However, there is strong evidence that clinical depression in adults is underrecognized and undertreated in medical practice (Pérez-Stable, Miranda, Muñoz, and Ying, 1990; Cleary, Goldberg, Kessler, and Nyez, 1982) and that treatment is often suboptimal (Keller, Lavori, Rice, Coryell, and Hirschfeld, 1986). It is evident that adults with depressive disorders, if they are treated at all, are often treated by internists and general practitioners, not mental health specialists (Regier, Goldberg, and Taube, 1978). This has prompted the Agency for Health Care Policy and Research to publish clinical practice guidelines on depression in primary care (Depression Guideline Panel, 1993a,b,c). Evidence has also accumulated that children with serious mental disorders in general and clinical depression in particular (Beardslee, Keller, Lavori, Staley, and Sacks, 1993; Keller, Lavori, Beardslee, Wunder, and Ryan, 1991) are also undertreated. In addition, Mexican-Americans and probably other minorities as well, significantly underutilize mental health services when in need of treatment for mental disorders in general. In the UCLA ECA study, utilization rates were low for the general population, at 22 percent, but for Mexican-Americans rates were only half that, or 11 percent (Hough, Landsverk, Karno, Burnam, Timbers, Escobar, and Regier, 1987).
A developmental perspective is necessary in the understanding of depressive disorders because the nature of clinical depression changes across the life span. Estimates of the rate of depressive disorders also vary depending on the instruments and diagnostic systems employed and on the samples studied. A useful technique is a structured interview scored according to standard diagnostic criteria, such as the DIS/DSM-III system used in the ECA study. ECA data suggest an overall lifetime disorder rate of 7.8 percent for all affective disorders (now classified as mood disorders in DSM-III-R), with major depression as the most common affective disorder, followed by dysthymia (with lifetime rates of 4.9 and 3.2, respectively) (Weissman, Bruce, Lief, Florio, and Holzer, 1991). Higher rates of depressive disorders have been reported when slightly broader diagnostic nomenclatures and semistructured interview schedules are used. For example, data collected from community samples using the Schedule of Affective Disorders and Schizophrenia suggest rates as high as 8 to 12 percent for men and 20 to 26 percent for women (Boyd and Weissman, 1981).
According to ECA data, the prevalence of major depression among the elderly living in the community is less than 3 percent. However, depressive symptoms occur in approximately 15 percent of community residents over 65 years of age. Thus clinical depression may have a different form in the elderly. The rates of major or minor depression