 |
 |
|
|
|
The epidemiology of late-life psychiatric disorders is comparatively new in psychiatric epidemiology and is most likely tied to changing demographic trends (1). Like the UK and the US, Canada has been characterized as an “aging” society because the proportion of adults aged 65 years and over has increased substantially relative to other age groups. While the overall population of Canada increased by 4% between 1996 and 2001, the proportion of those aged 65 years and over increased by 10.2%, to 13%. The question of whether mental illness declines or increases with age is complex. This is, in part, because it is tied to the challenges associated with measuring psychological disorder(s). Depression serves as a useful illustration. Studies that use symptom measures have shown a U-shaped relation between age and depression, with the lowest prevalence among people aged 45 to 50 years (2,3). These studies suggest that, among young adults (aged 20 to 24 years) and the very old (aged 80 years and over), depression has higher prevalence and is a significant health concern. A different pattern emerges, however, when we focus on studies that used DSM criteria for disorders. Several studies, including the ECA, have documented a negative linear relation between age and depression, with the lowest reported prevalence estimates among people aged 65 years and over (4). Using a large, representative sample of Canadians aged 15 years and older, Wade and Cairney (5) reported a reversed J-shaped relation between age and depression across successively older age cohorts. Depression declines until about age 75 years where a modest upturn occurs. These discrepancies may also reflect real differences in the manifestation of depressive illness in old age. Older adults may be more likely than younger adults to have subsyndromal or minor depression rather than MDD (6,7). There are also other possible explanations that may account for low prevalence estimates among older adults. Most studies that use DSM criteria for caseness report the prevalence of psychiatric disorders among community- dwelling older adults (aged 65 years and over) is lower than younger age groups (aged 18 to 64 years) (8,9). In the ECA studies, for example, pooled data from all sites show the 1-month prevalence of any disorder among older adults is 12.3%, compared with 17% in adults aged 18 to 24 years (8). This general pattern holds for virtually all disorders examined in the ECA study, except severe cognitive impairment, where prevalence estimates are highest among older adults (4.9%). Based on ECA data, the most prevalent disorders in old age are anxiety, phobia, and dysthymia (5.5%, 4.8%, and 1.8%, respectively). Bland and others report similar data for residents of Edmonton, Alberta (9). Both 6-month and lifetime estimates of disorders are consistently lower among respondents aged 65 years and over, compared with the whole population. For example, the 6-month prevalence of any disorder is 10.9% among older adults, compared with 17.1% for all adults in the study. Apart from cognitive impairments, dysthymia, phobias, and panic disorders are the most prevalent among the older respondents (3.3%, 3.5%, and 3.0%, respectively). When first published, data such as these differed from the consensus of the day that mental health problems were highly prevalent among older adults (10). Of particular surprise was the prevalence estimate of 0.9% for MDE (8). Many experiences associated with growing old (for example, loss of loved ones, chronic physical health problems, and the loss of valued social roles) are also associated with increased risk for psychopathology, especially depression (1,2). Despite greater exposure to these negative occurrences, older adults were less likely than younger adults to suffer from psychiatric disorders. Moreover, these findings were inconsistent with other data that showed a high use of psychotropic medication among the elderly, a proxy for the presence of psychiatric disorder (7). In the aftermath of the release of the ECA data, epidemiologists interested in aging and mental health have sought to account for this rather surprising finding. Hocking and her colleagues offer 2 explanations for the low prevalence estimates in older adults (1). The first is methodological and concerns reporting behaviour. It is not that older adults are less likely to have psychiatric disorders per se; rather, they are less likely than younger adults to report symptoms, more likely to forget them, and (or) more likely to express psychological, mood, or behavioural symptoms in somatic terms. Closely related to reporting is the issue of capture with conventional sampling techniques. Elderly people at greatest risk for psychiatric disorders (that is, the institutionalized, the homeless, and those suffering significant physical or cognitive impairment) are likely either to refuse to participate in surveys or to be excluded from the sampling frame altogether. While few would dismiss the contribution that methodology and sampling make to underestimating mental disorders in older adults, most epidemiologists treat the findings as real, not artifactual, and believe the explanation for the pattern of results is owing to period and (or) cohort effects (1,7). Proponents of the cohort interpretation argue that individuals born prior to 1920 (aged 65 years and over at the time of the ECA survey), for some yet to be determined reason, are simply psychologically healthier than individuals born after (or considerably earlier) that time period. Older adults report fewer psychiatric symptoms because they have always been healthier than younger cohorts. The period effect hypothesis is a more satisfying explanation because it identifies actual causal mechanisms. This hypothesis holds that older adults, at the time of the ECA survey, shared several significant historical events (for example, the Great Depression and Second World War) that had a profound impact on psychological development. In particular, many individuals who lived through these periods endured significant economic and social hardship. However, the same individuals also experienced significant improvements in their economic and social circumstances during the post Second World War period. Younger adults (for example, the “baby boomer” generation), conversely, have not experienced such dramatic changes in their standard of living and therefore are not as adept as their parents at handling adversities. This “baby boomer” generation, however, has set high expectations for itself in terms of material standard of living and, as a result of growing up during a population boom, has to compete for fewer social and economic resources. Together, these effects explain higher rates of disorder in younger, as opposed to older, cohorts. These differences are also explained as true age effects. In other words, psychiatric disorders are less common in older adults because psychological well-being improves with age. As noted earlier, this is at odds with gerontological and psychological theories (for example health, social roles, friends, and loved ones), and how they affect psychological well-being (2,11–15). However, stage theories of human development, which have also been influential in gerontology and the psychology of aging, support the notion that positive mental well-being accompanies aging (16–21). Levinson and others (21), for example, portray aging as a process of maturational unfolding. With age comes greater maturity and acceptance as individuals either master their environments or learn to accept limitations and change, hence perceiving situations in a more positive light than younger adults. Finally, the healthy survivor or selection explanation is a less sanguine version of the aging effect. Rates of psychiatric disorder are lower in older adults because only the healthiest members of a given age cohort survive to old age. This particular hypothesis has been used to account for narrowing socioeconomic status gaps in physical health with age (22), although there is no reason it cannot also be applied to psychiatric outcomes (23). There are few data regarding the epidemiology of depression and anxiety among the elderly in Canada. Moreover, there are certain issues specific to the Canadian context that have not been well examined. In particular, an exploration of possible differences in estimates of disorder between Anglophone and Francophone Canadians is an important, yet generally overlooked, area in psychiatric epidemiology. Previous research has found a significant difference in the prevalence of depression and in mean levels of psychological distress between French- and English-speaking Canadians (24). Whether this difference is owing to socio-cultural factors or an artifact of measurement (for example, translation problems) is not yet established. We know of no work, however, that has examined whether there are differences in mood and anxiety disorders, based on DSM criteria, between Anglophones and Francophones in this country. The purpose of this study was to use the CCHS 1.2 data to determine the prevalence of depression and other psychological disorders among people aged 55 years and over. MethodThe Survey The CCHS 1.2 interview (21) is a nationally representative, community mental health survey conducted by Statistics Canada (the national statistical agency) between May 2002 and December 2002. The target population included people aged 15 years or over, living in private occupied dwellings (98% of the population). Excluded were individuals living in health care institutions, on Indian reserves, on government- owned land, in one of the 3 northern territories, or in remote regions. Full-time members of the Armed Forces were sampled separately, and are not included in these analyses. One person aged 15 years or over was randomly selected from each eligible household. A significant effort was made to interview respondents in person at their place of residence (86% of cases). Interviews were conducted in English, French, Chinese, or Punjabi (as required or requested by the interviewee). From the initially selected 48 047 households, there was an 86.5% household-level response rate, and among responding households there was an 89.0% person-level response rate. The overall response rate was thus 77.0%, resulting in a total sample size of 36 984 respondents. In this survey, we focus on respondents aged 55 years and over (n = 12 792). Description of Variables The CCHS 1.2 interview is based on the WMH-CIDI. Well-trained lay interviewers using computer-assisted interviewing administered the survey. Study respondents were assessed for demographic variables, psychiatric diagnoses, illness history, service use data, and past 12-month medication use. We collected data on each respondent’s age, sex, and immigrant status (defined as country of birth outside of Canada). Psychiatric Disorders The CCHS 1.2 evaluated respondents for 5 disorders: major depression, bipolar disorder (as indicated by the presence of one or more manic episodes), social phobia, agoraphobia, and panic disorder. All “diagnostic” algorithms follow DSM-IV criteria, with the exception of the duration requirement for a manic episode. The CCHS 1.2 asked only whether manic symptoms lasted “several days or longer,” whereas a duration of 7 days is required by DSM-IV unless hospitalization is necessary. Each disorder was assessed for the lifetime of each respondent and for the 12 months prior to the survey. In this study, we use 12-month indicators of probable caseness. “Any mood disorder” refers here to a respondent who met the criteria for either depression or mania. “Any anxiety disorder” denotes the presence of one or more of agoraphobia, panic disorder, and social phobia. Statistical Analysis The CCHS 1.2 used a multistage, stratified cluster design to select eligible households. To correct the potential bias resulting from this complex survey design, Statistics Canada recommends bootstrapping all tests according to a set of supplied replicate weights. All results presented here were produced with this approach and are therefore representative of the targeted population. Similarly, statistics dependent on SEs (including P values and CIs) are adjusted for survey design effects. We used WesVar 4.2 to bootstrap all results. Data for this study were obtained from the CCHS 1.2 master file maintained at the Statistics Canada Research Data Centre, Toronto, Ontario. ResultsAll Participants Figure 1 shows the prevalence rates of lifetime mood disorders separately for men and women from those aged 55 years to those aged 75 years and over. There is a monotonic decrease for both men and women ranging from a prevalence of 17.62% (SE = 1.34) for women aged 55 to 59 years to 4.81% (SE = 0.59) for those aged 75 years and over and a prevalence of 12.03% (SE = 1.27) for men aged 55 to 59 years to 4.49% (SE = 0.90) for men aged 75 years and over. Until age 74 years, the prevalence is higher among women than men. Over age 74 years, the prevalence is equivalent. Table 1 shows that, for the most part, the same pattern holds for the past-year prevalence of mood disorders, lifetime and past-year anxiety disorders, and all disorders (both lifetime and past-year). Where an upturn for those in the oldest age group does occur (any disorder and past-year mood disorder for men), it is slight and not statistically significant.
Anglophone Compared With Francophone The picture is somewhat more complex when anglophones are compared with francophones. As shown in Table 2, there is a higher lifetime prevalence for any disorder, mood disorders, and anxiety for anglophone women at all ages. However, the prevalence of mood disorders after age 74 years is higher for francophone men than women, although the SEs are relatively large because of the small number of people in these groups. In fact, the sample size for older francophone men was too small to estimate the prevalence of lifetime anxiety disorder.
The lifetime prevalence rates for mood and anxiety disorders decrease monotonically for anglophones and francophones of both sexes. However, looking at all disorders, the prevalence for francophone women increases from their early 60s to their late 60s before declining again, while there is a slight upturn for the oldest francophone men. Again, the large SEs make these fluctuations nonsignificant. Overall, the lifetime prevalence of mood disorders is higher for francophones than anglophones of both sexes. For the anxiety disorders, though, the rates for rancophone and Anglophone women are very similar, Francophone men report lower rates at all ages than do their anglophone counterparts. Immigration Status Among people who immigrated to Canada after the age of 18 years, those aged 55 to 64 years had lifetime and 1-year prevalence rates for any, mood, and anxiety disorders that were roughly one-half those for people who were born in Canada. This difference decreased with age, so the rates were nearly identical for those aged 75 years and over. Within the middle group (people aged 65 to 74 years), mood disorders were less frequent among immigrants than among native-born Canadians. DiscussionThe main conclusion from this study is that, for the most part, the prevalence of anxiety disorders, mood disorders, and any psychiatric disorder measured by the CCHS 1.2 decreases linearly between the ages of 55 and 75 years and over. The same pattern holds for men and women as well as for anglophones, francophones, and allophones. This supports Wade and Cairney, who found a linear decrease between the ages of 15 and 80 years (5). It differs from Mirowsky, Ross, and others, who report an upturn in the prevalence of depression after the age of 60 years (2). Because the CCHS 1.2 is a cross-sectional study, we cannot determine whether this decline is owing to age effects, a cohort effect, or a “survivor” phenomenon in that those who are less healthy—physically or psychologically—die at a younger age. Arguing for an age effect, and against a cohort one, is the consistency in the effect over different time periods using different surveys. One hypothesis offered to account for the lack of an upturn in the prevalence at older ages is that older adults suffering from depression may be in hospitals, long-term care facilities, or other settings in disproportionate numbers and thus were not included in this survey or others with similar sampling strategies. However, according to the NPHS, which was conducted in Canada in 1996 and included individuals from long-term care facilities, only 1% of people aged 65 to 69 years were in institutions, as were 1.7% of those aged 70 to 74 years, 4% of those aged 75 to 79 years, and 18.7% of those aged 80 years and over. Counting only individuals who labelled themselves “somewhat or more unhappy” in this survey (an extremely liberal definition of dysthymia), this constituted only 1.2% of people aged 65 years and over. Even if each of them were diagnosed with depression in the CCHS 1.2, it would result in a flattening of the curve at the older ages, not an increase in the prevalence. Consequently, it does not appear as if our results were attributed to a sampling bias owing to the underrepresentation of the institutionalized elderly. Another possible reason for the lower prevalence of disorders, especially depression, is that the CCHS 1.2 included only a limited number of diagnoses. The 5 that were included, however, are the most prevalent. In the National Comorbidity Survey (25), the prevalence of these disorders among people aged 18 to 54 years was 18.1%. Adding generalized anxiety disorder, posttraumatic stress disorder, dysthymia, and nonaffective psychoses increased this to only 20.1%. Thus it is unlikely that the CCHS 1.2 seriously underestimated the prevalence of anxiety or mood disorders in this population. The fact that we found francophone-anglophone differences in the reporting of anxiety and mood disorders in this population is intriguing and worthy of further consideration. Results from the NPHS also indicated that francophones were more likely to meet the criteria for major depression than English- speaking and other Canadians. It is not readily apparent why there are differences in disorder between French- and English- speaking Canadians. However, there are significant cultural differences between these groups, making language only a proxy for other sociocultural factors that may be driving these differences. We do not have room in this paper for further speculation, other than to comment that differences in the expression of emotions among cultural groups may be one possible avenue for further consideration. The differences may also be an artifact of measurement. Although the CIDI was translated and administered in French (and various other languages), the psychometric equivalence of the translated instruments has not been evaluated. The results of this study further confirm that, despite the negative changes that accompany old age—losses owing to death and physical limitations—the elderly are psychologically healthier than younger people. Although we are unable, with these data, to test possible explanations for the decline in disorders with age (that is, aging, period, and cohorts effects), the remarkable consistency of our results with earlier work in Canada (5,23) and elsewhere (25) suggests that the observed pattern is certainly not a chance finding. The robustness of the age effect suggests efforts should now focus on why disorder declines with age. Funding and SupportThis study was supported in part by Grant Number ACC-65889 from the Canadian Institutes for Health Research. AcknowledgmentsMembers of the Aging and Depression Group: Meshak Agbayewa, John Cairney, Marsha Cohen, David Conn, Betty Lin, Lynn McCleary, Sarah Romans, David Streiner, and Vincent Thomas References1. Hocking LB, Koenig HG, Blazer DG. Epidemiology and geriatric psychiatry. In: Tsuang MT, Tohen M, Zahner GEP, editors. Textbook in psychiatric epidemiology. New York (NY): Wiley-Liss; 1995. p 437B49. 2. Mirowsky J, Ross CE. Age and depression. J Health Soc Behav 1992;33:187B205. 3. Kessler RC, Foster C, Webster PS, House JS. The relationship between age and depressive symptoms in two national surveys. Psychol Aging 1992;7:119B26. 4. Myers JK, Weissman MM, Tischler GL, Holeer CE 3rd, Leaf PJ, Orvaschel H, and others. Six-month prevalence of psychiatric disorders in three communities: 1980 to 1982. Arch Gen Psychiatry 1984;41:959B67. 5. Wade TJ, Cairney J. Age and depression in a nationally representative sample of Canadians: a preliminary look at the National Population Health Survey. Can J Public Health 1997;88:297B302. 6. Blazer DG. Clinical features in depression in old age: a case for minor depression. Curr Opin Psychiatry 1991;4:596B9. 7. Blazer DG, Hughes DC, George LK. The epidemiology of depression in an elderly community population. Gerontologist 1987;27:281B7. 8. Regier DA, Rae, DS, Narrow WE, Kaelber CT, Schatzberg AF. Prevalence of anxiety disorders and their comorbidity with mood and addictive disorders. Br J Psychiat Suppl 1998;34:24B8. 9. Bland RC, Newman SC, Orn H. Prevalence of psychiatric disorders in the elderly in Edmonton. Acta Psychiatr Scand 1988,;77:57B63. 10. Goldstrum ID, Burns BJ, Kessler LG. and others. Mental health services use by elderly adults in a primary care setting. J Gerontol 1987;17:147B53. 11. Cumming E, Henry WF. Growing old: the process of disengagement. New York (NY): Basic Books; 1961. 12. Rosow I. Status and role change through the life cycle. In Binstock RH, Shanas E, editors. Handbook of aging and the social sciences. New York (NY): Van Nostrand Reinhold; 1985. 13. Mirowsky J. Age and the sense of control. Soc Psychol Q 1995;58:31B43. 14. Rodin J. Aging and health: effects of the sense of control. Science 1986;233:1271B6. 15. Cockerham WC. The aging society. Upper Saddle River (NJ): Prentice Hall; 1997. 16. Buhler C. The curve of life as studies in biographics. J App Psychol 1935;19:405B9. 17. Dannefer D. Adult development and social theory: a paradigmatic reappraisal. Am Sociol Rev 1984;49:100B16. 18. Erikson E. Childhood and society. New York (NY): Norton; 1963. 19. Erikson, E. The life cycle completed. New York (NY): Norton; 1982. 20. Jung CG. Psyche and symbol. New York (NY): Anchor; 1958. 21. Levinson DJ, Darrow CN, Klein EB, Levenson MH, McKee B. Seasons of a man’s life. New York (NY): Knopf; 1978. 22. House JS, Lepkowski JM, Kinney AM, Mero RP, Kessler RC, Herzog AR. The social stratification of aging and health. J Health Soc Behav 1994;35:213B34. 23. Cairney J. Age, socioeconomic status and the stress process [unpublished PhD dissertation] 2002 University of Western Ontario, Canada. 24. Cairney J, Krause N. The social distribution of psychological distress and depression in older adults. Journal of Aging and Health 2005. Forthcoming. 25. Kessler RC, McGonagle KA, Zhao S, and others. Lifetime and 12-month prevalence of DSM-IIIR psychiatric disorders in the United States: results from the National Comorbidity Survey. Arch Gen Psychiatry 1994;51:8B19. 26. Regier DA, Myers JK, Kramer M, Robins LN, Blazer DG, Hough RL, and others. The NIMH Epidemiologic Catchment Area Program: historical context, major objectives, and study population characteristics. Arch Gen Psychiatry 1984;41:934B41. Author(s)Manuscript received July 2005, revised, and accepted September 2005. 1. Professor, Department of Psychiatry, University of Toronto, Toronto, Ontario; Director, Kunin-Lunenfeld Applied Research Unit, Baycrest Centre for Geriatric Care, Department of Psychiatry, University of Toronto, Toronto, Ontario. 2.Assistant Professor, Department of Psychiatry, University of Toronto, Toronto, Ontario; Researcher, Health Systems and Research Consultation Unit,Centre for Addiction and Mental Health, Department of Psychiatry, University of Toronto, Toronto, Ontario. 3. Research Assistant,Centre for Addiction and Mental Health for the Aging and Depression Group, University of Toronto, Toronto, Ontario. Address for correspondence: Dr DL Streiner, Director, Kunin-Lunenfeld Applied Research Unit, Baycrest Centre for Geriatric Care, 3560 Bathurst Street, Toronto, Ontario, M6A 2E1 e-mail: dstreiner@klaru-baycrest.on.ca
1 | 2
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||