The answer to this question remains unclear. Community-based epidemiologic studies have reported rates of clinical depression in samples of older adults in the range of 1%-16%. As in the case for most disorders, prevalence rates for depression vary considerably depending on the sample studied and methods used. For example, Girling et al. (1) noted that studies using DSM-III-R diagnostic criteria generally yielded lower prevalence rates for major depressive disorder in the elderly. Studies that focused on symptoms of depression reported much higher prevalence rates than did studies using diagnostic procedures. Studies using the National Institute of Mental Health Diagnostic Interview Schedule (DIS) and DSM criteria have tended to generate prevalence rates for major depressive disorder on the order of 1%-3% (2-4), whereas studies using instruments designed specifically for use with the elderly have reported prevalence rates of 11%-16% for elderly subjects who have substantial depressive symptoms but who do not meet diagnostic criteria for major depression (5-7).There actually have been few studies of depression among the very elderly, even though this segment of the population is one of the fastest growing. The few studies that have been done report disparate prevalence rates for major depression, ranging from less than 3% to more than 12% (1,8).Snowdon (9) commented on this divergence of findings from epidemiologic investigations in 1990; he noted that the results of prevalence surveys of depression among the elderly can be grouped into low prevalence studies and high prevalence studies on the basis of measurement strategy. Studies that used the DIS and/or DSM criteria yielded much lower rates of depression than did those using other procedures. Snowdon argued that dementia, disability, physical illness, bereavement, loss of independence and security, and suicide all are much more common in old age; therefore, prevalence of depression should increase as well. However, others