Recent years have seen much debate about both the prevalence and the nature of anxiety and depression in older adults. On the one hand, some authors have suggested that older populations are characterized by surprisingly high levels of well-being and resilience, despite increasing losses and functional impairment (Staudinger and Fleeson, Reference Staudinger and Fleeson1996) and that the prevalence of mental illness, with the exception of dementia, decreases in late life (Jorm, Reference Jorm2000). Others have suggested that this is a spurious finding resulting from the methodological problems in obtaining accurate data for older adults (Beekman et al., Reference Beekman1998, Krasucki et al., Reference Krasucki, Howard and Mann1999; O'Connor, Reference O'Connor2006), with categorical diagnostic systems, such as the DSM-IV (American Psychiatric Association, 1994) and ICD-10 (World Health Organization, 1992) frequently cited as aggravating these difficulties (Palmer et al., Reference Palmer, Jeste and Sheikh1997).
The recognition of depression and anxiety
This guest editorial focuses on the challenges in recognizing and diagnosing anxiety disorders and depression in older populations. Problematic though the DSM-IV criteria are, they are not the only challenge to the recognition and accurate diagnosis of these disorders in older adults. Some of the difficulty lies in issues that are unique to older people. First, older adults may be less likely to report symptoms of depression or anxiety. This is likely to be a cohort effect, rather than related to aging per se: it has been suggested that the current cohort of older adults may be less comfortable than later generations in discussing emotions, and they tend to take a somewhat stoic attitude to life's vicissitudes (Pachana, Reference Pachana and Rieger2008). Secondly, it may be easier for older adults to avoid situations that evoke anxiety. For example, it may be more acceptable for an older adult who experiences a marked and persistent fear of social situations (Criterion A of the DSM-IV criteria for the diagnosis of Social Phobia) to avoid such occasions, as age-related stereotypes may reinforce the withdrawal of older people from social interaction. Such a person may, in fact, experience significant interference with social activities (Criterion E for the diagnosis of Social Phobia), but not come to clinical attention.
Thirdly, older people may also be less aware of the treatments available for anxiety or depression, particularly psychotherapeutic approaches (Woodward and Pachana, Reference Woodward and Pachana2009). This problem is aggravated by the shortage of psychologists willing to work with this population (Lee et al., Reference Lee, Volans and Gregory2003). Furthermore, there is evidence that general practitioners (GPs – also known as primary care physicians) may be less likely to enquire about psychological symptoms than physical symptoms in their older patients (Collins et al., Reference Collins1997). Finally, the language used by older adults may be different from that used by clinicians, with all these factors having the effect of reducing the recognition of depression and anxiety in older adults.
Quantitative or qualitative differences in the presentation of anxiety and depression?
Differences between younger and older adults in the presentation of depression are well documented, but there has been debate about whether these differences are quantitative or qualitative. Certainly, it does appear that older adults tend to experience symptoms too infrequently or in insufficient numbers to meet the DSM-IV (American Psychiatric Association, 1994) criteria for major depression. This has given rise to a number of terms to describe this type of “non-major” depression, such as “subsyndromal” (Flint, Reference Flint2002), “subthreshold” (Judd and Akiskal, Reference Judd and Akiskal2002) or “clinically significant non-major depression” (Lavretsky and Kumar, Reference Lavretsky and Kumar2002). While these constructs have been useful in stimulating debate about depression that does not meet standard criteria (Jeste et al., Reference Jeste, Blazer and First2005), they have also been criticized on a number of grounds. First, criteria for non-major depression are inconsistent: Judd and Akiskal (Reference Judd and Akiskal2002) defined subthreshold depression as having two or more simultaneous symptoms of depression present for most or all of the time for at least two weeks, while Lavretsky and Kumar (Reference Lavretsky and Kumar2002) stated that the term emerged to classify patients with fewer than five clinically significant symptoms.
Perhaps of greater importance is that terms such as “non-major” depression might suggest that this type of depression is not clinically important, which does not appear to be the case. A study by Chachamovich et al. (Reference Chachamovich, Fleck, Laidlaw and Power2008) assessed 4,316 respondents in 20 countries using the Geriatric Depression Scale (GDS; Yesavage et al., Reference Yesavage, Brink, Rose, Lum, Huang, Adey and Leirer1983) as a measure of depressive symptoms, and found that a GDS score as low as 2 was a predictor of lower quality of life and a more negative attitude toward aging. Moreover, depression was a stronger predictor of quality of life than other variables such as age, gender, marital status or educational level. Data from the Berlin Ageing Study have also confirmed the impact of what was termed “subthreshold depression” in old age on the ability to carry out activities of daily living (Wilms et al., Reference Wilms, Kanowski and Baltes2000).
Turning to qualitative differences between depression in older and younger age groups, Blazer (Reference Blazer2003) has suggested that a syndrome of “depression without sadness” may be more common in older adults. This is characterized by apathy and loss of interest, rather than subjectively experienced low mood. While not unique to older adults, it has been suggested that this presentation is more common in older than younger adults (Gallo et al., Reference Gallo, Rabins, Lyketsos, Tien and Anthony1997). Another term used is “depletion syndrome”, referring to a “depression-like” syndrome experienced by older adults and characterized by a preponderance of physical symptoms and withdrawal (Newman et al., Reference Newmann, Engel and Jensen1991).
Other unique features of the presentation of anxiety and depression in old age include the high level of comorbidity with physical illness and the well-documented difficulties that can arise in distinguishing psychological from physical symptoms (Koenig, Reference Koenig1998; Bryant et al., Reference Bryant, Jackson and Ames2009). Comorbidity with cognitive impairment and dementia also complicate the picture. Finally, unlike depression in younger age groups, it appears that clear etiological factors can be identified, at least for a subgroup of geriatric depressions. It is generally accepted that a valid distinction on the basis of etiology can be made between early (generally defined as before the age of 60) and late onset depression (Conway and Steffens, Reference Conway and Steffens1999; Bagulho, Reference Bagulho2002). Bagulho (Reference Bagulho2002) stated that the only well-replicated differences between early and late onset depression were indeed causative, with the underlying cause attributed to degenerative brain changes characterized by deep white matter hyperintensities and other vascular changes.
Thus, different lines of evidence converge to suggest that depression and anxiety are increasingly heterogeneous in old age, and are both experienced and expressed in ways that are different from their counterparts in younger populations. How then, do categorical diagnostic systems, such as ICD-10 and DSM-IV cope with these presentations that may be not only phenomenonologically but also etiologically diverse?
An answer to this question can be given by examining the current diagnostic criteria in DSM-IV, using Generalized Anxiety Disorder (GAD) as an illustrative example. This disorder is defined as being worry that is deemed to be excessive (Criterion A), difficult to control (Criterion B), and is associated with three or more (one in children) physiological symptoms (Criterion C), and causes impairment in social, occupational or other areas of functioning (Criterion D) (italics mine). All of these criteria are potentially problematic in older adults. First, many older people deny that worries are excessive, saying, for example, that a fear of going out at night is justifiable in the context of visual impairment. The physiological symptoms listed under Criterion C include fatigue, tension and poor sleep, for which there could be health-related causes, while difficulty concentrating could be attributable to age- or dementia-related cognitive impairments. Finally, clinicians may be less inclined to pursue enquiries regarding the functional consequences of the symptoms, in light of the examples given in the manual (“social and occupational”).
Possible ways forward
Even before work on DSM-V started formally in 2007, there was much discussion about the validity of the DSM-IV criteria for anxiety and depression in older adults, and as the pace of work on DSM-V accelerates towards the anticipated publication date of 2012, there will certainly be an increased level of debate about this issue. One approach to the problems outlined is to improve the existing framework without making radical changes. Thus, a number of authors have called for criteria for the DSM criteria for anxiety (Beekman, Reference Beekman2008) or depression (Jeste et al., Reference Jeste, Blazer and First2005) to be modified and made more sensitive to the needs of older patients.
Others have gone further and highlighted the inherent problems of a categorical system that imposes arbitrary boundaries to define case or non-case status (Flint, Reference Flint2002). Beekman (Reference Beekman2008) suggested that it may be more clinically useful, and consistent with empirical research, to lump disorders into broader categories, but split symptoms at a dimensional level. Flint (Reference Flint2002) has also advocated the use of a dimensional approach to the study of affective disorder in old age as this can “provide a framework for studying the complex interactions that exist between depression, anxiety, medical illness, cognitive impairment, personality and adverse psychosocial conditions” (Flint, Reference Flint2002, p. 231).
At the very least, a case can be made for making the text more age-friendly. Why, for example, are there currently two paragraphs of caveats with regard to the diagnosis of GAD in children, and no consideration of issues to guide clinicians in making accurate diagnoses in older adults? William Osler famously stated that “It is much more important to know what sort of a patient has a disease than what sort of a disease a patient has”. We would do well to remember the importance of the “sort of patient” in relation to developing diagnostic criteria for anxiety and depression in older adults, so that these take into account the complexities of the presentation of these disorders in this age group.
Conflict of interest declaration
Christina Bryant is a member of the DSM-V Late Life Anxiety Disorders Advisory Group.
