Systematic data capture reduces the need for source data verification: exploratory analysis from a phase 2 multicenter randomized controlled platform trial

Le résumé fourni par la source

BACKGROUND: The COVID-19 pandemic gave rise to clinical trials focused on systematic, accurate primary data capture, and reduced reliance on source data verification (SDV). Here, we report on a natural experiment that allowed us to assess the quality, cost, and impact of this approach compared to traditional SDV. METHODS: The I-SPY COVID trial (NCT04488081) was a multicenter, open-label, platform trial that employed a streamlined daily checklist, daily capture of labs and medications, and centralized monitoring to ensure accurate data collection in lieu of SDV. The trial enrolled 1,111 patients in 11 drug arms with severe COVID-19. After the trial arms were closed, extensive retrospective SDV was performed on 333 (30.0%) patients, including 10,101 of 44,486 (23%) electronic case report forms (eCRFs), allowing us to evaluate the impact of our strategy on data integrity, outcomes, and costs. RESULTS: We find that retrospective SDV results in changes to 0.36% (1,234 / 340,532) of data fields. It results in no changes to the type of outcome recorded (death, recovery, or censored), but changes in the day of recovery in 9 instances, by a median of 2 days (range 1-7). Two additional AEs are added during SDV that had not previously been captured. Costs associated with retrospective SDV of 23% eCRFs are 61,073 person-hours at a cost of $6.1 M. CONCLUSIONS: Extensive SDV does not change any results or conclusions of the I-SPY COVID trial, which was designed with a systematic strategy for data capture, monitoring, and safety. This strategy could improve the efficiency of clinical trials and eliminate the need for manual SDV.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.