The Challenge of Depression in Late Life: Title and subTitle BreakBridging Science and Service in Primary Care

There is an ongoing and unprecedented worldwide demographic transformation—the aging of the world's peoples. In the United States, the proportion of adults 85 years or older is growing so rapidly that in a few decades, 1 in 4 persons older than 65 years will be among the oldest old.¹ At the same time, the aged are becoming more ethnically diverse. Whether future cohorts of older persons will exhibit less disability than the current cohort can be debated, but there can be little doubt that global aging carries profound implications for social, economic, and health policy.

In the context of global aging, the common mental health conditions of late life such as dementia, delirium, and depression command attention because of their relationship to disability, diminished quality of life, and the demands they place on family members and other caregivers.² According to the World Health Organization, major depression was the fourth leading cause of disability worldwide in 1990, and it will soon be second only to heart disease as a cause of disability.³ In addition, the available evidence suggests that depression affects the development and course of cardiovascular disease.^{4 - 5} Even among older persons with symptoms not meeting full diagnostic criteria for a major depressive disorder, depression is associated with physical illness, functional impairment, and death.^{6 - 9} "Minor" depression is not so minor.

But the number of terms used to describe subsyndromal depression belies the difficulty of characterizing a syndrome that may lie outside of standard diagnostic categories.^{10 - 11} Major depression, minor depression, and dysthymia may not cover the entire diagnostic territory for older people. Certainly, heterogeneity of presentation along the cultural dimension has long drawn attention from anthropologists and others concerned with depression in non-Western cultures. Since depressed older adults sometimes deny feeling sad or depressed, patients and physicians alike are all too willing to ascribe symptoms to something other than depression. Depressive symptoms may be a marker for an underlying pathophysiologic process that is not yet fully understood, occurring in the context of the physical, social, and psychological changes of aging. For example, vascular changes in the central nervous system might play an etiologic role in depression for some older people.¹² Depression in older persons may differ qualitatively, not just quantitatively, compared with depression in younger persons. Nevertheless, a common conceptual framework to define subsyndromal depression is lacking.

For the most part, this heterogeneity of the clinical presentation of depression is played out not in the offices of psychiatrists and other mental health specialists but in the offices of primary care physicians. Older adults with depression commonly present to primary care physicians in the context of physical illness and not to specialists in mental health.¹³ Nevertheless, few studies have focused on whether pharmacologic or psychotherapeutic treatments designed for patients with major depression are effective for other symptom clusters, especially among older primary care patients.

In this issue of THE JOURNAL, Williams and colleagues¹⁴ report results from a multisite effectiveness clinical trial that had excellent participation rates and was carefully carried out in the primary health care setting. The goal of the study was to assess treatment strategies for depression in older adults who did not meet full standard criteria for major depression. The investigators compared the Problem-Solving Therapy–Primary Care (PST-PC), which is a brief treatment strategy adapted for delivery in primary care, paroxetine, and placebo along with active clinical management. In this study, in contrast to the research criteria for minor depression in the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition,¹⁵ at least 3 (not 2) symptoms in addition to mood disturbance had to be present nearly every day for 4 weeks (not just for 2 weeks) or more. The study participants reported significant mental and physical impairment at entry.

The authors found no differences between the 2 active treatments for reducing symptoms of depression, but they emphasized the findings that patients treated with paroxetine showed greater and more rapid symptom reduction than patients receiving placebo. Patients treated with PST-PC did not show more improvement compared with those taking placebo. Site differences in the outcomes of patients treated with PST-PC related to the experience of the therapist probably preclude any broader generalization about effectiveness. Findings related to mental health functioning were complex and could not have been anticipated. For instance, paroxetine improved the mental health functioning only of the patients with dysthymic disorder who were higher in functioning at baseline. On the other hand, the effects of both paroxetine and PST-PC on the mental health functioning of patients with minor depression were limited to those with the lowest baseline functioning. As in other studies of this kind, the remission rate in the placebo group was high.

What are the clinical implications of this report? First, this study should not be considered the final word on the effectiveness of psychotherapy for subsyndromal depression in late life. The investigators randomly assigned participants to receive either medication or psychotherapy but not both. Many older patients prefer counseling to taking another medication, and if antidepressant medication is required, the benefits may be greater when medication is combined with counseling. In any case, it is not news that older patients tend to be taking multiple medications and are at increased risk for adverse events due to medication. The task for the future is how to help primary care physicians and others who work in primary care to be more effective in providing counseling to their patients. Also, since a past history of depression was not an exclusion to participation in this study, it is possible that many of these patients could be more properly viewed as having major depression. In other words, some persons identified with minor depression have simply experienced incomplete remission of major depression.

Despite these limitations, the study deserves attention because many of the patients who received treatment would have been excluded from clinical trials conducted in specialty psychiatric settings because of physical comorbidity. In addition, patients seen in specialty mental health settings differ in attitudes, readiness to accept treatment, and in clinical characteristics that could prove decisive in clinical outcomes. Understanding these factors requires carrying out the kind of research in primary care settings exemplified by the investigation of Williams and colleagues. This study provides an opportunity to highlight 3 important themes: (1) the chronic nature of depression; (2) the public health perspective on mental health; and (3) addressing the mental health needs of patients in primary care.

The concept of depression as a chronic illness (like illnesses such as diabetes) has emerged from follow-up studies that show that persons who have had an episode of major depression are prone to have recurrent episodes.^{16 - 18} Older adults take longer to recover from depression and have shorter times to relapse than do younger persons.^{19 - 20} By some estimates, about 30% of depressed older adults remain chronically depressed.²¹ Because of their continuous relationship with older patients, primary care physicians can play a pivotal role in helping identify when depression interferes with functioning or adherence to medical treatment.²²

The second theme concerns the public health view of mental health. The public health model places emphasis on diversity in samples, use of treatment that involves mixed, not pure modalities, assessing outcomes beyond symptom counts (eg, to include functional status and quality of life), and participation of nonselected clinicians working in nonacademic venues.^{23 - 26} The emphasis of research is changing from an exclusive preoccupation with the question: "Does it work?" to "Why, for whom, and under what circumstances does it work?"²⁷ A central conclusion of the National Institute of Mental Health report Bridging Science and Service was the need to shift from efficacy studies (investigating the outcome of treatments rendered to carefully selected patients under rigorously monitored conditions) to effectiveness and dissemination studies (investigating treatments rendered in primary care practices).²³ This challenging agenda for research recognizes that to limit disability from depression in the population requires extending studies to venues other than the offices of mental health specialists.

The third theme looks to the future. Specialists in the mental health care setting cannot expect to translate what works there to the primary care setting without modification. In "bridging science and service," investigators must take the "ecology of primary care"²⁸ into account, an ecology that includes factors related to the patient and the health care system as well as to the physician. For example, depressed patients whom primary care physicians recognize and treat tend to be more functionally impaired and more willing to take antidepressant medication than depressed patients who are not identified as such by these clinicians.^{29 - 30} Merely placing mental health personnel into primary health care settings may be too simple a solution for such a complex issue. Team work is needed. It is not a question of integrating mental health care into the everyday tasks of primary health care—the best generalist physicians strive to do so despite the barriers.³¹ Real change may be more likely by learning to integrate the training experiences of physicians with mental health specialists and the training of mental health specialists with primary care physicians.

More than 30 years ago, Shepherd and colleagues³² concluded that the way to improve the treatment of mental disturbances in the community was to strengthen the therapeutic role of the primary care physician. Like many good studies, the investigation of Williams and colleagues is not the last word, but it brings fundamental issues to the fore for discussion in an arena vital to improving public mental health.