Associations between anxiety disorders, suicide ideation, and age in nationally representative samples of Canadian and American adults

https://doi.org/10.1016/j.janxdis.2014.09.005Get rights and content

Highlights

  • We examine anxiety disorders and suicide ideation in Canadian and American samples.

  • We examine whether age moderates the effect of anxiety disorders on suicide ideation.

  • Individual anxiety disorders are associated with increased odds of suicide ideation.

  • Comorbidity of anxiety and suicide ideation is similar for younger and older Canadians.

  • Comorbidity of anxiety and suicide ideation differs for younger and older Americans.

Abstract

Suicidal behaviors are of significant concern for the individuals displaying such behavior and for service providers who encounter them. Using nationally representative samples of Canadian and American adults, we aimed to examine: whether age moderates the relationship between having any anxiety disorder and suicide ideation (SI), the prevalence of SI among younger and older adults, and whether age and individual anxiety disorders were differentially associated with SI. Age moderated the relationship between any anxiety disorder and SI among Americans only. Past-year SI was less prevalent among older, compared to younger, adults; though, nearly every anxiety disorder was associated with increased odds of SI among younger and older Canadian and American adults after controlling for covariates. Effect sizes were particularly large for older American adults, but were coupled with large confidence intervals. Findings contribute to a growing literature suggesting that SI in the context of anxiety is a highly prevalent and complex mental health problem across the adult lifespan.

Introduction

Recurrent thoughts of suicide are a significant health concern for the individuals experiencing suicidal behavior and for service providers who encounter them, given that suicidal ideation (SI) is predictive of both suicide plans and attempts (Nock, Borges, Bromet, Alonso, et al., 2008). An extensive body of research points to mood disorders, substance use disorders, and to a lesser extent, anxiety disorders as significant risk factors for SI in the general population (Sareen et al., 2005a, Sareen et al., 2005b). However, less is known about how SI is associated with individual past-year anxiety disorders, age, and region of residence (i.e., whether individuals live in Canada or the United States).

Epidemiological studies in Canada and the United States continue to suggest that anxiety disorders cannot be overlooked as a correlate of suicidal behavior, though the mechanisms remain understudied. Across these epidemiological studies, nearly every lifetime anxiety disorder has been identified as a correlate of SI, sometimes independent of comorbid mood disorders and substance dependence, in younger and middle-aged adults (Cougle et al., 2009, Nepon et al., 2010, Sareen et al., 2005a). Additionally, in a community sample of Canadian adults, individuals who had lifetime specific phobia or panic disorder had significantly increased odds of attempted suicide (Dyck, Bland, Newman, & Orn, 1988). Thus, there is a dearth of research on the degree to which individual past-year anxiety disorders are uniquely associated with SI in nationally representative lifespan samples.

Rates of SI tend to decrease among older adults (Nock, Borges, Bromet, Alonso, et al., 2008), with estimated prevalence rates ranging from 2.2% to 16.5% (Awata et al., 2005, Corna et al., 2010, Jorm et al., 1995, Yen et al., 2005), depending on the time point of survey measurement. Suicide attempts also decrease with age, but the rate of completed suicide is highest in late life (Conwell, 2001), where the ratio of attempted to completed suicide is 4:1, compared to a range of 8:1–33:1 in the general population (Conwell and Thompson, 2008, Conwell et al., 2011). This startling discrepancy highlights the importance of understanding the correlates of SI among older adults for early detection and prevention of completed suicide.

Despite the burgeoning body of research examining anxiety and SI in the general adult population, anxiety disorders have received significantly less attention as a correlate of geriatric SI compared to major depressive disorder (MDD). SI has been linked to social phobia, panic disorder, and depression among a nationally representative sample of Canadian older adults (Corna et al., 2010). However, research has yet to examine the relationship between past-year SI and additional past-year anxiety disorders (e.g., post-traumatic stress disorder [PTSD], generalized anxiety disorder, and specific phobia) and whether this relationship is affected by age. Thus, there is a paucity of population-based research that has considered age as a potential moderator in the relationship between a broad range of anxiety disorders and SI while also controlling for co-occurring mental disorders.

To our knowledge, no studies to date have examined the relationship between anxiety disorders and SI among younger and older adults across developed nations. The World Health Organization (WHO) World Mental Health Surveys have documented slightly lower rates of SI in developed versus developing countries (Borges et al., 2010). However, the WHO study collapsed across developed countries, and did not include Canada. Thus, it is unclear whether the relationship between anxiety disorders and SI differs between the United States and Canada, which are geographically proximal and culturally similar, albeit with very different healthcare systems.

The present study employed nationally representative samples of Canadian and American adults to address gaps in the current understanding of correlates of SI. Our primary objective was to determine whether age moderates the relationship between having any anxiety disorder and SI and, if so, to stratify our samples into younger versus older adults. Our second objective was to examine the prevalence of SI among younger and older adults. Our third objective was to examine whether age and individual anxiety disorders are differentially associated with SI, and whether these relationships differ by country of residence, while statistically controlling for sociodemographic characteristics, and comorbid major depressive disorder and substance dependence.

Section snippets

Participants

We used data from the public use Canadian Community Health Survey (CCHS) Cycle 1.2 (we refer to respondents as “Canadians”), and the CPES (Collaborative Psychiatric Epidemiology Surveys; we refer to respondents as “Americans”). Interviewers in both the Canadian and American surveys were extensively trained by their research teams, and received additional training to increase their sensitivity to mental health issues and potentially sensitive questions (Beland et al., 2001, Pennell et al., 2004

Results

We first examined whether age moderated the effect of any anxiety disorder on SI. Our age by any anxiety interaction term significantly predicted SI in the American survey (OR = 4.02, 95% confidence interval [CI] = 1.78–9.07). The interaction term was not significant in the Canadian survey (OR = 1.45, 95% CI = 0.79–2.63). However, our findings from the American dataset provided justification to stratify our logistic regression analyses by age group to gain an understanding of age differences across

Discussion

To our knowledge, this was the first study to show strong and significant associations between individual past-year anxiety disorders and past-year SI, controlling for sociodemographics, depression and substance dependence, in two nationally representative samples of younger and older adults. Past-year any anxiety disorder was differentially associated with past-year SI depending on respondent age and country of residence. Results of logistic regressions revealed large effect sizes in the

Conclusions

The current study investigated the relationship between past-year anxiety disorders and past-year SI among Canadian and American adults. Despite the fact that the prevalence of mood and anxiety disorders generally decreases in later life, individual anxiety disorders were significantly and strongly associated with SI among both younger and older adults. By shedding light on the relationship between SI and anxiety disorders across the lifespan, American and Canadian health care providers may be

Conflicts of interest and sources of funding

RE is supported by a Research Manitoba Graduate Studentship and a Manitoba Graduate Scholarship; JE is supported by a Research Manitoba Graduate Studentship; JS is supported by a Research Manitoba Chair Award. The authors have no disclosures to report.

Acknowledgements

The authors would like to thank Dr. Tracie Afifi for her assistance with later drafts of the manuscript. The Canadian Community Health Survey (CCHS) was conducted and funded by Statistics Canada, and the Collaborative Psychiatric Epidemiology Surveys (CPES) were conducted and funded by the National Institute of Mental Health (NIMH). We thank Statistics Canada, the NIMH, and field representatives who administrated the CCHS and CPES interviews and made the surveys available for researchers. We

References (40)

  • L. Bailie et al.

    Data quality assurance for the Canadian Community Health Survey

    (2002)
  • D.W. Baker et al.

    Lack of health insurance and decline in overall health in late middle age

    New England Journal of Medicine

    (2001)
  • Y. Béland et al.

    Sample design of the Canadian Mental Health and Wellbeing Survey

  • G. Borges et al.

    Twelve-month prevalence of and risk factors for suicide attempts in the World Health Organization World Mental Health Surveys

    Journal of Clinical Psychiatry

    (2010)
  • Y. Conwell

    Suicide in later life: a review and recommendations for prevention

    Suicide and Life-Threatening Behavior

    (2001)
  • L.M. Corna et al.

    Suicide ideation in older adults: relationship to mental health problems and service use

    Gerontologist

    (2010)
  • R.J. Dyck et al.

    Suicide attempts and psychiatric disorders in Edmonton

    Acta Psychiatrica Scandinavica

    (1988)
  • J. Fawcett et al.

    Suicide: a four-pathway clinical-biochemical model

    Annals of the New York Academy of Sciences

    (1997)
  • J.M. Haro et al.

    Concordance of the Composite International Diagnostic Interview Version 3.0 (CIDI 3.0) with standardized clinical assessments in the WHO World Mental Health Surveys

    International Journal of Methods in Psychiatric Research

    (2006)
  • S.G. Heeringa et al.

    Sample design and sampling methods for the Collaborative Psychiatric Epidemiology Studies (CPES)

    International Journal of Methods in Psychiatric Research

    (2004)
  • View full text