Assessing the Reporting Quality of Randomized Controlled Trials on Acupuncture for Acute Ischemic Stroke Using the CONSORT Statement and STRICTA

Objective To evaluate the reporting quality of randomized controlled trials (RCTs) on acupuncture for acute ischemic stroke. Methods Six databases including The Cochrane Central Register of Controlled Trials (CENTRAL, Issue 4, 2005), MEDLINE (1966 to December 2005), EMbase (1984 to December 2005), China National Knowledge Infrastructure (CNKI, 1994 to December 2005), China Biomedicine Database disc (CBMdisc, 1980 to December 2005), VIP (a full text issues database of China, 1989 to December 2005) were searched systematically. Handsearch for further references was conducted. Language was limited to Chinese and English. We identified 74 RCTs that used acupuncture as an intervention and assessed the quality of these reports against the Consolidated Standards for Reporting of Trials (CONSORT) statement and Standards for Reporting Interventions in Controlled Trials of Acupuncture (STRICTA) . Results In regard to the items in the CONSORT statement, 54 (73%) RCTs described baseline demographic and clinical characteristics in each group. Twenty-six (35%) mentioned the method of generating the random sequence, with 4 (5%) using a computer allocation. Only 6 (8%) RCTs had adequate allocation concealment, with 5 RCTs using sealed opaque envelopes and 1 RCT using centralized computer allocation. Only 8 (11%) RCTs used blinding, including 5 described as double-blind. Four (5%) RCTs reported the sample size calculation and 5 (7%) RCTs reported that an intention-to-treat (ITT) analysis. In regard to the items in STRICTA , only 4 (5%) RCTs reported the numbers of needles inserted. In 35 (47%) RCTs the needle type was reported, but only 26 (35%) mentioned the depths of insertion. Only 1 (1%) RCT mentioned the length of clinical experience and 6 (8%) RCTs reported the background of the acupuncture practitioners, but none stated the duration of their training. Conclusion The reporting quality of RCTs of acupuncture for acute ischemic stroke was low. The CONSORT statement and STRICTA should be used to standardize the reporting of RCTs of acupuncture.