Objective: To determine the rate of functional decline in a large cohort of patients with Huntington's disease (HD) followed at 43 sites by the Huntington Study Group (HSG). Methods: The annual rate of functional decline was measured using the Total Functional Capacity Scale (TFC) and the Independence Scale (IS) in 960 patients with definite HD followed prospectively for a mean of 18.3 months. All patients were rated with the Unified Huntington's Disease Rating Scale (UHDRS). Sample size calculations for hypothetical clinical trials were calculated. Results: A factor analysis of the UHDRS at baseline yielded 15 factors accounting for 77% of the variance. The TFC score declined at a rate of 0.72 units/year (standard error [SE] 0.04) and the IS score declined at a rate of 4.52 units/year (SE 0.23). Lower TFC score at baseline, indicating more severe impairment, was associated with less rapid annual decline in TFC score, perhaps reflecting the floor effect of the scale. The annual rate of decline for 575 patients with baseline TFC scores of 7 to 13 was 0.97 (SE 0.06), was 0.38 (SE 0.08) for 270 patients with baseline TFC scores of 3 to 6, and was 0.06 (SE 0.1) for 101 patients with TFC scores of 0 to 2. In multivariate analysis (n = 960), longer disease duration and better cognitive status at baseline were associated with a less rapid rate of decline in TFC score, whereas depressive symptomatology was the only factor associated with more rapid decline on the IS score. Age at onset of HD, sex, weight, and education did not affect decline on either score. Conclusions: The comparable rates of decline on the TFC and the IS scores with other published studies suggest that these estimates of functional decline are representative of HD patients who are evaluated at HSG research sites. In longitudinal analysis, longer disease duration and better neuropsychological performance at baseline were associated with a less rapid rate of decline in TFC score, whereas depressive symptomatology at baseline was associated with a more rapid decline in the IS score. These rates of functional decline and the covariates that modify them should be considered in estimating statistical power and designing future therapeutic trials involving HD patients with early or moderately severe disease.