J. Moncrieff, S. Wessely, R. Hardy, Meta-analysis of trials comparing antidepressants with active placebos, Brit J Psychiatry, 1998, 172, 227-231. with Commentary (D. Healy) pp 232-234.
This important review examines the effects of unblinding in clinical trials, seeking to establish more clearly the efficacy of antidepressants. It examines the results of nine reasonably well control trials in which the efficacy of antidepressants was compared with an "active" placebos, atropine. (Atropine is not recognised as an antidepressant, but mimics the characteristic "dry mouth" [anticholinergic] effects of tricyclic and other ADs.)
"Only two of the nine studies examined produced effect sizes which showed a consistent significant difference in favour of the active drug" suggesting that "unblinding effects may inflate the efficacy of antidepressants in trials using inert placebos". Authors acknowledge the several limitations of the trials they examined, therefore "these conclusions must remain tentative". Nevertheless, the results give "cause for concern about the potential effects of unblinding in psychiatric trials" and suggest the need to identify and use safe active placebos to improve "the validity of antidepressant research". Regulatory agencies take note.
See also: Greenberg & Fisher, 1989