Qualité méthodologique des études contrôlées randomisées. Une étude sur 5 ans de 7 Journaux d’Orthopédie selon les critères CONSORT, accessibles en Chine

Abstract

BackgroundIn recent years, the number of randomized controlled trials (RCTs) in the field of orthopaedics is increasing in Mainland China. However, randomized controlled trials (RCTs) are inclined to bias if they lack methodological quality. Therefore, we performed a survey of RCT to assess: (1) what about the quality of RCTs in the field of orthopedics in Mainland China? (2) Whether there is difference between the core journals of the Chinese Department of orthopedics and Orthopedics Traumatology Surgery & Research (OTSR). Material and methodsThis research aimed to evaluate the methodological reporting quality according to the CONSORT statement of randomized controlled trials (RCTs) in seven key orthopaedic journals published in Mainland China over 5years from 2010 to 2014. All of the articles were hand researched on Chongqing VIP database between 2010 and 2014. Studies were considered eligible if the words “random”, “randomly”, “randomization”, “randomized” were employed to describe the allocation way. Trials including animals, cadavers, trials published as abstracts and case report, trials dealing with subgroups analysis, or trials without the outcomes were excluded. In addition, eight articles selected from Orthopedics Traumatology Surgery & Research (OTSR) between 2010 and 2014 were included in this study for comparison. The identified RCTs are analyzed using a modified version of the Consolidated Standards of Reporting Trials (CONSORT), including the sample size calculation, allocation sequence generation, allocation concealment, blinding and handling of dropouts. ResultsA total of 222 RCTs were identified in seven core orthopaedic journals. No trials reported adequate sample size calculation, 74 (33.4 %) reported adequate allocation generation, 8 (3.7 %) trials reported adequate allocation concealment, 18 (8.1 %) trials reported adequate blinding and 16 (7.2 %) trials reported handling of dropouts. In OTSR, 1 (12.5 %) trial reported adequate sample size calculation, 4 (50.0 %) reported adequate allocation generation, 1 (12.5 %) trials reported adequate allocation concealment, 2 (25.0 %) trials reported adequate blinding and 5 (62.5 %) trials reported handling of dropouts. There were statistical differences as for sample size calculation and handling of dropouts between papers from Mainland China and OTSR (P<0.05). ConclusionThe findings of this study show that the methodological reporting quality of RCTs in seven core orthopaedic journals from the Mainland China is far from satisfaction and it needs to further improve to keep up with the standards of the CONSORT statement. Level of evidenceLevel III case control.