We examined a sample of individuals in the Nebraska State Parkinson's Disease Registry (NSPDR) to determine what proportion meets standard criteria for Parkinson's disease (PD). The NSPDR was established in 1996 in an effort to understand the high prevalence of PD in Nebraska. Only minimal demographic data are included for each entrant. Subjects enter the NSPDR by means of diagnosing physicians, pharmacists dispensing anti-PD medications and the patients themselves. A series of 356 registrants diagnosed between 1997 and 2001 were contacted and invited to participate in a case-control study. Medical records were reviewed by a single abstractor in a standard manner. When patients consented, history was filled in by interview. A subset of patients were examined by a movement disorders specialist, who assigned all patients a probability of PD. Where sufficient information was available, 78% of registrants were confirmed to have PD (i.e., percent probability > 50%), including 83% of the patients previously diagnosed by a neurologist. Tremor was an initial symptom in 72% of confirmed versus 39% of excluded cases, and resting tremor was present in 86% of those that were confirmed. The most frequent reasons for exclusion were drug-induced Parkinsonism, multiple systems atrophy, vascular disease, and essential tremor. Use of the NSPDR for epidemiologic study requires careful review of the data set before assignment of cases. When histories are compiled in a standardized, comprehensive manner, the necessity for actual patient examinations can be minimized.