anger, hostility, and aggression among Iraq and Afghanistan war veterans reporting PTSD and subthreshold PTSD. J Trauma Stress 2007; 20(6): MILITARY MEDICINE, Vol. 177, April