Abstract
Background
Alcohol use disorders are a serious public health concern among soldiers. Although deployment-related exposures have been linked with alcohol use disorders in soldiers, less is understood about the link between modifiable, civilian stressors and post-deployment alcohol use disorders.
Purpose
To (1) compare the influence of civilian stressors and deployment-related traumatic events and stressors on post-deployment alcohol use disorders among Army National Guardsmen primarily deployed to Afghanistan and Iraq; and (2) evaluate whether civilian stressors influence a different set of alcohol use disorder phenotypes than deployment-related traumatic events and stressors.
Methods
A cohort of Ohio National Guard soldiers was recruited in 2008–2009 and interviewed three times over 3 years. The analytic sample included Ohio National Guard soldiers who had been deployed by 2008–2009, had participated in at least one follow-up wave, had reported consuming at least one alcoholic drink in their lifetime, and had non-missing data on alcohol use disorders (n=1,095). Analyses were conducted in 2013.
Results
In a model including measures of civilian stressors and deployment-related traumatic events, only civilian stressors (OR=2.07, 95% CI=1.46, 2.94) were associated with subsequent alcohol use disorder. The effects of civilian stressors were only present among people with no history of alcohol use disorder.
Conclusions
Independent of deployment-related exposures, post-deployment civilian stressors are associated with the onset of alcohol use disorder among reserve-component soldiers. Concerted investment to address daily civilian difficulties associated with reintegration into civilian life may be needed to prevent new cases of alcohol use disorders among returning military personnel.
Introduction
Alcohol use disorders (AUDs) are a concern among reserve-component U.S. soldiers returning from deployment. While 6.8% of the U.S. population manifested alcohol abuse or dependence in 2012,1 14% of reserve-component soldiers experienced alcohol abuse.2 To address this public health problem, modifiable determinants of AUDs among reserve-component soldiers must be identified.
Exposure to life-threatening situations during and after combat3–7 and military sexual harassment8,9 are associated with AUDs.6,7,10,11 The role of modifiable civilian stressors has not been investigated and may particularly affect reserve-component soldiers who return to civilian communities and employment upon return from deployment.12,13
This study focused on the Army National Guard and asked two questions. First, what is the relative influence of civilian stressors and deployment-related traumatic events and stressors on post-deployment AUDs among Guardsmen primarily deployed to Afghanistan and Iraq? Second, do civilian stressors differentially influence new onset versus recurrence of AUDs as compared to deployment-related traumatic events and stressors? By investigating the specific impact of civilian versus deployment-related exposures on AUD onset versus recurrent phenotypes, the study aimed to identify targets for intervention.
Methods
Ohio Army National Guard (ONG) members who served in June 2008–February 2009 (N=12,225) were contacted by mail, and a final sample of 2,616 were recruited, with a response rate of 43.2%. Similar to the ONG, the sample was predominantly male and white. The sample was slightly older than the ONG, and approximately half were married (Appendix 1). For this study, respondents were excluded if they had not been deployed by Wave 1 or declined to report deployment status, had no follow-up data, never consumed alcohol, or were missing data on AUD timing (final analytic sample=1,095, Appendix 2).
Respondents were interviewed by telephone in December 2008–November 2009, and twice yearly thereafter. The Case Western Medical Center, University of Toledo, and Columbia University IRBs approved the study.
Measures
AUD was assessed at each wave using the Mini International Neuropsychiatric Interview, following DSM-IV criteria.14 A concurrent reappraisal by clinicians found fair agreement (κ=0.21–0.37), low-moderate sensitivity (0.4–0.6), and high specificity (0.80–0.81).
Civilian stressors related to the most recent deployment were assessed using a 12-item list (e.g., job loss, Appendix 3).15 Respondents were classified as exposed to any versus no stressors.
Baseline measures of deployment-related events included: (1) combat-related traumatic events (e.g., receiving incoming fire); (2) post-battle traumatic events (e.g., seeing severely wounded enemy soldiers after battle); and (3) sexual harassment (e.g., unwanted groping) (Appendix 4).16 Combat-related and post-battle traumatic events were included as tertiles (“low”=0 events, “medium”=1–2 events, and “high”= ≥3 events). Sexual harassment was classified as any versus none.
Other variables considered included age, gender, marital status, race, household income, education, family history of substance use, enlisted status, and AUD in the previous wave.17–19
Statistical Models
A generalized linear mixed model with a random intercept was used to estimate SEs in the presence of repeated assessments over time.20 To identify potential confounders, a model including demographic characteristics, prior AUD, and study wave (Waves 2 and 3 versus Wave 1) was estimated.
The first question asked about the relative influence of civilian stressors and deployment-related traumatic events and stressors on post-deployment AUDs. Stressors and traumatic events were included in separate crude and adjusted models. Adjusted models included significant demographic predictors of AUD, prior AUD, and study wave. A final model included all stressors and traumatic events. Likelihood ratio tests were used to compare model fit.
The second question asked whether civilian stressors influence a different set of AUD phenotypes than deployment-related traumatic events and stressors. Models included the stressor/traumatic event of interest, prior AUD, and the interaction of the two measures as covariates. The same modeling steps were followed as those used to answer the first question.
The Proc Glimmix procedure in SAS, version 9.2 (SAS Institute Inc., Cary NC) was used to fit models. Missing data on the covariates was addressed through listwise deletion (2–4 cases dropped). Analyses were conducted in 2013.
Results
More than half (59.8%) of soldiers experienced combat-related traumatic events, 35.9% of soldiers experienced civilian stressors, and 17.0% reported sexual harassment during the most recent deployment. The past-year AUD prevalence was 13.2%, 7.1%, and 5.2% at Waves 1–3, respectively (sample characteristics listed in Appendix 5).
In separate adjusted models, experiencing at least one civilian stressor and sexual harassment were associated with higher odds of AUD; combat-related and post-battle traumatic events were marginally associated with AUD (Table 1). In a final model, only civilian stressors were associated with AUD (associations between individual items and AUD described in Appendices 6–9).
Table 1.
Deployment-related and civilian events associated with alcohol use disorders among Ohio National Guard members
| Crude modelsa | Single exposure–adjusted modelsb | Fully adjusted modelc | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||
| OR | 95% CI | p-value | OR | 95% CI | p-value* | LRT* | OR | 95% CI | p-value | LRT** | |
| Civilian stressors | <0.01 | <0.01 | <0.01 | <0.01 | 0.40 | ||||||
| None | ref | ref | ref | ||||||||
| Any | 3.81 | 2.40, 6.06 | 2.25 | 1.61, 3.16 | 2.07 | 1.46, 2.94 | |||||
| Peri-deployment events | |||||||||||
| Combat | 0.02 | 0.08 | <0.01 | 0.53 | <0.01 | ||||||
| Low | ref | ref | ref | ||||||||
| Medium | 1.44 | 0.81, 2.53 | 1.27 | 0.83, 1.91 | 1.04 | 0.66, 1.65 | |||||
| High | 2.14 | 1.27–3.61 | 1.56 | 1.07, 2.28 | 1.28 | 0.79, 2.10 | |||||
| Post-battle | 0.01 | 0.08 | <0.01 | 0.88 | <0.01 | ||||||
| Low | ref | ref | ref | ||||||||
| Medium | 1.83 | 1.05, 3.20 | 1.46 | 0.97, 2.18 | 1.11 | 0.70, 1.78 | |||||
| High | 2.04 | 1.19, 3.52 | 1.48 | 1.00, 2.20 | 1.02 | 0.61, 1.69 | |||||
| Sexual harassment | 0.01 | 0.03 | <0.01 | 0.16 | <0.01 | ||||||
| None | ref | ref | ref | ||||||||
| Any | 2.21 | 1.27, 3.85 | 1.58 | 1.04, 2.39 | 1.36 | 0.89, 2.08 | |||||
Note: Boldface indicates statistical significance (p<0.05).
Crude models include only the civilian stressor or peri-deployment traumatic event and no other variables.
Single exposure–adjusted models include a single civilian stressor or peri-deployment traumatic event, plus alcohol use disorder history at the prior wave, study wave, age, gender, and marital status.
The fully adjusted model includes civilian stressor or the peri-deployment traumatic event, alcohol use disorder history at the prior wave, and are adjusted for all other civilian stressors and peri-deployment events, wave, age, gender, and marital status.
p-value for an LRT, which compares a crude model including only the civilian stressor or peri-deployment traumatic event to a model also adjusted for alcohol use disorder history at the prior wave, study wave, age, gender, and marital status.
p-value for an LRT, which compares a single exposure–adjusted model to a fully adjusted model with all stressors and peri-deployment events included.
LRT, likelihood ratio test
Table 2 shows the associations between exposures to civilian stressors and deployment-related traumatic events, as well as the odds of AUD onset and recurrence. In a final model, civilian stressors and high exposure to combat-related traumatic events were associated with AUD onset.
Table 2.
Deployment-related and civilian events associated with alcohol use disorder onset and recurrence, Ohio National Guard members
| Crude modelsa | Single exposure–adjusted modelsb | Fully adjusted modelc | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
|
| |||||||||||
| OR | 95% CI | p-value* | OR | 95% CI | p-value* | LRT** | OR | 95% CI | p-value* | LRT*** | |
| Civilian stressors | 0.03 | 0.03 | <0.01 | 0.03 | 0.02 | ||||||
| For recurrent cases of alcohol use disorder | 1.62 | 1.11, 2.36 | 1.47 | 1.00, 2.17 | 1.52 | 1.02, 2.27 | |||||
| For new onset cases of alcohol use disorder | 3.02 | 1.97, 4.63 | 2.84 | 1.85, 4.37 | 2.41 | 1.54, 3.77 | |||||
| Peri-deployment events | |||||||||||
| Combat | <0.01 | <0.01 | <0.01 | <0.01 | <0.01 | ||||||
| For recurrent cases of alcohol use disorder | |||||||||||
| High combat exposure | 0.99 | 0.64, 1.53 | 0.83 | 0.53, 1.31 | 0.80 | 0.45, 1.44 | |||||
| Medium combat exposure | 0.93 | 0.57, 1.50 | 0.85 | 0.51, 1.40 | 0.79 | 0.47, 1.37 | |||||
| For new onset cases of alcohol use disorder | |||||||||||
| High combat exposure | 2.90 | 1.73, 4.83 | 2.84 | 1.70, 4.74 | 2.08 | 1.09, 3.97 | |||||
| Medium combat exposure | 1.69 | 0.94, 3.02 | 1.86 | 1.04, 3.34 | 1.49 | 0.78, 2.84 | |||||
| Post-battle | 0.04 | 0.02 | <0.01 | 0.03 | <0.01 | ||||||
| For recurrent cases of alcohol use disorder | |||||||||||
| High post-battle exposure | 1.06 | 0.62, 1.79 | 0.86 | 0.50, 1.46 | 0.93 | 0.51, 1.71 | |||||
| Medium post-battle exposure | 1.23 | 0.71, 2.11 | 1.00 | 0.58, 1.73 | 1.02 | 0.58, 1.79 | |||||
| For new onset cases of alcohol use disorder | |||||||||||
| High post-battle exposure | 2.52 | 1.47, 4.33 | 2.51 | 1.46, 4.32 | 1.13 | 0.60, 2.14 | |||||
| Medium post-battle exposure | 1.99 | 1.13, 3.49 | 2.07 | 1.18, 3.63 | 1.22 | 0.67, 2.23 | |||||
Note: Boldface indicates statistical significance.
The crude models include the civilian stressor or the peri-deployment traumatic event, alcohol use disorder history at the prior wave, and their interaction term(s).
The single exposure–adjusted models include the civilian stressor or the peri-deployment traumatic event, alcohol use disorder history at the prior wave and their interaction term(s), and are adjusted for wave, age, gender, and marital status.
The fully adjusted models include civilian stressor or the peri-deployment traumatic event, alcohol use disorder history at the prior wave, and their interaction term(s), and are adjusted for all other civilian stressors and peri-deployment events, wave, age, gender, and marital status.
p-values are for the interaction term(s).
p-values for an LRT, which compares the single exposure interaction–adjusted model to the crude single exposure–interaction model.
p-values for an LRT, which compares the fully adjusted interaction model to the single exposure interaction–adjusted model.
LRT, likelihood ratio test
Discussion
Civilian stressors during deployment, including family disruption, and stressors following deployment, including changes in employment status, problems with health insurance, and legal problems, are commonplace in military families.21 Related characteristics, such as divorce and lower family income have been linked with alcohol misuse in military samples.22,23 Civilian stressors may place an added burden upon military personnel who are already affected by deployment-related traumatic events.24–26
The association of deployment-related traumatic events and civilian stressors with AUDs was specific to AUD onset, suggesting that adult-onset AUDs may constitute a different phenotype from recurrent AUDs.27,28 Previous research has identified two “types” of people with AUD:27–29 one characterized by later onset, greater malleability to the social environment, use of alcohol for its anti-anxiety effects, and lower severity of alcohol dependence symptoms, and another characterized by early onset, high genetic diathesis, use of alcohol for its euphoric effects, and a chronic treatment history.
Study findings must be taken in context of its limitations. First, the results may be subject to recall bias. Participants who report developing alcohol problems may be more likely to remember experiencing deployment-related and civilian traumatic events and stressors.30 However, prospective data on AUDs allows us to temporally separate reports of traumatic events and stressors from reports of AUD. Second, although the study sample was comparable to the ONG population, it was not possible to determine whether the final analytic sample was comparable to the ONG members that would have met our inclusion criteria. Third, soldiers interviewed shortly after their deployment may have been misclassified, as increased rates of psychiatric symptoms tend to arise 3–4 months after deployment.31 Fourth, clinical reappraisal of the AUD measure found fair agreement with the self-report measure, low–moderate sensitivity, and high specificity. Hence, the study may have misclassified a number of Guardsmen as not meeting criteria for AUD. If true, these sources of misclassification would have led to an underestimation of new onset cases of AUD, biasing results toward the null.
Independent of the traumatic events and stressors faced during deployment, soldiers who experience civilian stressors are at greater risk of onset of AUDs. With more than 1.6 million service members deployed in support of war efforts Operation Enduring Freedom, Operation Iraqi Freedom, and Operation New Dawn, the need to develop a more effective reintegration response that addresses such civilian stressors is critical.
Supplementary Material
Acknowledgments
Dr. Cerdá designed the study, interpreted the data, drafted the article, and provided final approval of the version to be published. Ms. Richards analyzed and interpreted the data, contributed to the article draft, and provided final approval of the version to be published. Mr. Cohen assisted with acquisition and analysis of the data, revised the article for important intellectual content, and provided final approval of the version to be published. Dr. Calabrese, Dr. Liberzon, and Dr. Tamburrino conceived and designed the original study, acquired the data, revised the article for important intellectual content, and provided final approval of the version to be published. Dr. Galea conceived and designed the original study, acquired the data, contributed substantially to the manuscript drafts, and provided final approval of the version to be published. Dr. Koenen designed the study, interpreted the data, drafted parts of the article, and provided final approval of the version to be published.
Design and conduct of the original study, data collection, management, and analysis were supported by grants from the Department of Defense: Nos. W81XWH-07-1-0409, W81XWH-08-2-0650, and W81XWH-08-2-0204, as well as a grant from the National Institute of Mental Health: No. MH082729. Analysis and interpretation of the data, as well as preparation, review, and approval of the manuscript were supported by a grant from the National Institute on Drug Abuse: No. DA030449 and a grant from the National Institute on Mental Health: No. MH093612.
Footnotes
No financial disclosures were reported by the authors of this paper.
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
References
- 1.Substance Abuse and Mental Health Services Administration. Results from the 2012 National Survey on Drug Use and Health: summary of national findings. Rockville MD: Substance Abuse and Mental Health Services Administration; 2013. [Google Scholar]
- 2.Riddle JR, Smith TC, Smith B, et al. Millennium cohort: the 2001–2003 baseline prevalence of mental disorders in the U.S. military. J Clin Epidemiol. 2007;60(2):192–201. doi: 10.1016/j.jclinepi.2006.04.008. [DOI] [PubMed] [Google Scholar]
- 3.Wright KM, Foran HM, Wood MD, Eckford RD, McGurk D. Alcohol problems, aggression, and other externalizing behaviors after return from deployment: understanding the role of combat exposure, internalizing symptoms, and social environment. J Clin Psychol. 2012;68(7):782–800. doi: 10.1002/jclp.21864. [DOI] [PubMed] [Google Scholar]
- 4.Wilk JE, Bliese PD, Kim PY, Thomas JL, McGurk D, Hoge CW. Relationship of combat experiences to alcohol misuse among U.S. soldiers returning from the Iraq war. Drug Alcohol Depend. 2010;108(1–2):115–21. doi: 10.1016/j.drugalcdep.2009.12.003. [DOI] [PubMed] [Google Scholar]
- 5.Koenen KC, Lyons MJ, Goldberg J, et al. Co-twin control study of relationships among combat exposure, combat-related PTSD, and other mental disorders. J Trauma Stress. 2003;16(5):433–8. doi: 10.1023/A:1025786925483. [DOI] [PubMed] [Google Scholar]
- 6.Griffith J, West C. The Army National Guard in OIF/OEF: relationships among combat exposure, postdeployment stressors, social support, and risk behaviors. Appl Prev Psychol. 2010;14(1–4):86–94. [Google Scholar]
- 7.Hassija C, Jakupcak M, Maguen S, Shipherd J. The Influence of combat and interpersonal trauma on PTSD, depression, and alcohol misuse in U.S. Gulf War and OEF/OIF women veterans. J Trauma Stress. 2012;25(2):216–9. doi: 10.1002/jts.21686. [DOI] [PubMed] [Google Scholar]
- 8.Gradus JL, Street AE, Kelly K, Stafford J. Sexual harassment experiences and harmful alcohol use in a military sample: differences in gender and the mediating role of depression. J Stud Alcohol Drugs. 2008;69(3):348–51. doi: 10.15288/jsad.2008.69.348. [DOI] [PubMed] [Google Scholar]
- 9.Hankin CS, Skinner KM, Sullivan LM, Miller DR, Frayne S, Tripp TJ. Prevalence of depressive and alcohol abuse symptoms among women VA outpatients who report experiencing sexual assault while in the military. J Trauma Stress. 1999;12(4):601–12. doi: 10.1023/A:1024760900213. [DOI] [PubMed] [Google Scholar]
- 10.Jacobson I, Ryan M, Hooper T, et al. Alcohol use and alcohol-related problems before and after military combat deployment. JAMA. 2008;300(6):663–75. doi: 10.1001/jama.300.6.663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Kehle SM, Ferrier-Auerbach AG, Meis LA, Arbisi PA, Erbes CR, Polusny MA. Predictors of postdeployment alcohol use disorders in National Guard soldiers deployed to Operation Iraqi Freedom. Psychol Addict Behav. 2012;26(1):42–50. doi: 10.1037/a0024663. [DOI] [PubMed] [Google Scholar]
- 12.Thomas JL, Wilk JE, Riviere LA, McGurk D, Castro CA, Hoge CW. 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):614–23. doi: 10.1001/archgenpsychiatry.2010.54. [DOI] [PubMed] [Google Scholar]
- 13.Milliken CS, Auchterlonie JL, Hoge CW. Longitudinal assessment of mental health problems among active and reserve component soldiers returning from the Iraq war. JAMA. 2007;298(18):2141–8. doi: 10.1001/jama.298.18.2141. [DOI] [PubMed] [Google Scholar]
- 14.Sheehan DV, Janavs J, Baker R, et al. MINI— Mini International Neuropsychiatric Interview—English version 5.0. 0—DSM-IV. J Clin Psychiatry. 1998;59:34–57. [PubMed] [Google Scholar]
- 15.Galea S, Tracy M, Norris F, Coffey S. Financial and social circumstances and the incidence and course of PTSD in Mississippi during the first two years after Hurricane Katrina. J Trauma Stress. 2008;21(4):357–68. doi: 10.1002/jts.20355. [DOI] [PubMed] [Google Scholar]
- 16.King LA, King DW, Vogt DS, Knight J, Samper RE. Deployment risk and resilience inventory: a collection of measures for studying deployment-related experiences of military personnel and veterans. Mil Psychol. 2006;18(2):89–120. [Google Scholar]
- 17.Bray RM, Hourani LL. Substance use trends among active duty military personnel: findings from the U.S. Department of Defense Health Related Behavior Surveys, 1980–2005. Addiction. 2007;102(7):1092–101. doi: 10.1111/j.1360-0443.2007.01841.x. [DOI] [PubMed] [Google Scholar]
- 18.Bray RM, Marsden ME, Peterson MR. Standardized comparisons of the use of alcohol, drugs, and cigarettes among military personnel and civilians. Am J Public Health. 1991;81(7):865–9. doi: 10.2105/ajph.81.7.865. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Ferrier-Auerbach AG, Kehle SM, Erbes CR, Arbisi PA, Thuras P, Polusny MA. Predictors of alcohol use prior to deployment in National Guard Soldiers. Addict Behav. 2009;34(8):625–31. doi: 10.1016/j.addbeh.2009.03.027. [DOI] [PubMed] [Google Scholar]
- 20.Vonesh EF, Carter RL. Mixed-effects nonlinear regression for unbalanced repeated measures. Biometrics. 1992;48(1):1–17. [PubMed] [Google Scholar]
- 21.Figley CR. Coping with stressors on the home front. J Soc Issues. 1993;49(4):51–71. [Google Scholar]
- 22.Foran HM, Smith Slep AM, Heyman RE. Hazardous alcohol use among active duty Air Force personnel: identifying unique risk and promotive factors. Psychol Addict Behav. 2011;25(1):28–40. doi: 10.1037/a0020748. [DOI] [PubMed] [Google Scholar]
- 23.Iversen A, Waterdrinker A, Fear N, et al. Factors associated with heavy alcohol consumption in the U.K. armed forces: data from a health survey of Gulf, Bosnia, and era veterans. Mil Med. 2007;172(9):956–61. doi: 10.7205/milmed.172.9.956. [DOI] [PubMed] [Google Scholar]
- 24.Gallo LC, Matthews KA. Understanding the association between socioeconomic status and physical health: do negative emotions play a role? Psychol Bull. 2003;129(1):10–51. doi: 10.1037/0033-2909.129.1.10. [DOI] [PubMed] [Google Scholar]
- 25.Hobfoll SE. Conservation of resources. A new attempt at conceptualizing stress. Am Psychol. 1989;44(3):513–24. doi: 10.1037//0003-066x.44.3.513. [DOI] [PubMed] [Google Scholar]
- 26.Jacobson DE. Types and timing of social support. J Health Soc Behav. 1986;27(3):250–64. [PubMed] [Google Scholar]
- 27.Chen YC, Prescott CA, Walsh D, et al. Different phenotypic and genotypic presentations in alcohol dependence: age at onset matters. J Stud Alcohol Drugs. 2011;72(5):752–62. doi: 10.15288/jsad.2011.72.752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Babor TF, Hofmann M, DelBoca FK, et al. Types of alcoholics, I. Evidence for an empirically derived typology based on indicators of vulnerability and severity. Arch Gen Psychiatry. 1992;49(8):599–608. doi: 10.1001/archpsyc.1992.01820080007002. [DOI] [PubMed] [Google Scholar]
- 29.Cloninger CR. Neurogenetic adaptive mechanisms in alcoholism. Science. 1987;236(4800):410–6. doi: 10.1126/science.2882604. [DOI] [PubMed] [Google Scholar]
- 30.Koenen KC, Stellman SD, Dohrenwend BP, Sommer JF, Jr, Stellman JM. The consistency of combat exposure reporting and course of PTSD in Vietnam War veterans. J Trauma Stress. 2007;20(1):3–13. doi: 10.1002/jts.20191. [DOI] [PubMed] [Google Scholar]
- 31.Bliese PD, Wright KM, Adler AB, Thomas JL, Hoge CW. Timing of post-combat mental health assessments. Psychol Serv. 2007;4:141–8. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.