Reporting of medication adherence in randomized controlled trials of pharmacological interventions for SARS-CoV-2: a cross-sectional analysis

BackgroundAdherence to pharmacological interventions in clinical trials is crucial for correct estimation of beneficial and adverse effects, including trials of SARS-CoV-2. The Template for Intervention Description and Replication (TIDieR) - a 12-item extension of the Consolidated Standards of Reporting Trials (CONSORT) reporting guidelines - includes two items (11 and 12) that address intervention adherence reporting in trial publications. ObjectiveTo assess compliance with TIDieR items 11 and 12 of randomised controlled trials (RCTs) of interventions in SARS-CoV-2 infection published in five selected journals during 2021. MethodsWe assessed SARS-CoV-2 pharmacological RCTs published in the Annals of Internal Medicine, The BMJ, JAMA, The Lancet, and The New England Journal of Medicine in 2021 for compliance with TIDieR items 11 and 12. Item 11 was assessed in two parts: 11a--how intervention adherence was assessed; 11b--if any strategies were used to maintain or improve how intervention adherence was maintained or improved. Item 12 assessed the extent to which the intervention was delivered as planned. We calculated raw adherence and proportional (weighted) adherence for pharmacological and comparator interventions where available. ResultsWe found 75 eligible RCTs, of which 28 (37%) reported results related to SARS-CoV-2 vaccinations. Compliance with items 11a and 12 could be assessed in 71 of these 75. Of those 71 RCTs, 37 (52%, 95% confidence interval 40-64%) were compliant with reporting of item 11a. Seven RCTs had a strategy to assess compliance with item 11b, and only three (43%, 9-82%) of those complied with item 11b reporting. Of the 71 RCTs, 70 complied with reporting of item 12. Only one of the 71 RCTs (1.4%, 0-7.6%) fully complied with TIDieR items 11a, 11b, and 12. Compliance varied across journals. ConclusionsRCTs of SARS-CoV-2 pharmacological interventions published in high-impact medical journals complied variably with reporting of intervention adherence, even though the journals endorse CONSORT. The implications for interpretation, application, and replication of findings based on these publications warrant consideration.