BACKGROUND:To evaluate the quality of the literature addressing traditional Chinese medicine for treating Parkinson's disease.DATA SOURCE:A computer-based online search of Chinese publications from January 2001 to July 2008 was conducted in Chinese Biology Medical Disc Database and China National Knowledge Infrastructure. Search key words were Parkinson's disease, integrated traditional Chinese and Western medicine, traditional Chinese medicine therapy, and Chinese herb therapy.DATA SELECTION:Articles describing randomized, controlled trials and quasi-randomized, controlled trials were included. Literature quality was assessed using the criteria-Systematic evaluation of clinical literature related to treatment of Parkinson's disease with traditional Chinese medicine. This included methodology, interventions in the treatment/control group, evaluation criterion of outcomes, and frequency.MAIN OUTCOME MEASURES:Evaluation criterion of outcomes (various score methods and evaluation scales), methodological quality, and frequency distribution were all measured.RESULTS:A total of 33 articles with randomized, controlled trials were included. Of these, six described a random method, and the remaining did not describe random allocation methods or random sequence generation methods. None of the studies estimated sample size. Case descriptions of withdrawal and loss to follow-up were unclear. Both the Unified Parkinson's Disease Rating Scale and Webster scale were used in the eligible studies as evaluation criteria.CONCLUSION:There are no high-quality studies that address traditional Chinese medicine therapy and integrated traditional Chinese and Western medicine for treating Parkinson's disease in China. Eligible studies were not performed in accordance with Consolidated Standards of Reporting Trials statement or Standards for Reporting Interventions in Controlled Trials of Acupuncture criteria, and the literature quality was low. The presently used criteria for evaluating therapeutic effects do not completely assess outcomes of traditional Chinese medicine for treating Parkinson's disease. The identification of precise outcomes should be verified using randomized, controlled studies with adequate controls and proper designs.