Read an Excerpt
Handbook of Assessment in Clinical Gerontology
By Peter A. Lichtenberg
Copyright © 2010 Elsevier Inc.
All right reserved.
Chapter One Assessment of Depression and Bereavement in Older Adults
Barry A. Edelstein, Lisa W. Drozdick, Caroline M. Ciliberti Department of Psychology, West Virginia University Morgantown, WV, USA, Clinical Assessment, Pearson, San Antonio, TX, USA
This chapter addresses the assessment of older adult depression and bereavement. The assessment of depression in older adults can be complicated due to age-related differences in the presentation of depression, comorbid medical and mental health problems, and age-related changes in cognitive functioning. Moreover, available assessment instruments may have less utility with older adults, either because they were developed with younger adults, or because they were developed to meet diagnostic criteria that may not be appropriate for older adults (see Jeste, Blazer, & First, 2005). Consequently, clinicians may be failing to identify depression adequately in older adults and to identify and treat older adults with subsyndromal or minor depression, which involves considerable disability but is not formally recognized as a clinical disorder.
This chapter addresses both depression and bereavement because loss is often a significant contributor to and risk factor for depression, and adults face increasing losses as they move through older adulthood. Depression is a normal response to a significant loss. The depression can last for a considerable amount of time and be functionally debilitating. Bereavement is one of the more significant risk factors for the first onset of depression and recurrent depression in older adults (Bruce, 2002). In light of the clinical significance of bereavement, its increasing likelihood over the lifespan, and the paucity of assessment literature addressing the topic, we have included the assessment of bereavement in this discussion of late-life depression assessment.
EPIDEMIOLOGY OF LATE-LIFE DEPRESSION
Symptoms of depression tend to be approximately as prevalent in late life as in mid-life (Blazer, 2003). The frequency of depressive symptoms among the oldest old appears higher than among younger adults, although factors other than age (e.g., greater proportion of women, increased cognitive impairment, lower socioeconomic status, greater physical disability) may account for the difference (Blazer, 2003). The prevalence of clinically significant symptoms of depression ranges from 8–16% among community-dwelling older adults (Blazer, 2003). The prevalence of major depression in community-dwelling older adults ranges from approximately 1–4% (Beekman, Copeland, & Prince, 1999). Prevalence estimates for minor depression among community-dwelling older adults range from approximately 4–13%, with the highest estimate found in the Netherlands (Beekman et al., 1995). With minor and major depression combined, Steffens, Fisher, Langa, Potter, and Plassman (2009) found an overall prevalence of 11.19%, with the prevalence being similar for community-dwelling older men and women.
Prevalence estimates of major depression vary across settings, with increases in prevalence as one moves from outpatient to inpatient settings. The prevalence of major depression among older adults seen in primary care settings ranges from 5–10% (Lyness et al., 2002; Schulberg et al., 1998). Among hospitalized older adults, prevalence rates of major depression range from 10–12% (Blazer, 1994; Koenig, Meador, Cohen, & Blazer, 1988). Prevalence estimates for major depression among long-term care residents are even higher, ranging from 12.4% to 14.4% (Parmalee, Katz, & Lawton, 1989; Teresi, Abrams, Holmes, Ramirez, & Eimicke, 2001).
These epidemiological findings must be tempered by the questionable adequacy of our current diagnostic system for older adults (see discussion below) and the finding of different relations between age and depression across studies. Researchers have noted varied relations between age and depression, including negative linear, curvilinear, and positive linear relations (Nguyen & Zonderman, 2006). Nguyen and Zonderman suggest that the differences in relations can be attributed, in part, to the nature of the assessment measures employed. Measures of depressive symptoms reveal a negative linear relation or positive curvilinear relation. Such relations suggest fewer symptoms of depression as one ages, or increased symptoms among younger and older adults when compared with an intermediate age group. The authors note that when diagnostic measures of major depression are used, there tends to be a positive linear or negative curvilinear relation between age and depression. Thus, major depression increases with age, or is lower among younger and older groups when compared with an intermediate age group.
CONCEPTUAL APPROACHES TO ASSESSMENT
The assessment paradigm employed by the clinician determines the assessment methods and instruments employed, the questions addressed, and the integration and use of the assessment results (Edelstein, Martin, & Koven, 2003; Edelstein & Koven, in press). Haynes and O'Brien (2000) have defined an assessment paradigm as "a set of principles, beliefs, values, hypotheses, and methods advocated in an assessment discipline or by its adherents" (p. 10). Two conceptually distinct paradigms are the traditional (e.g., trait-oriented, psychodynamic) and the behavioral (e.g., behavior—analytic, cognitive—behavioral). One can distinguish between traditional and behavioral paradigms through an examination of how each explains or accounts for behavior. More traditional approaches tend to emphasize an individual's dispositional characteristics (see Mischel, 1968) or hypothetical constructs (e.g., anxiety, depression), which are inferred from the individual's self-reports and observed behavior (Edelstein, Woodhead, Bower, & Lowery, 2006). Such approaches to psychopathology often include exploration of an individual's feelings or affective states.
Behavioral approaches tend to be more contextual and emphasize descriptions of environmental conditions under which the behavior of interest is more or less likely to occur. A behavioral account of an individual's behavior involves a description of the conditions under which the behavior occurs (see Edelstein & Koven, in press). More emphasis is placed on the variables controlling the behavior of interest, and less emphasis is placed on characteristics of the individual. It is important to note that behavioral approaches do not discount the role of cognitions or private events; however, they do not consider cognitions to have causal efficacy. Cognitions are treated as any other behavior, whether observable or not. For the purposes of the present discussion, emphasis is placed on behavioral assessment that relies primarily on direct observation of overt behaviors.
One might also distinguish between traditional and behavioral approaches by considering the distinction between nomothetic and idiographic approaches to personality assessment (see Allport, 1936). Traditional approaches are more closely aligned with a nomothetic approach, which involves an examination of the commonalities among individuals. This approach underlies classification systems such as the Diagnostic and Statistical Manual—Fourth Edition (DSM-IV) (American Psychiatric Association, 1994). In contrast, behavioral approaches are more similar to the idiographic approach, which is used to ascertain the uniqueness of an individual.
Traditional and behavioral approaches and instruments are often combined. For example, one might administer a self-report depression inventory and examine the individual item responses to gain an individualized understanding of the individual's mood. The total score on the instrument may also be compared with that of a normative sample to enable one to make a judgment about whether the individual's score is below or above a cutoff score that signals a clinical level of depression. Each approach has its strengths and weaknesses.
As one moves from cognitively intact to cognitively impaired older adults, one must rely more on the direct observation of behavior, because the self-reported experiences of depression become unreliable and/or invalid, and eventually unavailable as cognitive skills diminish. Nomothetic assessment instruments must now depend upon the inferences of clinicians based on direct observations of the older adult. Mood and other symptoms must now be inferred from overt behavior. The question of why an individual is reporting particular experiences and engaging in particular behaviors can no longer be answered by questioning that individual. The nomothetic and idiographic assessment methods tend to converge on the observation of behavior.
DEFINITION AND DIAGNOSTIC ISSUES
Jeste et al. (2005) cogently argued that age-appropriate diagnostic criteria are needed for the major DSM psychiatric diagnoses. This issue is particularly salient for the diagnosis of late-life depression, as older adults often present a different array or profile of symptoms than younger adults (Caine, Lyness, King, & Connors, 1994; Fiske & O'Riley, 2008). For example, older adults are less likely than younger adults to report suicidal ideation (Blazer, Bachar, & Hughes, 1987), guilt (Gallo, Rabins, & Anthony, 1999; Musetti et al., 1989; Wallace & Pfohl, 1995), and dysphoria (Gallo, Anthony, & Muthen, 1994; Gallo et al., 1999). In contrast, older adults are more likely than younger adults to report hopelessness and helplessness (Christensen et al., 1999), somatic symptoms (Gallo et al., 1994), psychomotor retardation (Gallo et al., 1994), weight loss (Blazer et al., 1987), and loss of appetite (Blazer et al., 1987).
The issue of whether somatic symptoms should be considered among the diagnostic criteria for depression among older adults has been somewhat controversial (see Norris, Snow-Turek, & Blankenship, 1995), in part because of the overlap of symptoms of physical disease and somatic symptoms of depression (e.g., low energy, sleep disturbance, diminished appetite and sexual drive). The frequency and severity of medical conditions increase with age and can lead to many of the somatic symptoms included in the diagnosis of depression (e.g., weight loss or gain, insomnia, fatigue). Moreover, depression is frequently comorbid with physical illness and cognitive dysfunction, both of which increase with age.
Several studies reported increased endorsement of somatic symptoms of depression with increasing age and suggest removing somatic symptoms from self-report measures of depression (e.g., Barefoot, Mortensen, Helms, Avlund, & Schroll, 2001; Berry, Storandt, & Coyne, 1984; Bolla-Wilson & Bleecker, 1989; Goldberg, Breckenridge, & Sheikh, 2003; Mahurin & Gatz, 1983). However, many studies suggest that removing somatic items from assessment instruments may result in decreased sensitivity to depression in older adults (Drayer et al., 2005; Kirmayer, 2001; Norris, Arnau, Bramson, & Meagher, 2004). Moreover, somatic symptoms cannot always be attributed to physical disease (Gatz & Hurwicz, 1990; Olin, Schneider, Eaton, Zemansky, & Pollock, 1992; Wagle, Ho, Wagle, & Berrios, 2000). There is evidence to suggest that while changes in appetite and sexual drive may not be indicative of depression among older adults, the remaining somatic symptoms are indicative (Nguyen & Zonderman, 2006; Norris et al., 2004). Clinicians should consider assessing somatic symptoms of depression, although caution should be used when interpreting results obtained in individuals for whom medical issues may be contributing to results.
Several other factors can complicate the assessment of late-life depression, including the onset of symptoms. The time of onset of the first depressive episode of major depression may be related to the nature of depression symptoms (Jeste et al., 2005). The symptoms of first-onset, late-life depression (after age 60) may be different from depression that occurs early in life (before age 60) and recurs in late life (Brodaty et al., 2001).
Older adults can experience symptoms of depression that do not meet criteria for a depression diagnosis, yet are associated with psychosocial and functional impairment similar to that associated with major depression (Beekman et al., 1995; Hybels, Blazer, & Pieper, 2001; Lavretsky, Kurbanyan, & Kumar, 2004). Various authors have argued that depression should be conceptualized along a continuum of severity (e.g., Rapaport et al., 2002), with major depressive disorder at one end, subsyndromal depression at the other end, and minor depression in the middle (e.g., Hybels et al., 2001; Lavretksy et al., 2004). Although subsyndromal depression is not currently classified, minor depression appears in the appendix of DSM-IV. Subthreshold depressions are of particular importance for older adults, as their prevalence increases with age (Judd, Schettler, & Akiskal, 2002). Moreover, in an examination of older primary care patients, Lyness, King, Cox, Yoediono, and Caine (1999) found that the prevalence of subsyndromal depression exceeded that of major depression, minor depression, and dysthymia.
There are two additional presentations of depression that do not meet criteria for major depression, dysthymia, or minor depression, and are thought to be more common in older adults. The first is termed "depression without sadness," (Gallo, Rabins, Lyketsos, Tien, & Anthony, 1997) in which individuals present with symptoms of depression (e.g., hopelessness, worthlessness, thoughts of death or suicide) but do not report sadness or loss of interest or pleasure in formerly enjoyed activities. Even though these individuals fail to meet DSM-IV criteria for depression, they are at risk for functional disability, psychological distress, cognitive impairment, and death (Gallo & Rabins, 1999). Similarly, Newmann, Engel, and Jensen (1991) characterized a "depletion syndrome" with symptoms of loss of appetite, lack of interest, thoughts of dying, and hopelessness (see also Adams, 2001).
Olin, Katz, Meyers, Schneider, and Lebowitz (2002) have argued that the depression that occurs with Alzheimer's disease is different from other depressive disorders and have proposed provisional criteria for "depression of Alzheimer's disease." They suggest that the depression observed with Alzheimer's patients is different from "depression due to a general medical condition." The outcome of the authors' proposal remains to be seen. Mayer et al. (2006) compared three rating scales for use as outcome measures in treatment trials of "depression of Alzheimer's disease." The Cornell Scale for Depression in Dementia (CSDD) (Alexopoulos, Abrams, Young, & Shamoian, 1988), particularly the mood subscale, appeared to be the best choice for measuring the effects of treatment. The CSDD is discussed later in this chapter.
The racial and ethnic diversity of older adults in the United States is expected to increase with the growing population (U.S. Census Bureau, 2008a), with minority populations expected to become the majority in 2042. This is particularly important to appreciate, as prevalence estimates of depression vary across racial and ethnic groups. Some authors (e.g., Mui, Burnette, & Chen, 2002) have argued that the prevalence data for some racial and ethnic groups are biased due to the low acceptability of measures used to report symptoms in minority populations, and various socio—cultural factors (e.g., tolerance of symptoms). Moreover, norms are often not available for racial and ethnic groups, and psychometric properties are often not available for racial and ethnic groups on assessment instruments that were developed with Caucasian samples. Moreover, there appear to be cross-cultural differences in the expression of depression symptoms (Futterman, Thompson, Gallagher-Thompson, & Ferris, 1997). For example, Japanese individuals tend to report interpersonal complaints, whereas Chinese individuals tend to present somatic symptoms (Krause & Liang, 1992). However, whether these cultural differences are exhibited through old age is unclear at this time. Readers are cautioned to carefully consider the available normative data for racially or ethnically diverse clients, to avoid stereotypes when considering culturally specific information in the assessment process. Interested readers are referred to Sue and Sue (2007), who provide helpful information on working with racially and ethnically diverse populations. Many of their suggestions are particularly useful for understanding perspectives on mental health care by first generation immigrants, and those who maintain a strong cultural identity.
Excerpted from Handbook of Assessment in Clinical Gerontology by Peter A. Lichtenberg Copyright © 2010 by Elsevier Inc.. Excerpted by permission of Academic Press. All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.