Abstract
Background: Earlier studies indicating that operational deployment affects health behaviours among military personnel and veterans generally lacked final conclusiveness due to cross-sectional or retrospective design.
Aim: The aim of this study is to review longitudinal studies investigating whether military service, in particular operational deployment, affects health behaviours, specifically alcohol misuse, smoking, eating disorders and obesity.
Methods: A MEDLINE database search was performed, using relevant keywords and MESH terms. The US Millennium Cohort study website was used to obtain the list of relevant publications. Only studies with prospective longitudinal cohort design, conducted on military or veteran populations of developed countries serving after the Vietnam War and investigating health behaviours and health markers such as excessive drinking, smoking, disordered eating and body weight were included.
Results: Six studies fulfilled the inclusion criteria, three that resulted from the US Millennium Cohort study and three that investigated other military populations.
Deployment with combat exposure was the most significant factor affecting health behaviours of military personnel.
Excessive drinking among US military personnel increased significantly in those deployed that were exposed to combat, especially among Reserve and National Guard members and in the youngest age groups, but was not affected by deployment without combat exposure. Among British military personnel, total alcohol consumption increased with time, was higher for those deployed compared to non-deployed, and highest in those who experienced war related stress.
Smoking in the US military increased among those deployed, particularly among those with prolonged and multiple deployments or with combat exposure. Among British military personnel, smoking rates declined.
Body weight increased for the majority of US military personnel, but disordered eating was reported only among deployed women with combat exposure. An increase in body weight was also reported in the Belgian military.
Conclusion: Generally, it was combat exposure, not deployment in general, that had affected health behaviours. As hazardous health behaviours may affect negatively on physical readiness and re-deployability of military personnel, preventive measures should concentrate on those subgroups that are most vulnerable.
Keywords: Military personnel, veterans, deployment, longitudinal study, health behaviours
Conflict of interest: The authors declare no conflict of interest
Introduction
Recent studies present compelling evidence that military deployment with combat exposure negatively affects the mental health of deployed personnel, especially the incidence of post-traumatic stress disorder (PTSD) 1. There is also a strong indication that combat exposure affects health behaviours 2, 3 , which will in turn affect physical readiness, health and health care needs in the future.
Alcohol abuse, increased smoking and disordered eating often serve as maladaptive coping mechanisms after traumatic events. Alcohol abuse is highly correlated with PTSD and other psychological disorders that may occur after stressful and traumatic events, such as those associated with war 2, 3. The rate of heavy drinking is estimated at 20% in the general US military population, but can be as high as 30% among younger groups (aged 18 – 25 years old) 4. Recent cross-sectional studies report an increased rate of excessive drinking and associated disorders in those returning from deployment 1.
The excessive use of alcohol has many implications on public health. Globally, nearly 4% of all deaths are related to excessive alcohol consumption, and can reach 9% among younger groups (aged 15-29 years old). Most alcohol-related deaths result from injuries, cancer, cardiovascular diseases and liver cirrhosis 5. The family and social consequences of excessive drinking are also extensive 6.
Cross-sectional studies found increased post-deployment smoking rates among previous non smokers and increased daily cigarette intake among smokers 7. This finding is important as smoking is the leading preventable cause of death and the long-term damaging health consequences associated with this health behaviour are well established 8, 9. In the US, more than 400,000 people die each year due to smoking, with $167 billion spent in annual health-related economic losses 10. In Australia, smoking was the single risk factor responsible for the greatest disease burden, around 12% of the total burden in Australian males. Alcohol use, physical inactivity and obesity are responsible for a further 6.6%, 6% and 4% of the burden of disease, respectively 11.
Although these data relate to the civilian, not military Australian populations, they emphasise the importance of health behaviours in determining physical readiness, re-deployability and future health care needs in the military and veteran populations.
Epidemiological studies show an increasing rate of obesity among whole populations including young people 12. The obesity epidemic has serious implications for recruitment and retention of Defence Force personnel 13 and places an additional strain on the health care needs of veterans and their families 14.
Recently, an increasing number of Australian Defence Force members have been involved in multiple deployments, with Australian troops taking part in conflicts in Iraq and Afghanistan and playing a significant peacekeeping role in the Pacific region. Consequently, increasing emphasis has been placed on investigating the effects of these deployments on the health of military personnel. Military forces have certain health and fitness standards aimed at selecting and maintaining individuals that are best suited to the physical demands of military service. Any knowledge regarding the effects of military deployment and military specific exposures could therefore allow for better preparation for the ensuing consequences of deployment.
A systematic review of prospective longitudinal cohort studies performed in the military was undertaken to investigate the often raised question of whether military service, in particular operational deployment, results in a higher risk of chronic illness among military personnel and veterans. This wider review, performed for the Centre for Military and Veterans’ Health investigated mental 15 and physical health outcomes.
The aim of this study is to review longitudinal studies investigating whether military service, in particular operational deployment, affects health behaviours, specifically alcohol misuse, smoking, eating disorders and obesity.
Methods
The MEDLINE database was searched using relevant keywords and MESH terms. Three separate searches were performed. In search 1, the relevant studies were retrieved from the US Millennium Cohort study website. Search 2 combined terms: Military Personnel (MESH), longitudinal study (MESH and keyword) and Health (MESH and keyword). Search 3 combined terms: Veterans(MESH), longitudinal study (MESH and keyword) and Health (MESH and keyword). The terms of the search were purposely broad and not restricted by the outcome, due to the search being performed for a wider CMVH review, which also investigated mental 15 and physical health outcomes. The references were downloaded to an EndNote library and duplicates were removed. The search was performed in July 2010.
Study design. Only longitudinal prospective cohort studies were included. Retrospective longitudinal studies and longitudinal panel studies were excluded. There are distinct benefits of prospective longitudinal cohort studies over cross-sectional and retrospective studies. Prospective longitudinal studies can distinguish between short-term and long-term phenomena, contribute to establishing causative associations between exposure and outcome, and minimise recall and selection biases that are often influenced by exposure and/or health outcome.
Outcomes. To be included in the present review, the studies had to investigate health behaviours such as alcohol drinking, smoking, disturbed eating patterns and health markers such as body weight in military or veteran populations.
Cohorts. At inception, the included cohort had to be in active military service after the Vietnam War.
The references were assessed for relevance, based on the examination of titles and abstracts. The flow of citations examined in the course of this review has been presented in Figure 1.
The quality of the studies was assessed using study design checklists based on recommendations by Sanderson et al. for the appraisal of observational studies 16. Quality criteria included cohort size, sample selection, follow-up rate and duration, outcome and exposure measurement bias, type of analysis, clarity of the results and adjustment for confounders. Each of the seven appraisal questions was assigned a score of 2 and 0 for “YES” and “NO” answers and 1 point for “Can’t tell or mixed answer”. Only cohorts above 1000 participants were assigned points for the size. The maximum number of points in the appraisal score was 14. Studies with scores of 13-14, 10-12 and 7-9 were considered to be, respectively, of very good, good and moderate quality, and those below 7 points, of low quality.
Results
There were 6 studies that fulfilled the inclusion criteria. Four studies investigated drinking and smoking behaviour 17-20, and 2 studies investigated body weight and disturbed eating patterns 21, 22. Three studies resulted from the US Millennium Cohort study 17, 18, 21 and three studies investigated other military populations 19, 20, 22.
The main results of the included papers are presented in the text below. Additional details are presented in Table 1.
The Jacobson et al. (2008) US Millenium Cohort study investigated whether deployment with combat exposures was associated with changes in drinking behaviour 17. Drinking was estimated at baseline and follow-up from the number of drinks consumed on each day of the week before completing the questionnaire. Heavy drinking for men was defined as more than 14 drinks consumed per week and binge drinking as 5 or more drinks consumed per drinking occasion; for women 7 drinks per week constituted heavy drinking and 4 per drinking occasion, binge drinking. The rates of new-onset of heavy weekly drinking, binge drinking and alcohol-related problems were compared between deployed with and without combat exposures and non-deployed personnel. The comparisons were performed separately for Reserve, National Guard and active duty personnel. There were no differences between drinking behaviour of those deployed without combat exposure and non-deployed. Significant increases of new onset for all three outcomes of drinking behaviour was seen in deployed with combat exposure compared to non-deployed. However, these outcomes differed between service components. Combat exposed members of Reserve and National Guard members showed increases on all three measures, while for active duty personnel, only binge drinking was increased. The youngest members of the cohort (those born in 1980 and later) were at the highest risk for all alcohol-related outcomes compared with more mature age groups. These findings suggest that Reserve and National Guard personnel and younger service members who deploy with reported combat exposures are at increased risk of new-onset heavy weekly drinking, binge drinking, and alcohol-related problems.
The Smith et al. (2008) US Millenium Cohort study investigated three measures of smoking behaviour, namely new smoking among never-smokers, smoking recidivism among past smokers and change in daily smoking among smokers in relation to military deployment 18. The rates of all three measures increased in those deployed compared to non-deployed. Among those deployed, those with combat exposures were at greater risk of initiating or resuming smoking compared to those without combat exposures. Deployment longer than 9 months and multiple deployments were independently associated with post-deployment smoking recidivism. Among those who smoked at baseline, deployment was not associated with changes in daily amount smoked.
Drinking and smoking behaviour in relation to deployment was also investigated in a study of almost 1000 British military personnel 19. Cigarette smoking declined during the three years of the study for the total sample and there was no evidence that deployment and combat exposures were associated with a change in the number of cigarettes smoked. However, there was an increase in total alcohol consumption and in the prevalence of binge drinking. Binge drinking in the UK study 19 was defined as 7 or more drinks per occasion, twice per week, and 5 or more drinks per occasion once per week in the US study 17. The increase in both outcomes was greater in those deployed compared to non-deployed, in particular in those who thought they might be killed or who experienced hostility from civilians and were most recently deployed.
In the study of about 700 Swedish military conscripts followed up for 30 years of civilian life, mental health at young age was predictive of smoking status at middle age 20. In particular, smoking cessation at 30–50 years of age was associated with indicators of mental well-being measured at 18 years of age. Approximately half of the young smokers had quit by the time they were resurveyed in middle age. Persistent heavy smokers were more likely to have indicators of poor mental health measured at a young age than non-smokers or quitters. Men who would subsequently be successful at smoking cessation reported better mental health and a lower prevalence of childhood mental health indicators at age 18 than persistent heavy smokers.
Disordered eating and changes in body weight were investigated in male and female participants of the Millennium Cohort study 21. Deployment was not significantly associated with new-onset disordered eating in women or men. Among those deployed, women reporting combat exposures were more likely to report new-onset disordered eating or to lose 10% or more of their body weight compared with deployed women without combat exposures. Among deployed women, combat exposures but not deployment itself represents a risk factor for developing eating problems and weight loss.
In Belgium peacekeepers followed for 14 years, median BMI increased by about 1 BMI unit (kg/m2), and there was an increase in the prevalence of excessive body weight and obesity from age 40 years or more 22.
All of the included studies were of adequate quality, with three US Millennium Cohort studies evaluated as very high quality (see Table 1).
Discussion
Excessive alcohol use has been a problem for many members of the Defence Forces over the years. Cross-sectional studies in the US and the UK have shown that baseline levels of drinking in the Armed Forces were higher than in the general population 23, 24. Increased alcohol use has been suggested as one possible explanation for previously unexplained increases in injury mortality subsequent to deployment 25. The Jacobson study found a significantly increased risk for new-onset heavy weekly drinking, binge drinking, and other alcohol related problems among Reserve/ Guard personnel deployed with reported combat exposures compared with non-deployed Reserve/Guard personnel. Interestingly, in the active duty personnel, only binge drinking was significantly increased. As a possible explanation for this discrepancy, authors cite inadequate training and preparation of civilian soldiers for the added stresses of combat exposures faced during deployment; increased stress among individuals and their families having to transition between military and civilian occupational settings; military unit cohesiveness; and reduced access to support services, including family services, health and physical fitness programs, and ongoing prevention programs in civilian communities 17. These explanations are in agreement with results of cross-sectional study that assessed the occupational factors and deployment experiences associated with heavy drinking in regular UK servicemen deployed to Iraq 26. Personnel whose role in theatre was outside their training or experience, and who experienced poor in-theatre unit leadership were more likely to be heavy drinkers. The results of the Jacobson study 17 clearly indicate that it was combat exposure, not a deployment itself that resulted in the increased alcohol use among US troops.
Among UK military personnel 19, the total alcohol consumption was greater in those deployed compared to non-deployed. Although there were methodological differences in the criteria for assessing binge drinking, in agreement with the results of the US Millennium cohort study, the greatest increase of drinking was seen among those exposed to war-related stress.
It is unclear whether these patterns of drinking will continue in the future. These questions will be answered when the US Millennium cohort study publishes the results of the third wave of assessments. However, it can be speculated that the patterns of drinking will be inter-related with post-deployment experiences and personal susceptibilities to the disorder. Such a finding was reported in a review of retrospective studies of alcohol use in the UK Armed Forces, where heavy drinking was associated not only with deployment stress (being deployed to Bosnia), but also with family circumstances (being unmarried/separated/divorced) and personal characteristics (poorer subjective physical and mental health) 27. The findings of all studies on alcohol imply that preventive measures should concentrate on those subgroups that are most vulnerable to adverse health behaviours.
In the US, the smoking rate among military personnel is much higher than in the civilian population. In 2005, the rate of smoking among military personnel was 32% compared to 21% in the civilian population, and the rate was found to be increasing in recent years 28. The increasing rates are explained by recent findings of the US Millennium Cohort study showing that deployment is associated with smoking initiation and, more strongly, with smoking recidivism, particularly among those with prolonged deployments, multiple deployments, or combat exposures 18.
Recent smoking trends among the UK military were different from those observed in the US forces. The Hooper study 19 showed that smoking was less prevalent in the UK Armed Forces than in the general population. Additionally, cigarette smoking rates declined during the three years of the study and there was no evidence that deployment and combat exposures were associated with a change in the number of cigarettes smoked. This was unexpected, as lower ranks are recruited from lower socioeconomic groups, and a military environment has been thought to encourage smoking 29. The differences in smoking behaviours in UK and US military forces may be related to cultural differences between the countries and to differences in the terms of deployment. In the US, the antismoking campaign is undermined by a pricing policy that allows discounted tobacco products to be sold in the military commissaries 30. The standard US Army deployment to Iraq and Afghanistan is longer than the comparable UK deployment, 12 and 6 months “boots on the ground”, respectively 31, 32. The rates of PTSD among US deployed personnel 33 are higher compared to UK deployed personnel 34. As smoking behaviour is co-morbid with PTSD, it may result in a decreased need for this maladaptive coping behaviour among UK military personnel compared to US military.
Civilian studies indicated that smoking behaviour was found to be related to mental well-being and personality traits. In Swedish conscripts followed up for 30 years of civilian life, mental health at a young age was predictive of smoking status at middle age 20. Although all participants in this study started up as military conscripts, after finishing their 2-year long military service they exited to the civilian life for the rest of the follow up period. In the New Zealand Dunedin birth cohort followed up for 24 years, negative emotionality and personality trends, such as higher aggression and alienation measured at earlier age predicted smoking at later assessment, while higher self-control and traditionalism predicted non-smoking later 35. It should be noted that it is only by inference that we assume that psychological factors that affect behaviour are similar in both civilian and military populations.
The US Millennium cohort study 21 did not find disordered eating or changes in body weight in the general population of deployed women or men. However, women exposed to combat were more likely to report new-onset disordered eating or to lose a significant amount of weight compared to deployed women without combat exposures. Among deployed women, combat exposures but nor deployment itself represents a risk factor for developing eating problems and weight loss. As these findings do not apply to the male group of US military personnel, it appears that disordered eating and changes in body weight are generally more common among female military personnel. This study is in an agreement with civilian studies that report a higher proportion of eating disorders in women compared to men. Based on the US National Survey sample of over 9000 adults, lifetime prevalence estimates of anorexia nervosa, bulimia nervosa, and binge eating disorder are 0.9%, 1.5% and 3.5% among women, and 0.3% ,0.5% and 2.0% among men 36. While anorexia and bulimia are 3 times less frequent in men, binge eating is not only more prevalent for both sexes but the rate of overeating among men is comparable to the rate among women.
The US Millennium cohort study of Jacobson et al. (2009) 21 uses self-reported data on weight and height. Other studies have shown that the overweight but otherwise healthy survey participants have a tendency to slightly overestimate their height and underestimate their weight 37 38. Participants with an eating disorder appear to be more accurate in reporting their weight 39 40, but still have a tendency to minimise their weight problems. Thus, the use of self-reported data in the Jacobson et al. (2009) study could potentially skew the data toward slight under-reporting of the problem rather then over-reporting; however, in such a large sample it should not significantly affect evaluation of changes over time in the development of new-onset disordered eating.
In the whole sample of the US Millennium Cohort study 21, the average weight gain in men and women between the baseline and follow-up was 2.1 kg and 2.7 kg, respectively, which represents a 2.6% increase among men and a 4.1% increase among women. Only approximately 33% of women and 48% of men reported maintaining a stable weight between baseline and follow-up, regardless of deployment status. It is not clear whether this trend will continue at the present pace. For example, in male Belgian military personnel, who were deployed to at least one peacekeeper mission, the median body weight increased at a much slower rate, by about 1 BMI unit (kg/m2) in total during the 14 years of follow up 22. This is in agreement with comparative data on international obesity rates, which indicate there is a much higher proportion of obese adults in the US, the UK and Australia compared to continental Europe (US: 30.6%, UK: 23%, Australia: 21.7%, Belgium: 11.7% in 2002) 41. As obesity is a known risk for many diseases, the strain on healthcare in the military may increase in parallel to that observed in the general population.
A recent systematic review of obesity in military populations found a general lack of information available to address the issues of obesity and subsequent health in military personnel, obesity status and work performance (absenteeism and discharge), and obesity status and physical performance 13. The authors concluded that it is not currently possible to report on the implications of obesity for recruitment, training and workforce maintenance in the military. The limited available data suggested that excessive body weight and obesity per se are not necessarily reasons for preclusion from military service, but efforts are required to evaluate the extent of the current and future risk for chronic obesity-related disease and to assess physical fitness 13.
Limitations of the review
This review is limited by the low number of longitudinal studies investigating health behaviours in the military. Although results and conclusions drawn from the US Millennium Cohort studies represent a high level of evidence, the study is ongoing and the long term effects from the third wave of assessments are yet to be published. Additionally, direct generalisation of results from the US Millennium Cohort to military populations of other countries may be limited by cultural and national differences and different terms of deployment.
Key points/Summary conclusions
- The key finding from these studies was that it was combat exposure, not deployment in general that was the most significant factor affecting health behaviours of military personnel.
- Alcohol drinking did not increase in the whole sample of deployed US military personnel, but drinking increased in those with combat exposure, especially among Reserve and National Guard and the youngest age groups. Among British military personnel, alcohol consumption increased in those deployed compared to non-deployed, especially among those who experienced war related stress.
- Smoking was significantly increased by deployment, particularly prolonged deployments or multiple deployments and combat exposures in the US military, but not in the British military, where smoking declined with time.
- A trend towards increased body weight was seen in the US and Belgian military, but disordered eating was seen only among deployed women with combat exposure.
- Hazardous health behaviours may have a negative effect on physical readiness and re-deployability of military personnel and increase future health care needs in the military and veteran populations.
- Preventive measures should concentrate on those subgroups that are most vulnerable to adverse health behaviours, especially those that were exposed to combat.