For the first time in its history, the board of the Dutch College of General Practitioners (GPs) decided in November 2004 that the multidisciplinary guideline about the management of patients with mood disorders is not acceptable for Dutch GPs. Their criticism concerned the lack of transparency in the process of making the guideline, and its overly positive tone about the effects of antidepressants.
What went wrong? In 1999, various Dutch groups of health professionals, including GP representatives of the Dutch College, started enthusiastically to participate in the guideline process. However, we could not agree on whether or not to re-evaluate the evidence about antidepressants versus placebo. The psychiatrists involved claimed that no such comparison was needed as the drugs are already registered. However, the GP participants preferred to take a fresh look at the evidence.
Most antidepressants are prescribed by GPs Citation[1], and serious concerns have been raised lately about the methodological quality of antidepressant studies. What were those concerns? A brief and far from exhaustive overview follows:
Major quality criteria, such as i) blinding, ii) prevention of selection bias, iii) concealment of allocation and iv) independent outcome assessment Citation[2], Citation[3], are insufficient Citation[4]. i) The obvious side effects of antidepressants, especially tricyclics, are likely to have “unblinded" many placebo-controlled trials using ordinary inert placebos. “Active” placebos, which contain substances that mimic the specific side effects of antidepressants, allow better blinding, and show better effects Citation[5]. ii) Selection bias: e.g., before randomization, placebo responders were excluded from most trials Citation[6]. iii) Psychiatrists randomized patients themselves in most trials, and iv) also mostly evaluated outcomes themselves Citation[4].
Analysis techniques: intention-to-treat (ITT) analyses are now considered state of the art, are more conservative and are therefore more real life Citation[2] but have been seldom used in drug placebo trials. These would decrease the risk difference from 28% to 19% Citation[7]. Several other techniques, such as selective reporting of positive outcomes while negative ones are ignored, may also have inflated differences.
Time: over the years, placebos have become more effective Citation[7], Citation[8].
Many depression-rating scales contain items such as sleep and agitation, which are likely to respond to the non-specific sedative effects of antidepressant drugs.
Setting effects: few studies have been performed in primary care Citation[4].
Publication bias is difficult to rule out, as almost all studies were sponsored by pharmaceutical industries Citation[4], but it may not play a very large role Citation[9], Citation[10].
Exclusion: relevant groups such as children and older people were mostly excluded from trials Citation[4].
Sustained response equivalence: antidepressants and placebos did not differ substantially in the longer term as many of the patients were close to, or at the end of, their episode Citation[11], Citation[12].
For the future, perhaps a European perspective would help. Regulatory bodies and guideline committees have a narrow, national focus. To counter the dominant influence of international pharmaceutical companies, and to acquire more resources, the European Agency for the Evaluation of Medicinal Products (EMEA) may become a powerful ally. Funding a large, high-quality general practice trial of antidepressants versus placebo is perhaps a good idea.
References
- van Marwijk HW, Bijl D, Ader HJ, de Haan M. Antidepressant prescription for depression in general practice in The Netherlands. Pharm World Sci 2001; 23: 46–9
- Moher D, Schulz KF, Altman DG. The CONSORT statement: revised recommendations for improving the quality of reports of parallel-group randomised trials. Lancet 2001; 357: 1191–4
- Schulz KF, Chalmers I, Hayes RJ, Altman DG. Empirical evidence of bias. Dimensions of methodological quality associated with estimates of treatment effects in controlled trials. JAMA 1995; 273: 408–12
- MacGillivray S, Arroll B, Hatcher S, Ogston S, Reid I, Sullivan F, et al. Efficacy and tolerability of selective serotonin reuptake inhibitors compared with tricyclic antidepressants in depression treated in primary care: systematic review and meta-analysis. Br Med J 2003; 326: 1014–17
- Moncrieff J, Wessely S, Hardy R. Active placebos versus antidepressants for depression. Cochrane Database Syst Rev 2004; CD003012.
- Lee S, Walker JR, Jakul L, Sexton K. Does elimination of placebo responders in a placebo run-in increase the treatment effect in randomized clinical trials? A meta-analytic evaluation. Depress. Anxiety 2004; 19: 10–9
- Stolk P, Ten Berg MJ, Hemels ME, Einarson TR. Meta-analysis of placebo rates in major depressive disorder trials. Ann Pharmacother 2003; 37: 1891–9
- Walsh BT, Seidman SN, Sysko R, Gould M. Placebo response in studies of major depression: variable, substantial, and growing. JAMA 2002; 287: 1840–7
- Williams JW, Jr, Mulrow CD, Chiquette E, Noel PH, Aguilar C, Cornell J. A systematic review of newer pharmacotherapies for depression in adults: evidence report summary. Ann Intern Med 2000; 132: 743–56
- Anderson IM. Meta-analytical studies on new antidepressants. Br Med Bull 2001; 57: 161–78
- Storosum JG, Elferink AJ, van Zwieten BJ, van den BW, Huyser J. Natural course and placebo response in short-term, placebo-controlled studies in major depression: a meta-analysis of published and non-published studies. Pharmacopsychiatry 2004; 37: 32–6
- Storosum JG, Elferink AJ, van Zwieten BJ, van den BW, Gersons BP, van Strik R, et al. Short-term efficacy of tricyclic antidepressants revisited: a meta-analytic study. Eur Neuropsychopharmacol 2001; 11: 173–80