Improving Evidence‐Based Management of Depression for Older Americans in Primary Care: If Not Now, When?

Depression is present in approximately 6% to 10% of older patients who present to primary care physicians, including 20% to 40% of those with such common medical conditions as hypertension, diabetes mellitus, cardiovascular disease, osteoarthritis, and neurodegenerative disorders. Depression adds further to the caregiving burden of family members,1 and it is well recognized that it adversely affects adherence to prescribed medical regimens. Strong evidence links depression to greater morbidity and mortality,2, 3 higher levels of health services utilization and costs,4-6 and poorer health-related quality of life, independent of disease severity.7 Primary care providers (PCPs) view the diagnosis and treatment of late-life depression as among their many other responsibilities.8 Of the estimated 9.8 million visits that older Americans made to office-based providers for depression in 2001 to 2002, nearly two-thirds were to PCPs.9 Thus, it is critical to understand how they treat depression in this population. Treatment strategies that have been proven effective and can be widely disseminated have been developed for use in primary care,10, 11 including late-life populations,12, 13 and may even reduce mortality.14 Unfortunately, they have yet to be broadly implemented in routine care. In this issue of the Journal of the American Geriatrics Society,15 Ming Tai-Seale and colleagues provide important insights into the “black box” process of depression care through their analyses of videotapes recorded during actual primary care visits. Although others have also recorded patient encounters and used sophisticated techniques to analyze patterns of communication,16-18 to the best of our knowledge this is the first report to focus specifically on the mental health care of older adults. The investigators recorded encounters between 367 patients aged 65 and older who presented to their PCPs (N=35) for routine care at three geographically dispersed and heterogeneous practice settings (academic group practice, suburban managed care, and inner-city fee for service). Although they required patients and physicians to provide informed consent to participate and used a Medical Outcomes Study 36-item Short Form Survey Mental Component Summary (SF-36 MCS) score to define depression (<42) instead of a more-formal depression rating scale or case-finding instrument, and the presence of their recording device could have affected the content of the encounters, nonetheless, a discussion of a mental health topic occurred in just 22% of visits and 29% of visits for patients with scoring <42 on the SF-36 MCS. Moreover, these discussions lasted a median of 2 minutes and varied only slightly according to practice setting, according to provider and patient sex, according to race, or according to sex and race concordance of patient and provider. Although the investigators did not provide data on the total length of the encounter, if the mean duration of the office encounter is estimated as 15 minutes or less, then the proportion of the encounter time devoted to mental health care was quite low. Qualitative analysis of several typical instances showed difficulty in diagnosing depression and with executing evidence-based treatment plans such as decreasing access to firearms (the method of suicide for most older males),19 using low doses and short courses of antidepressants, and failing to measure symptom-based changes. The current findings may also be relevant to research on disparities according to sex and race in access to mental health services. In patients with chronic and severe depression, men are typically underrepresented in receiving depression care even though they have a greater risk of suicide completion than females, whereas African Americans are overrepresented.20 This inequality may reflect, in part, that African Americans are less likely to have a discussion of mental health during primary care office visits because of historical mistrust of the healthcare system,21 the added stigma of mental illness among African Americans,20 or doctor–patient communication problems such as fewer patient-centered statements22 and less physician discussion of mental health concerns, as reported by Tai-Seale et al.15 How likely then are PCPs to provide quality depression care, given the multiple competing demands made on their time to provide patient-centric, evidence-based, preventive, acute and chronic illness care—and given the clinical ecology of primary care and the typical 15-minute visit? The data presented by Tai-Seale and colleagues raise anew the pressing need for structural changes in primary care to allow implementation of collaborative care strategies for major depression, particularly in older patients. Given the difficulties in implementing these models of depression treatment into routine primary care,23 one pragmatic alternative is to raise awareness among patients, PCPs, medical specialists, and payers of the adverse effect of depression on comorbid medical illness. This approach builds upon the realities that older patients are more likely than younger patients to have coverage for health insurance and pharmaceutical benefits, older patients are more likely to have a comorbid chronic medical condition, many large-scale integrated healthcare delivery systems have already implemented collaborative care programs for treating patients with such chronic high-cost medical conditions as congestive heart failure and kidney failure,24 providers will have added incentive to recognize and then address nonadherence to recommended medical care related to depression25-27 as use of provider “report cards” and pay-for-performance arrangements becomes more widespread,28 and guideline-appropriate depression treatment may be an effective “secondary prevention” strategy for treating the underlying medical condition.14, 29, 30 Studies like Prevention of Suicide in Primary Care Elderly—Collaborative Trial12 and Improving Mood-Promoting Access to Collaborative Treatment Study13 have taught us that providing evidence-based depression care management to elderly primary care patients doubles the rate of treatment response and reduces the burden of depressive symptoms, including suicidal ideation, more rapidly than PCPs' “usual care” for depression that is enhanced by case identification and communication of depression diagnosis with the PCP. Specifically, by training or placing medical or mental health nurses or social workers into primary care practices (with mental health specialist supervision and backup as needed), case identification and provision of on-target, on-time consultation to PCPs can be enhanced, together with treatment engagement and adherence. Depression in old age can be successfully treated, and recovery can be maintained over long periods of time with implementation of chronic disease-management strategies.31 These are straightforward in the majority of cases and amenable to primary care and are not expensive compared with most medical procedures. Furthermore, similar telephone-based approaches for treating anxiety disorders in primary care have also proven to be effective.32 Although additional qualitative research remains necessary to identify effective counseling strategies that PCPs can use to assist depressed and anxious33 older Americans, it is past time to advocate for policy changes and to provide resources for care of the millions of older depressed Americans who seek treatment in primary care settings annually. Clinicians need to partner with patients and care givers to make it happen. Conflict of Interest: Supported in part by National Institute of Mental Health Grant P30 MH071944, the John A. Hartford Foundation Center of Excellence in Geriatric Psychiatry, and the University of Pittsburgh Medical Center Endowment in Geriatric Psychiatry. Dr. Charles Reynolds receives research support from GSK, Forest, Pfizer, Lilly, and BMS. Author Contributions: CFR: First author. BR: Reviewing of drafts. CFT: Participated in writing of editorial and reviewing drafts. MC: Participated in reviewing drafts. Sponsor's Role: None.