Psychiatry and the Oldest Old

Geriatric psychiatry came of age during the past two decades in the United States (1). The American Board of Psychiatry and Neurology now offers a certificate of added qualifications in the specialty, and multiple journals have emerged (such as the American Journal of Geriatric Psychiatry) devoted specifically to old age psychiatry. Nevertheless, the theoretical constructs and empirical study of psychiatric problems among the oldest old have lagged behind the field in general. Most research efforts focus on specific disease entities, such as Alzheimer’s disease or major depression, seeking pure types of the disease for phenomenological, epidemiologic, and biomedical investigation. Treatment studies recruit clearly delineated cases among the young old, as illustrated by the investigations from the Pittsburgh group (2) into the effectiveness of a combination of psychotherapy and medication for the treatment of major depression (2).

In this article, I focus on geriatric syndromes (as opposed to specific psychiatric disorders) and the potential of these geriatric syndromes to lead to functional impairment, frailty, and ultimately failure to thrive. I propose that psychiatrists who treat the oldest old rejoin our colleagues in geriatric medicine by emphasizing health-related quality of life (specifically functional status) and a comprehensive, interdisciplinary approach to assessment and management of psychiatric disorders.

Neglect of the Oldest Old in Geriatric Psychiatry Investigations

The oldest old are neglected in geriatric psychiatric investigations for a number of reasons. First, they are less available for study except in institutional settings. Ambulatory-based clinical trials are extremely difficult to field for people 85 years old or older compared with those 65–74 years old, and the subjects of such studies are not representative of the majority of the oldest old. In addition, recruitment from institutional settings often conflicts with busy ward routines and transfers from one setting to another (3). Second, multiple coexisting diseases, both physical and psychiatric, often render the study of pure forms of a disease impossible. One answer to the question, Who is the typical geriatric psychiatric patient? is, Think of your oldest, sickest, most complicated and frail patient (4), not the patient usually included in geriatric psychiatric investigations, especially clinical trials. Finally, the boundaries of psychiatry and medicine become inextricably blurred at the most advanced ages. Despite the variability in function across the oldest old, once people at an advanced age experience a serious medical and/or psychiatric illness, physical, psychiatric, and social impairments coalesce and cascade, often resulting in the condition described by geriatricians as frailty (5).

The neglect of the oldest old in clinical studies, however, does not mean that geriatric psychiatrists ignore these patients. The assessment and management of the oldest old in hospitals and long-term care settings is the core of the practice of geriatric psychiatry.

Who Are the Oldest Old?

Demographics

In the United States, the term “oldest old” was coined in 1984 to highlight the fact that, as a group, individuals 85 years old or older (although 80 years of age has also be suggested as a threshold), identified by demographers as growing far faster than previous projections, would dramatically increase the burden on Medicare and Medicaid. The oldest old are the fastest growing age group in the United States; approximately 2.9% of the population was 80 years old or older in 1993, and this figure is projected to be 4.3% in 2025 (6). Life expectancy at age 65 is 15.5 years for men and 19.1 years for women, so reaching age 80 is becoming the usual rather than the exceptional. Compared with the “young old,” the oldest old are more likely to be female, to experience more poverty, to have less education, and to receive far more federal transfer payments. Over 50% of nursing home residents in the United States are 80 years old or older, representing a cost of $30 billion per year (7).

Functional Status and Quality of Life

Understanding of the physical and psychiatric well-being of the oldest old requires a broader concept of health. Patrick and Erickson (8), among others, have described this concept as health-related quality of life, which they defined as “the value assigned to duration of life as modified by the impairments, functional states, perceptions and social opportunities that are influenced by disease, injury, treatment or policy.” The core concepts (and metrics) of health-related quality of life include symptoms and subjective complaints (e.g., depression rating scales); diagnoses (e.g., coronary heart disease); physiological function (e.g., blood pressure); psychological and cognitive function (e.g., performance on the Mini-Mental State [9]); physical function (e.g., activity restrictions and impairment in self-care); general perceptions of health (e.g., self-rated health); and social function (e.g., nature and extent of social interaction).

Function across many but not all measures declines with increasing age, yet there is wide variation. Most people over the age of 85 maintain independent function. For example, in a community sample of adults 85 years old or older in East Boston (not including residents of long-term care facilities) (10), 71% of the men and 55% of the women could bathe without assistance, 82% of the men and 76% of the women could dress without assistance, 74% of the men and 59% of the women could walk without assistance, and 87% of the men and 80% of the women could use the toilet without assistance. In a racially mixed sample in North Carolina using the same measures (11), African American men generally performed better than white men and African American women performed less well than white women. The oldest old who are living independently in the community, regardless of race and socioeconomic status, are usually just that—independent.

Projections of life expectancy and functional independence into the twenty-first century for the oldest old are encouraging. In a widely cited and debated study, demographers Manton and Vaupel (12) reported that, for people 80 years old or older, life expectancy is greater in the United States than in Sweden, France, England, and Japan and is increasing. They suggested the reason is that elderly Americans are receiving better health care than elderly citizens of other developed countries (the reader can readily understand why the report is controversial). Singer and Manton (13) found an accelerating decline in disability from 1982 to 1994 among the oldest old (this age group is becoming more independent as we enter the twenty-first century). They went so far as to suggest that, if function continues to improve at the current rate for the oldest old, the support ratio (the ratio of economically active persons aged 20–64 to the number of chronically disabled persons 65 years old or older) will remain at or above its 1994 value of 22:1 (when the Hospital Insurance Trust Fund was in fiscal balance) until year 2070, despite the rising proportion of the oldest old. If accurate, this is great news for Social Security and Medicare.

Other Markers of Health and Well-Being

Self-rated health among the oldest old also varies but is generally good. In one study (14), 70% of individuals 80 years old or older rated their overall health as good or very good, although physical complaints were frequent. For example, the frequency of mild to severe pain has been estimated at 86% in people 85 years old or older (15). Although the perception of health may generally be good, that perception does not necessarily translate into health-promoting behaviors. In a health promotion survey, the oldest old were less likely to seek cancer screening, less likely to exercise, and less likely to comply with a low cholesterol diet. When asked about their behavior, sample respondents stated that they had lived a sufficient period of time and were no longer interested in disease prevention and the prolongation of life. They were more interested in the health-related quality of their lives (16). Clinicians must therefore take care not to interpret less interest in disease prevention as an inevitable reflection of depression or other psychopathological processes in this age group. They must also recognize that health-related quality of life for the oldest old is a key factor determining their desire for a prolonged life.

Cognitive function also varies widely among the oldest old. Compared with younger persons, mean declines in a number of functions have been documented. When education was controlled for, it was shown that nondemented people 85 years old or older perform less well than persons 65–84 years old on verbal and nonverbal memory, psychomotor/executive tasks, and category verbal fluency (17, 18). In contrast, there usually are no significant differences between the oldest old and younger subjects on tests of language, such as the Boston Naming Test, and tests of reading (17, 18).

Personality remains remarkably consistent into advanced age. One study of 150 of the oldest old (19) revealed that almost all respondents thought they were still essentially the same person they were in their middle years. Perceived continuity was related to a positive affect but not to either recent disruptive events or mortality. One expression of this stability is the finding that for most of the oldest old there is a growing coherence of their life story (20). Nevertheless, this stability of personality and coherence of a life story can be disrupted by stressful life events, especially physical health impairment. Whereas the feeling that one can master one’s health in late life enhances continuity, the loss of health (or the loss of a spouse’s health) dramatically interrupts the sense of coherence (21).

Living Arrangements

A study conducted in East Boston (22) found that 31% of men and 53% of women 85 years old or older residing in the community lived alone and that another 46% and 7%, respectively, lived with a spouse. Over 80% of these elders maintained at least monthly contact with one or more of their children. Although only about 5% of individuals 65 years old or older are living in nursing homes at any given time, over 20% of those 85 years old or older live in nursing homes (23). For those who remain in the community, their participation in outside activities decreases by about 33% from the late 60s to the late 80s. For example, in East Boston the participation by men in religious meetings at least once a week dropped from 28% to 17% and for women from 43% to 13% (22).

The oldest old exhibit wide variability in function. Most continue to live and function independently, and some remain independent well into their 90s. Centenarians, for example, number more than 100,000 worldwide, and many function independently. Studies in Sweden (24, 25) found that only 27% of centenarians were diagnosed with dementia, 25% were living in their own home, and 52% managed their activities of daily living with little if any assistance. Chronological age is a poor indicator of biological, psychological, social, and functional aging (and therefore health-related quality of life) for any given elder despite robust correlations between age and function overall.

A common vulnerability to a cascade of physical and psychiatric problems with the onset of any single serious physical or psychiatric illness underlies the variability, however. This vulnerability is reflected in the increased likelihood of comorbidity, the association of psychiatric illness with decline in multiple functions and health-related quality of life, and frailty, a syndrome observed all too frequently in the oldest old.

The Pathway to Frailty

Comorbidity

Comorbidity is the rule rather than the exception in late life and is tied inextricably to functional decline across multiple measures with increasing age. Comorbidity of depression with a medical illness, for example, is associated with poorer physical, mental, and social functioning (the ingredients of a substandard quality of life) than depression only or physical illness only in all age groups (26). Not only does the frequency of comorbidity increase with age, the impact of multiple conditions has a more adverse effect with advancing age.

Data from the National Health Interview Survey (27) revealed that, for individuals 70 years old or older, the rates of new confinement to bed or chair for those with two or more chronic diseases (such as arthritis, heart disease, stroke, diabetes, cancer, cataracts, and broken hip) are only slightly higher after 2 years than for all individuals 70 years old or older (24% versus 17%). Among individuals 80 years old or older, however, 42% of those with two or more chronic diseases were newly confined to a bed or chair, a significantly higher figure than that for all individuals 80 years old or older. Comorbid psychiatric disorders are also frequent. For example, among people with Alzheimer’s disease, between 15% and 29% meet criteria for major depression and 40% altogether experience at least mild depression (28–30).

Two large groups of people with comorbid illnesses among the oldest old make up the vast majority of those treated by psychiatrists. The first is those who experience comorbid depression and dementia (usually Alzheimer’s disease and/or vascular dementia). The association between depression and vascular lesions in the brain has led to the description of vascular depression as a new disease (31). The second largest group of elders with comorbid illnesses is composed of those who experience comorbid depression and physical illness; these patients are well described in studies of depression and functional impairment (32). Comorbidity is the rule rather than the exception among the oldest old and is a prime risk for a spiral downward to frailty.

The impact of psychiatric symptoms on physical function has been well described in the literature, especially for depression (33–35). In a recent report from a large community survey of older adults (32), depression increased the risk for activities of daily living disability and mobility disability over 6 years by 67% and 73%, respectively, in controlled analyses. Less physical activity and fewer social contacts among depressed individuals further explains the increased disability risk. Physical immobility and social isolation in turn increase disability over time, which further decreases mobility and social interaction, placing the person at risk for physical, psychological, and social impairment.

Frailty and Failure to Thrive

The end result of this downward spiral in health-related quality of life is the syndrome described by geriatricians as frailty and failure to thrive. Some estimate that, after the age of 85, nearly half of those living in the community are frail despite their apparent functional well-being (27, 36, 37). The manifestation of frailty most agreed upon is a constellation of weight loss, weakness, fatigue, inactivity, decreased food intake, and depression. Physical signs that accompany these symptoms include sarcopenia, balance and gait abnormalities, deconditioning, and decreased bone mass (27, 37). Failure to thrive specifies the end stage of frailty and is marked by continued and unchecked weight loss, severe muscle wasting, apathetic depression, and a host of physiological abnormalities, including hypoalbuminemia, low creatinine, anemia, decubitus ulcers, and ultimately death.

The following case report illustrates the vicious cycle downward in health-related quality of life to frailty precipitated by an acute depressive episode (23).

A Comprehensive, Functional Approach to Assessment

Geriatric Syndromes

The coincidence of multiple problems (comorbidity), as illustrated in the case report, often produces a blurring of diagnostic categories, nonspecific presentations, and courses of psychiatric disorders in the oldest old (4). Attempting to fit the oldest old into the Procrustean bed of our usual diagnostic categories is often of little benefit when developing a treatment plan. For this reason, geriatricians and geriatric psychiatrists have defined a series of specific symptom constructs known as geriatric syndromes (38). An emphasis on syndromes in the treatment of the oldest old takes into account multiple causes of a cluster of symptoms, the functional impairment that accompanies these symptoms, and the need for a multimodal approach to improve health-related quality of life and prevent frailty. In geriatric medicine, emphasis on syndromes has created a paradigm shift in the management of incontinence, falls, constipation, infection, and immobility. This paradigm shift has advanced evidenced-based management strategies and the development of realistic quantitative outcome measures as well as estimates of cost effectiveness.

Incontinence of urine is a useful example. The syndrome of incontinence is common; 15%–30% of older adults are incontinent of urine (39). Even though incontinence may not be cured, it can almost always be managed so that the elder is more comfortable, the spiral of functional decline is interrupted, and the patient’s health-related quality of life is improved. Caretakers are also less burdened when the patient improves. The etiology of incontinence is complex in the oldest old, including coexisting urological (e.g., bladder capacity), neurological (e.g., impaired bladder contractility), psychological (e.g., dementia), and functional (e.g., fecal impaction) causes (40).

Diagnostic evaluation of urinary incontinence, therefore, is focused not only on obtaining a comprehensive history and reviewing the most frequent causative factors but also on assessing function. Simple urodynamic tests (e.g., observation of voiding and the cough test for stress incontinence) as well as more complex tests (e.g., urine flowmetry) document the nature and extent of the functional impairment. In addition, a careful history is taken to assess the impact of the incontinence on the daily activities of the patient and her or his family. The critical feature of the syndromal approach to incontinence, however, is the implementation of a multimodal treatment plan focused on improving function (in this case, restoring continence).

For example, stress incontinence is treated with pelvic muscle exercises, α-adrenergic agonists, estrogen, and, at times, periurethral injections. Outcome is assessed not in terms of a particular physiological measure but, rather, in terms of the frequency of incontinent episodes per day and the quality of life of the incontinent elder.

I have suggested previously (38, 41) that the geriatric syndromes should include the psychiatric syndromes of acute confusion, memory loss, insomnia, suspiciousness and agitation, hypochondriasis, anxiety, and depression. As with incontinence, multiple causes should be sought for each of these syndromes in the oldest old, assessment should focus especially on function, and management should be directed to improving function and health-related quality of life as well as relieving symptoms. Some symptoms cannot be relieved in the oldest old, but the elder’s health-related quality of life can be enhanced by comprehensive assessment of the geriatric syndrome and implementation of an multimodal approach to management. In addition, multiple intervention strategies can break the vicious cycle downward through frailty to failure to thrive.

Comprehensive Geriatric Assessment

The complexity of medical and psychiatric problems among the oldest old, coupled with the vulnerability to frailty and failure to thrive, prompted the development of comprehensive geriatric assessment (42–45). Comprehensive geriatric assessment was instituted with the notion that a systematic evaluation of vulnerable or frail older persons by a team of health professionals could uncover treatable health problems and functional impairment. The components of comprehensive geriatric assessment include 1) identifying the patient at risk through careful screening; 2) comprehensive assessment in a geriatric care context; 3) focus on function and quality of life; 4) development and implementation of a comprehensive treatment plan; and 5) monitoring response to the treatment plan (43, 46). The core team usually includes a physician, nurse, and social worker. Many geriatric psychiatrists, in the absence of a geriatrician, have acquired the expertise to assess and manage the medical as well as the psychiatric problems encountered with the oldest old. Therefore, a core team of a psychiatrist and nurse practitioner/physician assistant with ready availability of a social worker and geriatrician is both necessary and sufficient for comprehensive geriatric assessment. The setting for comprehensive geriatric assessment may be a step-down unit in a hospital, an ambulatory clinic, or the home of the elder.

Comprehensive geriatric assessment has been subjected to a number of efficacy and cost-benefit analyses. For example, over a 1-year follow-up, frail elderly persons who received an integrated social and medical case management intervention experienced a 5% improvement in their physical health function, compared with a 13% decline among control subjects (45).

Depression in the Oldest Old— Focus on Health-Related Quality of Life

A call to focus on health-related quality of life in the assessment and management of geriatric psychiatric syndromes in the oldest old is unfortunately necessary. In a hurried clinical world characterized by managed care and fragmented delivery of health services, psychiatrists have withdrawn in part from such practice. For example, in four well-known geriatric psychiatry textbooks published in the United States (47–50) (one co-edited by me), only one chapter appears on comprehensive assessment as a strategy for patient care, and no studies are cited that report the efficacy of comprehensive assessment compared with usual assessment. (Prominent chapters on comprehensive geriatric assessment in textbooks of geriatric medicine are replete with recommendations based on empirical studies.)

The syndrome of depression is useful as an example of this approach. Many other syndromes could just as easily be used as examples, such as late-onset schizophrenia with suspiciousness and agitation, memory loss secondary to Alzheimer’s disease, and delirium. I have chosen to focus on depression as an example because of the frequency of the problem, the tendency to treat the problem in the oldest old with medications alone, and the potential for reversing the downward spiral to frailty with appropriate treatment.

Screening for Depression in the Oldest Old

Depressive symptoms are more frequent among the oldest old than in the young old living in the community (over 20% compared with less than 10%), but the higher frequency is explained completely by factors associated with aging, such as a higher proportion of women, more physical disability, more cognitive impairment, and lower socioeconomic status (51). When these factors are controlled for, there is no relationship between depressive symptoms and age (52). The rate of major depression in the oldest old is lower than for persons in midlife but may be somewhat higher than for the young old, usually estimated at 2%–5%. (Good estimates from community samples are difficult to obtain because the oldest old are underrepresented in these samples [52, 53].) The 1-year incidence of clinically significant depressive symptoms is high in the oldest old, reaching 13% in those 85 years old or older; the incidence of major depression is around 1.5%, similar to the rate in younger age groups (54–56).

Depression has been associated with disability among the oldest old in a number of studies (32, 57, 58), illustrating the link between depression and frailty. This association is not limited to major depression alone but to a range of depressive symptoms. Therefore, most people with clinically significant depressive symptoms among the oldest old do not meet criteria for a diagnosis of major depression but experience comorbid physical and/or cognitive impairment and are at risk for a decline in functional status.

Screening for depression in hospital settings and primary care practices as well as in institutions is the critical first step in comprehensive geriatric assessment. Among the self-administered screening instruments available, the Geriatric Depression Scale (59, 60) and the Center for Epidemiologic Studies Depression Scale (61) are the most frequently used. If the patient cannot participate in a self-report assessment, interviewer-rated scales, such as the Hamilton Rating Scale for Depression (62), can be administered by the nurse practitioner/physician assistant trained in its use. Screening, however, is not directed to making a diagnosis of major depression but, rather, to documenting moderate to severe symptoms of depression regardless of the cause.

Comprehensive Assessment of Depression

If the patient screens positive for moderate to severe depressive symptoms, comprehensive assessment should be undertaken in most cases with the oldest old. Many will view this recommendation as excessive, especially in the era of managed care and managed Medicare. Nevertheless, effectiveness as well as cost effectiveness should be paramount if we are to advance our care of the oldest old and reduce the risk of frailty and failure to thrive. Comprehensive assessment is time intensive but not procedure intensive. The assessment should, if possible, take place in a setting devoted to such assessments—such as a geriatric psychiatry or geriatric medicine clinic. For the remainder of this article, I will describe an approach to assessment management in a geriatric psychiatric clinic staffed by a psychiatrist and nurse practitioner/physician assistant with ready access to a geriatrician and social worker (Table 1).

Once the medical and psychiatric histories are obtained, the focus of the assessment should be on health-related quality of life, and much of the assessment can be done by the nurse practitioner/physician assistant. A number of standardized scales are available to assess functional status from self-report (and reports from family members) (63–65). The use of standardized assessment tools provides the metrics for monitoring the progress of therapeutic intervention. Some examples will suffice. The nurse practitioner/physician assistant can assess the severity of depressive symptoms and cognitive impairment using standard screening scales (9, 62). The patient’s ability to communicate should be documented (66). Physical function can be assessed directly by asking the patient to stand and sit, reach above her or his head, walk across the room, perform fine motor movements such as writing a sentence, etc. The activity should be timed, and the elder’s ability to perform the activity should be determined. The patient can also be asked about his or her ability to perform routine activities of daily living (43) (Table 1).

Social function can be assessed by determining the limitations in usual roles and satisfaction with the support network (8, 11). The perception by the elder of poor social support has been found to be a powerful predictor of poor health and mental health outcomes (67). General health perceptions are also central to the assessment of health-related quality of life. The perception of poor physical health, even when objective measures of health are controlled for, is a strong predictor of poor health and mental health outcomes (68).

Following a physical examination, the psychiatrist usually determines that a blood screen and chemical screen will be sufficient laboratory examination. Brain imaging and psychological testing are rarely necessary during the initial comprehensive geriatric assessment of the acutely depressed elder, but they should be ordered if specifically indicated. Findings from such assessments will alter the approach to management. For example, if the depressive symptoms are complicated by memory loss and hypertension (suggestive of vascular dementia), then magnetic resonance imaging is indicated. During the midst of a moderate to severe depression, psychological testing is rarely indicated. In other words, a comprehensive assessment that focuses on health-related quality of life is not procedure intensive; therefore, costs primarily reflect time spent with the elder and his or her family.

Management of Depression in the Oldest Old

Therapy must proceed across multiple domains simultaneously. Management of the depressed oldest old frequently begins with securing the social support necessary to maintain the elder in the community. Referral to a social worker, optimally one who works closely with the clinic, is not as critical for the depressed patient as it is for the patient with memory loss. Nevertheless, short-term caretaking may not be easily arranged during the acute stage of the depression, and the social worker can inform and connect the family with short-term services. Not infrequently, however, depression is in part a symptom of frailty that has gone undetected and untreated. In such cases, families almost always will require outside services over a longer period of time if the elder is to remain in the home (most elders and their families desire initially to reestablish independent living in the home by a transition through in-home care). Managing in-home care is difficult at best for families, even those families with the financial resources to purchase such care. The social worker can provide a much-needed reality check for patients and their families so that the services needed by the depressed elder are implemented and continue without interruption.

The treatment of depression in the oldest old is targeted at improving health-related quality of life in which remission of depressive symptoms is key but not sufficient. Medications are a central component of symptom remission. The new-generation antidepressants, especially the selective serotonin reuptake inhibitors (SSRIs), have greatly improved the safety of antidepressant therapy in the oldest old. The advantage may be more apparent than real, however. For example, in a large study of nursing home residents, there was little difference in rate of falls between those treated with tricyclic antidepressants and those treated with SSRIs (69). Although it is not clear that efficacy is improved, the new-generation medications are more likely to be taken for sufficient duration and in adequate doses.

Overall efficacy of antidepressant therapy in the oldest old appears to be similar to that of the young old, but there are few good studies of the oldest old in terms of treatment response. In a recent article (70), short-term response to a combination of nortriptyline and interpersonal psychotherapy was good, but the oldest old had more recurrences during the first year of maintenance therapy than the young old. Comorbid illness and other factors that are known to decrease treatment effectiveness, rather than age, probably account for most age differences in treatment response. For example, people with a frontal lobe syndrome, more frequent in the oldest old but not inevitable, respond less well to antidepressant therapy (71).

Antidepressant therapy should usually be started at a dose about one-half that given to persons in midlife, e.g., fluoxetine 10 mg/day; paroxetine 10 mg/day; sertraline 25 mg/day; nefazodone 100 mg/day; and citalopram 10 mg/day. Ultimate doses will range widely, even in the oldest old. There are few studies that provide clear dose recommendations for the elderly, but the injunction to “begin low and increase slowly” is well taken.

The adverse side effects that are most disturbing to patients throughout the life cycle are the ones most disturbing to the oldest old—agitation, sleep disturbance, sedation, loss of appetite, nausea, and occasional anticholinergic effects. These side effects are more frequent in the oldest old than the young old, yet they remain relatively infrequent (exact comparisons were not found in the literature). Some side effects that are less frequent at younger ages emerge as major concerns with the oldest old. Hyponatremia may occur in 25% of elderly subjects taking SSRIs (72). The cause may be the syndrome of inappropriate antidiuretic hormone secretion (73).

Psychotherapy with the oldest old/frail elderly has been investigated even less than antidepressant therapy and not usually by psychiatrists. One group (74) investigated interpersonal counseling with a group of medically ill depressed elderly patients and demonstrated improvement according to Geriatric Depression Scale scores compared with a control group. The treatment effect did not improve physical or social functioning, however. In another study (75), cognitive behavior therapy was found to reduce depressive symptoms but not to improve function in a group of hospitalized stroke patients. Despite the value of interpersonal therapy along with antidepressant therapy in the young old (2), there is little evidence that psychotherapy is of value for improving function in the oldest old, even though mood did improve in the few controlled trials published.

Reports of behavioral interventions focused on improving social interactions and functional status in the frail elderly are rare, yet behavioral interventions should prove the most important adjunct to pharmacological treatment of depression in the frail elderly. Behavioral interventions range from prescribing regular group activities, physical exercise (such as walking), and training for increased independent function (physical and occupational therapists are most valuable for such training). In actuality, reports of comprehensive interventions, mostly from countries other than the United States, by either institution-based or community-based psychogeriatric teams, usually involve combinations of drug therapy, supportive psychotherapy, work with the families of the depressed elders, and a healthy dose of behaviorally oriented interventions.

Although few in number, studies of comprehensive interventions have demonstrated their value. In a meta-analysis of 14 studies of the effectiveness of outreach teams to depressed elderly in the community (76), the efficacy was comparable to that found in younger adults, but the dropout rates were high. In another study (77), home care by an interdisciplinary team for depressed and frail elders led to improvement in 58% of the group treated by the interdisciplinary team compared with 25% in a control group. These studies are preliminary and all too few. The yield in terms of reducing the downward spiral of frailty by such a comprehensive approach to the treatment of the depressed oldest old could be enormous, not only in relieving suffering but also in ultimately reducing health care costs.

Conclusions

Comprehensive, interdisciplinary assessment and therapy were the cornerstones of the treatment of psychiatric disorders across the life cycle 30 years ago. With the emergence of geriatric psychiatry during the 1970s in the United States, many initial studies focused on interdisciplinary and comprehensive therapy (78–80). At that very time, however, psychiatry was moving to a medical model, and pharmacological treatment of specific psychiatric disorders dominated the maturation of geriatric psychiatry during the 1980s and 1990s. I find it ironic that our colleagues in geriatric medicine have captured the vision of a comprehensive, functional approach to the assessment and treatment of both medical and psychiatric syndromes in the oldest old. I believe the time has come for geriatric psychiatry to rejoin geriatric medicine so that we may recapture our roots and deliver optimal care to the oldest old.

TABLE 1

Enlarge table