The Association Between U.S. Army Enlistment Waivers and Subsequent Behavioral and Social Health Outcomes and Attrition From Service

Transcription

1 MILITARY MEDICINE, 178, 3:261, 2013 The Association Between U.S. Army Enlistment Waivers and Subsequent Behavioral and Social Health Outcomes and Attrition From Service M. Shayne Gallaway, PhD*; COL Michael R. Bell, MC USA ; Christine Lagana-Riordan, PhD*; David S. Fink, MPH*; Charles E. Meyer, ME*; Amy M. Millikan, MD* ABSTRACT Soldiers granted enlistment waivers for medical concerns, misconduct, or positive alcohol/drug tests may or may not be associated with an increased likelihood of negative behavioral outcomes. Soldiers in the population examined (n = 8,943) who were granted enlistment waivers from 2003 to 2008 were significantly more likely to subsequently be screened for alcohol/substance abuse, test positive for illicit substances, or receive an Army separation for behavioral misconduct. These associations were highest among Soldiers granted waivers for nonlawful alcohol/drug violations. Soldiers granted waivers for felony offenses and serious nontraffic violations were significantly less likely to separate from the Army compared with Soldiers not granted enlistment waivers. *U.S. Army Institute of Public Health, Behavioral and Social Health Outcomes Program, 5158 Blackhawk Road, Building E-1570, Gunpowder, MD Uniformed Services University of the Health Sciences, 4301 Jones Bridge Road, Bethesda, MD The work submitted conforms to all governmental regulations and discipline appropriate professional ethical standards. This manuscript has not been published elsewhere and has not been submitted simultaneously for publication in another outlet. Some of these findings were briefly summarized in a Department of the Army Medical Command Report. A summary of this work was presented at the Armed Forces Health Protection Annual Conference in Albuquerque, NM, August, The views expressed in this article are those of the authors and do not reflect official policy or position of the Department of the Army, the Department of Defense, the U.S. Government, or any of the institutional affiliations listed. doi: /MILMED-D INTRODUCTION When considering an applicant for service in the U.S. Army, recruiters examine several indicators, including Armed Forces Qualification Test scores, demographics, and criminal history. Enlistment waivers may be granted for applicants who fail to meet one or more criteria. Enlistment waivers are granted for a number of different reasons, including age at enlistment, married with two or more dependents, unmarried with dependents, medical concerns, misconduct (a single incident or pattern of behavior), or positive alcohol/drug tests at a Military Entrance Processing Station. Enlistment waivers for misconduct and/or positive alcohol/drug tests are referred to as moral conduct waivers. Because of the difficulties associated with keeping the force at strength during a time of war (i.e., reduced accessions and increased attrition), there is a perception the Army relaxed standards for enlistment and increased the number of waivers granted for medical and conduct/drug use problems. 1 Changes in accession standards have been anecdotally implicated with the observed increased in previous negative behavioral health outcomes occurring among Soldiers, 2 such as alcohol/drug use, domestic violence, and attrition related to misconduct, and most likely varies based on the types of waivers granted. Although only one peer-reviewed study 3 could be identified that has assessed the association between specific waivers and subsequent negative behavioral and/or social health outcomes, there are a number of nonpeer-reviewed studies (e.g., technical reports, postgraduate theses) that have examined this association. 4 9 Service members granted specific enlistment waivers have been shown to be more likely to have negative outcomes, such as a medical evaluation board (MEB) 3 or attrition for reasons related to poor moral character, 4 10 but they also have been shown to be more likely to complete basic training, have lower rates of personality disorders, less likely to attrite from service early, and be more likely to reenlist. 4,8 Previous population-wide enlistment waiver studies that examined recruits enlisting before 2001 were reviewed, but considered not comparable to the current study population. No previous studies were identified that examined the likelihood of negative behavioral and social health outcomes during military service, such as positive drug tests, interaction with substance abuse programs, or domestic violence. The objective of this study was to evaluate whether Soldiers granted specific types of moral and medical enlistment waivers were more likely than other Soldiers to experience negative behavioral and social health outcomes during their military service and/or attrition from the Army. We hypothesized that Soldiers granted moral waivers would be associated with an increased likelihood of subsequent behavioral and social health outcomes during their military career and would be more likely to separate (attrite) from the Army for negative reasons. METHODS Data and Procedures All Soldiers who enlisted in the Army between 2003 and 2008 and served in either of two brigade combat teams (BCTs) of MILITARY MEDICINE, Vol. 178, March

2 interest (n = 8,843) were identified for this study by the Armed Forces Health Surveillance Center (AFHSC). Data were obtained from the AFHSC (demographics and military characteristics), U.S. Army Accessions (description of enlistment waivers and any subsequent attrition), Army Center for Substance Abuse Programs (ACSAP) (Army Substance Abuse Program (ASAP) screening for alcohol and/or drugs), Drug and Alcohol Management Information System (DAMIS) (positive illicit drug tests), and the Army Central Registry (ACR) (history of perpetrating domestic violence and/or child abuse/neglect). Data were then linked using a standardized nonidentifying study key developed by the AFHSC to ensure the data remained unidentified to the study team. Measures Accession waivers are considered by each service for applicants with a disqualifying medical or moral condition. Moral waivers for enlistment require documentation that a meritorious case exists to warrant an exception to the standards established (Army Regulation ). 15 Many factors are used to determine eligibility for waivers, including education, Armed Forces Services Vocational Aptitude Battery (ASVAB) scores, references, and involvement in the community, among others. The waiver process serves as a filter to screen for potential applicants, who have been rehabilitated; are good risks from a moral standpoint; and possess a documented, meritorious waiver request (Army Regulation ). 15 For this study, in addition to examining Soldiers granted any type of waiver, four specific classes of waivers were examined: serious nontraffic violations, juvenile or adult felony violations, nonlawful violations involving alcohol or drugs (to include positive drug tests), and medical waivers for mental conditions (International Classification of Diseases, 9th Revision, Clinical Modification ). Serious nontraffic violations can include, but are not limited to, conviction or adverse disposition for carrying of weapon on school grounds; an act of violence including threats against any school faculty members; domestic battery/violence not resulting in a qualifying Lautenberg conviction; conviction of other adverse disposition for driving while intoxicated, driving under the influence, or driving while impaired; and possession of marijuana or drug paraphernalia. Subsequent to accession, negative behavioral and social health outcomes of interest were examined, including alcohol and/or drug problems, domestic abuse, and attrition. Soldiers were defined as having a problem with alcohol or drugs during their military career if they tested positive for an illicit substance or if they were screened at the ASAP clinic for the use of alcohol or drugs. Soldiers were defined as perpetrating domestic abuse if there was documented history of substantiated neglect and/or physical, sexual, or emotional abuse of a spouse or child. Attrition from the Army was also examined. Attrition was defined as any Soldier who separated from the Army for any reason (except death) and is categorized as: all causes, behavior, mental, and performance. Attrition for behavioral reasons was further defined for Soldiers who separated from the Army for behavioral misconduct or Uniformed Code of Military Justice (UCMJ) violations. Data Analysis Enlistment waivers granted to Soldiers in this population of two BCTs from 2003 to 2008 were examined to determine whether there was an increasing trend from 2003 to Adjusted odds ratios (AORs) (and 95% confidence intervals [CIs]) were calculated for the association between Soldiers granted waivers and subsequent negative behavioral or social health outcomes and/or attrition from the Army. Odds ratios were adjusted to account for the hierarchical nature of the data (Soldiers are nested within battalions, which are nested within brigades). Nested data are likely to be correlated, thus Soldiers in a specific battalion or brigade are more likely to behave like other Soldiers in the battalion or brigade than they are to act like Soldiers in another battalion or brigade. Thus, generalized linear mixed modeling (GLMM) was utilized to properly account for the intraclass correlation. The random effects in the model were battalions nested within brigades. Initially, the only fixed effect was the exposure variable (i.e., specific waivers), and the resulting odds ratios and CIs were adjusted for the correlated nested nature of the data. Additionally we added to the GLMM, demographic factors (gender, ethnicity [White, African-American, Hispanic, other], marital status while assigned to the BCT [single, married, divorced, other], age at entry into service [<21, 21 25, 26 30, 30+ years], home of record region at enlistment [midwest, northeast, south atlantic, south central, west, other]) and other potential covariates of interest (grade [enlisted, commissioned officer], time in service [<1, , , 5+ years]) to compute the covariate AORs and associated CIs for each specific waiver type. For each of the outcomes assessed, covariate-adjusted models to include all covariates with a plausible bivariate association (p < 0.15) were developed, backward deleting the least significant variable. Model building and variable selection procedure was based on a commonly used approach. 16 If after removing the least significant variable, other covariate regression coefficients changed by more than 15%, the nonsignificant covariates were included in the model as a confounder. When all remaining covariates had a p-value less than 0.05 or had been identified as a confounder, a covariate-adjusted model resulted. All analyses were completed using SAS, v9.1 (Cary, North Carolina), GLMM procedures. RESULTS The current study population consists of 8,943 Soldiers who joined the U.S. Army from 2003 to 2008 and were assigned to the two BCTs of interest. Upon enlistment, this population was primarily male, White, under 21 years of age, and from no particular region on the United States; while in the BCT, 262 MILITARY MEDICINE, Vol. 178, March 2013

3 TABLE I. Demographic and Military Characteristics of Study Population (N = 8,843) TABLE III. Number and Percent of Subsequent Outcomes Among Soldiers in the Study Population (8,843) n % Gender Male 7, Female Race/Ethnicity Caucasian/White 6, African-American 1, Hispanic Other Marital Status (While in Unit) Single 5, Married 3, Divorced Other Age at Entry to Army (Years) <21 6, , > Home of Record Region Midwest 1, Northeast 1, South Atlantic 1, South central 1, West 2, Other 3, Time in Service (Years) < , , , , , Selected demographic characteristics of the total Army accessions recruits ( , n = 714,679) for comparability are as follows: male (88%), Caucasian/White (78%), and less than 21 years of age at entry to Army (74%). Data Source: Accession Medical Standards Analysis and Research Activity 2010 Annual Report. 17 TABLE II. Number and Percent of Soldiers Granted Enlistment Waivers in the Study Population (N = 8,843) n % Any Enlistment Waiver 1, Alcohol/Drug Nonlawful Violation Juvenile/Adult Felony Offense Serious Nontraffic Violation Medical Waiver for Mental Health Concern n % Either ASAP Screening or a Positive Drug Test 1, for An Illicit Substance ASAP Screening 1, Positive Drug Test for An Illicit Substance Substantiated Domestic Abuse Attrition From Army for Any Reason 2, Attrition From Army for UCMJ Violations or Behavioral Misconduct the majority of Soldiers were single (Table I). Comparatively, the study sample demographics were quite similar to all Army accessions from 2004 to 2008 (n = 714,679) with respect to gender, age, and race. 17 The percent of Soldiers in the current study population entering the Army from 2003 to 2008 who were granted medical and moral waivers increased significantly between 2003 (6.8% and 12.4%, respectively) and 2008 (9.2% and 19.7%, respectively), with the largest increase observed in the percent of moral conduct waivers. The increasing trend of moral waivers was driven by serious nontraffic violations, which increased significantly among Soldiers from 2003 (5.7%) to 2008 (14.2%). Overall, about 1 in 5 Soldiers (21.7%) in the current study population were granted some type of enlistment waiver; 7.5% were granted a waiver for a serious nontraffic violation (Table II). The number of Soldiers in the current study population who had drug and/or alcohol concerns during their military career is indicated by the percent who were screened for alcohol and/or drug problems at ASAP (14.3%) and/or tested positive for illicit drugs (7.5%). Soldiers with a substantiated case of domestic abuse were rare (1.5%). Nearly a third of this population had separated from the Army at the time of the study (32.0%), a much smaller percent specifically for misconduct or behavioral problems (4.8%) (Table III). Soldiers granted any type of enlistment waiver were significantly more likely to test positive for an illicit substance (AOR = 1.5, 95% CI ), to be screened at ASAP for alcohol/drug abuse (AOR = 1.4, 95% CI ), or to attrite from the Army for UCMJ violations or behavioral misconduct (AOR = 1.6, 95% CI ). Soldiers granted waivers for nonlawful alcohol or drug violations or serious nontraffic offenses were significantly associated with the greatest likelihood of a positive test for an illicit substance (AOR = 3.3, 95% CI ; AOR = 1.8, 95% CI , respectively) or being screened at ASAP for alcohol/drug abuse (AOR = 2.9, 95% CI ; AOR = 1.6, 95% CI , respectively) during their military career. Soldiers who were granted waivers for nonlawful alcohol or drug violations were also more likely to leave the Army because of UCMJ violations or behavioral misconduct (AOR = 2.9, 95% CI ) (Table IV). Soldiers granted waivers for serious nontraffic violations and/or juvenile/adult felonies were significantly less likely to separate from the Army for UCMJ violations or behavioral misconduct (AOR = 0.8, 95% CI ; AOR = 0.6, 95% CI , respectively). Enlistment waivers were not significantly associated with subsequent domestic abuse in this population. DISCUSSION The percent of Soldiers enlisting in the current study population who were granted medical and conduct waivers increased MILITARY MEDICINE, Vol. 178, March

4 TABLE IV. AORs and 95% CIs for the Association Between Specific Enlistment Waivers and the Risk for Subsequent Behaviors and/or Attrition ASAP Screening Illicit Drug Test ASAP Screening or Illicit Drug Test Substantiated Domestic Abuse Attrition (All Causes) Attrition (Misconduct/Behavior) AOR 95% CI AOR 95% CI AOR 95% CI AOR 95% CI AOR 95% CI AOR 95% CI Any Enlistment Waiver * * * * Alcohol/Drug Nonlawful Violation * * * * Juvenile/Adult Felony Offense * Serious Nontraffic Violation * * * * Medical Waiver for Mental Health Associations assessed for confounding from the following covariates: gender, ethnicity, marital status while assigned to BCT, age at entry into service, home of record region at enlistment, military grade, and time in service. *p < significantly from 2003 to 2008, with the largest increase seen in the percent of moral conduct waivers. During this same time, there was an increase in the percent of Soldiers accessing into the overall Army with medical and moral waivers from 2003 (12%) to 2008 (20%) (US Army Recruiting Command Waivers [by Component] Briefing, Army G-1, Department of Military Personnel Management, EOM February 2009). Soldiers in the current population examined who were granted enlistment waivers were significantly more likely to subsequently be screened at ASAP, test positive for an illicit substance, or separate from the Army for misconduct behavior. These associations were highest among Soldiers granted specific moral waivers for nonlawful alcohol/drug violations. Soldiers granted waivers for felony offenses and serious nontraffic violations were significantly less likely to separate from the Army when compared with Soldiers who were not granted enlistment waivers. The results are similar to previous studies, which examined enlistment waivers for misconduct and drug/alcohol use. 4 6,9,10 The increase in subsequent ASAP screening, illicit drug tests, and attrition for misconduct/behavior among Soldiers granted waivers for alcohol/drug nonlawful violations was most striking. Other reports have also shown Soldiers granted waivers for drugs were significantly more likely to subsequently attrite within 36 to 48 months of accession. 4,9,10 A population assessment of first-term attrition among all U.S. Army Soldiers who enlisted from 2000 to 2006 (n > 400,000) reported recruits granted waivers for drugs were twice as likely to attrite for substance use compared with nonwaivered recruits, though the absolute magnitude was relatively small since less than 7% of all attrition discharges were for drug/ alcohol problems. 5 An analysis comparing the likelihood of 48-month attrition and the average time to sergeant (E5) promotion determined that being granted a waiver for drugs/alcohol was associated with both relatively high 48-month attrition and lower probability of getting promoting fast to Sergeant. 6 A study completed by the RAND Arroyo Center 9 reported recruits with conduct or drug waivers had significantly increased patterns of misconduct, drug abuse, court martial, and commission of serious crimes. As observed in other large population assessments, 5,10 serious nontraffic waivers were the most common moral waiver granted to Soldiers in the current study population from 2003 to The largest increase in waivers observed in the current study population was also waivers for serious nontraffic violations. Soldiers granted a waiver for a serious nontraffic violation were associated with a significant increased likelihood of subsequent screening at ASAP and positive drug tests for an illicit substance. Soldiers granted serious nontraffic violation waivers were not significantly more likely to attrite from the Army for any reason or specifically for UCMJ violations or behavioral misconduct. With the exception of attrition for negative behavioral reasons, no previously published studies have assessed behavioral and social health outcomes subsequent to accession with waivers. Malone and Carey 6 also observed a null association between waivers for serious nontraffic violation waivers and attrition for behavioral reasons, but observed that Soldiers with these waivers had faster promotion rates than Soldiers without waivers. Distifeno 5 observed significantly lower early attrition rates (180 days), but significantly higher full-term attrition rates for any reason among Soldiers with serious nontraffic waivers. Positive outcomes were not specifically examined in the current study; however, we did observe a significantly decreased likelihood of subsequent overall attrition among recruits granted waivers for felony offenses ( 40%) or serious nontraffic violations ( 20%) compared with recruits not granted enlistment waivers. Other studies have also showed that early attrition rates (i.e., within 1 year of accession) are typically lower among waivered recruits compared with nonwaivered recruits. 5,10 RAND Arroyo Center (2009) 9 reported that recruits with conduct or drug waivers had significantly higher success rates in basic training, better entry level performance/conduct, and were more likely to receive good conduct medals. A study commissioned by the House Armed Services Committee 4 concluded that new recruits with moral, medical, or other waivers were less likely to drop out of basic training, had lower rates of personality disorders, reenlisted in higher numbers, were promoted to sergeant 4 months faster, and were more likely to receive medals of 264 MILITARY MEDICINE, Vol. 178, March 2013

5 valor than new recruits not given waivers. Awareness of the lack of comparable job opportunities outside of the military for persons with criminal backgrounds 18 may factor into a reduced likelihood for attrition for these Soldiers. Alternatively, it is possible that the factors that contributed to engaging in criminal behavior or misconduct have been impacted by the Army core values (i.e., duty, respect, selfless service, honor, integrity, personal courage) and/or camaraderie associated with being a Soldier. In this population, medical enlistment waivers for mental health reasons were not associated with subsequent negative behavioral health outcomes or attrition from the Army. One previous study found that the relative risk of undergoing an MEB among U.S. Navy and Marine personnel receiving medical waivers for mental health reasons was the same as Sailors and Marines receiving any type of waiver, but MEBs specifically for mental health reasons were 5 times more likely. 3 No other studies were identified which specifically examined mental health waivers and subsequent behavioral and social health outcomes or attrition. Although few studies have directly explored the impact of moral conduct and mental health waivers, Soldiers themselves often report negative opinions about these waivers and Soldiers who receive these waivers. Previous studies have shown, during focus groups, that active duty Soldiers commonly report perceptions that recruitment standards have been reduced in recent years to sustain force strength for Operation Iraqi Freedom/Operation Enduring Freedom conflicts 19,20, and many believe the practice of granting enlistment waivers to recruits with mental health problems, criminal backgrounds, and substance abuse histories has led to a less qualified force with weak-minded or bad Soldiers. 20 Soldiers believe that the consequences of increased enlistment waivers include an increase in discipline problems, behavioral health concerns, and relationship problems. One study found that Soldiers believed relaxed recruitment standards, and the increase in enlistment waivers, particularly felony waivers, was contributing to acts of violence and aggression within Army populations. 19 However, to date, there is no evidence (published or unpublished) to support the assertion that units with higher levels of risk have more high-risk substandard Soldiers than other units. The Army-wide percent of accessions with a moral conduct waiver decreased from fiscal year (FY) 2007 (15%) to FY 2008 (13%) (U.S. Army Recruiting Command Waivers [by Component] Briefing, Army G-1, Department of Military Personnel Management, EOM February 2009), and because of increased accession standards and requirements for new recruits starting in FY 2009, 21 the percent of new recruits granted enlistment waivers should continue to decrease. Limitations The findings of this study should be considered in conjunction with the following limitations. This analysis based on data available for Soldiers assigned to 2 BCTs may not be representative of all Army Soldiers or all Soldiers with waivers. The time to follow-up for the current study population entering the Army from 2003 to 2008 differs by year of accession and thus, although we controlled for the year of accession in the analysis, Soldiers with less follow-up time are likely to bias the association observed between waivers and subsequent behavioral outcomes and attrition toward the null. This is important to note because previous studies have shown that recruits with waivers typically do well early in their first enlistment 5,10 and also because in this population we observed a large increase in waivers (particularly serious nontraffic waivers) among recruits entering in 2007 to Behavioral outcomes among recruits entering the broader Army around this time with waivers may be similar or more likely than those observed in the current study. In this study, we were unable to ascertain data on combat medals, promotions, and details of basic training success, which might provide indicators of successful outcomes among recruits granted waivers. Likewise, our study population size did not lend itself to assessing rates of early attrition. Future studies should also consider other factors occurring before, or at the time of enlistment (e.g., undiagnosed mental health concerns, learning disabilities, body mass index), which may be associated with, or predictive of, the outcomes examined in this study. Conclusions The percent of Soldiers enlisting in the current study population who were granted medical and conduct waivers increased significantly from 2003 to 2008, with the largest increase seen in the percent of moral conduct waivers. Soldiers granted waivers for alcohol/drugs or serious nontraffic offenses were significantly more likely to test positive for an illicit substance or be screened at ASAP for alcohol/drugs during their military career, and Soldiers granted waivers for alcohol/drugs were also significantly more likely to attrite from the Army for misconduct/ucmj violations. Accession policies that may have increased the number of Soldiers with waivers could have selected a population with more preexisting risk factors for negative behavioral health outcomes. Army leadership should consider these findings to inform decision making about changes to waiver policy and to anticipate the potential need for additional support services and other second-order effects associated with any such changes. REFERENCES 1. Alvarez L: Army giving more waivers in recruiting. The New York Times, February 14, Available at 02/14/us/14military.html?pagewanted=all; accessed November 14, Army G-1 Suicide Prevention Program: US Army health promotion, risk reduction and suicide prevention report Available at Report_2010_v00.pdf; accessed December 1, MILITARY MEDICINE, Vol. 178, March

6 3. Krahl PL, Litow FK: Mental health disorders in a cohort of U.S. Navy and Marine Corps personnel receiving waivers of medical accession standards. Mil Med 2009; 174: Baldor LC: Army recruits with bad pasts reviewed. Associated Press, April 30, Available at articles/2008/04/30/army_recruits_with_bad_pasts_reviewed/; accessed November 14, Distifeno C: Effects of Moral Conduct Waivers on First-Term Attrition of U.S. Army Soldiers. Naval Postgraduate School Thesis, Monterey, CA, Available at 6. Malone L, Carey N: Waivered recruits: an evaluation of their performance and attrition risks. CNA Report No. D A4, Available at 7. Putka DJ, Noble CL, Becker DE, Ramsberger PF: Evaluating moral character waiver policy against service member attrition and in-service deviance through the first 18 months of service, Available at Putka DJ, Stricklan WJ: A comparison of the FY03 and FY99 first term attrition study cohorts. Arlington, VA, US Army Research Institute for the Behavioral and Social Sciences, Available at 9. RAND Arroyo Center: Recruiting policies and entry level standards-wavier analysis, Available at /RAND_AR7147.pdf; accessed April 30, Sahin F, Ayhan S: The Next Best Alternative to an Ideal Recruit: Attrition Characteristics of Recruits with Waivers and Low Educational Credentials in the U.S. Army. Naval Postgraduate School Thesis, Monterey, CA, Available at Bohn D, Schmitz E: Waiver policy attrition. Arlington, VA, Commander Navy Recruiting Command, Available at Connor JW: The Effects of Pre-Service Criminal History on Recruit Performance in the U.S. Navy. Naval Postgraduate School Thesis, Monterey, CA, Available at GetTRDoc?AD=ADA331671; 13. Etcho L: The Effect of Moral Waivers on First-Term, Unsuitability Attrition in the Marine Corps. Naval Postgraduate School Thesis, Monterey, CA, Available at GetTRDoc?AD=ADA309309; 14. Frabutt AW: The Effects of Pre-Service Legal Encounters on First-Term Unsuitability Attrition in the U.S. Navy. Naval Postgraduate School Thesis, Monterey, CA, Available at GetTRDoc?AD=ADA307674; 15. Department of the Army: Army Regulation : Active and Reserve Components Enlistment Program. Available at Hosmer D, Lemeshow S: Applied Logistic Regression. New York, NY, Wiley-Interscience, Walter Reed Army Institute of Research (WRAIR) Accession Medical Standards Analysis and Research Activity (AMSARA): Attrition and Morbidity Data for FY 2009 Accessions, Available at accessed July 8, Holzer HM, Raphael S, Stoll MA: Perceived criminality, criminal background checks, and the racial hiring practices of employers. J Law Econ 2006; 49: Gallaway MS, Lagana-Riordan C, Millikan AM, the Behavioral and Social Health Outcomes Program: Assessment of reintegration and behavioral health risk at Joint Base Lewis-McChord, Washington, September 2010 January 2011 (Epidemiological Consultation Report No. 14-HK-0DHW-10), Available at accessed October 7, Millikan AM, Bell MR, Gallaway MS, et al: Investigation of homicides at Fort Carson, Colorado, November 2008 May 2009 (Epidemiological Consultation Report No. 14-HK-OB1U-09), Available at accessed July 9, Tyson AS: Army more selective as economy lags. The Washington Post, April 19, Available at content/article/2009/04/18/ar html?hpid=topnews?xid= rss-page; accessed April 30, MILITARY MEDICINE, Vol. 178, March 2013