THE PARKINSONS-DISEASE-QUESTIONNAIRE (PDQ-39) - DEVELOPMENT AND VALIDATION OF A PARKINSONS-DISEASE SUMMARY INDEX SCORE

Objectives: to briefly outline the development and validation of the P arkinson's Disease Questionnaire (PDQ-39) and then to provide evidence for the use of the measure as either a profile of health status score s or a single index figure. Design: the PDQ-39 was administered in two surveys. a postal survey of patients registered with local branches o f the Parkinson's Disease Society of Great Britain (n = 405) and a sur vey of patients attending neurology clinics for treatment for Parkinso n's disease (n = 146). Data from the eight dimensions of the PDQ-39 we re factor-analysed. This produced a single factor on the data from bot h surveys. Outcome measures: the eight dimensions of the PDQ-39 and th e new single index score-the Parkinson's disease summary index (PDSI), together with clinical assessments (the Columbia rating scale and the Hoehn and Yahr staging score). Results: in the postal survey 227 pati ents returned questionnaires (58.2%). All 146 patients approached in t he clinic sample agreed to take part. Higher-order principal-component s factor analysis was undertaken on the eight dimensions of the PDQ-39 and produced one factor on both datasets. Consequently it was decided that the scores of the eight domains could be summed to produce a sin gle index figure. The psychometric properties of this index were explo red using reliability tests and tests of construct validity. The newly derived single index was found to be both internally reliable and val id. Discussion: data from the PDQ-39 can be presented either in profil e form or as a single index figure. The profile should be of value in studies aimed at determining the impact of treatment regimes upon part icular aspects of functioning and well-being in patients with Parkinso n's disease, while the PDSI will provide a summary score of the impact of the illness on functioning and well-being and will be of use in th e evaluation of the overall effect of different treatments. Furthermor e, the PDSI reduces the number of statistical comparisons and hence th e role of chance when exploring data from the PDQ-39.