 |
 |
|
|
|
Depression and suicidality (ideation and nonfatal behaviours) contribute strongly to premature mortality and are costly, disabling conditions internationally (1–7). Thus the widespread uptake of effective interventions for individuals who are suicidal or who have depression could have a major public health impact. A core symptom of depression is suicidal ideation. As such, suicidal ideation and behaviours may arise from depression or other conditions. There is a large body of evidence concerning the benefits of treating depression. However, there is an ongoing controversy about the relative merits of prescribing the newer antidepressants to those who have depression. While the newer antidepressants are less toxic if taken in overdose, suicidality may arise or become more intense for some (at least initially) (8–11). Even less is known about treatment effectiveness for those who are suicidal (12–14). Until recently, there has been a lack of systematic evidence and guidelines about how to treat those who are suicidal apart from, or in addition to, concurrent conditions such as depression (15,16). As knowledge grows about which treatments work best for whom, access to effective interventions can be promoted through knowledge of the patterns of mental health services used. Therefore, the purpose of this study is to determine the size of depression and suicidal populations, the overlap between these populations, and the associated patterns of mental health service use. The design of the CCHS 1.2 (17) overcomes several methodological issues present in previous population-based studies of depression, suicidality, and mental health service use. The overlap in these populations and related service use can be examined because the full sample aged 15 years and over were asked about suicidality rather than only those who passed the depression screen (18,19). As well, more current use (12-month compared with lifetime recall) of different types of mental health serivce providers can be depicted, given the large sample size. There have been only a few studies of the types of mental health service providers seen by persons with depression and (or) suicidality in the general population. Whereas people with a history of depression or suicidality were more likely to report mental health service use than not, little could be concluded about patterns in providers seen specific to one or both of these populations owing to variations in the methods and reduced statistical power (20–24). Further, the patterns of providers seen in these other countries may not extend to Canada, where medical care is publicly funded and universally available to the full population. Materials and MethodsA more thorough description of the CCHS 1.2 is provided in the September 2005 special issue of The Canadian Journal of Psychiatry (17). This study received approval from the Research Ethics Board of St Michael’s Hospital, and data access was granted through the Social Sciences and Humanities Research Council and Statistics Canada. Dependent Variable: Provider Group Contacted Most Often for Mental Health ReasonsIn the CCHS 1.2, respondents were asked a series of questions about contact(s) with different types of providers for “help with emotions, mental problems or use of alcohol or drugs.” For each provider type seen in the past 12 months, the respondent was asked to think of the provider she or he talked to most often. Ambulatory care providers contacted most often were categorized in a hierarchical fashion on the basis of physician specialty, the rationale being that, of the 7.8% of the sample who reported on a specific provider, 73.7% included a physician. Persons whose reports included a psychiatrist were classified as being in the psychiatrist group (1.65%). In this group, 61.4% were seen only by physicians, with about equal proportions having seen only a psychiatrist and having seen both a psychiatrist and another physician. Those whose reports included a family doctor, a general practitioner, or another medical doctor (but not a psychiatrist) were classified as being in the primary medical care group (4.07%). A total of 67.6% of this group comprised physicians only. Those whose reports included a psychologist, nurse, social worker or counsellor, psychotherapist, or another nonmedical professional (but no physicians) were classified as being seen by the other provider only group (2.05%). Each provider group was compared with persons reporting no mental health contacts. Those who reported an inpatient mental health stay in the past 12 months (0.51%) were examined separately. Independent Variables: Depression and SuicidalityWe used the CIDI (17,25,26) to assess the presence of a DSM-IV major depression in the past 12 months. Thoughts of death are included in the symptom count to determine whether a respondent meets the symptom criteria portion in the CIDI diagnostic algorithm for major depression. In theory, the diagnosis of depression could depend on whether people have suicidal thoughts. This, however, was not the case. Of those who met the necessary symptom count and who had death thoughts, 92.9% continued to meet the necessary symptom count when the death thought symptom was removed. On the basis of previous surveys and question sensitivity, suicidal ideation, plans, and attempts were asked about as experience A, B, or C, and respondents filled out their answers in a booklet. Suicidality was identified in terms of the following questions: have you ever “seriously thought about committing suicide or taking your own life,” “made a plan for committing suicide,” “attempted suicide or tried to take your own life” (in the 12 months prior to the interview or age when the experience last occurred). Owing to technical problems in the questionnaire (that is, skip patterns in the depression module), about 5% of respondents were not directly asked about suicidality in the 12 months before the interview. For many, it was possible to fill in these missing values, based on other questions asked in the depression module. Statistical Analyses We examined initial analyses of cross-tabulations of respondent characteristics (depression, suicidal thoughts, and suicide attempts) and provider groups contacted most often for mental health reasons. Owing to small cells, suicide attempts were collapsed into the suicidal thoughts group. We quantified the associations between 1) depression (vs no depression) and provider groups and 2) suicidality (vs no suicidality) and provider groups, using multinomial logistic regression (27,28). Models estimated crude ORs of use and their respective 95%CIs to compare each of the 3 provider groups (psychiatrist, primary care physician, and other provider only) with the no mental health contacts group. A multivariate multinomial logistic regression model was fit first with the main effects of depression and suicidality; we then included the interaction between these terms. We used the statistical software package SAS (29) to carry out all analyses. Estimates are weighted, and variances are bootstrapped to reflect the complex sampling design (30,31). ResultsDepression and Suicidality In the 12 months prior to the survey, 4.8% of Canadians suffered from depression. Almost 4% reported suicidal ideation, and less than 1% reported a suicide attempt. Among those who reported suicidal ideation, 37.2% had depression. Among those reporting a suicide attempt, 56.1% had depression. When suicide thoughts and attempts were combined, 36.7% of the suicidal population had depression. This proportion increased somewhat when a lifetime depression was assessed (an increase of 13.8%). Among those with depression in the previous 12 months, 28.6% reported suicidal ideation, and 6.0 % reported a suicide attempt. When suicide thoughts are combined with ideation, 28.7% of the depression subjects were suicidal. Provider Group Contacted Most Often for Mental Health Reasons Because the overlap between suicidality and depression was partial, it is possible that there would be variations seen in the provider groups, depending on whether they had one or both conditions. In Table 1, the proportions of use are shown for persons with suicidality but not depression, depression but not suicidality, both depression and suicidality, and neither condition. For persons with depression and (or) suicidality, the proportions of use (row percents) were all higher than for those with neither condition. Those who had both conditions were most likely to report contact or have an inpatient mental health stay (61.6%). The next group most likely to report this use was the group with depression but not suicidality (50.8%); next were those with suicidality but not depression (28.6%). Those with neither condition were the least likely to report this use (5.5%). This pattern was not apparent in the only other provider group. There was little variation in whether one had depression and (or) suicidality in the other provider only group, and the use of this group was less frequent (2.1%), compared with the physician provider groups (5.7%).
In the multivariate model where depression and suicidality were included together as main effects, each term was independently associated with use; however, the associations with use for those with suicidality were greatly attenuated after we adjusted for depression. When we added the interaction term to this model, it was statistically significant. Therefore, Table 2 presents the ORs and 95%CIs from this model to further describe the patterns of use. Figure 1 illustrates these ORs with a graph.
The direction of each association with use was positive and statistically significant, as each OR exceeded 1, and the 95%CIs did not include 1. The highest ORs of use occurred for those with depression (suicidal or not) in the psychiatrist and primary care physician groups. The ORs of the use of these provider groups were considerably lower for those who were suicidal (but had no depression). The use of the physician provider groups was distinguished by whether one had depression or not. Conversely, the use of the other provider group only was not differentiated by depression or by suicidality. There was little difference in the ORs of use for the other provider group. For those who were suicidal (but who had no depression), there was little difference in which provider group was seen. DiscussionTo our knowledge, this is the first study to document the unique patterns of provider groups seen by people who are suicidal and (or) who have depression in Canada. This information comes at an important time, namely, during a period when there has been enormous growth in access to newer antidepressants (32) and in concerns about the role of these agents in relation to suicidality. Although the suicidal subjects were more likely to make contact than not, their pattern of contact depended on whether they had a concurrent depression. Those who had a concurrent depression were most likely to contact a provider group, and these provider groups were most likely to be physicians. Suicidality, in and of itself, did not exert a strong effect on the contact of any particular provider group. Before we discuss the findings further, several interpretative cautions are necessary. The CCHS 1.2 is a cross-sectional survey, and therefore, estimates of suicidality and depression in the population are based on the prevalence of the conditions over a period of time. Prevalence estimates tend to overrepresent people with a longer duration and (or) more frequent recurrences of these conditions. Individuals with severe and acute depression and (or) suicidality might not have been included in the study because of premature mortality or because they resided in settings not included in the sample design. Another important consideration is that, when data are cross-sectional, it is not known whether the patterns of use observed are a consequence of the condition(s) or whether the condition is a consequence of the patterns of use. For example, while it may seem appropriate that persons with depression and suicidality were most likely to report contact with the psychiatrist provider group, it is not known whether this pattern is a reflection of a lack of response to treatment and (or) delays in treatment. Cross-sectional data cannot directly answer these questions. From a planning perspective, it is noteworthy that, of the 34% of the suicidal people who used the ambulatory provider groups, about 2 of 3 were not seen by the psychiatrist provider group. Nevertheless, the estimates of use reported by those with depression or suicidality in the past 12 months may actually be inflated. About 75% of these individuals were also currently at a high level of distress. Two studies conducted in Ontario found that people at a high level of distress reported more use than was recorded in their administrative health care records (33,34). The evidence from these studies was consistent with a recall bias implying that those who are currently distressed work harder to recall their use, compared with those who are not currently distressed. This differential recall could upwardly bias depression and suicidality ORs of provider group use compared with no contact group. Since those who had depression or who were suicidal were similar in their level of distress, comparisons of use between these conditions were not likely altered by recall bias. The main question this study raises concerns why suicidal people are much less likely to report mental health contacts with the physician provider groups unless they have a concurrent depression. The assessment of need for mental health services occurs at multiple and interacting levels of society. While there is an ongoing debate about how to best define and measure the need for mental health services for the purposes of resource allocation, there tends to be consensus that key dimensions of need are having a mental illness, illness duration, disability, distress associated with the illness, and the risk of harm to self and others (35–37). Previous studies (19) and preliminary analyses of this sample suggest that many of those who are suicidal would have met one or more of these criteria, regardless of depression. It is posited that depression is viewed as more treatable, and therefore, ongoing contact is encouraged. There has been widespread marketing of antidepressants for the treatment of depression, and physicians are able to prescribe these medications. Because many symptoms of depression are somatic in nature, with perhaps less stigma, such symptoms might have prompted initial contact with a physician. For those who are suicidal but who do not meet criteria for depression, there is less certainty about the effectiveness of antidepressants. Cognitive problem-solving therapies can benefit adults who have made suicide attempts (38), but less is known for youths (39) or those who have suicidal thoughts. In any case, such therapies require highly trained personnel, who are not always in supply or insured. Accordingly, treatment options for people who are suicidal but who do not have depression may be limited, and those who are suicidal may not seek treatment or may delay treatment. Providers may not invest in developing new treatments and discourage ongoing contact. In the absence of alternatives, suicidal persons may become sicker and (or) take their lives. If they seek treatment later, the severity of their illness may dictate a higher level of care, and the illness may be much more difficult to treat. Thus perhaps the question does not concern whether suicidal people are in need but, rather, the proportion of the population affected and likely to (directly or indirectly) benefit from treatment, the monetary costs of providing the treatment, and the risk of adverse outcomes from treatment (40). Funding and SupportDr Rhodes is supported by a Career Scientist Award from the Ontario Ministry of Health and Long-Term Care. AcknowledgementsWe thank Veronica Yei and Ron Gravel from Statistics Canada for their assistance with this project. The research and analysis are based on data produced by Statistics Canada and the opinions expressed do not represent the views of Statistics Canada. The opinions, results, and conclusions are those of the authors, and no endorsement by the Ministry is intended or should be inferred. References1. Vos T, Mathers C. The burden of mental disorders: a comparison of methods between the Australian burden of disease studies and the Global Burden of Disease study. Bulletin of the World Health Organization 2000;78:427–38. 2. Langlois S, Morrison P. Suicide deaths and suicide attempts. Health Reports 2002;13(2):9–22. 3. Clayton D, Barceló A. The cost of suicide mortality in New Brunswick, 1996. Chronic Diseases in Canada 1999;20(2):1–10. 4. Simon GE, Barber C, Birnbaum H, Frank RG, Greenberg P, Rose R, and others. Depression and work productivity: the comparative costs of treatment versus nontreatment.J Occup Environ Med 2001;43:2–9. 5. Rice D, Miller L. Health economics and cost implications of anxiety and other mental disorders in the United States. Br J Psychiatry 1998(Suppl 34):4–9. 6. Stephens T, Joubert N. The economic burden of mental health problems in Canada. Chronic Diseases in Canada 2001;22(1):1–7. 7. Hirschfeld R, Montgomery S, Keller M. Social functioning in depression: a review. J Clin Psychiatry 2000;61:275. 8. Gunnell D, Ashby A. Antidepressants and suicide: what is the balance of benefit and harm? BMJ 2004;329:34–8. 9. Geddes J. Selective serotonin reuptake inhibitors. Remain useful drugs which need careful monitoring. BMJ 2004;329:809–10. 10. Brent D. Antidepressants and pediatric depression–the risk of doing nothing. New Engl J Med 2005;351:1598–1601. 11. Jick H, Kaye J, Jick S. Antidepressants and the risk of suicidal behaviors. JAMA 2004;292:338–43. 12. Ernst C, Goldberg J. Antisuicide properties of psychotropic drugs: a critical review. Harv Rev Psychiatry 2004;12:14–41. 13. Rhodes A, Links P. Suicide and suicidal behaviours: implications for mental health services. Canadian J Psychiatry 1998;43:785–91. 14. Hawton K, Arensman E, Townsend E, Bremner S, Feldman E, Goldney R. Deliberate self harm: systematic review of efficacy of psychosocial and pharmacologic treatments in preventing repetition. BMJ 1998;317:441–7. 15. Gaynes B, West S, Ford C, Frame P, Klien J, Lohr K. Screening for suicide risk in adults: a summary of the evidence for the US Preventive Services Task Force. Ann Intern Med 2004;140:822–37. 16. APA. Practice guideline for the assessment and treatment of patients with suicidal behaviors. Arlington (VA): American Psychiatric Association; Nov 2003. 17. Gravel R, Béland Y. The Canadian Community Health Survey of Mental Health and Well-Being. Can J Psychiatry 2005;50:573–9 18. Moscicki EK, O’Carroll PW, Rae DS, Locke BZ, Roy A, Regier DA. Suicide attempts in the Epidemiologic Catchment Area Study. Yale J Biol Med 1988;61:259–68. 19. Pirkis J, Burgess P, Dunt D. Suicidal ideation and suicide attempts among Australian adults. Crisis 2000;21(1):16–25. 20. Bijl R, Ravelli A. Psychiatric morbidity, service use, and need for care in the general population: Results of the Netherlands Mental Health Survey and Incidence Study. Am J Public Health 2000;90:602–7. 21. Goldney R, Fisher L, Wilson D, Cheok F. Major depression and its associated morbidity and quality of life in a random, representative Australian community sample. Aust N Z J Psychiatry 2000;34:1022–9. 22. Parslow R, Jorm AF. Who uses mental health services in Australia? An analysis of data from the National Survey of Mental Health and Wellbeing. Aust N Z J Psychiatry 2000;34:997–1008. 23. Marino S, Gallo JJ, Ford D, Anthony JC. Filters on the pathway to mental health care, I. Incident mental disorders. Psychol Med 1995;25:1135–48. 24. Pirkis J, Burgess P, Meadows G, Dunt D. Suicidal ideation and suicide attempts as predictors of mental health service use. Med J Aust 2001;175:542–4. 25. Kessler R, Wittchen H-U, Abelson J, McGonagle K, Schwarz N, Kendler K, and others. Methodological studies of the Composite International Diagnostic Interview (CIDI) in the US National Comorbidity Survey (NCS). Int J Methods Psychiatr Res 2001;7(1):33–55. 26. Wittchen HU. Reliability and validity studies of the WHO Composite International Diagnostic Interview (CIDI): a critical review. J Psychiatr Res1994;28(1):57–84. 27. Hosmer D, Lemeshow S. Applied logistic regression. New York: John Wiley and Sons; 1989. 28. Allison P. Multinomial logit analysis. Logistic Regression Using the SAS System: Theory and Application. Cary (NC): SAS Institute; 2000. p 111–31. 29. SAS Institute Inc. SAS Release 8.2 for Unix Environments. Cary (NC): SAS Institute; 2000. 30. Rao J, Wu C, Yue K. Some recent work on resampling methods for complex surveys. Survey Methodology, Statistics Canada. Cat 12-001 1992;18:209–17. 31. Rust K, Rao J. Variance estimation for complex surveys using replication techniques. Stat Methods Med Res 1996;5:283–310. 32. Hemels M, Koren G, Einarson T. Increased use of antidepressants in Canada: 1981–2000. Ann Pharmacother 2002;36:1375–9. 33. Rhodes A, Lin E, Mustard C. Self-reported use of mental health services versus administrative records: should we care? Int J Methods Psychiatr Res 2002;11:125–33. 34. Rhodes A, Fung K. Self-reported use of mental health services versus administrative records: care to recall? Int J Methods Psychiatr Res 2004;13:165–75. 35. Mechanic D. Is the prevalence of mental disorders a good measure of the need for services? Health Affairs 2003;22(5):8. 36. Regier DA. Mental disorder diagnostic theory and practical reality: an evolutionary perspective. Health Affairs 2004;22(5):21. 37. Andrews G, Henderson S. Unmet need in psychiatry. Cambridge (UK): University Press; 2000. 38. Townsend E, Hawton K, Altman DC, Arensman E, Gunnell D, Hazell P, and others. The efficacy of problem-solving treatments after deliberate self-harm: meta-analysis of randomized controlled trials with respect to depression, hopelessness and improvement in problems. Psychol Med 2001;31:979–88. 39. Gould M, Greenberg T, Velting D, Shaffer D. Youth suicide risk and preventive interventions: a review of the past 10 years. J Am Acad Child Adolesc Psychiatry 2003;42:386–405. 40. Mechanic D. Policy challenges in improving mental health services: some lessons from the past. Psychiatr Serv 2003;54:1227–32. Author(s)Manuscript received May 2005, revised and accepted August 2005. 1. Research Scientist, Suicide Unit, St Michael’s Hospital, Toronto, Ontario; Assistant Professor, Department of Psychiatry, University of Toronto, Toronto, Ontario. 2. PhD Candidate, Department of Public Health Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario; Research Coordinator, Suicide Studies Unit, St Michael’s Hospital, Toronto, Ontario. 3. Assistant Professor, Department of Public Health Sciences, Faculty of Medicine, University of Toronto, Toronto, Ontario. Preliminary results were presented to the Canadian Academy of Psychiatric Epidemiology; 2004 October. Address for correspondence: Dr A Rhodes, Suicide Studies Unit, Suite 1010f, 2 Shuter Wing, St Michael’s Hospital, 30 Bond St, Toronto ON, M5B 1W8 e-mail: rhodesa@smh.toronto.on.ca
1 | 2
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||