Screening Soldiers in Outpatient Care for Mental Health Concerns

Transcription

1 MILITARY MEDICINE, 173, 1:17 24, 2007 Screening Soldiers in Outpatient Care for Mental Health Concerns COL Gregory A. Gahm, MS USA*; Barbara A. Lucenko, PhD ABSTRACT Significant recent effort has been directed toward screening and describing military populations in relation to deployment. Missing from these recent efforts is information describing screened mental health symptoms for the population of active duty military that are seen for mental health services. This article presents mental health screening data for 2,882 soldiers seeking services at a military facility outpatient behavioral health clinic. Screening positive for multiple symptom domains was common, with 60% of the sample screening positive for more than one clinical symptom domain. Post-traumatic stress disorder and depression were among the most commonly identified disorders, followed by alcohol abuse. This screening data, gathered using measures similar to those used in published deployment-related screening efforts, suggest differences that exist between the clinical population and the overall military population, providing some insights into the rates of clinical symptomatology within the military health system and providing a point of comparison for population- screening efforts. Clinical implications include the importance of provider awareness to the high rates of comorbidity across symptom domains. INTRODUCTION Behavioral health (BH) problems are a major component of the overall health and utilization of health care services for military and Veteran s Administration (VA) beneficiaries. A recent analysis of VA patient records indicated that possible mental disorders were reported for 26% of veterans who experienced combat in Iraq and Afghanistan. 1 Active duty soldiers, particularly those who experienced combat, are at increased risk of mental health difficulties, particularly posttraumatic stress disorder (PTSD), depression, and anxiety. 2 4 Mental disorders have been shown to be a primary source of disability and separation from the military, with a reported 6-month attrition rate of 45% for those hospitalized for mental health diagnoses. 5 The accurate detection of mental disorders is also critical for their impact on health and well-being. Untreated depression, PTSD, and other anxiety disorders have a documented association with reduced well-being and functioning comparable to physical problems such as heart disease. 6 The accurate identification of mental disorders across domains is thus critical within the military. How to best accomplish this is open to debate. Health screening has been used as a preventive mechanism in various areas of medicine. Such screening allows for assessment of risk, early intervention, and, in some cases, Department of Psychology, Madigan Army Medical Center, MCHJ-CP ( ), 9040 Fitzsimmons Drive, Tacoma, WA A version of this study was presented at the International Society for Traumatic Stress Studies Annual Meeting, November 2 5, 2005, Toronto, Ontario, Canada. The opinions or assertions contained herein are the views of the authors and are not to be construed as the official views of the Department of Defense. The investigators have adhered to the policies for protection of human subjects as prescribed in 45 CFR 46. This manuscript was received for review in June The revised manuscript was accepted for publication in March prevention of symptoms. Some areas in which screening has made a major impact are early cancer detection (e.g., mammography for women over 40), prenatal and infant health screening (e.g., Down s syndrome, phenylketonuria), 7 alcoholism and depression, 8,9 and mental health concerns in the wake of mass disaster. 10,11 Similar screening approaches have been attempted for screening soldiers for mental health difficulties following combat. 12 Standardized screening, at varying levels of detail, has been implemented across the Department of Defense (DoD) both immediately pre- and postdeployment and 3 to 6 months postdeployment. It has not, as of yet, been implemented in BH clinics. The information gathered in BH clinics tends to cover similar domains to that gathered on deployment-related screenings. It is often gathered in response to requirements of regulatory and credentialing agencies and rules. However, the actual BH assessment screening conducted in military clinical settings is generally not standardized and does not cover a broad array of screening domains. Screening tools, when used, may not be drawn from the most current, empirically based tools available. There is no standardized mental health screening process that is applied across either primary care or mental health clinics in the DoD. The use of standardized measures for mental health screening in a consistent manner can impact the accuracy of identification of behavioral health problems. In one study that assessed the ability of primary care providers to identify service members with potential depressive symptoms, 13 Air Force physicians were able to correctly identify only 21% of those flagged for major depression by the Patient Health Questionnaire (PHQ) depression module. Additionally, when patients are seen in multiple clinics or programs for BH difficulties (e.g., substance abuse clinic, family violence program, outpatient BH clinic), there is frequently a lack of MILITARY MEDICINE, Vol. 173, January

2 FIGURE 1. BHSI outpatient clinic administration. communication between providers and no efficient process for sharing data, reports, or basic assessment information. With data not typically shared across specialties, this often results in duplicated efforts between providers and services, and ultimately is an added response burden for patients. Consistent and standardized screening procedures are therefore warranted. Ideally, efforts to identify soldiers with mental health difficulties would also be tailored in scope and design to support specific settings and populations while maintaining a consistent core of standardized information. It is assumed that soldiers who present, or are referred, to mental health represent the more severe end of the psychopathology continuum. However, little is known about how soldiers who present to mental health clinics score on screening measures. What is known about these populations tends to be based on trainee or Vietnam veterans. Additionally, while PTSD may be a widely accepted diagnosis for soldiers who have experienced combat trauma, the rates among those presenting to mental health are not known. Furthermore, PTSD has been shown to be commonly comorbid with other diagnoses such as depression and alcohol abuse. 3 In a study of 200 Vietnam veterans, it was noted that PTSD was the most common diagnosis for both clinical (53%) and nonclinical (17%) groups, that comorbid diagnoses were common, and that PTSD was rarely diagnosed without an additional diagnosis of major depression, panic or phobic disorder, and/or alcohol abuse. 14 Soldiers seeking care may not accurately convey the variety or severity of their symptoms without the use of standardized screening tools. A published study on screening of soldiers for mental health problems showed that less than half of those soldiers meeting scale criteria for a mental disorder expressed interest in receiving help. Findings and recommendations from general population or postdeployment screening efforts may therefore not apply in a clinical setting, in which the need for assessment or care has already been determined. One research effort that has used screening to identify soldiers needing care is that of the U.S. Army Medical Research Unit Europe (USAMRU-E), which developed a screening program for the purpose of identifying deployed soldiers in need of services specifically for PTSD, depression, and alcohol use. In one published study of this effort, soldiers were screened in garrison (n 790) and in two deployment locations (Albania, n 1,000; Kosovo, n 3,520) during the time frame 1996 to Findings from their work included that junior-enlisted personnel were more likely to screen positive for mental health problems, their Active Army samples screened positive at higher rates than their National Guard or Reserve samples, women were more likely to exceed PTSD and depression criteria in garrison but not in deployed environments, and males in garrison were more likely than women to exceed alcohol criteria. Longer time deployed was associated with positive screenings for males but not females. 12 USAMRU-E has more recently conducted in depth screening and laid the groundwork to validate a mental health screening process. 15,16 The USAMRU-E efforts screen for depression, traumatic stress, relationship problems, anger, and alcohol misuse, and also cover postcombat sleep disturbances. USAMRU-E is conducting research to determine the necessity for assessing all of these domains on a primary screen, with the three possible options listed as: (1) asking soldiers if they want to see a counselor; (2) using one brief measure of overall distress; and (3) screening the domains listed above. Sensitivity or percentage of true positives identified was reported as low for the single item (27%) and the overall distress scale (21%), and was much higher (80%) when specific domain scales for depression, PTSD, alcohol, and desire to speak to counselor item were included as a single screen (87% specificity or true false negatives with this process). Although these findings provide suggestions as to measures that could be used in a mental health clinic and rates that would likely be seen in nonclinical active duty populations, they do not specifically address the methodology or results of screening in mental health settings. To address the current needs of soldiers and behavioral health providers, we developed a screening instrument from multiple validated preexisting standardized scales to use as a standard screening in a military hospital outpatient BH clinic. This study reviews the measure and existing data from this effort to describe the mental health characteristics identified through screening of soldiers seeking care during time of war in a military outpatient mental health setting. METHODS Participants Data for this study were gathered from clinical records of active duty Army soldiers who entered an outpatient BH 18 MILITARY MEDICINE, Vol. 173, January 2008

3 TABLE I. Screening Demographics for BH Clinic and Overall Army 40 BH Clinic (N 2,882) n (%) Army G-1 Fiscal Year 2004 n (%) Sex Male 2,313 (80.6) 421,608 (85.0) Female 555 (19.4) 72,683 (15.0) Age (years) 25 1,176 (42.3) 202,266 (41.9) (21.7) 104,081 (21.1) (24.1) 136,703 (27.6) (12.0) 51,241 (10.4) Rank E1 E4 1,723 (60.5) 221,758 (44.8) E5 E9 1,016 (35.7) 188,388 (38.1) Officer 109 (3.8) 78,773 (15.9) Warrant officer 0 (0.0) 2,021 (0.4) Race/ethnicity Caucasian 1,780 (62.5) 197,069 (60.0) African American 364 (12.8) 112,204 (23.0) Hispanic 346 (12.1) 50,912 (10.0) Other 358 (12.6) 34,106 (7.0) Education Some HS 159 (5.5) 4,147 (0.8) HS graduate/some college 2,430 (84.4) 379,763 (76.4) BA/BS 271 (9.4) 92,587 (18.7) Other/unknown 71 (0.7) 17,861 (3.6) HS, high school. clinic located in an Army medical treatment facility on a large military base. The research protocol for this project was approved by the medical treatment facility Department of Clinical Investigation. The Behavioral Health Screening Instrument (BHSI) was administered as part of standard clinic business practice. This screen consists of standardized measures covering multiple domains, including demographics and military information, psychosocial history, current symptoms and behavior, and previous stressors (e.g., child abuse). All military personnel presenting for intakes are asked to complete the BHSI. During the time this study data were collected, this was completed via paper form in the waiting room with clinic staff TABLE II. members collecting completed forms and scanning them via bubble-sheet scanning software, checking for accuracy, and correcting duplicate responses when prompted. This process served to populate a database, which is programmed to generate reports based on scale scores and critical item responses. Reports are then included in patient files for clinician review. (Fig. 1). All new cases seen in a military outpatient mental health clinic between June 2003 and July 2005 were considered for analysis. Data were available for 2,882 patients. All participants were patients who presented to the clinic via a number of referral sources. Clinic management statistics suggested the following breakdown of referral sources: selfreferral (64%); pre- or postdeployment referral (22%); discharge from inpatient psychiatry (3%); command-directed evaluations (e.g., fitness for duty, 5%); emergency room (3%); Reserve Officer Training Course (2%); and medical evacuation (1%). Completion of the instrument is routinely followed by a thorough assessment by a behavioral health provider. Of the sample, 62.4% reported having been deployed to a war or combat zone. Branch of service for these soldiers was primarily Army (97.8%). Military status was categorized as Active (67.9%), Active Guard (11.6%), Guard (8.0%), Active Reserve (6.8%), and Reserve (5.7%). Average years in the military was 7.12 (SD 6.59; median 5.00), average total months deployed was (SD 12.27), and average months in current unit reported was (SD 19.67). Other self-reported demographics for the sample, and for the overall Army, are presented in Table I. Measures The BHSI covers similar domains to that of the USAMRU-E screening tool, including depression, PTSD, relationship problems, anger, and alcohol misuse. Additionally, several scales of the BHSI, such as the Primary Care PTSD (PC- PTSD) Screen, the Brief Symptom Inventory (BSI) Hostility Scale, and the Quality of Marriage Index, directly map to those of the USAMRU-E tool. 12 In addition to the clinical scales described below, the Clinical Scales Intercorrelations and Internal Consistency Coefficients Mean SD Depression Anxiety PTSD Panic Alcohol QMI Hostility Support Depression Anxiety a 0.80 PTSD a a 0.82 Panic a a a 0.92 Alcohol a a a a 0.87 QMI a a a 0.97 Hostility a a a a a a 0.87 Support a a a a a b a 0.93 Coefficient reliability estimates for each scale are the last (rightmost) values in each row. a Correlation is significant at the 0.01 level (two tailed). b Correlation is significant at the 0.05 level (two tailed). MILITARY MEDICINE, Vol. 173, January

4 BHSI includes items to address basic demographics, military information, (e.g., current rank, years in military, months deployed), and psychosocial history (e.g., mental health treatment, physical health). Coefficient and Kuder Richardson 20 coefficients were calculated for each clinical scale and are presented in Table II. Internal consistency coefficients for all clinical scales were above Alcohol Use Disorders Identification Test (AUDIT) The AUDIT is a self-report measure of alcohol use intended for screening of alcohol use disorders. The measure was empirically derived as part of a World Health Organization six-country effort to develop a test to identify a broad spectrum of problem drinkers and was introduced in a primary care setting. 9 Scores are generated by summing the 10 four-point item values for a possible total of 40. A score of 8 or higher indicates hazardous or harmful consumption. Scores of 13 for women and 15 for men indicate possible alcohol dependence and 20 indicates probable alcohol dependence. 9 This measure was modified to include a question does not apply due to deployment response option for use when respondent alcohol consumption behavior was changed due to the unavailability or an order not to consume alcohol. Soldiers reporting that they do not drink were instructed to skip the remaining AUDIT items. Including the AUDIT as part of an overall health screen and slight word changes have not been found to impact the accuracy of the instrument. 17 PC-PTSD The PC-PTSD 18 is a four-item self-report screening instrument for PTSD. The PC-PTSD was validated with a VA patient sample. Using a PC-PTSD cutoff score of 3 and the Clinician Administered PTSD Scale 19 as the gold standard for PTSD diagnosis, the PC-PTSD has demonstrated acceptable sensitivity and specificity, with a correlation of 0.83 with Clinician Administered PTSD Scale diagnoses. The four items are the same PTSD items found on DoD Postdeployment Health Screening Form (DD 2796), routinely administered to soldiers following deployment. The scale has been shown to be comparable in sensitivity and specificity to the well-known PTSD checklist 20,21 in assessing soldiers for PTSD. 22 The USAMRU-E group found that sensitivity for the PTSD scale was 73% when using a two-item cutoff and 46% when using the three-item cutoff. 23 Patient Health Questionnaire Several scales of the BHSI are based upon the Primary Care Evaluation of Mental Disorders (PRIME-MD) PHQ. 24 The PRIME-MD was developed by authors of the Structured Interview for DSM-IV, 25 along with primary care physicians, and was based upon existing epidemiological research and psychiatric nosology. 26 In a validation study, the PRIME-MD was administered to 1,000 patients from four primary care locations, including 303 from Walter Reed Army Medical TABLE III. Gender Differences in BHSI Clinical Scale Scores BHSI Scale Gender n Mean SD t df Panic Female a 2,763 Male 2, PHQ-9 Female b 2,734 Male 2, AUDIT Female b 1,953 Male 1, Anxiety Female b 2,727 Male 2, Unit support Female b 2,626 Male 2, QMI c Female a 1,881 Male 1, PTSD Female ,788 Male 2, Hostility Female a 2,815 Male 2, a Value of p b Value of p c Includes only those indicating spouse or significant other. Center. A primary care study of 3,000 patients, including 431 who had mental health diagnostic interviews, revealed sensitivity and specificity similar to that of the original, physicianadministered instrument. When assessing its perceived value to physicians, 87% found the diagnostic information somewhat or very useful and 88% found it somewhat or very comfortable for patients. 24 Agreement between PRIME-MD-based diagnoses and mental health practitioners for modules used in the BHSI was generally good. 24 Patient Health Questionnaire 9 The PHQ-9, a version of the PRIME-MD depression module, is a self-administered scale used to assess depression. 24 There are nine items based on the Diagnostic and Statistical Manual of Mental Disorders (Fourth Edition; DSM-IV; 1994) diagnostic criteria. The PHQ-9 has two scoring methods, one for diagnostic purposes and one for measuring severity. Items were scored as follows for the past 2-week time period: not at all 0; several days 1; more than half the days 2; and nearly every day 3. For diagnostic scoring, if five of the nine symptom items were endorsed for at least more than half the days, including symptoms of depressed mood and/or anhedonia, major depressive disorder was suggested. For major depressive disorder, the PHQ-9 has 73% sensitivity, 98% specificity, and 93% overall accuracy. 24 PHQ Anxiety and Panic The PHQ anxiety module is similar to the depression module with responses ranging from 0 for not at all to 2 for more than half the days. To suggest the presence of panic disorder, all five panic symptom questions must be endorsed. One study of 499 outpatients found the PHQ sufficiently sensitive (75%) in detecting the presence of panic disorder, with increased sensitivity when the algorithm was modified to require endorsement of only three (86%) or two (91%) panic items MILITARY MEDICINE, Vol. 173, January 2008

5 70% 60% 50% 40% 30% 20% 10% 0% 61% 38% 44% 15% 23% Depression Anxiety PTSD Panic Alcohol Use 33% Hostility FIGURE 2. Soldiers screening positive for mental health difficulties (N 2,882). Depression: PHQ-9 score of 5 (mild through severe); Applying the diagnostic scoring algorithm, 42.6% screened positive for major depressive disorder and 18.7% for other depression (total n 1769, 61.3% positive depression screens). Applying the symptom severity algorithm, 82.5% screened positive for depression (21.9% mild, 22.7% moderate, 21.6% moderately severe, and 16.3% severe). Anxiety: 3PHQ anxiety items endorsed for more than half the days; PTSD: PC-PTSD score of 2 3; panic: all PHQ panic scale items endorsed; alcohol: AUDIT score of 8 or more; hostility: BSI HOS Scale, 2 items or if the single item, having urges to beat, injure, or harm someone was endorsed for often or very often in the past week. Brief Symptom Inventory The BSI 28,29 is a shorter version of the widely accepted Symptom Checklist-90 Revised 30 from which the Hostility (HOS) scale is comprised of five items. Test-retest reliability is reported as 0.81, based on 2-week interval retesting of 60 nonpatients. Correlation between HOS scales on the BSI and the SCL-90-R (i.e., multiple forms reliability) was 0.99 for 565 psychiatric outpatients. Quality of Marriage Index (QMI) The QMI 31 is an empirically derived, six-item scale designed to be brief and focused on a single construct. The QMI has good established concurrent validity with the Kansas Marital Satisfaction Scale. 32 The six suggested QMI items with the highest factor loadings and correlations that met semantic criteria (as per Ref. 31) are included in the BHSI. Possible scores range from 1 to 36, with higher scores reflecting more relationship satisfaction. Deployment Social Support Information about perception of support from unit, leaders, and the military during deployment was drawn from the Deployment Risk and Resilience Inventory, 33 which was developed as a research measure to evaluate factors related to long-term health for veterans. The entire Deployment Risk and Resilience Inventory includes 14 scales that may be used together or individually. Internal consistency reported by the scale authors in a study of veterans was Additionally, deployment social support scores were negatively correlated with measures of mental health such as PTSD, depression, and anxiety. RESULTS All statistics were generated using SPSS version 13 for Windows (Chicago, Illinois). Descriptive statistics for group demographics can be viewed in Table I. Mean symptom scale scores can be viewed in Table II. Gender differences were assessed and are presented in Table III. Average mean scores were significantly higher for women on the following clinical scales: panic, depression, anxiety, and hostility. Men had significantly higher scores of alcohol misuse and reported higher levels of both unit support and relationship satisfaction than women (all p 0.01). There were no gender differences detected on the mean PTSD score. To further describe the sample, percentages of individuals meeting screening cutoff criteria for particular domains are presented in Figure 2. For the total sample, 89.4% screened positive for at least one domain of behavioral health difficulty. Limiting this list to diagnostic clinical scales (depression, anxiety, panic, PTSD, and alcohol use), the number of soldiers screening positive for one (23.2%), two (27.3%), three (23.2%), four (10.4%), and all five domains (1.7%) is displayed in Figure 3. DISCUSSION AND CONCLUSIONS In this exploratory investigation of existing screening data for soldiers seeking mental health services, we described findings on clinical scales and for demographic and psychosocial domains. Given that this was a clinical sample, it was expected that most of the sample would screen positive for some mental health symptoms. We found that 89.4% screened positive for at least one symptom domain, with most (62.6%) screening positive for multiple domains. Thus, soldiers entering treatment for BH difficulties endorsed symptoms for multiple domains as opposed to simply depression or PTSD which might have been their identified presenting problem. This finding is consistent with that for Vietnam veterans, 14 suggesting that active duty soldiers clinical needs are also multifaceted and complex. The presence of comorbid symptoms among this population has implications for future clinic developments and additional research. Much of the existing PTSD treatment outcomes research does not sufficiently take into account the common co-occurrence of additional symptoms and disorders. 34 This suggests a need to further explore multiple domains to develop treatment approaches that address overlapping symptoms (e.g., anger, PTSD, and depression), as opposed to the traditional model of single diagnosis-focused 70% 60% 50% 40% 30% 20% 10% 0% 14% 23% 27% 23% 10% FIGURE 3. Percentage of soldiers screening positive for 0, 1, and multiple clinical symptom domains (N 2,882). 2% MILITARY MEDICINE, Vol. 173, January

6 specialty groups. This would also argue against the model of separating out clinics by specialty and instead supports an integrated clinic concept. What this does not address, however, are the differences in treatment efficacy that may be found through the single specialty area model. Additional or co-occurring behavioral health difficulties have been linked to increased utilization of health care, 24 an issue that could have further implications for soldier readiness and military primary care utilization. Gender differences were in the direction expected for clinical scales, with women higher on scales of mood and anxiety. They were also higher on the measure of hostility, a difference that may reflect item overlap of this measure with general distress. There were no differences between men and women on the measure of PTSD which could reflect a higher rate of exposure to traumatic events by the men in this sample or a lack of sensitivity of this measure. Neither explanation can be evaluated with these data. Both men and women reported low levels of marital satisfaction, although the mean QMI score was significantly higher for men. The overall QMI scores were significantly low compared to those reported in a study of civilian depressed patients, 35 suggesting that soldiers seeking care for mental health difficulties are less satisfied with current relationships than the general civilian clinical population. The stresses of military service to include deployment have been suggested elsewhere to impact on marriage and divorce rates which may be reflected in these data. It should be noted that the clinic in which these data were collected does not explicitly offer marital therapy. The clinical sample was generally representative of the overall Army in terms of demographics, with some overrepresentation of women, younger, and lower-ranking service members. Most soldiers seeking services in the outpatient clinic (62.4%) had also been deployed. It is possible that deployment experiences may have exacerbated preexisting conditions such as substance abuse and personality disorders, as well as potentially impacting on their marital relationship. Additional exploration of such associations are areas for future research. Conditions and experiences during deployment may be also associated with varying levels of discomfort, random violence, and/or chronic fear and strain. 36 A recent study of VA administrative data indicated higher rates of PTSD for veterans serving in Army and Marine ground units than for Navy and Air Force members, suggesting the possibility that more severe combat experiences are a factor in the development of mental health difficulties for this population. 1 The Army and Marines also, however, represent a different population demographic distribution than the other services (i.e., Air Force and Navy) suggesting that these factors may also account for the differences. It would therefore be important to gather detailed information pertaining to deployment experiences in future studies of psychological symptoms among active duty military. The existing literature presents a clear link between military experiences and PTSD symptoms, as well as barriers to care for service members. 2 A recent study also demonstrated that 35% of a large sample of soldiers accessed mental health services within 12 months of returning from Iraq. 37 What has not been addressed empirically to date is which factors predict help-seeking and access to care among soldiers with mental health difficulties who have experienced combat. Referrals to a variety of services along a continuum of care may allow soldiers to seek services within their individual comfort zones. For example, the availability of a range of outreach psychoeducational, voluntary group treatment, individual psychotherapy, and internet-based interactive PTSD services, with multiple points of contact, may increase the probability that soldiers will initiate and continue needed services. When examining the value and appropriateness of this screening process, we looked to recent suggestions 38 for population level psychological screening in the military, some of which apply to screening in a military behavioral health clinic. Specifically, their recommendations include that the screening tools must be clinically, socially, and ethically acceptable, simple, precise and validated, and benefits from the screening program should outweigh potential harms. The screening tools used in the BHSI are wellestablished mental health screening measures and certainly meet the first two criteria. The assumed benefit of screening is that it assists clinicians in providing the highest quality and most appropriate level of care possible. This area warrants further study in the form of a formal program evaluation with outcome measures such as long-term service utilization. One limitation to the current study is the lack of baseline, predeployment levels of functioning for soldiers who have been deployed. A longitudinal investigation would allow for more definitive conclusions to be drawn regarding the sources of symptomatology for clinical groups with and without particular combat experiences. A prospective approach taken with soldiers would allow for the impact of specific experiences, particularly those related to combat, to be partialed out for this population. Additional detailed information on deployment experiences would have been helpful in these analyses, however, they were not part of the screening tool at the time of this data gathering but have been added to render more complete information for providers. The DoD-wide screening at predeployment and 90 to 180 days following return from deployment the Post Deployment Health Risk Assessment (PDHRA) process represents one effort that will allow for longitudinal data comparisons. The screening that is described in this article has undergone modifications to allow it to map directly onto that screening in addition to expanding the scope of the screening that is conducted at the PDHRA for the soldiers at this military base to allow us to answer some of these questions in the future. Lack of formal diagnostic information has been listed as a limitation for several studies using questionnaires. 39 Documentation of patient diagnosis and diagnostic procedures would allow further interpretation of the data, as well as analyses of risk variables as they pertain to comorbidity. A 22 MILITARY MEDICINE, Vol. 173, January 2008

7 future effort will address this validation requirement of the screening process both for the PDHRA and for the clinic screener. In other settings, such as postdeployment and in theater screening programs, screening ensures that soldiers at risk for severe difficulties such as suicidal behavior and severe pathology are identified and treated before crisis situations. In a clinical setting, such tools allow for timely and more comprehensive information that can be quickly communicated to providers, suggesting more detailed assessment of particular symptom domains or endorsed critical items. Given the high correlation between mental health disorders and separation from military service (45%), 5 this could become a crucial concern for the military to address in the near future. Standardized assessments may or may not play a role in this process. It is unclear how many soldiers in distress do not present for treatment and how different screening and referral approaches could impact this process. The implementation of a valid and reliable mental health screening process will facilitate identification of soldiers whose responses warrant more intensive exploration in multiple realms. In a military setting, such a tool allows for timely and more comprehensive information that can be quickly communicated to providers in the context of referrals. BH difficulties cover a wide range of symptoms and domains that may be appropriate for treatment by a variety of services and approaches such as individual or group psychotherapy, case management, and drug and alcohol treatment. Comprehensive screening findings in a military outpatient mental health setting are therefore essential in guiding clinician assessment and treatment efforts. It is assumed that such an approach enhances efficiency, reduces redundancy, and ensures that all critical domains are addressed. Further research is warranted to assess the efficacy of deploymentrelated screening programs, as well as those implemented in clinical settings. Validation studies will be a first step and necessary prerequisite for this work. In addition to findings pertaining to sensitivity and specificity of the scales for this particular population, it is also important to compare effectiveness of screening to other approaches to determine a best-practices model for the DoD. ACKNOWLEDGMENTS Research was funded in part by the U.S. Army Medical Research and Material Command, Military Operational Medicine Program. We are grateful to Army Behavioral Health Technology Office Staff Matt Rein and Chris Allen for their patience and technical expertise. Special thanks go to the population of soldiers who are the topic of this study for their sacrifices and contributions. REFERENCES 1. Kang HK, Hyams KC: Mental health care needs among recent war veterans. N Engl J Med 2005; 352: Hoge CW, Castro CA, Messer SC, McGurk D, Cotting DI, Koffman RL: Combat duty in Iraq and Afghanistan, mental health problems, and barriers to care. N Engl J Med 2004; 351: Kessler RC, Sonnega A, Bromet E, Hughes M, Nelson CB: Posttraumatic stress disorder in the National Comorbidity Survey. Arch Gen Psychiatry 1995; 52: Stimpson NJ, Thomas HV, Weightman AL, Dunstan F, Lewis G: Psychiatric disorder in veterans of the Persian Gulf War of 1991: systematic review. Br J Psychiatry 2003; 182: Hoge CW, Toboni HE, Messer SC, Bell N, Amoroso P, Orman DT: The occupational burden of mental disorders in the U.S. military: psychiatric hospitalizations, involuntary separations, and disability. Am J Psychiatry 2005; 162: Schonfeld WH, Verboncoeur CJ, Fifer SK, Lipschutz RC, Lubeck DP, Buesching DP: The functioning and well-being of patients with unrecognized anxiety disorders and major depressive disorder. J Affect Disord 1997; 43: United States Preventive Services Task Force: Screening for suicide. In: Guide to Clinical Preventive Services, Chap 50, Ed 2, pp Baltimore MD, Williams & Wilkins, Greenfield SF, Reizes JM, Muenz LR, Kopans B, Kozloff RC, Jacobs DG: Treatment for depression following the 1996 National Depression Screening Day. Am J Psychiatry 2000; 157: Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M: Development of the Alcohol Use Disorders Identification Test (AUDIT): WHO Collaborative Project on Early Detection of Persons with Harmful Alcohol Consumption-II. Addiction 1993; 88: Jordan NN, Hoge CW, Tobler SK, Wells J, Dydek GJ, Egerton WE: Mental health impact of 9/11 Pentagon attack: validation of a rapid assessment tool. Am J Prev Med 2004; 26: Smith R, Katz C, Holmes A, et al: Mental health status of World Trade Center rescue and recovery workers and volunteers New York City, July 2002 August MMWR Morb Mortal Wkly Rep 2004; 53: Wright KM, Huffman AH, Adler AB, Castro CA: Psychological screening program overview. Milit Med Oct 2002; 167: Hunter CL, Hunter CM, West ET, Kinder MH, Carroll DW: Recognition of depressive disorders by primary care providers in a military medical setting. Milit Med 2002; 167: Green BL, Lindy JD, Grace MC, Gleser GC: Multiple diagnosis in posttraumatic stress disorder: the role of war stressors. J Nerv Ment Dis 1989; 177: Wright KM, Bliese PD, Adler AB, Hoge CW, Castro CA, Thomas JL: Screening for psychological illness in the military. JAMA 2005; 294: Wright KM, Thomas JL, Adler AB, Ness JW, Hoge CW, Castro CA: Psychological screening procedures for deploying U.S. Forces. Milit Med 2005; 170: Reinert DF, Allen JP: The Alcohol Use Disorders Identification Test (AUDIT): a review of recent research. Alcohol Clin Exp Res 2002; 26: Prins A, Ouimette P, Kimerling R, et al: The primary care PTSD screen (PC-PTSD): development and operating characteristics. Prim Care Psychiatry 2003; 9: Blake DD, Weathers FW, Nagy LM, et al: The development of a Clinician-Administered PTSD Scale. J Trauma Stress 1995; 8: Weathers FW, Keane TM, Davidson JR: Clinician-administered PTSD scale: a review of the first ten years of research. Depress Anxiety 2001; 13: Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM: The PTSD Checklist (PCL): Reliability, Validity, and Diagnostic Utility. International Society for Traumatic Stress Studies (ISTSS), October 24 27, San Antonio, TX. 22. Bliese P, Wright K, Adler A, Thomas J, Hoge C: Screening for traumatic stress among re-deploying soldiers. USAMRUE Heidelberg, Germany, U.S. Army Research Unit Europe, Bliese PD, Wright KM, Adler AB, Hoge CW, Prayner R: Post-Deployment Psychological Screening: Interpreting and Scoring the DD Form MILITARY MEDICINE, Vol. 173, January

8 2900. Research Report Heidelberg, Germany, USAMRU-E, Spitzer RL, Kroenke K, Williams JB: Validation and utility of a selfreport version of PRIME-MD: the PHQ Primary Care Study: Primary Care Evaluation of Mental Disorders: Patient Health Questionnaire. JAMA 1999; 282: First MB, Spitzer RL, Gibbon M, Williams JBW: Structured Clinic Interview for DSM-IV Personality Disorders, (SCID-II). Washington, DC, American Psychiatric Press, Inc., Hahn SR, Kroenke K, Williams JBW, Spitzer RL: Evaluation of mental disorders with the PRIME-MD. In: Handbook of Psychological Assessment in Primary Care Settings. Edited by Maruish ME. Mahwah, NJ, Lawrence Elbaum Associations, Lowe B, Grafe K, Zipfel S, et al: Detecting panic disorder in medical and psychosomatic outpatients: comparative validation of the Hospital Anxiety and Depression Scale, the Patient Health Questionnaire, a screening question, and physicians diagnosis. J Psychosom Res 2003; 55: Derogatis LR, Cleary PA: Confirmation of the dimensional structure of the SCL-90-R: a study in construct validation. J Clin Psychol 1977; 33: Derogatis LR: BSI Brief Symptom Inventory: Administration, Scoring, and Procedures Manual, Ed 4. Minneapolis, MN, National Computer Systems, Derogatis LR, Savitz KL: The SCL-90-R and Brief Symptom Inventory (BSI) in primary care. In: Handbook of Psychological Assessment in Primary Care Settings, pp Edited by Maruish ME. Mahwah, NJ, Lawrence Elbaum Associations, Norton R: Measuring marital quality: a critical look at the dependent variable. J Marriage Fam 1983; 45: Calahan CA: Correlations of scores on the Kansas Marital Satisfaction Scales and the Quality Marriage Index. Psychol Rep 1996; 78: King DW, King LA, Vogt DS: Manual for the Deployment Risk and Resilience Inventory (DRRI): A Collection of Measures for Studying Deployment-Related Experiences of Military Veterans. Boston, MA, National Center for PTSD, Spinazzola J, Blaustein M, van der Kolk BA: Posttraumatic stress disorder treatment outcome research: the study of unrepresentative samples? J Trauma Stress 2005; 18: Heyman RE, Sayers SL, Bellack AS: Global marital satisfaction versus marital adjustment: an empirical comparison of three measures. J Fam Psychol 1994; 8: National Center for PTSD: Iraq War Clinician Guide, Vol 2004, Ed 2. Boston, MA, National Center for PTSD, Hoge CW, Auchterlonie JL, Milliken CS: Mental health problems, use of mental health services, and attrition from military service after returning from deployment to Iraq or Afghanistan. JAMA 2006; 295: Rona RJ, Hyams KC, Wessely S: Screening for psychological illness in military personnel. JAMA 2005; 293: Orsillo SM, Weathers FW, Litz BT, Steinberg HR, Huska JA, Keane TM: Current and lifetime psychiatric disorders among veterans with war zone-related posttraumatic stress disorder. J Nerv Ment Dis 1996; 184: Office of Army Demographics: Army Profile FY-04. Washington, DC, U.S. Army, MILITARY MEDICINE, Vol. 173, January 2008