engleski

ADEQUACY OF REGISTRATION AND RESULTS REPORTING OF RANDOMIZED CONTROLLED TRIALS IN CLINICALTRIALS.GOV AND PUBLICATIONS

Aim: To assess effectiveness of legislative initiatives to stimulate registration of trial results by adherence to protocol and results reporting, changes to registry and publication data, and characteristics of adverse event reporting in randomized controlled trials (RCT) after introduction of the Food and Drug Administration Amendment Act (FDAAA). Study Design and Setting: Study 1: observational cohort of registered FDAAA-covered RCTs in ClinicalTrials.gov between 2009 and 2012 and publications. Study 2: trials of antidepressants, analgesics/anesthetics, anti-inflammatories, antineoplastics, enzyme inhibitors, hypoglycemics, neuromusculars, antidepressants, anti-allergics, and anti-infectives. Results: Secondary outcomes at initial and last registration were often omitted during registration. RCT registration changes mostly involved scientific title. Inclusion criteria omission was most common in publications. Inferential statistical methods for primary and secondary outcomes mismatched between registry and publication for many RCTs. Few trials with zero serious adverse events (SAEs) in registry published them as non-occurring. MedDRA was mostly used for SAEs and other adverse events. We found omitted medical terminology sources. A same lay term defined up to 3 different adverse events. Conclusions: Discrepancies remain relatively high between registered and published outcomes as well as in the reporting adverse events even after 10 years after registration policy implementation. Underreporting and inconsistencies in the reporting of data from RCTs seriously undermine transparency of clinical trials and need immediate attention of all stakeholders in health research.

ClinicalTrials.gov; randomized controlled trials; side effects

nije evidentirano