OUP user menu

Comment on: ‘Antidepressant medications and other treatments of depressive disorders: a CINP Task Force report based on a review of evidence’

R. H. Belmaker M.D.
DOI: http://dx.doi.org/10.1017/S1461145708008705 577-578 First published online: 1 June 2008

The report of the CINP Task Force ‘Antidepressant medications and other treatments of depressive disorders’ (CINP Task Force, 2007) represents a scholarly and comprehensive review on the huge literature on antidepressants. As such, it is within the burgeoning modern tradition of evidence-based medicine. However, as a CINP report several points mentioned in the review could use additional emphasis.

(1) The CINP is an organization promoting interaction between research psychopharmacologists, clinical psychopharmacologists and the pharmaceutical industry. A review done by a strict computer search, evidence-based approach can be biased by the fact that pharmaceutical company funding favours controlled trials of new patented compounds and that alternative and non-patented treatments often have small numbers and inferior methodology (Turner et al., 2008). The CINP is committed to attempting to correct this bias by searching out promising new treatments of depression anywhere in the world, including unpatented natural remedies that may have potential such as herbal remedies, Hypericum, inositol, omega-3, etc.

(2) It is well known that the placebo response in depression studies is high and the treatment effects are limited. In large studies the treatment–placebo difference may be highly significant but the treatment effect may be small. Of course the size of the treatment effect can be a question of whether the glass is half empty or half full. Some recent scholarly reviews in our field have sometimes emphasized the tremendous usefulness of antidepressants; others have emphasized the limits of efficacy as well as time delay with present antidepressants and the need therefore to develop new antidepressants (Lacasse and Leo, 2005). Some in our field have felt that the limited placebo–treatment difference in controlled studies of antidepressants is a function of the rigors of the clinical trial situation often including fixed dose and limited ability of the clinician to add other treatments. However, the recent STAR*D trials looking at efficacy of antidepressant in clinical situations closer to real life have revealed disappointing results. Moreover, it is clear that clinical research recruits patients with depression severities of at least moderate degree with clear DSM criteria for major depression and often exclude comorbidity (Belmaker and Agam, 2008). However, epidemiological surveys of depression often include milder depressions or depressions with complex comorbidity that may respond less well to antidepressants than those patients in clinical trials. Thus the CINP needs to be in favour of careful, proper use of antidepressants and not merely increased use of antidepressants. This is especially true in those areas of the world where antidepressants will probably be prescribed by family physicians. Clinical practice often suggests high response rates when antidepressants are given for minor depressions or problems in living. However, these illnesses have such a high placebo response rate that the clinical experience by the individual physician of his patient having recovered is a poor indicator of the true effectiveness of giving the drug.

(3) The CINP is an international organization and economic conditions vary greatly on our planet, despite the fact that scientific communication has improved and that science is global. The CINP has a responsibility to those populations for whom expensive modern medicines are far beyond their means. An electronic search of controlled studies does not always take this into account. We must avoid a situation where we are recommending medications but that our recommendations will only be relevant for the small upper-middle class elite in Third World countries. We must take into account that some of the non-patented older medications may be almost equally good and we should clearly discuss that pharmacoeconomic cost benefit varies in different countries (Patel et al., 2007).

Conflict of interest has become a major issue in the public mind with regard to all kinds of research and perhaps especially antidepressant research. The present method of disclosure may not completely take care of the problem. While as scientists we may feel that funding sources do not influence our scientific judgement, considerable empirical evidence shows that studies funded by a particular company are more likely to yield results that favour that company's interests. Experience in research shows that the same data can be analysed in more than one way and many are increasingly uncomfortable in being a co-author on studies where one centre may have enrolled a small percentage of the patients but where the overall data analysis was done internally at the pharmaceutical company and was not made available to all the authors. A survey of current literature in our field cannot take this unquantifiable factor into account. However, the public clearly is taking it into account and we must make our results cautious until we are sure that the studies we review, and their relative weighting and results, are not influenced by subtle conflicts of interest.



Statement of Interest



View Abstract