The British Journal of Psychiatry
Efficacy of drug treatment in obsessive-compulsive disorder. A meta-analytic review.
M Piccinelli, S Pini, C Bellantuono, G Wilkinson


BACKGROUND A review of the efficacy of antidepressant drug treatment in patients with obsessive-compulsive disorder (OCD), using a meta-analytic approach.

METHOD Randomised double-blind clinical trials of antidepressant drugs, carried out among patients with OCD and published in peer-reviewed journals between 1975 and May 1994, were selected together with three studies currently in press. Forty-seven trials were located by searching the Medline and Excerpta Medica-Psychiatry data bases, scanning psychiatric and psychopharmacological journals, consulting recent published reviews and bibliographies, contacting pharmaceutical companies and through cross-references. Hedges' g was computed in pooled data at the conclusion of treatment under double-blind conditions or at the latest reported point of time during this treatment period. For each trial, effect sizes were computed for all available outcome measures of the following dependent variables: obsessive-compulsive symptoms considered together; obsessions; compulsions; depression; anxiety; global clinical improvement; psychosocial adjustment; and physical symptoms.

RESULTS Clomipramine was superior to placebo in reducing both obsessive-compulsive symptoms considered together (g = 1.31; 95% CI = 1.15 to 1.47) as well as obsessions (g = 0.89, 95% CI = 0.36 to 1.42) and compulsions (g = 0.79; 95% CI = 0.34 to 1.24) taken separately. Also, selective serotonin re-uptake inhibitors (SSRIs) as a class were superior to placebo, weighted mean g being respectively 0.47 (95% CI = 0.33 to 0.61), 0.54 (95% CI = 0.34 to 0.74) and 0.52 (95% CI = 0.34 to 0.70) for obsessive-compulsive symptoms considered together, and obsessions and compulsions taken separately. Although on Y-BOCS the increase in improvement rate over placebo was 61.3%, 28.5%, 28.2% and 21.6% for clomipramine, fluoxetine, fluvoxamine, and sertraline respectively, the trials testing clomipramine against fluoxetine and fluvoxamine showed similar therapeutic efficacy between these drugs. Finally, both clomipramine and fluvoxamine proved superior to antidepressant drugs with no selective serotonergic properties.

CONCLUSION Antidepressant drugs are effective in the short-term treatment of patients suffering from OCD; although the increase in improvement rate over placebo was greater for clomipramine than for SSRIs, direct comparison between these drugs showed that they had similar therapeutic efficacy on obsessive-compulsive symptoms; clomipramine and fluvoxamine had greater therapeutic efficacy than antidepressant drugs with no selective serotonergic properties; concomitant high levels of depression at the outset did not seem necessary for clomipramine and for SSRIs to improve obsessive-compulsive symptoms.