OBJECTIVE: To assess the quality of randomized controlled trials (RCTs) on Compound Danshen Dripping Pill, Compound Danshen Tablet, Suxiao Jiuxin Pill, Tongxinluo Capsule, and isosorbide dinitrate (ISDN) in treating coronary heart disease angina. METHODS: RCT reports were retrieved from CNKI, Wanfang Data, PubMed, China Master Theses Full-text Database, Chinese Electronic Periodical Services, Social Sciences Citation Index, Science Direct, Cambridge Journals Online, and EBM Reviews. Qualities of RCTs were assessed according to Jadad scale, M scale, CONSORT 2010, and CONSORT for herbal medicine. Kendall correlations between basic study characteristics and qualities of included RCTs were analyzed using R statistical software. RESULTS: Eighty-eight RCTs were included. The medians (means, 95% CIs) were 2.00 (2.09, [1.81,2.37]) for the Jadad scale, 4.00 (3.52,[2.95, 4.09]) for the M scale, 15.00 (15.39, [13.87, 16.91]) for the CONSORT 2010, and 19.00 (19.06,[17.16, 20.96]) for the CONSORT for herbal medicine. Only 9.09% (8/88) of RCTs were of high quality (Jadad score >2). These RCT quality measures were not correlated with individual items of reporting, such as sample sizes and follow-up periods. CONCLUSIONS: The quality of RCTs on Chinese patent medicine compared with ISDN was not improved from 1997 to 2009. It is urgent to improve the design of RCTs and the report quality.
OBJECTIVE: To assess the quality of randomized controlled trials (RCTs) on Compound Danshen Dripping Pill, Compound Danshen Tablet, Suxiao Jiuxin Pill, Tongxinluo Capsule, and isosorbide dinitrate (ISDN) in treating coronary heart disease angina. METHODS: RCT reports were retrieved from CNKI, Wanfang Data, PubMed, China Master Theses Full-text Database, Chinese Electronic Periodical Services, Social Sciences Citation Index, Science Direct, Cambridge Journals Online, and EBM Reviews. Qualities of RCTs were assessed according to Jadad scale, M scale, CONSORT 2010, and CONSORT for herbal medicine. Kendall correlations between basic study characteristics and qualities of included RCTs were analyzed using R statistical software. RESULTS: Eighty-eight RCTs were included. The medians (means, 95% CIs) were 2.00 (2.09, [1.81,2.37]) for the Jadad scale, 4.00 (3.52,[2.95, 4.09]) for the M scale, 15.00 (15.39, [13.87, 16.91]) for the CONSORT 2010, and 19.00 (19.06,[17.16, 20.96]) for the CONSORT for herbal medicine. Only 9.09% (8/88) of RCTs were of high quality (Jadad score >2). These RCT quality measures were not correlated with individual items of reporting, such as sample sizes and follow-up periods. CONCLUSIONS: The quality of RCTs on Chinese patent medicine compared with ISDN was not improved from 1997 to 2009. It is urgent to improve the design of RCTs and the report quality.