Patterns of outcome measurement in Parkinson's disease clinical trials

Neuroepidemiology. 2000 Mar-Apr;19(2):100-8. doi: 10.1159/000026244.


The study examines the pattern of use and clinimetric properties of clinical endpoints used in randomized trials for Parkinson's disease (PD). Randomized drug trials for PD were identified through a Medline search conducted from January 1966 until August 1998. The endpoints used in these trials were abstracted. Reports examining the clinimetric properties of the disease-specific scales used in these trials were also abstracted. Data regarding the consistency, accuracy, discrimination and feasibility of scales used in at least 10% of trials were determined. One hundred and thirty-seven articles met the inclusion criteria; 70.8% of trials used some clinical scale for PD as an endpoint. The Unified Parkinson's Disease Rating Scale (UPDRS) was the most commonly used scale (32.8%). Factors independently associated with the use of the UPDRS included: the study location in the US, mean age of subjects over 62.7 years and publication after 1994. The UPDRS was more thoroughly studied and superior in most clinimetric domains compared to scales developed earlier. Few studies included generic health status (2.9%) or cognitive measures (16.8%) as secondary endpoints. There have been definite improvements in the area of disease-specific measurement in PD trials since the introduction of the UPDRS. The results of studies that used instruments with poor or unreported clinimetric properties should be critically interpreted.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Aged
  • Feasibility Studies
  • Humans
  • Middle Aged
  • Parkinson Disease / therapy*
  • Randomized Controlled Trials as Topic
  • Retrospective Studies
  • Treatment Outcome