Correlates of Mental Health Problems in Canadian Armed Forces Veterans – 2013 Life After Service Survey

Mental Health Findings from the 2013 Life After Service Survey

Executive Summary

Citation: Thompson JM, Sweet J, VanTil L, Poirier A, MacKinnon K. Correlates of Mental Health Problems in Canadian Armed Forces Veterans – 2013 Life After Service Survey. Charlottetown PE: Research Directorate, Veterans Affairs Canada. Research Directorate Technical Report. Charlottetown. 14 September 2016.

This document is the third and last in a series initiated in 2014 to comprehensively analyze the mental health findings from the 2013 Life After Service Studies survey (LASS 2013). The LASS 2010 and 2013 surveys were the first in-depth looks of the well-being of CAF Veterans (former CAF members) living in the general Canadian population. The surveys used large samples of Veterans who were released from service from 1998 and identified using the Department of National Defence Human Resources Database. The LASS 2010 survey included Regular Force (RegF) Veterans who were released in 1998-2007, and the LASS 2013 survey included RegF Veterans who were released in 1998-2012 and Primary Reserve Force (ResF) Veterans who were released in 2003-2012. The goal of this body of work is to provide evidence to inform policy, programming and services aimed at enhancing and maintaining the well-being of CAF Veterans.

The first of the three reports from this project covered findings from descriptive analyses of the mental health findings. That report was an in-depth analysis of the prevalence of mental health indicators and identification of characteristics of subgroups in whom the unadjusted odds of self-reported diagnosed mental health conditions were higher (Thompson et al. 2015).

The second report described a method for combining three brief mental health measures used in LASS 2013 into a composite measure of “mental health problems” (MHPs): self-reported diagnosed conditions, the K10 measure of psychological distress and the PC-PTSD primary care screener for posttraumatic stress disorder (Thompson et al. manuscript in submission).

This third and final report completes the 2014 plan for the analysis of LASS 2013 mental health data by reporting on multivariable regression analysis of the mental health problems findings in LASS 2013 using the new composite MHPs measure. The objectives were to (1) assess the relative importance of factors associated with mental health problems in recent-era CAF Veterans using multivariable regression modelling; (2) assess the association of MHPs with difficult adjustment to civilian life adjusting for factors at the time of the survey using multivariable regression modelling; and (3) identify implications for policy, programming, services and further research.

The conceptual framework for this project is based on the domains of global well-being concept (Thompson et al. 2013). Global well-being is measured subjectively and objectively across the seven key domains of employment or other meaningful activity, finances, health, life skill/preparedness, social integration, housing and physical environment, and cultural and social environment. Global well-being in each domain fluctuates across the life course in response to determinants from all the domains, including policies, programs and services. The theory states that well-being has “multidimensional” influences. For example, that well-being in terms of health is influenced by employment, finances, social support, life skills, physical environment and social environment. Similarly, that well-being in terms of employment or other meaningful activity is influenced by health and the other well-being domains.

Main Findings and Implications

The table below summarizes main findings and related implications in two areas: policy, programs and services on the one hand, and further research on the other. There are implications for both population health and individual care. The findings in this report inform further analysis of the LASS 2013 survey data to provide greater insight into the mental health and well-being of CAF Veterans and inform researchers planning new studies to close gaps in knowledge of mental health in Veterans.

The findings cluster into key themes:

  • Well-being multidimensionality. The univariable and multivariable analyses confirm that mental health problems and difficulty in adjustment to civilian life are associated with multiple, inter-related domains of well-being. A key implication for policy, programs and services is that attention to multiple domains is necessary rather than focus on only one at a time.
  • Avoiding siloes. This observation comes from finding associations between determinants of well-being, with two key implications: (1) addressing both physical health and mental health rather than one or the other, and (2) focusing policy and program development more inclusively than only one domain of well-being without considering interactions with other domains.
  • Screening. The analyses identified risk indicators marking subgroups in whom mental health problems or difficult adjustment to civilian life are more common. These findings can be used to refine screening systems.
  • Targeting limited resources. The risk indicators can be used to target limited mental health resources to subgroups with the greatest needs.
  • Understanding mental health problems in CAF Veterans. The findings will be of value to policy and program developers and service providers by providing insights into mental health problems in CAF Veterans who were released from service since 1998.
  • Paradigm shift from “medical release”. Self-reported difficult adjustment to civilian life was strongly associated with mental health problems regardless of the type of release from service. This important finding shifted focus from only the medically released to more broadly including all with mental health problems, regardless of type of release.
Implications of mental health findings from LASS 2013 for Regular Force and deployed Reserve Force Veterans who released since 1998.
Finding Implications for Policy, Programs, Services Implications for Research
While a majority of these Veterans had no significant mental health problems (symptoms or diagnosed conditions), 39% did. About a fifth (22%) had mild to moderate mental health problems and about an eighth (16%) had severe ones. Supports provision of a hierarchy of services to meet the broad range of needs across degrees of severity of mental health problems in the Veteran population. There is need for research on the longitudinal natural history of mental health problems across the transition to life after service.
The 39% prevalence of mental health problems is larger than the previously reported 24% with diagnosed mental health conditions. Informs and supports the ongoing development of mental health services and well-being supports for CAF Veterans. The use of brief symptom measures together with measures of diagnosed conditions provides clearer picture of mental health problems.
Mental health problems were associated with a variety of socioeconomic, military, health and disability characteristics. Supports the multidimensional provision of services and VAC’s domains of well-being approach. Need for further research on the determinants of mental health during adjustment to civilian life.
Mental health problems (diagnosed conditions and subthreshold or undiagnosed symptoms) were not associated with male or female sex but, in prior analyses, the odds of self-reported diagnosed mental health conditions were 55% greater in women than men (Thompson et al. 2015). Together with research in civilian populations, it is possible that male Veterans might be less likely to seek diagnosis and treatment for their mental health symptoms than females. Further work is need on gender differences in the determinants of mental health problems and treatment-seeking.
Mental health problems were most likely in the middle age groups centered 40-49 years in all regression models, as in analyses of other LASS well-being indicators including difficult adjustment to civilian life, disability and suicidal ideation. Points to the need for programs targeting mental health in middle-aged Veterans. At this age, Veterans released from service in mid-career are dealing with multiple stressors in multiple well-being domains: employment, finances, raising families and emerging chronic physical health conditions. Additional work is needed on the determinants of mental health problems in these working-age Veterans during adjustment to civilian life, and impacts of MHPs in other domains of their well-being.
Chronic physical health conditions and chronic pain were consistently associated with mental health problems. Furthermore, 84% of those with mental health problems and 90% of those with diagnosed mental health conditions also had chronic physical health conditions. Reinforces (again) the importance of breaking down siloes between mental and physical health care and attending to both when they co-occur. There are important knowledge gaps in the complex causality issues for co-occurring physical and mental health problems. Longitudinal research studies are required that follow CAF members through MCT into later life.
Chronic pain was associated with mental health problems independently of chronic physical conditions, although the strength of association attenuated when adjusting for socioeconomic factors and physical conditions. Chronic pain in these Veterans has complex associations with factors in multiple domains of well-being, including both physical and mental health problems. Requires treatment and recovery supports from multiple domains. There are complex, incompletely understood causality factors for chronic pain, and more work is required to find effective and safe treatments.
Most releasing CAF members are of working age and seek employment in the workforce. There was a strong unadjusted association between not being employed and having mental health problems. Chronic mental and physical health problems and related impairments play roles in unemployment and precarious employment and need to be considered in programming to support well-being in the employment domain. Causality can be bidirectional: having MHPs can make it difficult for a person to get or keep a job, but difficulty getting or keeping a job can exacerbate or trigger MHPs.
The association between low income and mental health problems persisted in the models until the life stress, life satisfaction, mastery and social support variables were added. Hypothesis: Veterans are more able to handle low income if not distressed or impacted in other domains of well-being, unless the low income state persists and becomes a determinant of poorer well-being in other domains. This finding reinforces the importance of subjective well-being (distress) in those with low income. Suggests the importance of addressing determinants of mental health across multiple well-being domains to reduce stress and improve sense of mastery (control over one’s life). The relationships between low income, mental health problems and factors in other well-being domains remain unclear. Test the hypothesis that easing stressors in other well-being domains and improving a sense of mastery improves mental health.
Adding rank removed the association between low education and mental health problems. However, rank and education were correlated: most NCMs had less than university degree education attainment while most commissioned officers had university degrees. It is well known in civilian studies that lower socioeconomic status is associated with health problems, and there are socioeconomic differences between lower and higher ranks. This finding informs both screening and tailoring services to different subgroups of Veterans. While there is evidence of socioeconomic gradients in health among military ranks in other countries, the issue has not been well studied in this country.
Prior LASS analyses demonstrated an association between NCM rank and chronic physical health conditions. In this study, NCM rank at release was associated with higher unadjusted odds of mental health problems but the association with rank disappeared when adjusting for physical conditions, chronic pain and socioeconomic factors. This finding suggests that having physical health problems +/- chronic pain and socioeconomic disadvantages explains the association of rank with mental health problems, supporting programming that addresses multiple determinants of well-being. The complex interrelated roles of mental health, physical health and socioeconomic factors in well-being require further study in CAF Veterans, with a view to understanding how best to support well-being in this population.
Army service had higher odds of mental health problems than Navy or Air Force service. Useful in screening for supports. Review of published studies of serving CAF member well-being and longitudinal studies across transition would help to identify life course factors associated with mental health specific to the three service elements.
High life stress, poor life satisfaction, low mastery and low social support were all associated with mental health problems and when added to the models diminished the strengths of association with other factors. This finding suggests that screening to identify those in distress would assist in targeting limited resources. The bidirectional causality between mental health and these factors remains unclear, particular with respect to the efficacy of interventions such as enhancing mastery.
Self-reported difficult adjustment to civilian life was strongly associated with mental health problems regardless of type of release from service. The prior focus on the medically released should be broadened to include those who were not medically released. Informs refinement of the transition interview and risk assessment procedures. It is unclear why self-reporting difficult adjustment to civilian life was so strongly associated with mental health problems.
Self-reported difficult adjustment to civilian life was associated with risk indicators across multiple domains of well-being, including at least health, employment, finances and social integration. Informs screening to target programs and services for those more likely to experience difficult adjustment, including the transition interview. In conjunction with other sources of information, this finding suggests that attention to well-being across multiple domains, particularly mental health and its determinants, is likely to factor in successful military-civilian transition outcomes. The difficult adjustment question requires further quantitative and qualitative validation in order to understand what respondents are considering when they answer the question.
Self-reported difficult adjustment to civilian life was associated with all adult age groups less than 60. Our hypothesis is that adults less than 60 are more likely than younger and older people to be dealing with multiple stresses simultaneously, for example raising families, education/career challenges, financial challenges, and emerging chronic physical health problems. Informs development of good transition and post-transition supports for Veterans of working age who are more likely to experience difficulty adjusting to civilian life. Longitudinal studies would help to inform understanding of factors that contribute to easy or difficult transitions to civilian life.

Interpretation Guidance

The findings in this report describe factors associated with mental health problems in CAF Veterans.

Study strengths:

  • Objective identification of former CAF members to construct the sampling frame, using the DND human resources database.
  • Use of self-report measures common to Canadian general population studies.
  • Use of the new composite mental health problem measure extends understanding of mental health in these Veterans by identifying subgroups with self-reported diagnosed mental health conditions across a range of symptom levels, and mental health symptoms beyond those who report diagnosed conditions.
  • The multivariable regression analyses account for the joint effects of several factors together in assessing associations with mental health problems (controls for confounding).
  • The use of ordinal regression modelling enables understanding of associations across a range of severity mental health problems and degrees of difficulty in adjustment to civilian life rather than just presence/absence.
  • The LASS surveys include Veterans living in the general population who are not using VAC services, in addition to those who are.

Study limitations:

  • These findings apply to the population as a whole, not necessarily to individuals. A factor with low prevalence or low odds of mental health problems could be highly significant for some individuals.
  • Be cautious about drawing conclusions about the presence of "risk" and "protective" factors. LASS 2013 was a point-in-time, cross-sectional survey, therefore causal conclusions cannot be made from this study alone, including the effects of military service or DND/CAF/VAC programs on Veterans' later life courses. The study identifies “risk indicators” (factors describing subgroups in whom mental health problems are more common) but not necessarily “risk factors” (factors that cause mental health problems).
  • The adjusted regression models depend on available data. Although this allows for identifying potential confounding in associations between measured factors and mental health problems or difficult adjustment to civilian life, there is always the possibility that important factors were not measured and therefore not included in the models.
  • Findings cannot be generalized to all Veterans because the study included only those who released in 1998-2012 (Regular Force) and 2003-2012 (deployed Reserve Force).
Date modified: