Assessing the quality of reports about randomized controlled trials of scalp acupuncture combined with another treatment for stroke

• Main Authors: Young-Nim You, Myung-Rae Cho, Jae-Hong Kim, Ju-Hyung Park, Gwang-Cheon Park, Min-Yeong Song, Jin-Bong Choi, Jae-Young Han
• Format: Article
• Language: English
• Published: BMC 2017-09-01
• Series: BMC Complementary and Alternative Medicine
• Subjects: Randomized controlled trials; Scalp acupuncture; Stroke; Consort; STRICTA
• Online Access: http://link.springer.com/article/10.1186/s12906-017-1950-6

Abstract Background This study was designed to assess the quality of reporting on randomized controlled trials (RCTs) of scalp acupuncture for the treatment of stroke. Methods The following 8 databases were systematically investigated from their inception to December 2015: PubMed, Embase, Cochrane Library, China National Knowledge Infrastructure, National Institute of Informatics Scholarly and Academic Information Navigator, National Digital Science Library, Korean Traditional Knowledge Portal, and Korean Studies Information Service System. RCTs utilizing scalp acupuncture as an intervention for stroke were selected, and the quality of reports was assessed based on the Consolidated Standards of Reporting Trials 2010 statement (CONSORT) and Standards for Reporting Interventions in Controlled Trials of Acupuncture 2010 (STRICTA). For each study, the overall quality score (OQS) of 13 CONSORT items, a combined key methodological index score (MIS) of 5 CONSORT items, and the OQS of 17 STRICTA items were measured. Results The original reports of 63 RCTs were ultimately obtained, and the median CONSORT OQS was 7 (minimum 2, maximum 11). Particularly, the items ‘trial design’, ‘sample size’, ‘ancillary analyses’, and ‘harms’ had a positive rate of less than 10%. The median MIS was 1 (minimum 0, maximum 5), with ‘allocation concealment and implementation’ and ‘intent-to-treat analysis (ITT) analysis’ having a positive rate of less than 10%. The median STRICTA OQS was 11 (minimum 6, maximum 14), and only the items ‘sample size’ and ‘intent-to-treat analysis’ were reported, with a positive rate of less than 10%. The mean CONSORT OQS increased by approximately 0.81 for each 5-year period in which manuscripts were published (95% confidence interval: 0.43 to 1.19; p < 0.001). No variable was significantly associated with MIS in the ordinal regression model. Conclusion The quality of reports on RCTs investigating scalp acupuncture treatment for stroke was moderate to low. Furthermore, reporting of some items was either insufficient or inadequate in the majority of studies. In order to improve and standardize the quality of RCTs investigating scalp acupuncture for stroke, CONSORT and STRICTA guidelines should be utilized more frequently.