Quality Indicators for the Care of Depression in Vulnerable Elders
Published in: Journal of the American Geriatrics Society, v. 55, no. S2, Oct. 2007, p. S302-S311
Posted on RAND.org on January 01, 2007
Major depressive disorder is present in approximately 3% of men and 5% to 9% of women. Although the prevalence of depression in the older population is believed to be lower than in the general population, primary care providers recognize depression in only one-third to one-half of cases, and an even smaller proportion receive appropriate care.1 Depression is a costly disease, with estimated costs to society of at least $65 billion annually in 1998 dollars in direct and indirect costs, yet treatment for depression is effective, improves overall quality of life, and can be cost effective. Depression in this paper connotes major depression, which the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) defines as a period of at least 2 weeks during which a patient experiences mood disturbances that last most of the day (feeling sad or loss of interest in pleasurable activities) and experiences at least four other symptoms. Most clinical practice guidelines only address treatment of major depression. Dysthymia (chronic depression) is a different diagnosis defined as a period of at least 2 years during which a patient experiences depressed mood or loss of interest in pleasurable activities most of the day, for more days than not, and experiences at least two of six other symptoms. Some studies have shown that treatment for dysthymia is efficacious, and some guidelines state that treatments for major depression also apply to dysthymia, but many people who have depressive symptoms do not meet the full criteria for major depression or dysthymia. The appendix of DSM-IV includes a research diagnosis of minor depression that is defined as two to four symptoms of major depression during a 2-week period. At the time of publication, there was a consensus that there was not enough information about minor depression to include it as a formal diagnosis, so it was classified as Depressive Disorder Not Otherwise Specified. Over the past few years, research in minor depression has increased in terms of descriptions of disability and treatment. For example, an analysis of the Epidemiologic Catchment Area Study data demonstrated that patients with minor depression had a greater risk of lost days from work as well as a greater risk of developing major depression than those without depression. Studies of older patients with minor depression have indicated that they have a greater risk of disability and a greater risk of developing major depression than those without depression. Neuroimaging studies show similar neuroanatomical changes in elderly patients with minor and major depression. Other studies have shown that minor depression responds to medication, psychotherapy, and quality improvement programs using psychotherapy and medication management. Because minor or subthreshold depression is a new area of research and studies of the benefits of treatment have not yet been incorporated in evidence-based guidelines, this article will not focus on minor depression. This monograph presents quality indicators (QIs) to assess the treatment of vulnerable older patients with major depression or dysthymia. It does not address subsyndromal depression. Because many primary care clinicians do not differentiate between subtypes of depression, it is assumed that, if a diagnosis of depression is noted in the chart, then the QIs apply; the indicators do not apply if the clinician notes that the patient has minor depression, subthreshold depression, or mild depression. METHODS: A total of 173 articles were considered in this review; 31 identified through a Web search, 54 identified through reference mining, and 73 from reference mining the Assessing Care of Vulnerable Elders-1 monograph. Fifteen articles were added after peer review. RESULTS: Of the 20 potential QIs, the expert panel process judged 19 to be valid, and the panel added one. One indicator was rejected. The literature summaries that support each of the indicators judged to be valid in the expert panel process are described.