Evaluation of consistency in adverse event reporting between trial registry and publications in COVID-19 pharmacological intervention trials

INTRODUCTION: Complete and consistent reporting of adverse events (AE) affects decisions in clinical practice. We conducted a cross-sectional study aiming to assess transparency and completeness of reported adverse events from randomized clinical trials (RCTs) on pharmacological interventions, including biologicals, to treat Coronavirus disease 2019 (COVID-19) registered on ClinicalTrials.gov on or after January 1, 2020, and updated on or before May 31, 2021, along with corresponding publications. AIM: We aimed to assess the completeness and consistency of adverse event and all-cause mortality reporting between ClinicalTrials.gov and corresponding publications. METHOD: We conducted a cross-sectional comparison of safety reporting between ClinicalTrials.gov registry and peer-reviewed publications of RCTs investigating COVID-19 pharmacological interventions, including biologicals. Two authors evaluated RCTs to reach κ ≥ 0.80. RESULTS: A total of 68 trials were assessed for discrepancies in adverse event and all-cause mortality data. Thirty-one (46%) were industry-funded, and 44 (65%) were double-blind randomized clinical trials. Forty-nine (72%) publications had discordant counts of serious adverse events (SAE) descriptions, 35 (51%) had discrepancies in the number of patients affected by SAE, and 11 (16%) omitted all-cause mortality compared to the records in the ClinicalTrials.gov registry. CONCLUSION: Discrepant reporting of AEs and essential trial data was high in trials on COVID-19 therapeutics.