Faculty Opinions recommendation of Antidepressant drug effects and depression severity: a patient-level meta-analysis.

Publisher Website
Google Scholar
Abstract
This is a meta-analysis of six studies of antidepressants versus placebo in major depressive disorder. The interesting but controversial conclusion is that differences in favor of antidepressants are not “substantial” until the depression at outset of the episode is rated as severe on the Hamilton Rating Scale for Depression (HRSD). Clinicians should more carefully consider the severity of a major depressive episode as well as perform more careful assessments of whether the mood disturbance is actually a major depressive episode before prescribing antidepressant medications. This paper has already gained a good deal of media attention, and it will certainly attract more. It states that there are no real positive differences in improvement between active medication and placebo in mildly to moderately rated major depressive episodes until the episode at treatment baseline receives a “severe” rating on the HDRS. The authors found that the effect size of medications over placebo was “medium” for HDRS scores greater than 25 (the American Psychiatric Association rates an HDRS score greater than 23 as “very severe”), but the effect size was large for those patients with HDRS score 27 or greater. Three of the studies looked at the tricyclic antidepressant imipramine and three other studies looked at the selective serotonin reuptake inhibitor (SSRI) paroxetine, and while one might argue as to the medications chosen and the dosages of those medications, the results were the same despite the different antidepressants. This study raises many questions. Certainly, there will be many arguments with respect to the methodology of the meta-analysis and how and why certain studies were included and others excluded, but there are a number of important issues raised here. First, are we overdiagnosing depression, i.e. are we calling too many things major depression when they might be other clinical syndromes that present with depressed mood, sleep and appetite problems, fatigue, lack of concentration and thoughts of suicide (see refs {1-3} and ref {4}, on which Ken Silk is the author)? How did we arrive at the idea that antidepressants work for a wide variety of degrees of disturbances in mood, which has led to antidepressants being among the most frequently prescribed medications today for mild as well as severe depression? Has direct consumer advertising and publication bias of more positive than negative antidepressant studies in the literature been responsible for this over-expectation that antidepressants work for any kind of mood disturbance from sadness to melancholia {5}? Perhaps more importantly, we do not want this study to become some watershed study that may lead to people who need antidepressants not getting them. We, on one hand, may be overprescribing antidepressants to people who do not necessarily need them but, on the other hand, a recent study reveals that there are not enough treatment resources for all the people who may need treatment for depression (though not necessarily always pharmacotherapy), particularly among minorities {1}. As we consider health reforms as well as how to control rapidly expanding health care costs, we will need to be more certain that we expend resources in treatments that are effective. This will allow more efficient resource allocation and reallocation, and effective treatments may then become more available to more people who need them. This paper adds to that dialogue.

This publication has 0 references indexed in Scilit: