We evaluated the ability of nurse clinicians to assess parkinsonian signs in older persons with a modified version of the motor section of the Unified Parkinson's Disease Rating Scale (UPDRS). After completing a structured training protocol, three nurse clinicians and a neurologist with expertise in movement disorders administered a modified UPDRS to 75 older persons. The nurses repeated the assessment about 3 weeks later. Inter-rater agreement and short-term temporal stability were estimated for each item, the total modified UPDRS score, and for summary measures of bradykinesia, postural reflex impairment, rigidity, and tremor, and a global parkinsonian sign score. We performed our assessment in Catholic religious communities in the Chicago area, using consecutive subjects at four communities participating in the Religious Orders Study, a longitudinal, clinical-pathologic study of older persons. Our results showed that nurses were not a significant source of variability, with intraclass correlations exceeding 0.97 for all items, and they showed good to excellent agreement with the neurologist for most modified UPDRS items. Correlations between nurses and neurologist exceeded 0.90 for the total modified UPDRS, ranged from 0.76 to 0.95 for the four parkinsonian domain scores, and exceeded 0.90 for the global parkinsonian sign score. Nurses showed fair to good agreement with themselves over the 3-week interval for most modified UPDRS items. Correlations over the 3-week interval exceeded 0.90 for the total modified UPDRS score, ranged from 0.70 to 0.95 for the four domain scores, and exceeded 0.90 for the global parkinsonian sign score. Ratings of parkinsonian signs by nurse clinicians corresponded closely to those of a neurologist with expertise in movement disorders and showed good inter-rater agreement and temporal stability. With appropriate training, nurse clinicians can reliably administer the modified UPDRS.