Comparative efficacy and tolerability of pharmacological interventions for the acute treatment of obsessive-compulsive disorder across the lifespan: a systematic review and network meta-analysis

Pharmacological treatment is recommended for obsessive-compulsive disorder.

Several options are available, yet little is known on comparative effects of individual medications. We aimed to elucidate the efficacy and tolerability of pharmacotherapy for people with obsessive-compulsive disorder, across the lifespan.

We did a systematic review and network meta-analysis comparing short-term pharmacological interventions for obsessive-compulsive disorder. We searched multiple databases, including MEDLINE, Embase, PsycInfo, and the Cochrane Central Register of Controlled Trials, for literature on licensed or recommended pharmacological treatments for obsessive-compulsive disorder, without age restrictions, from database inception up to March 11, 2026.

We included double-blind, randomised controlled trials with outcomes between 8 and 16 weeks. Two reviewers independently and in duplicate extracted outcomes, with a quality check done by a third reviewer.

Primary outcomes were efficacy, defined as change from baseline score, and tolerability, defined as the proportion of participants discontinuing treatment due to adverse events. Separately for children and adolescents (aged 0-17 years) and adults (aged ≥18 years), we calculated standardised mean differences (SMDs) and odds ratios (ORs) with random-effects pairwise and network meta-analysis.

The certainty of evidence was assessed using the Confidence in Network Meta-Analysis (CINeMA) framework. People with lived experience of obsessive-compulsive disorder were involved in the entire review process, including protocol development, writing, and interpretation of the findings.

The study was preregistered in Open Science Framework (identifier 4V3GS). We included 52 randomised controlled trials with 7316 participants.

Nine trials were included for children and adolescents (N=822; median age 12·5 years [IQR 11·6-13·2]; 352 [55%] male and 285 [45%] female among 637 participants with reported sex distribution). 43 trials were included for adults (N=6494; median age 36·4 years [35·0-37·9]; 3105 [49%] male and 3200 [51%] female among 6305 participants with reported sex distribution).

Data on ethnicity were reported in 15 (29%) of 52 randomised controlled trials, using inconsistent definitions. In children and adolescents, clomipramine was more efficacious than placebo (SMD -1·19, 95% CI -1·73 to -0·64, CINeMA low), as were sertraline (-0·52, -0·80 to -0·02, low), fluoxetine (-0·43, -0·76 to -0·10, low), and fluvoxamine (-0·39, -0·71 to -0·07, low).

Clomipramine was more efficacious than fluoxetine (SMD -0·75, 95% CI -1·39 to -0·11; very low), fluvoxamine (-0·79, -1·43 to -0·16; very low), paroxetine (-0·87, -1·51 to -0·22; very low), and sertraline (-0·87, -1·51 to -0·22; very low). Only sertraline was less tolerable than placebo (OR 4·51, 95% CI 1·36 to 14·96; low).

In adults, clomipramine (SMD -0·71, 95% CI -0·88 to -0·54; high), fluvoxamine (-0·63, -0·83 to -0·43; high), paroxetine (-0·51, -0·73 to -0·29; moderate), sertraline (-0·49, -0·70 to -0·28; moderate), escitalopram (-0·43, -0·85 to -0·02; moderate), and fluoxetine (-0·37, -0·6 to -0·08; moderate) were more efficacious than placebo, and clomipramine was more efficacious than fluoxetine (-0·34, -0·65 to -0·04; low). Clomipramine (OR 4·65, 95% CI 3·27 to 6·60; moderate), fluvoxamine (2·73, 1·77 to 4·20; moderate), fluoxetine (2·52, 1·27 to 4·98; moderate), sertraline (2·23, 1·49 to 3·34; moderate), and paroxetine (1·99, 1·36 to 2·93; moderate) were less tolerable than placebo.

Clomipramine was less tolerable than all active comparators (OR range 1·70 to 3·50; moderate) except fluoxetine (moderate) and venlafaxine (low).

Clomipramine showed the largest estimated efficacy for obsessive-compulsive disorder but had a lower tolerability than SSRIs.

The evidence of its superiority over individual SSRIs was more pronounced in children and adolescents. Our findings suggest that the trade-off between efficacy and tolerability should be discussed within an individualised, shared, decision-making framework.