Essay

Mental and Physical Health Status and Alcohol and Drug Use Following Return From Deployment to Iraq or Afghanistan Susan V. Eisen, PhD, Mark R. Schultz, PhD, Dawne Vogt, PhD, Mark E. Glickman, PhD, A. Rani Elwy, PhD, Mari-Lynn Drainoni, PhD, Princess E. Osei-Bonsu, MSPH, PhD, and James Martin, PhD

The mental and physical health of veterans returning from war zone deployment is of substantial concern to the public as well as military leaders and civilian policymakers.1

Although most veterans return from deployment without suffering long-term consequences, a significant number experience serious psycho- logical harm. Recent research on the Gulf War and Operation Enduring Freedom/Operation Iraqi Freedom (OEF/OIF) veterans confirmed increased risk for mental health problems, in- cluding posttraumatic stress disorder (PTSD), depression, suicidality, neuropsychological defi- cits, and alcohol and drug use.2---7 These disor- ders have implications for individual and unit readiness; physical, social, and emotional health of veterans and their families; and the Veterans Health Administration and other health care systems in which veterans obtain services.4,7,8

Mental health problems are also often com- pounded by problems with physical health.9

Recent conflicts resulted in over 46 000 soldiers wounded in action, some with serious and debilitating conditions, such as chronic pain, traumatic brain injury, and high risk for cardio- vascular disease.10---13

The objectives of our study were to examine (1) mental and physical health symptoms and functioning, including posttraumatic stress symptoms (PTSS), and alcohol and drug use in a national sample of veterans within 1 year of returning from deployment in Afghanistan or Iraq; and (2) differences in mental and physical health and alcohol and drug use by gender, service component (Active, National Guard, other Reserve), service branch (Army, Navy, Air Force, Marines), and deployment operation (OIF or OEF). Examination of differences in mental health, alcohol use, and drug use is important for identifying those at high risk so that evidence-based interventions to prevent and treat serious disorders can be implemented

when indicated. Gender, service component, branch, and deployment operation are of in- terest because previous research identified differences in mental health, alcohol use, and drug use as a function of these factors. Con- tinued assessment of these issues is valuable, because as the focus and scope of military engagement changes over time, the sequelae of deployment may also change, resulting in different subgroups emerging as high risk.

Previous research suggested that women were at higher risk for mental health conditions including depression, anxiety, and PTSD after traumatic exposure, whereas men were at higher risk for substance use disorders.14---19

Based on these findings, we hypothesized that women would report more symptoms of de- pression and PTSD but less alcohol and drug use than men. Previous research found National

Guard and Reservists to be at higher risk for mental health, alcohol use, and drug use disor- ders than Active component personnel.20---23

Consequently, we hypothesized more mental health, alcohol use, and drug use problems among National Guard and Reservists than among Active component members. Several studies identified Army or Marine veterans as at higher risk for PTSD, depression, or alcohol misuse compared with Navy or Air Force vet- erans.16,22,24---27 Based on these findings, we hypothesized higher levels of PTSD, other men- tal health symptoms, alcohol use, and drug use among Army and Marines compared with Navy and Air Force veterans. Significantly more mental health symptoms were reported among OIF (Iraq) veterans compared with OEF (Afghanistan) veterans,2,4,11,28 leading us to hy- pothesize more mental health symptoms and

Objectives. We examined (1) mental and physical health symptoms and

functioning in US veterans within 1 year of returning from deployment, and (2)

differences by gender, service component (Active, National Guard, other Reserve),

service branch (Army, Navy, Air Force, Marines), and deployment operation

(Operation Enduring Freedom/Operation Iraqi Freedom [OEF/OIF]).

Methods. We surveyed a national sample of 596 OEF/OIF veterans, over-

sampling women to make up 50% of the total, and National Guard and Reserve

components to each make up 25%. Weights were applied to account for

stratification and nonresponse bias.

Results. Mental health functioning was significantly worse compared with the

general population; 13.9% screened positive for probable posttraumatic stress

disorder, 39% for probable alcohol abuse, and 3% for probable drug abuse. Men

reported more alcohol and drug use than did women, but there were no gender

differences in posttraumatic stress disorder or other mental health domains. OIF

veterans reported more depression or functioning problems and alcohol and

drug use than did OEF veterans. Army and Marine veterans reported worse

mental and physical health than did Air Force or Navy veterans.

Conclusions. Continuing identification of veterans at risk for mental health

and substance use problems is important for evidence-based interventions

intended to increase resilience and enhance treatment. (Am J Public Health.

2012;102:S66–S73. doi:10.2105/AJPH.2011.300609)

RESEARCH AND PRACTICE

S66 | Research and Practice | Peer Reviewed | Eisen et al. American Journal of Public Health | Supplement 1, 2012, Vol 102, No. S1

greater substance use among OIF than OEF veterans.

METHODS

We used an observational research design in which a national random sample of US OEF/ OIF veterans was surveyed within 12 months of returning from deployment. Surveys were mailed in November 2008, and data collection, which included repeat mailings and reminder phone calls, was completed by the end of 2009. We allowed up to 1 year for receipt of completed surveys.

Sampling Plan

The sample was stratified by service com- ponent (50% Active, 25% National Guard, and 25% other Reserve) and gender, with over- sampling of women to make up 50% of the total within each service component. Two thousand OEF/OIF veterans, meeting these stratification specifications and who returned from deployment between 3 and 12 months previously, were obtained from the Defense Manpower Data Center (DMDC). The DMDC is the Department of Defense central repository for personnel data, maintaining 230 secure and protected databases, websites, and pro- grams.29 A total of 167 individuals were de- termined to be ineligible because they had a military address outside the United States (n = 102), were redeployed (n = 52), or indi- cated that they had not been deployed in OEF/ OIF (n = 13). Of the 1833 eligible individuals, we confirmed that 1043 received the survey materials, and we obtained completed surveys from 598, although 2 surveys were received too late to include in the data analysis. Likely reasons for not receiving the survey included high mobility of the sample, many of whom were still on active duty and might have been deployed elsewhere, moved, or changed tele- phone numbers. Survey respondents came from every state except Wyoming, as well as the District of Columbia, Puerto Rico, and the US Virgin Islands.

Procedure

Prenotification letters and an “opt-out” postcard were sent to all eligible service mem- bers to inform them about the study. Anyone who returned the opt-out postcard within 2

weeks was not recontacted. Two weeks after the prenotification mailing, each eligible vet- eran was sent a cover letter, an information sheet detailing all elements of informed con- sent, the survey, and a preaddressed, postage paid return envelope. A modified Dillman method was used to enhance response rates.30

If completed surveys were not returned within 2 weeks, a reminder letter was sent. Up to 6 reminder telephone calls were made, followed by a repeat mailing to remaining nonrespondents. To further maximize response rates, an Internal Revenue Service (IRS) address search, available via a VA Environmental Epidemiology Service interagency agreement with the IRS, was initiated to obtain updated addresses for in- dividuals who could not be reached by mail or phone. This procedure was approved by the institutional review board. Those who returned a completed survey received a $30 gift card.

Measures

Established, reliable. and valid self-report instruments were used to assess mental and physical health symptoms and functioning, including PTSS, alcohol use, and drug use. Demographic characteristics including race and Hispanic ethnicity, length of deployment, service component, service branch, and de- ployment operation (OEF/OIF) were obtained from the DMDC. Additional demographic and descriptive information were obtained using self-reported questions used in previous VA research studies.31,32 Race/ethnicity was obtained for descriptive purposes and is presented in Table 1 based on self-report. Mental and physical health, alcohol use, and

drug use. Mental and physical health functioning, alcohol use, and drug use were assessed with the Veterans Rand-12 (VR-12),33,34 Behavior and Symptom Identification Scale (BASIS-24),35

PTSD Checklist Military Version (PCL-M),36

Alcohol Use Disorders Identification Test (AUDIT-C),37 and Drug Abuse Screening Test (DAST).38

The VR-12 is a brief version of the Veterans RAND-36, assessing 2 broad domains of men- tal and physical health functioning (Mental Component Score [MCS] and Physical Compo- nent Score [PCS]).34,39 The VR- and Short Form-12 are among the most widely used functional status measures in the world and have

been shown to be highly reliable and valid. The reliability estimate for the PCS-12 was 0.80, and for the MCS-12 was 0.76.

The PCL-M is a 17-item instrument derived from PTSD diagnostic criteria to assess PTSS severity among veterans. High test---retest and internal consistency reliability and good con- current validity have been reported.40 The BASIS-24 is a multidimensional mental health assessment instrument. We used 5 of the 6 BASIS-24 subscales: depression and functioning, interpersonal relationships, emotional lability, psychotic symptoms, and substance use. A sixth subscale assessing self-harm was excluded be- cause it was deemed inappropriate for use in a mailed survey in which risk of self-harm could not be followed up locally. Reliability of the subscales ranges from 0.77 to 0.91 with good concurrent and discriminant validity.35 BASIS- 24 was validated in a national sample of more than 5800 recipients of mental health or sub- stance abuse services, and was also used in other VA mental health studies.31,32,35

The AUDIT-C is a brief version of the 10-item AUDIT, a self-report measure designed to iden- tify individuals experiencing problems with al- cohol. AUDIT-C identified 90% of patients with alcohol abuse or dependence and 98% of patients with heavy drinking.37 The DAST-10 is a 10-item instrument designed to identify illegal drug use problems. It has high internal consis- tency reliability (a = 0.94), test---retest reliability (intraclass correlation coefficient = 0.71), and has been found to discriminate between outpatients with and without drug use disorders.38

Data Analysis

Sample weights. Two survey respondent weightings were applied to enhance the repre- sentativeness of the sample to the larger OEF/ OIF cohort. First, we computed sampling weights for the original sample of 2000 across each of the 6 strata from which service mem- bers were sampled (all combinations of men and women with Active Component, National Guard, and other Reserve). Second, we com- puted weights to account for survey nonre- sponse. This was done by performing a logistic regression on the initial sample of 2000 with “returned survey” (0/1) as the dependent variable, and age, gender, race, and service component as independent variables to esti- mate a probability of returning the survey for

RESEARCH AND PRACTICE

Supplement 1, 2012, Vol 102, No. S1 | American Journal of Public Health Eisen et al. | Peer Reviewed | Research and Practice | S67

each person in the sample. The reciprocal of these probabilities were the values of the second set of weights. The final set of weights was the product of an individual’s sampling weight and the nonresponse weight. These weights were used in the calculation of mental health and substance use scores. Data analysis. We used the t test to assess

differences in mental health and substance use by gender, and by deployment operation (OEF/OIF). Analysis of variance was used to assess differences by component and branch of service. All analyses incorporated sampling and nonresponse weights as previously described. In addition, for the PCL-M, AUDIT-C, and DAST-10, we used previously established cut scores to compare probable PTSD, alcohol, and drug use between gender, component, service, and deployment operation. Results based on cut scores are reported in the text but not in tables. Significance level adjust- ments for multiple testing were not performed, as these adjustments are not recommended for studies in which multiple hypotheses are tested.41---43

RESULTS

To identify differences between survey re- spondents and nonrespondents, we used de- scriptive data obtained from the DMDC to compare demographic and deployment charac- teristics of the 596 survey respondents with those of the 1237 nonrespondents. Survey re- spondents were more likely than were non- respondents to be women (v2 = 15.8; P < .001), older (t = –8.48; P < .001), National Guard or Other Reserve (v2 = 10.7; P = .005), and from the Air Force or Navy (v2 = 16.9; P = .001) than from the Army or Marine Corps. There were no differences in response rate as a func- tion of race, Hispanic ethnicity, or length of previous deployment.

Sample Characteristics

Table 1 presents demographic and deploy- ment characteristics of the 596 survey re- spondents. Fifty-eight percent were female; average age was 33.9 years; and 75% were White, 18% were African American, and 12% were Latino. Reflecting the oversampling of National Guard and other Reserve personnel, 58% were from these groups. Two thirds of the

sample were deployed most recently to Iraq. The majority (87.5%) were still in the military.

Mental and Physical Health, Alcohol Use,

and Drug Use

Addressing our first study objective, mean mental and physical health, alcohol use, and drug use scores are presented for the full sample in Table 1. The mean (SD) MCS score on the VR-12 was 40.56 (7.91), more than 1 SD below the general population mean of 50, whereas the mean (SD) PCS score was 49.51 (9.63), which was close to the general population mean.44 The mean (SD) PCL-M score was 31.06 (14.00); 13.9% of the sample met the established threshold score of 50, indicating a likely di- agnosis of PTSD.45 Mean BASIS-24 subscale and overall scores ranged from 0.56 to 1.52 and were consistently worse than BASIS-24 scores obtained from a national, nonclinical community sample, but better than scores reported for both veteran and nonveteran samples receiving out- patient mental health treatment (T. Idiculla, PhD, unpublished data, 2011).31,32,46 Using the VA- established AUDIT-C threshold for probable alcohol use disorder (‡ 5 for both genders),19

39.2% of the total sample screened positive for probable alcohol use disorder. Regarding drug abuse, 2.9% screened positive for drug abuse on the DAST-10. Gender differences. Contrary to our hypothe-

sis that women would report higher levels of depression and PTSS, there were no statistically significant gender differences on any of the mental health or PTSS measures (Table 2). However, our hypothesis regarding substance use was largely supported in that men reported higher mean AUDIT-C, DAST, and BASIS-24 substance use scores than women. Using the AUDIT-C cut score (‡ 5) indicating probable alcohol abuse, significantly more men (41.4%) than women (17.0%) screened positive (v2 = 12.3; P < .001), although there was no significant gender difference in the proportion of respondents who screened positive for drug abuse (3% of men and 2% of women). Deployment operation (OEF/OIF) differences.

In partial support of our hypothesis, OIF veterans reported significantly more difficulty in the areas of depression or functioning, overall mental health, and alcohol or drug use than did OEF veterans, as reflected in higher mean BASIS-24 scores in these areas, as well as

TABLE 1—Sample Characteristics for

Health Status, Alcohol and Drug Use

Following Return from Deployment to

Iraq or Afghanistan: United States,

2008–2009

Variablesa No. (%) or

Mean 6SD

Gender

Male 253 (42.4)

Female 343 (57.6)

Age, y

18–24 117 (19.6)

25–29 127 (21.3)

30–39 180 (30.2)

‡40 172 (28.9) Race

a

White 448 (75.2)

African-American 107 (18.0)

Asian 20 (3.4)

American Indian/Alaskan native 20 (3.4)

Native Hawaiian/Pacific Islander 10 (1.7)

Hispanic 70 (11.8)

Education

High school grad/GED/vocational

school

94 (15.8)

Some college/associates degree 287 (48.2)

Bachelors degree 137 (23.0)

Post-bachelors degree 74 (12.4)

Marital status

Single (never married) 174 (29.2)

Married/with partner 326 (54.7)

Separated 16 (2.7)

Divorced 79 (13.3)

Widowed 1 (<1)

Employment status b

Working for pay 495 (83.1)

Working as volunteer 31 (5.2)

Student 84 (14.1)

Homemaker 24 (4.0)

Not working but actively

looking for work

66 (11.1)

Not working and not looking 8 (1.3)

Retired 6 (1.0)

Unable to work 10 (1.7)

Time in military, y

<1 0 (0)

1–2 32 (5.4)

3–4 91 (15.3)

5–10 187 (31.4)

>10 y 285 (47.8)

Most recent deployment operation

OIF (Iraq) 401 (67.3)

OEF (Afghanistan) 181 (30.4)

Other 11 (1.8)

Continued

RESEARCH AND PRACTICE

S68 | Research and Practice | Peer Reviewed | Eisen et al. American Journal of Public Health | Supplement 1, 2012, Vol 102, No. S1

higher mean levels of alcohol use on the AUDIT-C and more positive screens for drug use (4% OIF vs < 1% OEF; v21 = 5.29; P = .02; Table 3). Component differences. The hypothesis

regarding poorer mental health and increased substance use among National Guard/Reserv- ists versus Active Component was not sup- ported. There were no significant differences among the components on any of the measures using mean scores or cut points for PTSD, alcohol use, or drug use (data not shown). Service branch differences. Supporting

hypotheses regarding service branch differ- ences, results indicated statistically significant differences on all of the measures (Table 4). Post hoc t tests indicated that Army and Marine veterans differed most from Air Force veterans, with statistically significant differences on 8 of the 11 measures. Army and Marine veterans indicated significantly poorer mental health functioning, including more PTSS, depression, and anxiety symptoms, higher alcohol and drug use, and more problematic relationships than did Air Force veterans. Army veterans indi- cated significantly higher emotional liability and psychotic symptoms than did Air Force veterans. Twenty-five percent of Marines and 15% of Army respondents screened positive for probable PTSD compared with 9.5% of Air Force and 5.9% of Navy veterans. Forty-seven percent of Army respondents and 45% of Marines screened positive for alcohol use compared with 26% of both Air Force and Navy respondents (v23 = 24.91; P < .001); 7.4% of Marines and 3.4% of Army respon- dents screened positive for drug use compared with < 1% of Air Force or Navy respondents (v23 = 10.99; P < .001).

DISCUSSION

This article extended previously published work in a number of ways. First, we obtained a broader sample than many previous studies, reporting results on a national stratified, ran- dom sample of Active and National Guard/ Reserve military personnel from all branches of service who were not necessarily enrolled in Department of Veterans Affairs (VA) health care. Second, to allow for comparisons by gender we oversampled women to make up 50% of the sample. Third, rather than using

brief screening tools, we included detailed, well-researched measures reflecting a wide range of mental health symptom and func- tioning domains, including PTSS, alcohol and drug use, and physical health functioning. Fourth, we applied 2 sets of weights to the data analysis, one to reflect the actual composition of the OEF/OIF force during the sampling interval, and another to account for nonre- sponse bias. Application of these weights pro- vided more accurate estimates of mental health functioning and substance use problems, thus increasing generalizability of results.

We found that OEF/OIF veterans indicated significantly worse mental health functioning than did the general population based on multiple measures. Alcohol use was also po- tentially problematic, with 39% screening positive for “probable alcohol abuse,” consid- erably higher than that reported by Hawkins et al.,19 based on mandated screening of VA outpatients. This discrepancy might suggest that the context of the VA clinical setting could lead veterans to underreport their drinking in routine screening, resulting in underestimates of the severity of the problem. By contrast, physical health functioning assessed by the VR-12 was comparable to the US general population mean.44 The relatively good physical health of our sample might be partly attributed to their youth and fitness compared with the general population.

Despite widespread publicity and concern about increasing rates of PTSD among OEF/ OIF veterans, probable PTSD for this sample (13.9%) was within the range reported by other researchers and lower than rates reported in some studies.2,4,28 A number of factors were identified as influencing reported rates of PTSD, including methodological factors (strictness of the definition, amount of time since return from deployment, sampling strategy, and response rates), as well as substantive factors (combat and other traumatic exposure).21,47,48

Differences in PTSS or PTSD and other mental health problems as a function of postdeployment assessment time interval might be especially important to monitor as there was conflicting evidence regarding the impact of time on these conditions.21 Kulka et al.49 reported that depression, anxiety, and other mental health concerns subsided over time, whereas other researchers reported that PTSD sometimes had

TABLE 1—Continued

Service component

Active 249 (41.8)

National Guard 170 (28.5)

Other Reserve 177 (29.7)

Branch of service

Army 344 (57.7)

Navy 101 (16.9)

Air Force 125 (21.0)

Marines 25 (4.2)

Total no. of deployments at time of survey

1 333 (55.7)

2 165 (27.7)

3 57 (9.6)

‡4 40 (6.7) Length of most recent deployment, d

£90 110 (18.5) 91–180 82 (13.8)

181–365 274 (46.0)

>365 123 (20.6)

Current military status

Enrolled 511 (85.7)

Discharged 73 (12.2)

Mental/physical health

PTSS (PCL-M) c

31.06 614.00

MCS (VR-12 mental health) d

40.56 67.91

PCS (VR-12 physical health) d

49.51 69.63

Depression/functioning (BASIS-24) e

0.95 60.80

Interpersonal relations (BASIS-24) e

1.17 60.78

Emotional lability (BASIS-24) e

1.52 60.97

Psychotic symptoms (BASIS-24) e

0.56 69.81

Overall score (BASIS-24) e

1.03 60.68

Alcohol/drug use

Substance abuse (BASIS-24) e

0.45 60.69

Alcohol Use (AUDIT-C) f

3.92 62.84

Drug Use (DAST) g

0.41 60.81

Note. AUDIT-C=Alcohol Use Disorders Identification Test; BASIS-24=Behavior and Symptom Identification Scale; DAST=Drug Abuse Screening Test; GED=General Equivalency Diploma; MCS= Mental Component Score; OEF/OIF=Operation Enduring Freedom/Operation Iraqi Freedom; PCL-M=PTSD Checklist Military Version; PCS= Physical Component Score; PTSD=posttraumatic stress disorder; PTSS=posttraumatic stress symptoms; VR-12= Veterans Rand-12. The sample size was n=596. a Missing values for each variable ranges between 0 and 12. b Percentage exceeds 100% because multiple categories may be endorsed. cPCL-M scores range from 1 to 68; higher scores indicate greater symptom severity. d MCS and PCS mean (SD)=50 (10). Their ranges are unbounded. Higher scores indicate better mental health. e BASIS-24 scores range from 0 to 4; higher scores indicate greater symptom or problem severity or frequency. fAUDIT-C scores range from 0 to 12; higher scores indicate greater risk of alcohol abuse. g DAST-10 scores range from 0 to 10; higher scores indicate greater risk of drug abuse.

RESEARCH AND PRACTICE

Supplement 1, 2012, Vol 102, No. S1 | American Journal of Public Health Eisen et al. | Peer Reviewed | Research and Practice | S69

a delayed onset, resulting in higher rates over time.21,50,51

Tests of our hypotheses regarding sub- group differences on mental health, PTSS, alcohol, and drug use yielded mixed results. Higher rates of alcohol use in men, and worse mental health, PTSS, alcohol, and drug use among Army and Marine Corps veterans were observed, suggesting that these veteran sub- groups were at higher risk for mental health problems. Although we were not aware of any significant differences in eligibility criteria for enlistment or commissioning in the different service branches that would lead to systematic predeployment differences among service personnel, demographic and cultural differ- ences among the service branches as well as different stressors might contribute to their increased risk.

Contrary to our hypothesis, we found no gender differences in mental or physical health symptoms or functioning. It was important to note, however, that women experienced less intense combat than men.52 Consequently, this finding did not preclude the possibility that women would be more vulnerable to combat stress when exposed at the same levels as men. To address this possibility more directly, another study based on this dataset examined whether associations between combat-related stressors and postdeployment mental health differed for women and men, and found no clinically signif- icant differences.52 The lack of significant gender differences might reflect improved training of female service members and the increased numbers of women currently in the military— 15% now compared with less than 3% during World War II—as well as the fact that women’s occupational roles in the military might be more similar to men’s roles than in the past. The higher proportion of women in today’s military might also provide social and emotional support that might increase their resilience.

Also contrary to our hypotheses, National Guard/Reserve Component veterans did not differ from Active component veterans with respect to mental or physical health functioning or substance use. The lack of differences between compo- nents might be because of changing expectations regarding deployment. As these wars continue, National Guard and other Reserve personnel are more likely to expect to be deployed. Conse- quently, they might be more psychologically and

TABLE 3—Differences in Mental and Physical Health Status, Alcohol and Drug Use by

Deployment Operation: United States, 2008–2009

Iraq (OIF; n=401),

Mean (SD)

Afghanistan (OEF; n=181),

Mean (SD) t P

Mental/physical health

PTSS (PTSD checklist) 31.69 (14.28) 30.54 (13.70) 0.89 £ .371 Depression/functioning (BASIS-24)

a 1.01 (0.84) 0.82 (0.74) 2.72 £ .007

Interpersonal relations (BASIS-24) a

1.21 (0.80) 1.07 (0.74) 1.93 £ .054 Emotional lability (BASIS-24)

a 1.57 (1.00) 1.42 (0.96) 1.68 £ .093

Psychotic symptoms (BASIS-24)a 0.54 (0.76) 0.60 (0.93) –0.70 £ .484 BASIS-24 overall scorea 1.07 (0.69) 0.93 (0.67) 2.37 £ .018 MCS (VR-12 mental health)b 40.69 (8.05) 40.24 (7.82) 0.63 £ .529 PCS (VR-12 physical health)b 49.40 (9.74) 49.76 (9.63) –0.41 £ .683

Alcohol/drug use

Alcohol use (AUDIT-C) c

4.17 (3.07) 3.51 (2.21) 2.97 £ .003 Substance abuse (BASIS-24) a 0.53 (0.74) 0.30 (0.53) 4.25 £ .001 Drug Use (DAST)

d 0.42 (0.88) 0.35 (0.63) 1.07 £ .286

Note. AUDIT-C=Alcohol Use Disorders Identification Test; BASIS-24=Behavior and Symptom Identification Scale; DAST=Drug Abuse Screening Test; MCS= Mental Component Score; OEF/OIF=Operation Enduring Freedom/Operation Iraqi Freedom; PCS=Physical Component Score; PTSD=posttraumatic stress disorder; PTSS=posttraumatic stress symptoms; VR-12=Veterans Rand-12. aBASIS-24 scores range from 0–4; higher scores indicate greater symptom or problem severity or frequency. b MCS and PCS mean (SD)=50 (10). Their ranges are unbounded. Higher scores indicate better mental health. c AUDIT-C scores range from 0–12; higher scores indicate greater risk of alcohol abuse. d DAST-10 scores range from 0–10; higher scores indicate greater risk of drug abuse.

TABLE 2—Gender Differences in Mental and Physical Health Status, Alcohol

and Drug Use: United States, 2008–2009

Variables Men (n=253), Mean (SD) Women (n=343), Mean (SD) t P

Mental/physical health

PTSS (PCL-M) a

31.17 (14.04) 29.97 (13.54) 1.04 £ .297 Depression/functioning (BASIS-24)

b 0.95 (0.80) 0.95 (0.85) 0.01 £ .989

Interpersonal relations (BASIS-24) b

1.18 (0.78) 1.11 (0.79) 1.08 £ .282 Emotional lability (BASIS-24)

b 1.51 (0.97) 1.60 (1.00) –1.09 £ .275

Psychotic symptoms (BASIS-24)b 0.57 (0.82) 0.45 (0.69) 1.89 £ .06 BASIS-24 overall scoreb 1.03 (0.68) 1.00 (0.69) 0.55 £ .582 MCS (VR-12 mental health)c 40.55 (7.78) 40.67 (9.11) –0.16 £ .87 PCS (VR-12 physical health)c 49.50 (9.64) 49.57 (9.55) –0.08 £ .937

Alcohol/drug use

Alcohol Use (AUDIT-C) d

4.06 (2.86) 2.54 (2.27) 6.99 £ .001 Substance abuse (BASIS-24)

b 0.47 (0.70) 0.24 (0.42) 4.74 £ .001

Drug Use (DAST) e

0.42 (0.82) 0.30 (0.63) 2.01 £ .045

Note. AUDIT-C=Alcohol Use Disorders Identification Test; BASIS-24=Behavior and Symptom Identification Scale; DAST=Drug Abuse Screening Test; MCS= Mental Component Score; PCL-M=PTSD Checklist Military Version; PCS=Physical Component Score; PTSD=posttraumatic stress disorder; PTSS=posttraumatic stress symptoms; VR-12=Veterans Rand-12. a PCL-M scores range from 1 to 68; higher scores indicate greater symptom severity. b BASIS-24 Scores range from 0–4; higher scores indicate greater symptom or problem severity or frequency. c MCS and PCS mean (SD)=50 (10). Their ranges are unbounded. Higher scores indicate better mental health. dAUDIT-C scores range from 0–12; higher scores indicate greater risk of alcohol abuse. eDAST-10 scores range from 0–10; higher scores indicate greater risk of drug abuse.

RESEARCH AND PRACTICE

S70 | Research and Practice | Peer Reviewed | Eisen et al. American Journal of Public Health | Supplement 1, 2012, Vol 102, No. S1

emotionally prepared and less vulnerable to deployment-related stresses. Similarly, the shift of combat operations from Iraq to Afghanistan might alter the balance between these 2 opera- tions in terms of associated risks.

This study had a number of limitations. First, because of the retrospective design of the study, we could not infer that the physical and mental health status of respondents reflected the impact of deployment. It was possible that predeployment variations might account for the postdeployment differences. Second, we used only self-reported measures of mental health and substance use, which might be subject to bias. To minimize potential bias, we implemented the survey at least 3 months after participants had returned from deployment, and we conducted the survey independently of the Department of Defense or the participant’s military unit. Thus, concerns about stigma re- garding reporting mental health problems and potential evaluation regarding fitness for duty were minimized. Third, our response rate of 57% was based on the number of potential participants that we confirmed received the survey, but was 33% of the eligible partici- pants, slightly lower than the 36% participation rate reported for the Millennium Cohort

Study.25 Although we weighted the data to address nonresponse, the weighting procedure assumed that the data were missing at random.53

It was possible that nonresponse might have depended on unobserved information. For ex- ample, if military personnel with poor mental health were less likely to return the survey, then our results would only apply to a somewhat healthier segment of the population.

From clinical, public health and policy per- spectives, the results of this study suggest a number of implications. First, continued multidimensional assessment of postdeployed veterans is valuable because previously iden- tified findings may change as contextual fac- tors, such as increasing numbers of women and the locus of combat, shift. Second, the context and timing of assessment (e.g., within one’s unit, immediately after returning from deployment) have been shown to affect results. Conse- quently, we should strive to implement assess- ments in as neutral a context as possible to minimize (1) potential response bias because of concern about discrimination associated with admitting mental health or substance use problems, and (2) failure to report problems to avoid further mental health evaluation that would delay return to home. Third, continuing

identification of those at highest risk for mental health and substance use problems, including examination of individual and deployment characteristics (e.g., race/ethnicity, education, income, combat exposure) is important for development and implementation of evi- dence-based interventions to increase resil- ience, as well as to enhance treatment when indicated. j

About the Authors Susan V. Eisen, Mark R. Schultz, Mark E. Glickman, A. Rani Elwy, Mari-Lynn Drainoni, and Princess E. Osei- Bonsu are with the Center for Health Quality, Outcomes and Economic Research (CHQOER), Edith Nourse Rogers Memorial Veterans Hospital, Bedford, MA. Susan V. Eisen, Mark E. Glickman, A. Rani Elwy, Mari-Lynn Drainoni, and Princess E. Osei-Bonsu are also with the Department of Health Policy and Management, Boston University School of Public Health, Boston, MA. Dawne Vogt is with the Women’s Health Sciences Division, National Center for PTSD, VA Boston Healthcare System, Boston, and the Division of Psychiatry, Boston University School of Medi- cine, Boston. James Martin is with Bryn Mawr College, Bryn Mawr, PA. Correspondence shoud be sent to Susan V. Eisen, PhD

Center for Health Quality, Outcomes & Economic Research (CHQOER), Edith Nourse Rogers Memorial Veterans Hospital, 200 Springs Road (152), Bedford, MA 01730 (e-mail: [email protected]). Reprints can be ordered at http://www.ajph.org by clicking the “Reprints/Eprints” link. This article was accepted November 28, 2011.

TABLE 4—Service Branch Differences in Mental and Physical Health Status, Alcohol and Drug Use: United States, 2008–2009

Army (n=343), Mean (SD) Air Force (n=125), Mean (SD) Marines (n=27), Mean (SD) Navy (n=101), Mean (SD) F P

Mental/physical health

PTSS (PTSD checklist)a 33.79 (14.76) 26.53 (12.67) 32.62 (14.48) 27.31 (10.01) 11.14 £ .001 Depression/functioning (BASIS-24)

a,b 0.99 (0.80) 0.71 (0.63) 1.18 (0.97) 0.93 (0.78) 6.26 £ .001

Interpersonal relations (BASIS-24) a,b

1.22 (0.77) 0.90 (0.69) 1.41 (0.85) 1.16 (0.72) 8.47 £ .001 Emotional lability (BASIS-24)

b,c 1.71 (1.05) 1.31 (0.91) 1.32 (0.67) 1.40 (0.95) 7.65 £ .001

Psychotic symptoms (BASIS-24) b,c

0.66 (0.87) 0.37 (0.67) 0.48 (0.62) 0.57 (0.86) 3.86 £ .009 BASIS-24 overall scorea,b 1.10 (0.68) 0.80 (0.59) 1.16 (0.71) 0.98 (0.65) 7.45 £ .001 MCS (VR-12 mental health)a,d 40.44 (8.04) 42.60 (7.83) 39.53 (6.99) 39.35 (7.92) 4.03 £ .007 PCS (VR-12 physical health)d,e 48.31 (9.89) 49.78 (10.07) 51.36 (9.33) 51.02 (7.98) 3.46 £ .016

Alcohol/drug use

Alcohol use (AUDIT-C) a,f

4.48 (3.13) 2.96 (2.02) 3.93 (2.63) 3.44 (2.59) 9.76 £ .001 Substance abuse (BASIS-24)

a,b 0.59 (0.79) 0.21 (0.35) 0.54 (0.74) 0.27 (0.44) 12.50 £ .001

Drug use (DAST) a,g

0.44 (0.83) 0.18 (0.45) 0.78 (1.18) 0.28 (0.47) 10.80 £ .001

Note. AUDIT-C=Alcohol Use Disorders Identification Test; BASIS-24=Behavior and Symptom Identification Scale; DAST=Drug Abuse Screening Test; MCS= Mental Component Score; PTSD= posttraumatic stress disorder; PTSS=posttraumatic stress symptoms; VR-12=Veterans Rand-12. aSignificant difference between Army and Marine and Air Force veterans. b BASIS-24 Scores range from 0–4; higher scores indicate greater symptom or problem severity or frequency. c Significant difference between Army and Air Force veterans. d MCS and PCS mean (SD)=50 (10). Their ranges are unbounded. Higher scores indicate better mental health. eSignificant difference between Marine and Navy and Army veterans. fAUDIT-C scores range from 0–12; higher scores indicate greater risk of alcohol abuse. gDAST-10 scores range from 0–10; higher scores indicate greater risk of drug abuse.

RESEARCH AND PRACTICE

Supplement 1, 2012, Vol 102, No. S1 | American Journal of Public Health Eisen et al. | Peer Reviewed | Research and Practice | S71

Contributors S. V. Eisen had overall responsibility and provided supervision for this study. S. V. Eisen, A. R. Elwy, M-L. Drainoni, J. Martin, and D. Vogt contributed to obtaining funding. S. V. Eisen, A. R. Elwy, D. Vogt, M. E. Glickman, M-L. Drainoni, and J. Martin contributed to the study concept, design, and acquisition of the data. S. V. Eisen, M. E. Glickman, and M. R. Schultz contributed to the statistical analysis. S. V. Eisen, M. R. Schultz, and P. E. Osei-Bonsu contributed to drafting the article. All au- thors contributed to the interpretation of data and to critical revision of the article for intellectual content.

Acknowledgments This research was funded by the Veterans Administra- tion Health Services Research and Development (HSR&D; grant IAC 06-259-2; S. V. Eisen, PI), by the Center for Health Quality, Outcomes & Economic Research, a VA HSR&D Center of Excellence, and by Women’s Health Sciences Division, National Center for PTSD, VA Boston Healthcare System.

We are most grateful to the veterans who participated in this study. We also thank Alexandra Howard, Patrick Furlong, and Nicole Del Vecchio for their assistance with data collection and management, with conference pre- sentation materials, and with article preparation.

Some of the results reported in this article were presented at the Annual Academy Health Conference in June 2010 and at the Annual VA HSR&D conference in February 2011.

Note. The views expressed in this article are those of the authors and do not necessarily represent the views of the Department of Veterans Affairs or any of the institutional affiliations listed.

Human Participant Protection This study was approved by the institutional review board of the Edith Nourse Rogers Memorial Veterans Hospital.

References 1. Smith TC, For the Millenium Cohort Study Team. The US Department of Defense Millennium Cohort Study: career span and beyond longitudinal follow-Up. J Occup Environ Med. 2009;51(10):1193---1201.

2. 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;351:13---22.

3. Southwick SM, Morgan MD, Nagy LM, et al. Trauma- related symptoms in veterans of Operation Desert Storm: a preliminary report. Am J Psychiatry. 1993;150(10): 1524---1528.

4. 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(9):1023---1032.

5. Sutker PB, Davis JM, Uddo M, Ditta SR. War zone stress, personal resources, and PTSD in Persian Gulf War returnees. J Abnorm Psychol. 1995;104(3):444---452.

6. Vasterling JJ, Proctor SP, Amoroso P, Kane R, Heeren T, White RF. Neuropsychological outcomes of army personnel following deployment to the Iraq war. JAMA. 2006;296(5):519---529.

7. Vasterling JJ, Proctor SP, Friedman MJ, et al. PTSD symptom increases in Iraq-deployed soldiers: comparison

with nondeployed soldiers and associations with baseline symptoms, deployment experiences, and postdeployment stress. J Trauma Stress. 2010;23(1):41---51.

8. Barrett DH, Doebbeling CC, Schwartz DA, et al. Posttraumatic stress disorder and self-reported physical health status among US military personnel serving du- ring the Gulf War period. Psychosomatics. 2002;43: 195---205.

9. Hoge CW, Terhakopian A, Castro CA, Messer SC, Engle CC. Association of posttraumatic stress disorder with somatic symptoms, health care visits, and absentee- ism among Iraq War veterans. Am J Psychiatry. 2007;164(1):150---153.

10. Department of Defense. Operation Enduring Free- dom/Operation Iraqi Freedom/ Operation New Dawn US casualty status. 10/17/2011. Available at: http:// www.defense.gov/news/casualty.pdf. Accessed October 18, 2011.

11. Kline A, Falca-Dodson M, Sussner B, et al. Effects of repeated deployment to Iraq and Afghanistan on the health of New Jersey Army National Guard Troops: implications for military readiness. Am J Public Health. 2010;100(2):276---283.

12. Hoge CW, McGurk D, Thomas JL, Cox AL, Engel CC, Castro CA. Mild traumatic brain injury in U.S. soldiers returning from Iraq. N Engl J Med. 2008;358(5): 453---463.

13. Cohen BE, Marmar C, Ren L, Bertentahl D, Seal KH. Association of cardiovascular risk factors with mental health diagnoses in Iraq and Afghanistan war veterans using VA health care. JAMA. 2009;302(5):489---492.

14. Bernhardt A. Rising to the challenge of treating OEF/OIF veterans with co-occurring PTSD and substance abuse. Smith Coll Stud Soc Work. 2009;79:344---367.

15. Haskell SG, Gordon KS, Mattocks K, et al. Gender differences in rates of depression, PTSD, pain, obesity, and military sexual trauma among Connecticut war veterans of Iraq and Afghanistan. J Womens Health (Larchmt). 2010;19(2):267---271.

16. Maguen S, Lucenko BA, Reger MA, et al. The impact of reported direct and indirect killing on mental health symptoms in Iraq war veterans. J Trauma Stress. 2010;23(1):86---90.

17. Tolin DF, Foa EB. Sex differences in trauma and posttraumatic stress disorder: a quantitative review of 25 years of research. Psychol Bull. 2006;132(6):959---992.

18. Vogt DS, Pless AP, King LA, King DW. Deployment stressors, gender, and mental health outcomes among Gulf War I veterans. J Trauma Stress. 2005;18(2): 115---127.

19. Hawkins EJ, Lapham GT, Kivlahan DR, Bradley KA. 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):147---153.

20. Jacobson IG, Ryan MAK, Hooper TI, et al. Alcohol use and alcohol-related problems before and after mili- tary combat deployment. JAMA. 2008;300(6):663---675.

21. Milliken CS, Auchterlonie JL, Hoge CW. Longitudi- nal assessment of mental health problems among active and reserve component. JAMA. 2007;298(18):2141--- 2148.

22. Schell TL, Marshall GN. Survey of individuals previously deployed for OEF/OIF. In: Tanelian T, Jaycox LH, eds. Invisible Wounds of War: Psychological and Cognitive Injuries, Their Consequences, and Services to

Assist Recovery. Santa Monica, CA: RAND Center for Military Health Policy Research; 2008:87---115.

23. Thomas JL, Wilk JE, Riviere LA, et al. Prevalence of mental health problems and functional impairment among active component and national guard soldiers 3 and 12 months following combat in Iraq. Arch Gen Psychiatry. 2010;67(6):614---623.

24. Bray RM, Olmsted KLR, Williams J, Sanchez RP, Hartzell M. Progress toward healthy people 2000 objectives among U.S. military personnel. Prev Med. 2006;42(5):390---396.

25. Smith TC, Ryan MA, Wingard DL, et al. New onset and persistent symptoms of post-traumatic stress disorder self reported after deployment and combat exposures: prospective population based US military cohort study. BMJ. 2008;336:366---371.

26. Jakupcak M, Tull MT, McDermott MJ, Kaysen D, Hunt S, Simpson T. PTSD symptom clusters in relation- ship to alcohol misuse among Iraq and Afghanistan war veterans seeking post-deployment VA health care. Addict Behav. 2010;35(9):840---843.

27. Wells TS, LeardMann CA, Fortuna SO, et al. A prospective study of depression following combat deployment in support of the wars in Iraq and Afghanistan. Am J Public Health. 2010;100(1):90---99.

28. Lapierre CB, Schwegler AF, LaBauve BJ. Posttrau- matic stress and depression symptoms in soldiers returning from combat operations in Iraq and Afghani- stan. J Trauma Stress. 2007;20(6):933---943.

29. Welcome to DMDC. Available at: https://www.dmdc. osd.mil/appj/dwp/index.jsp. Accessed November 17, 2011.

30. Dillman DA. Mail and Telephone Surveys: The Total Design Method. New York: Wiley; 1978.

31. Fasoli DR, Glickman M, Eisen SV. Predisposing characteristics, enabling resources and need as predictors of utilization and clinical outcomes for veterans receiving mental health services. Med Care. 2010;48:288---295.

32. Eisen SV, Bottonari KA, Glickman ME, et al. The incremental value of self-reported mental health mea- sures in predicting functional outcomes. J Behav Health Serv Res. 2011;38(2):170---190.

33. Ware J Jr, Kosinksi M, Keller SD. A 12-item short- form health survey (SF-12): construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34:220---233.

34. Iqbal SU, Rogers W, Selim A, et al. The Veterans RAND 12-item health survey (VR-12): what it is and how it is used. Available at: http://www.chqoer.research.va. gov/docs/VR12.pdf. Accessed November 17, 2011

35. Eisen SV, Normand SLT, Belanger A, Spiro A 3rd, Esch D. The Revised Behavior and Symptom Identifica- tion Scale (BASIS-24): reliability and validity. Med Care. 2004;42(12):1230---1241.

36. Weathers FW, Huska JA, Keane TM. The PTSD Checklist (PCL). Boston, MA: National Center for PTSD; 1991.

37. Bush K, Kivlahan DR, McDonell MB, Fihn SD, Bradley KA. The AUDIT alcohol consumption questions (AUDIT-C): an effective brief screening test for problem drinking. Ambulatory Care Quality Improvement Project (ACQUIP). Alcohol Use Disorders Identification Test. Arch Intern Med. 1998;158(16):1789---1795.

38. Cocco KM, Carey KB. Psychometric properties of the Drug Abuse Screening Test in outpatients. Psychol Assess. 1998;10(4):408---414.

RESEARCH AND PRACTICE

S72 | Research and Practice | Peer Reviewed | Eisen et al. American Journal of Public Health | Supplement 1, 2012, Vol 102, No. S1

39. Kazis LE, Miller DR, Clark JA, et al. Improving the response choices on the veterans SF-36 health survey role functioning scales: results from the Veterans Health Study. J Ambul Care Manage. 2004;27(3):263---280.

40. Weathers F, Ford J. Psychometric properties of the PTSD checklist (PCL-C, PCL-S, PCL-M, PCL-PR). In: Stamm BH, ed. Measurement of Stress, Trauma, and Adaptation. Lutherville, MD: Sidran Press; 1996:250---251.

41. O’Keefe DJ. Colloquy: should familywise alpha be adjusted? Against familywise alpha adjustment. Hum Commun Res. 2003;29(3):431---447.

42. Rothman KJ. No adjustments are needed for multiple comparisons. Epidemiology. 1990;1(1):43---46.

43. Perneger TV. What’s wrong with Bonferroni adjustments. BMJ. 1998;316(7139):1236---1238.

44. Selim AJ, Rogers W, Fleishman JA, et al. Updated U.S. population standard for the Veterans RAND 12-item Health Survey (VR-12). Qual Life Res. 2009;18(1):43---52.

45. Weathers FW, Litz BT, Herman DS, Huska JA, Keane TM. The PTSD Checklist (PCL): Reliability, Val- idity, and Diagnostic Utility. Paper presented at: Annual Meeting of International Society for Traumatic Stress Studies; October, 1993; San Antonio, TX.

46. Eisen SV, Gerena M, Ranganathan G, Esch D, Idiculla T. Reliability and validity of the BASIS-24 Mental Health Survey for Whites, African-Americans and Lati- nos. J Behav Health Serv Res. 2006;33:304---323.

47. Seal KH, Bertenthal D, Miner CR, Sen S, Marmar C. Bringing the war back home: mental health disorders among 103,788 US veterans returning from Iraq and Afghanistan seen at Department of Veterans Affairs facilities. Arch Int Med. 2007;167:476---482.

48. Ramchand R, Schell TL, Karney BR, Osilla KC, Burns RM, Caldarone LB. Disparate prevalence estimates of PTSD among service members who served in Iraq and Afghanistan: possible explanations. J Trauma Stress. 2010;23:59---68.

49. Kulka RA, Schlenger WE, Fairbank JA, Hough RL. Trauma and the Vietnam War Generation: Report of Findings from the National Vietnam Veterans Readjust- ment Study. New York: Brunner/Mazel; 1990.

50. Gabriel R, Neal LA. Lesson of the week: post- traumatic stress disorder following military combat or peace keeping. BMJ. 2002;324(7333):340---341.

51. Fear NT, Jones M, Murphy D, et al. What are the consequences of deployment to Iraq and Afghanistan on the mental health of the UK armed forces? A cohort study. Lancet. 2010;375(9728):1783---1797.

52. Vogt D, Vaughn R, Glickman ME, et al. Gender differences in combat-related stressors and their associ- ation with postdeployment mental health in a nationally representative sample of US OEF/OIF veterans. J Abnorm Psychol. 2011;120(4):797---806.

53. Rubin DB. Inference and missing data. Biometrika. 1976;63:581---592.

RESEARCH AND PRACTICE

Supplement 1, 2012, Vol 102, No. S1 | American Journal of Public Health Eisen et al. | Peer Reviewed | Research and Practice | S73

Copyright of American Journal of Public Health is the property of American Public Health Association and its

content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's

express written permission. However, users may print, download, or email articles for individual use.