Comparison of published orthopaedic trauma trials following registration in Clinicaltrials.gov

Abstract

Background: After the Food and Drug Administration Modernization Act of 1997, the registration of all clinical trials became mandatory prior to publication. Our primary objective was to determine publication rates for orthopaedic trauma trials registered with ClinicalTrials.gov. We further evaluated methodological consistency between registration and publication. Methods. We searched Clinical Trials.gov for all trials related to orthopaedic trauma. We excluded active trials and trials not completed by July 2009, and performed a systematic search for publications resulting from registered closed trials. Information regarding primary and secondary outcomes, intervention, study sponsors, and sample size were extracted from registrations and publications. Results: Of 130 closed trials, 37 eligible trials resulted in 16 publications (43.2%). We found no significant differences in publication rates between funding sources for industry sponsored studies and nongovernment/ nonindustry sponsored studies (p > 0.05). About half the trials (45%) did not include the NCT ID in the publication. Two (10%) publications had major changes to the primary outcome measure and ten (52.6%) to sample size. Conclusions: Registration of orthopaedic trauma trials does not consistently result in publication. When trials are registered, many do not cite NCT ID in the publication. Furthermore, changes that are not reflected in the registry of the trial are frequently made to the final publication. © 2011 Gandhi et al; licensee BioMed Central Ltd.