The Influence of Military Service Experiences on Current and Daily Drinking

Abstract Background: An extensive public health literature associates military service with increased alcohol consumption and problematic drinking. However, few well-controlled population-based studies compare alcohol use among nonveterans and veterans with diverse military service experiences, and no such study examines everyday drinking. Methods: We use population-representative data from the 2010 and 2011 Behavioral Risk Factor Surveillance System and distinguish four groups of men: nonveterans; non-combat veterans without a psychiatric disorder (PD) or traumatic brain injury (TBI); combat veterans without a PD or TBI; and veterans (non-combat and combat combined) with a PD and/or TBI. We estimate hierarchical multivariate logistic regression models of current drinking (N = 21,947) and daily drinking (among current drinkers; N = 11,491). Results from supplemental analyses are discussed. Results: Relative to nonveterans, non-combat veterans with no PD or TBI and veterans with a PD and/or TBI, respectively, but not combat veterans with no PD or TBI, are more likely to be current drinkers. Among current drinkers, non-combat and combat veterans with no PD or TBI, respectively, are less likely than nonveterans to be daily drinkers. Conversely, among current drinkers, veterans with a PD and/or TBI are more likely to be daily drinkers than nonveterans, non-combat veterans with no PD or TBI, and combat veterans with no PD or TBI. Conclusion: We document heterogeneous and countervailing influences of military service experiences on current and daily drinking. Results indicate that harmful military service experience may be associated with an increased risk of current, moderate daily drinking, which may represent a form of self-medication.

The U.S. Department of Veterans Affairs (VA) estimates that 4,743,108 veterans with a serviceconnected disability received VA Disability Compensation in 2018. 1 Increased alcohol consumption among veterans is associated with traumatic military service experiences and harms, 2,3 and may contribute to liver disease, social and functional impairments, and veteran suicide. 4 It is estimated that 13.6% of veterans have an alcohol use disorder 4 and many more are heavy or binge drinkers. [2][3][4] While veterans with a service-connected disability have priority access to VA healthcare services and use them to a substantial degree, 5 many veterans live with unmet needs for care.
To date, few studies have compared alcohol use among nonveterans and veterans with diverse military service experiences, and no study has examined everyday (daily) drinking. In our study, we focus on daily drinking because it may represent a form of self-medicating among veterans who were harmed by their military service. The possibility that some veterans self-medicate with alcohol as a substitute for seeking VA services or other professional healthcare is concerning because it may mean that veterans are not getting the healthcare and social supports they need to be well.
We use population-representative data collected from males age 18 and older from eight states that included the Veteran Heath module in the 2010 and 2011 Behavioral Risk Factor Surveillance System. For our analysis, we distinguished four groups who likely had different military service experiences: 1) nonveterans; 2) non-combat veterans without a psychiatric disorder (PD) or traumatic brain injury (TBI); 3) combat veterans without a PD or TBI; and 4) veterans (non-combat and combat combined) with a PD and/or TBI.
Based on health screenings prior to military service, we assume that the group of veterans with a PD and/or TBI includes a disproportionate number of veterans who were harmed by their military service. While controlling for a range of other factors that influence alcohol consumption, we compare the risk of current drinking and daily drinking among current drinkers, across these four groups. We define

KEY FINDINGS
• Veterans with a psychiatric disorder (PD) or traumatic brain injury (TBI) are more likely than nonveterans to be current, daily drinkers. • Non-combat and combat veterans without a PD or TBI are less likely than nonveterans to be current, daily drinkers. • Daily drinking may represent self-medicating among veterans who were harmed by their military service. • Daily self-medicating with alcohol may serve as a substitute for receipt of health services, which may increase some veterans' risk for unmet needs for military service-related health care and alcohol problems. current drinking as drinking at least one alcoholic beverage in the past 30 days. We define daily drinking as drinking at least one alcoholic beverage each day of the past 30 days. A range of supplemental analyses inform and nuance our interpretation of observed patterns.

Differential Influences of Military Service
We found differential influences of military service experiences on current and daily drinking. Overall, non-combat veterans without a PD or TBI and veterans with a PD and/or TBI, respectively, were more likely to be current drinkers (i.e., drank at least one alcoholic beverage in the past 30 days), relative to nonveterans. However, combat veterans without a PD or TBI were similar to nonveterans with respect to current drinking. Not all veterans drink or drink more than nonveterans.
Given our definitions, only current drinkers can be daily drinkers. Thus, daily drinking is examined among the subpopulation of current drinkers. As shown in Figure 1, harmful military service experience (indicated by the presence of a PD and/or TBI) is associated with a higher likelihood of daily alcohol consumption (i.e., drinking at least one alcoholic beverage each day for the past 30 days). Specifically, veterans with a PD and/or TBI are significantly more likely than the other three groups to be current daily drinkers. For example, veterans with a PD and/or TBI are about two times more likely than nonveterans to be current daily drinkers. Interestingly, non-combat veterans and combat veterans without a PD or TBI, respectively, are significantly less likely than nonveterans to be current daily drinkers. Our findings suggest that there is a more complicated relationship between daily alcohol consumption and military service than one might have assumed. Some veterans are more likely than nonveterans to be current daily drinkers, while other veterans are less likely.

Harmful Military Service Experiences Matter
Not all veterans drink alcohol or drink alcohol daily, but some do. Generalizations about drinking behaviors among veterans must account for differences in military service experiences. Specifically, trauma and harm resulting in a psychiatric disorder (PD) and/or traumatic brain injury (TBI) seem to play a critical role in the drinking behaviors of veterans. In our study of current daily drinking, we conducted a supplemental analysis that used a dichotomous (yes or no) indicator of veteran status rather than the four categories used for our main analyses. Contrary to our main findings, these supplemental results showed no significant differences in the current daily drinking habits between veterans and nonveterans. In this supplemental analysis, the experiences of veterans with and veterans without a PD and/or TBI were canceling each other out when combined into one large, undifferentiated group. The use of a dichotomous measure of veteran status leads to the incorrect generalization that there is no difference between the drinking habits of veterans and nonveterans. It obscures the possibility that veterans who were harmed by their service might be drinking daily as a form of selfmedication. It also hides the fact that most veterans (i.e., non-combat veterans and combat veterans without a PD or TBI) are less likely than nonveterans and veterans with a PD and/or TBI to be current daily drinkers. Both of these findings are relevant for policy, practice, and future research.

Recommendations for Policy and/or Practice
Current daily drinking among veterans may be a form of self-medicating due to service-related trauma or harm. However, it may also be a form of self-medicating across a range of contexts that are unrelated to military service. Researchers should more systematically study current daily drinking as a potential form of self-medicating.
Researchers must move beyond dichotomous measures of "ever versus never" served in the military and collect better data on specific military service experiences and their consequences. Only then will we be able to directly evaluate the circumstances under which everyday drinking functions as a form of selfmedicating for veterans. The analyses we present in this paper build upon prior research examining military service experiences and inadequate sleep using the same data and approach. 6 In that study, we documented the same pattern that is depicted in Figure 1. Inadequate sleep was higher among veterans with a PD and/or TBI and lower among non-combat and combat veterans, respectively, without a PD or TBI than among nonveterans. In both studies, we would have missed important policy-and practicerelevant associations between military service experiences and health behaviors if we had used more commonly available measures.

Data and Methods
The Behavioral Risk Factor Surveillance System is an annual, population-based survey of all 50 states, Washington, D.C., Guam, the U.S. Virgin Islands, and Puerto Rico. In each location, the standard core survey is administered to participants along with one or more optional topic modules. We use data from 2010 and 2011 because those were the only years the Veteran Health Module was included. The analytic sample included participants from states that included the Veteran Health Module: Alaska . We focus on men because very few women reported combat exposure. Weighted multivariate logistic regression models were used to generate the estimates reported in this brief. For full methodological details and findings, please see the peer-reviewed publication. 7