Beliefs that exposure with response prevention (ERP) is excessively distressing and will result in client dropout from treatment are commonly-cited reasons for clinicians not providing evidence-based treatment. This meta-analysis examined treatment attrition for ERP for youth with obsessive compulsive disorder (OCD) compared with other treatment modalities. A systematic literature search identified 11 randomized controlled trials (RCTs) comparing ERP to active or waitlist control conditions, 9 comparing pharmacotherapy to control, and 3 comparing ERP to pharmacotherapy for youth with OCD. Attrition rates were low for ERP (10.24%) compared to pharmacotherapy (17.29%), active control (e.g., relaxation, metacognitive therapy; 20.63%), and pill placebo (23.49%). ERP had lower risk of attrition compared to active control conditions (RR = 0.60; p = .02), and was not significantly different to waitlist (RR = 0.80; p = .59). In head-to-head trials, there was no difference between the risk of attrition from ERP and pharmacotherapy (RR = 1.26; p = .74). Of the pharmacotherapy trials, risk of attrition from serotonin reuptake inhibitors treatment was not significantly different compared to placebo (RR = 0.94; p = .76), with no difference between antidepressants and clomipramine (p = .19). Attrition from ERP was primarily for logistical reasons, compared to lack of efficacy for relaxation and/or adverse reactions from pharmacotherapy. Attrition from ERP is low, and is generally lower than non-ERP interventions. Given favorable attrition and efficacy data, there is little justification for appropriately-trained clinicians not to offer ERP as a first-line treatment for youth with OCD. © 2019 Wiley Periodicals, Inc.