For all controlled studies that bear upon the analgesic-effect of antidepressants, a meta-analysis was conducted to get an estimation of the effect size and to get a sight on the possible modes of action. For this purpose both content and methodological variables were coded for each study. The mean size of the analgesic effect is 0.63. This means that the average patient getting an antidepressant has less pain than 74% of the patients getting a placebo. However because of selective drop-out and the use of psychometri-cally inadequate pain scales, this may be an over-estimation. Real analgesic qualities of antidepressive agents seem to offer the most plausible explanation for the effect, but the importance of serotonin reuptake blocking is not confirmed. A main objection, however, is that in all studies the depression variable was insufficiently controlled at inclusion. Because both the estimation of the effect size as the explanation of the effect are confounded with methodological doubts, it is concluded that further studies need to be done to overcome these difficulties.
