OBJECTIVE: Subtle signs and symptoms of Huntington disease (HD) are often present before impairments reach a point where the neurologic disease is manifest and a diagnosis must be considered. The objective is to examine the prognostic significance of these early clinical signs and symptoms regarding time until unequivocal clinical HD diagnosis. METHODS: We analyzed longitudinal data from 218 at-risk but healthy participants in the Huntington Study Group database who had either normal motor examination results or minimal soft motor signs at first observation. This group was followed periodically in HD clinics for up to 4.5 years. We used survival analysis to examine predictors of time until HD diagnosis. RESULTS: Diagnostic prediction was significantly improved using specific, nonredundant items from the Unified Huntington's Disease Rating Scale. When a movement disorder specialist initially had a global impression of "soft signs" present, cumulative relative risk of diagnosis was 4.68 times greater at 1.5 years of follow-up and 3.58 at 3 years. A neuropsychological test pattern with psychomotor speed 1 SD worse than a semantic knowledge measure increased cumulative risk by 1.99 times at 1.5 years and 1.81 at 3 years. Finally, reports of various subjective HD symptoms increased 3-year relative risk by 2.6 to 3.4. CONCLUSIONS: Findings demonstrate that neuropsychological performance and both the clinician rating and the patient subjective perception of motor difficulties contribute nonredundantly to a prediction of Huntington disease diagnosis. These findings may have implications for prognostic assessment of persons at risk and eventually assist with early interventions. ©2007AAN Enterprises, Inc.