Accountability for the Psychological Costs of Military Service

The extent to which the military has led developments in mental health is often a forgotten story. ¹ A current example is the series of papers ^{2 –6} in this edition that analyze the 2002 and 2013 studies of the Canadian Armed Forces (CAF). Here, epidemiology is used to provide answers to complex questions about the manifestations of psychiatric disorders and their changed patterns over time. These findings need to be considered in the context of a responsive series of service delivery innovations made by the CAF that would seldom occur in the civilian mental health systems. Such research is extremely difficult to do in civilian settings because of the fragmented management and funding streams and the lack of accountability that public scrutiny demands of the military.

This research was only possible because the stability of a consensus research design using valid and reliable measures over a decade allowed these longitudinal comparisons of CAF and civilian populations. ³ This is an important matter in the light of the recent changes in the Diagnostic and Statistical Manual of Mental Disorders (DSM), Fifth Edition, diagnostic criteria for posttraumatic stress disorder (PTSD) whose application will disrupt similar longitudinal programs in the future. Further debate is required about these changes as the validity of changed criteria in DSM-5 have been questioned, particularly in military populations due to the diagnostic discordance with DSM-IV criteria. ⁷

A key finding between the CAF in 2002 and 2013 was the significant increase in the past-year prevalence of PTSD, generalized anxiety disorder (GAD), and panic disorder. ⁴ This appears to have been due to the burden of the Afghanistan deployments and despite the major service improvements put in place by the CAF in the interim. ² Evidence about the impact of these service improvements came from better reported treatment experiences in the CAF in 2013, compared with 2002. ⁵ The finding of a greater prevalence in 2013 of major depressive disorder, GAD, alcohol dependence, and suicidal attempts, compared with the general population, ³ may reflect the moral and ethical burdens associated with military service. These not only arise in combat, ⁸ but also occur on disaster and humanitarian missions, as this kind of exposure to human suffering is not common in civilian life. ⁹

The articles in this issue highlight the strengths of epidemiology to provide insight into the impact of war in a representative sample of an entire armed force. Larson et al ¹⁰ used a random representative sampling process in contrast to a range of other studies that have examined specific cohorts of military personnel, where different sampling methods have contributed to debate about differences in mental health outcomes of Iraq deployments between the United States and the United Kingdom. ¹¹ Despite the improved mental health services in the CAF, the findings of the higher rates of disorder in people deployed to Afghanistan is an important finding informing the debate about the consequence of combat exposure. The increased rates of disorder between 2002 and 2013, the time that covered the CAF’s deployments to Afghanistan, were related to increased combat exposure, in the context of the reported increased prevalence of general exposure to traumatic stress across this decade of study. ⁴ The increased incidence of panic disorder is of particular interest as this may represent subsyndromal PTSD where the pattern of distress on exposure to triggers is present without avoidance and numbing required to get a PTSD diagnosis. This observation supports the use of a subsyndromal PTSD diagnosis, as panic disorder is probably indicative of a significant future risk of PTSD.

The analyses about the risk modelling of the vulnerability factor of child abuse that antecede military service, in contrast to the impact of Afghanistan deployment, are challenging. ⁶ The greater overall attributed risk to child abuse was significantly greater for any disorder (population attributable fraction [PAF] = 29.6%) than the Afghanistan mission (PAF = 9.6%). ⁶ In contrast to the risks of any disorder, Afghanistan deployment contributed 31.5% to the risk for PTSD, a substantially greater effect than for depression, ⁵ highlighting that the population attributable risk of combat varies significantly between different disorders. Further, these findings need to be contrasted to another study ¹² of deployed CAF personnel where clinicians judged that 71% of people in care had a disorder that was attributed to the mission. Hence, there is a caution about interpreting epidemiological modelling in making causal decisions in clinical settings, such as pension applications, because of the need to account for the role of the necessary proximal exposures in the onset of disorder, in contrast to the contribution of background risk factors. ¹²

The comparison of the CAF with the general population between 2002 and 2013 highlights the complexity of interpreting epidemiological findings. ³ Military populations should have better health outcomes despite the greater rates of child abuse found in people in the CAF ³ due to the selection of a healthy work force. For example, in Australia during the Vietnam War, more than 50% of the possible conscripts were rejected on health grounds. As a consequence, Australian Vietnam veterans, despite the increases in illness associated with combat exposure, ¹³ more than 30 years after the end of the Vietnam War, this cohort had a 6% lower mortality, compared with the Australian population from which they were drawn. ¹⁴ Hence, the comparisons with the Canadian community of the CAF in 2002 and 2013 are likely to represent an underestimate of the burden of military service due to the healthy warrior effect. Those remaining in the military are the healthy survivors of the system with the injured and ill likely to being discharged and replaced by healthy recruits—factors that should create a bias to better mental health of current CAF members. In contrast, the comparative Canadian community, apart from immigration and death, is a stable population between 2002 to 2013. These differences need to be highlighted to prevent the minimization of the psychological cost of military service, in public discussion. At times, media concerns about the rates of suicide and psychiatric disorder in military populations are counteracted with the response that the rates are similar to the general population.

The longevity of service of people in the CAF, highlighted by the median duration since deployment to Afghanistan, was 5 years. ⁶ It is probable that those who had disabling psychological injuries in the aftermath of combat have been discharged. Hence, the true cost of deployment to Afghanistan is not characterized by these studies as they do not include the entire cohort of the people who were deployed to Afghanistan, excluding potentially the most at-risk group. A further factor that may minimize the psychological impact of combat is that combat can have a significant delay in the emergence of disorder, and the long-term costs of the Afghanistan deployments are as yet to be fully manifested. ¹⁵

The planning, execution, and track record of publication of the military mental health outcomes from the CAF is exemplary. The research infrastructure that has allowed a longitudinal program of systematic analysis of the previous surveys ² has depended on a well-coordinated research program that is continuously funded within the military. It highlights the importance of military organizations being open to audit and scrutiny to counter the, at times sensational, media accounts about the costs of military service. The excellent response rates in these studies demonstrate how adequate funding that ensures participant tracking and uses face-to-face interviews is necessary for optimal research outcomes.

Finally, epidemiological studies need to be integrated with studies of neurobiology that help define the underpinnings of the adverse impacts of the exposures associated with military service. For example, the known impact of prior trauma exposure on glucocorticoid receptor expression could, in part, account for the mechanism of how child abuse and victimization may modify the risk of developing PTSD in combat. ¹⁶ These findings and the approach this body of work demonstrates, need to be applied to the emergency services that have not been studied with the same rigour. These people are also put at risk protecting the community and need similar proactive service as those provided to the military. The research program and clinical developments in the CAF exemplified how military psychiatry continues to lead developments in mental health.