Postmyocardial infarction mortality in relation to depression: a systematic critical review

The aim of this review was to survey the literature on depression in patients with myocardial infarction to assess the methodological quality and to test whether depression leads to an increased postmyocardial infarction mortality. Medline, Psycinfo, and www.UMI.com were searched, and researchers were contacted in the autumn of 2003. Thirty-one articles were reviewed. Only seven articles scored above a predefined level of 75% for acceptable quality. The articles lack description of non-responders, recall period for depressive symptoms, validation of applied instrument on target population, and sample size large enough to show differences between groups. The prevalence rates of depression ranged from 1.6 to 50%. In eight articles, a diagnostic test was applied, in the rest of the studies, questionnaires were used. The prevalence of depression was highest in those using patient-completed questionnaires. A significant positive association was shown between depression and postmyocardial infarction mortality in 15 studies, a non-significant association in 14, and in two articles, this was not reported. In articles with data collection starting after 1994, a non-significant relation tended to be reported. The studies were generally not of acceptable quality. They lacked sufficient power to show differences in stated end points between groups. Application of non-validated instruments caused large differences in prevalence rates of depression. Future studies should include a minimum of 1,000 patients, use a validated instrument, re-examine the patients, and describe participants and non-participants in detail.