Test-retest reliability of an epidemiological instrument for Parkinson's disease.

Reliability testing was conducted on a survey instrument for a genetic epidemiological study of Parkinson's disease (PD). Information on demographics, medical history, places of residence, and occupations was collected. The instrument was administered to 11 pairs of siblings discordant for PD and then re-administered after 4 weeks. Twenty of 22 (91%) subjects completed and returned the retest questionnaire. Overall reliability of the instrument was acceptable (kappa = 0.88). This instrument could be used to elicit reliable information from similar profiled subjects in future investigations.