A multicenter, randomized, placebo-controlled, double-blind study was conducted to evaluate the efficacy of pregabalin in preventing relapse of generalized anxiety disorder (GAD) after response to short-term treatment. Outpatients (n=624) with GAD for > or =1 year received open-label pregabalin (450 mg/day) for 8 weeks and, if a clinical response was observed, were randomized to receive either pregabalin (450 mg/day; n=168) or placebo (n=170) for 24 weeks. The primary efficacy parameter was time to relapse. Among responders to open-label acute treatment with pregabalin, time to relapse of GAD was significantly longer for patients treated with pregabalin compared with placebo (P<0.0001). Fifty per cent of the placebo group had relapsed by day 23, and at study endpoint, 65% had relapsed. In the pregabalin group, only 42% had relapsed by study end. Total attrition during double-blind treatment was somewhat higher on pregabalin compared with placebo (21.4 vs. 15.3%); attrition owing to adverse events (AEs) was also somewhat higher on pregabalin (6.0 vs. 2.4%). AEs were relatively low in the double-blind phase; only three AEs occurred with an incidence of more than 5% on pregabalin and placebo, respectively: infection (14.9 vs. 11.2%), headache (10.1 vs. 11.2%), and somnolence (6.0 vs. 0%). No safety concerns were identified with long-term treatment. The study indicates that pregabalin is an effective treatment for the prevention of relapse in patients with GAD.