Treatment of Depressive Conditions in Later Life: Title and subTitle BreakReal-World Light for Dark (or Dim) Tunnels

Long predicted age-demographic shifts are now a reality. The "graying" of the US population will accelerate in the next few years as the first cohorts of the baby boom generation reach the chronological milestones traditionally used to define older age. So it is appropriate that increased attention be paid to improving the quality of life for seniors. The article by Ciechanowski et al¹ in the current issue of THE JOURNAL and a recently published article by Bruce et al² do just that by presenting findings from effectiveness trials for depression in the elderly. These studies represent important advances and warrant consideration of their broader context.

Later life depression is a major public health problem. Depressive conditions in elderly persons are common, leading causes of functional disability and powerful risk factors for mortality from general medical conditions as well as suicide.^{3 - 5} Pharmacological and psychotherapeutic treatments for geriatric major depression are well supported empirically,^{3 - 4 ,6} although until recently much of this evidence stemmed from efficacy studies conducted in mental health settings. Even so, too few elderly persons with depression receive adequate therapy.^{4 ,7}

Further complicating the picture is the heterogeneity of clinical states encompassed by the term depression. A large body of evidence from epidemiological studies demonstrates that most elders with clinically significant depressive symptoms do not meet diagnostic criteria for major depressive disorder.^{8 - 12} However, the cumulative functional morbidity of these so-called lesser conditions actually exceeds that of major depression among the elderly.^{12 - 13} Various diagnostic criteria have been proposed to capture these states, using terms such as minor, subsyndromal, or subthreshold depression. Some of these patients may have partially remitted (or partially recurrent) major depressive episodes, as it has become clear that such waxing and waning of symptoms is characteristic of the course of major depression.¹⁴ Other patients have a long-recognized condition known as dysthymic disorder, which consists of chronic depressive symptoms at a lower level of severity than major depression.^{15 - 16}

However, the majority of older depressed adults have minor or other subsyndromal depressions that do not fit into the established categories of major depression or dysthymic disorder, with point prevalences ranging from 10% to 20% in the community or primary care settings and from 15% to 25% or greater in more medically ill groups, such as those residing in nursing homes.^{3 - 4 ,6 ,8 - 12} Data regarding longitudinal outcomes and biological, psychosocial, and functional correlates increasingly support the notion that minor and subsyndromal depressions are part of a spectrum of depressive illness severity.^{17 - 18} In other words, while major depression is a useful diagnostic construct, it may not be pathogenetically distinct from less-severe forms; thus, the dividing line, while allowing reliable diagnosis, may be an arbitrary one.

In the midst of such diagnostic complexities, clinicians must make recommendations to patients based on the best available evidence on treatment outcomes. However, as recently as mid-2002, the evidence base for treatment of less-than-major depression in the elderly was limited¹⁹ ; the few extant randomized controlled trials focusing on elderly patients suggested that antidepressant medication or depression-specific counseling methods had relatively modest benefit.^{19 - 20} Subsequently, THE JOURNAL published important findings from the IMPACT (Improving Mood–Promoting Access to Collaborative Treatment) study by Unützer et al.²¹ IMPACT's on-site collaborative care model, using a depression care manager to support antidepressant medication treatment or brief psychotherapy, was effective in improving symptomatic and functional outcomes for older primary care patients with major depression or dysthymic disorder treated in the primary care setting.

The recent publication of PROSPECT (Prevention of Suicide in Primary Care Elderly: Collaborative Trial) in THE JOURNAL by Bruce and colleagues² reported on an important multisite intervention study that adds substantially to such prior work. In this randomized controlled trial, the intervention group received care based on a clinical algorithm, including antidepressant medication or, for those declining medication, the offer of a brief individual therapy called interpersonal psychotherapy. Depression care managers located on site in the various primary care practices operationalized the treatment management. Further distinguishing this study was its inclusion of patients with either major depression or "clinically significant" minor depression, the latter defined by a more stringent application of criteria proposed in the appendix to the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition.^{16 (pp719-721)} Because of the high rates of suicide in older adults and the primacy of depression as a risk factor,⁵ suicidal ideation was a key outcome of interest. The PROSPECT intervention substantially reduced suicidal ideation at rates comparable with that seen in prior efficacy studies and in specialty mental health clinical settings. Intervention group patients also experienced significant improvements in depressive symptom severity. Such findings are impressive for a study with methods designed to favor generalizability to real-world practices.

Of note, the reduction in overall depressive symptoms for the minor depression subgroup did not differ significantly from that in the control group, save for the relatively few whose initial symptoms included suicidal ideation. This lack of intervention-control difference for minor depression might support recommendations to focus depression-specific care management on those with major depression. However, the nonsignificant improvement in minor depression might be due to the difficulty of detecting a "signal" in a group that was (by definition) less symptomatic initially and in a group that, as in prior studies, showed substantial improvement even when assigned to the usual care control condition. It is possible that improvement in the minor depression usual care group resulted from improved physician recognition of depression, but other factors may be involved. Prior data also show that milder types of depression "respond" well to control conditions, which in effect often amount to a nonspecific psychosocial intervention.¹⁹ An unresolved question is whether an emphasis on specific psychosocial treatment might demonstrably improve the lives of patients with less-than-major depressions.

In this light, the article by Ciechanowski and colleagues¹ published in this issue of THE JOURNAL is particularly intriguing. The authors report findings from a randomized controlled trial of a home-based program called PEARLS (Program to Encourage Active, Rewarding Lives for Seniors) to treat minor depression or dysthymic disorder in seniors recruited through community agencies. The intervention group received a form of brief psychotherapy known as problem-solving treatment, modified to emphasize physical activity and increased socialization. Treatment was not solely psychosocial; for participants who did not improve in the first few weeks, a depression care management team made recommendations to the primary care physician about diagnostic evaluation and antidepressant medication therapy. The intervention improved symptomatic and functional states substantially, with outcomes comparable with those from primary care–based studies that included patients with major depression. The PEARLS intervention, while different than PROSPECT, similarly lends itself to adaptation and implementation across a variety of settings, including community agencies in the case of PEARLS.

The PROSPECT and PEARLS studies represent sophisticated, high-quality science using flexible, mixed-modality interventions with broad, ethnically diverse participant groups. Both studies also reflect the increasing research attention in geriatric depression beyond traditional mental health subspecialty settings.^{3 - 4 ,22} These findings also provide evidence-based hope for the millions of elderly persons living in the dark tunnel of major depression or the only slightly less dim tunnels of "lesser" depressions.

Of course, more work needs to be done. Interventional research on depression in later life must continue to test rigorously the application of existing treatment strategies to patients in a variety of medical and community settings. At the same time, studies of risk factors and potential moderating factors will help identify patients with less-than-major depressions who are at highest risk for persistent or worsening symptoms and disability. However, existing treatments help most but not all patients, and some of those helped do not achieve full remission easily. Studies testing pathogenetic models involving factors across the biopsychosocial systems continuum^{23 - 26} have the goal of defining etiologically distinct subgroups that might respond preferentially to specific existing or to-be-developed treatments.

Unfortunately, there are numerous barriers to the delivery of mental health care for older adults, even for traditional services,^{27 - 28} let alone for innovative methods such as on-site care managers or home-based programs. Among these barriers are disparities in Medicare reimbursement for depression and other mental illnesses compared with "physical" disorders. The current system can only be described as discriminatory and, in many cases, results in prohibitive costs for elders. To turn the implications of studies such as PROSPECT and PEARLS into reality for older adults will require the application of their results, and concomitant demonstrations of favorable cost-benefit analyses,^{29 - 30} to the changing of social policy and health care payment and delivery systems.³¹ The well-being of an aging society demands meeting these challenges.