Current issues of ACP Journal Club are published in Annals of Internal Medicine


Meta-analysis: Most elderly patients with depression recover or improve

ACP J Club. 1991 Jan-Feb;114:22. doi:10.7326/ACPJC-1991-114-1-022

Source Citation

Cole MG. The prognosis of depression in the elderly. Can Med Assoc J.1990;143:633-9.



To determine the prognosis of depression in the elderly.

Data sources

Original research articles published from January 1980 to May 1989 were identified using MEDLINE, Psychological Abstracts, and Mental Health Abstracts using the following keywords: prognosis, depression, and elderly. Bibliographies from identified articles were searched by hand.

Study selection

Criteria for study selection were: original research study published in English or French after 1950, inclusion only of patients over 60 years of age, sample of at least 25 patients, mean follow-up of 1 year or more, and description or classification of patients' mental state during the follow-up period.

Data extraction

To determine validity, the methods and design of each study were assessed using the following criteria: formation of an inception cohort, description of referral pattern, completion of follow-up, development and use of objective outcome criteria, blind outcome assessment, and adjustment for extraneous prognostic factors.

Main results

10 studies involving 990 patients met the selection criteria. Although most studies completed follow-up for > 80% of patients, all 10 studies had serious methodologic flaws. For example, only 1 study included only patients with onset of their first episode of depression in old age. Only 3 outcome categories were consistent across most of the studies: well, relapse with recovery, and continuously ill. When the results were combined for the 5 studies that followed patients > 24 months, 27% studied remained well after recovery, 34% had relapses with recovery, 10% remained continuously ill, and 29% had other (for example, death) or unknown outcomes. Poor prognosis was associated with cognitive impairment, physical illness, and more severe depressive symptoms. Social factors such as marital status and living circumstances were not associated with prognosis.


Most studies on the prognosis of depression in the elderly are flawed; however, these studies suggest that prognosis is relatively good, with over 60% of patients becoming well without relapse or having relapses that are successfully treated.

Source of funding: Not stated.

Address for article reprint: Dr. M. G. Cole, Department of Psychiatry, St. Mary's Hospital, 3830 Lacombe Avenue, Montreal, Quebec H3T lM5, Canada.


Community surveys indicate that elderly persons commonly report depressive symptoms, although major depression is relatively less common (1). In this review, the analysis indicates that all the studies have methodologic flaws. Because most studies lack inception cohorts and fail to report referral patterns, they provide limited evidence to estimate prognosis and inadequate information to determine the applicability of the reported findings to patients in a given physician's practice. The analysis also was unable to distinguish between those with depressive symptoms alone, dysthymia, or major depression. Although solid evidence of prognostic differences between these groups may be lacking, these differences cannot be discounted. Cole's findings do suggest that the natural (or in some cases, the "treated") history of depression in the elderly is generally favorable. The burden of proof is on those who believe otherwise.

Albert Sui, MD, MSPH
University of California at Los AngelesLos Angeles, California, USA


1. Blazer D, Hughes DC, George LK. The epidemiology of depression in an elderly community population. Gerontologist. 1987;27:281-7.