OBJECTIVES: To carry-out a cost-effectiveness analysis (CEA) of the effect of Pregabalin (PGB) versus usual care in refractory out-patients with Generalized Anxiety Disorder (GAD) treated according with usual medical practice in Mental Health Centers in Spain. METHODS: CEA was carried-out by means of a nested case-control design which used secondary data extracted at random from a 6-month cohort prospective observational study (the ADAN study) conducted to ascertain cost-of-illness in GAD (DSM-IV criteria). Refractory subjects were those who claimed of persistent symptoms of anxiety and showed suboptimal response (HAM-Anxiety scale≥16) after a course of standard doses of any anti-anxiety drug, alone or in combination, for 6 months. At baseline, patients could switch to PGB (monotherapy or add-on) or to usual care, alone or in combination, which were considered cases and controls respectively. Health outcomes included quality-adjusted life years (QALYs) gain by EQ-5D. A 3rd-payer perspective was chosen for the CEA analysis in year 2009. Sensitivity analysis was performed by bootstrapping techniques (10000 re-samples were obtained) in order to obtain a cost-effectiveness acceptability curve. RESULTS: A total of 858 refractory subjects (mean baseline HAM-A; score 25.7) were identified; 429 in each of the PGB and UC group. Compared with UC, PGB was associated with significant higher QALY gain after 6 months of treatment; 0.1225+0.0995 vs 0.1009+0.0938 (p=0.001), but significantly increased healthcare costs; €1,240.6+1,407.4 vs €931.4+940.2 (p<0.0001). The deterministic ICER ratio was €14,287 per QALY gained with a re-sampling ICER of €15,804 (95 CI: 15,064-16,545). The 90% of re-samples fold below €25,052 threshold and 95% ≤ €30,000. CONCLUSIONS: Despite the design of the primary study, this evaluation could suggest that pregabalin may be cost-effective in comparison with usual care in refractory out-patients with Generalized Anxiety Disorders managed in Mental Health Centers under usual medical practice in Spain.