Methods for the efficient and accurate detection of parkinsonism are essential for epidemiological studies. We sought to determine whether parkinsonism could be detected by a neurologist from a videotaped assessment and whether neurobehavioral methods (motor, cognitive, and sensory) discriminated between patients with Parkinson's disease (PD) and controls. Fifteen patients with mild PD (Hoehn and Yahr I-III) were compared to 15 age-, sex-, and education- matched controls. Each participant underwent a videotaped neurological examination (based on the Unified Parkinson's Disease Rating Scale, UPDRS), administered by a trained technician, and reviewed by a neurologist, as well as a series of neurobehavioral tests. The neurologist identified PD patients with 86% sensitivity and 100% specificity. Among the neurobehavioral tests, finger tapping, combined with one or more among olfaction, visual contrast sensitivity, or Paired Associates Learning, correctly classified 90%, or more, of subjects. Individual psychological tests did not discriminate reliably between groups. We conclude that videotaped assessments of parkinsonism or objective tests of motor and sensory function can accurately detect patients with PD. Both approaches have potential for identifying PD cases, but the latter may be more efficient for screening.