Quality of randomized controlled trials reporting in the treatment of sarcomas.

Abstract

Randomized controlled trials (RCTs) represent the best evidence in oncology practice. The aim of this study was to assess the reporting quality of sarcoma RCTs and to identify significant predictors of quality. Two investigators searched MEDLINE for pediatric and adult bone and soft tissue sarcoma RCTs published between January 1988 and December 2008. The quality of each report was assessed by using a 15-point overall reporting quality score based on 15 items from the revised Consolidated Standards of Reporting Trials (CONSORT) statement (overall quality score [OQS] range, 0 to 15 points). Concealment of allocation, appropriate blinding, and analysis according to intention-to-treat principle were assessed separately because of their crucial methodologic importance by using a 3-point key methodologic index score (MIS; range, 0 to 3). We retrieved 72 relevant RCTs that included 16,029 patients. The median OQS was 9.5. Allocation concealment, blinding, and analysis by intent to treat were reported only in 21 (29%), nine (12.5%), and 23 (32%) of the 72 RCTs, respectively. The median MIS was 1 with a minimum of 0 and a maximum of 2. On multivariate analysis, publication after 1996 and high impact factor remained independent and significant predictors of improved OQS. The sole variable associated with improved MIS was the publication of chemotherapy-only trials. Although the overall quality of sarcoma RCTs reporting has improved over time, reporting of key methodologic issues remains poor. This may lead to biased interpretation of sarcoma trial results.