To revalidate the Freezing of Gait Questionnaire (FOG-Q), patients with Parkinson's disease (PD) were randomly assigned to receive rasagiline (1 mg/day) (n = 150), entacapone (200 mg with each dose of levodopa) (n = 150), or placebo (n = 154). Patients were assessed at baseline and after 10 weeks using the FOG-Q, Unified Parkinson's Disease Rating Scale (UPDRS), Beck Depression Inventory (BDI), and Parkinson's Disease Questionnaire (PDQ-39). FOG-Q dimensionality, test-retest reliability, and internal reliability were examined. Convergent and divergent validities were assessed by correlating FOG-Q with UPDRS, BDI, and PDQ-39. Comparisons between FOG-Q item 3 and UPDRS item 14 were also made. Principal component analysis indicated that FOG-Q measures a single dimension. Test-retest reliability and internal reliability of FOG-Q score was high. FOG-Q was best correlated to items of the UPDRS relating to walking, general motor issues, and mobility. Correlations between baseline and endpoint suggested that FOG-Q item 3 is at least as reliable as UPDRS item 14. At baseline, 85.9% of patients were identified as "Freezers" using FOG-Q item 3 (> or =1) and 44.1% using UPDRS item 14 (> or =1) (P < 0.001). FOG-Q was a reliable tool for the assessment of treatment intervention. FOG-Q item 3 was effective as a screening question for the presence of FOG.