J Am Coll Cardiol. 2026 Jul 28;88(4):437-450. doi: 10.1016/j.jacc.2026.05.028.
ABSTRACT
BACKGROUND: Public health emergencies require rapid generation, synthesis, and translation of clinical evidence into practice guidelines; yet, systematic synthesis of the challenges and lessons from these processes remains limited.
OBJECTIVES: The purpose of this study was to examine how evidence on anticoagulation in COVID-19 was generated and translated into clinical guidance, using randomized controlled trials (RCTs) as a case study, and to contextualize these processes with an individual participant data meta-analysis (IPDMA).
METHODS: Systematic searches identified RCTs comparing therapeutic- vs nontherapeutic-dose anticoagulation in noncritically ill patients hospitalized for COVID-19. Trial characteristics, evidence accumulation, and associated guideline recommendations were summarized over time. An IPDMA was performed to estimate summary treatment effects using mixed-effects logistic regression.
RESULTS: Evidence evolved from an early coordinated platform RCT (preprint May 2021) to subsequent RCTs published between June 2021 and July 2023. In exploratory counterfactual sequential IPDMA, the pooled treatment effect on organ support or death became statistically significant in May 2021, approximately 13 months after first patient enrollment (adjusted OR: 0.73; 95% CI: 0.58-0.91; 6 RCTs, n = 3,944). Guideline recommendations shifted from uniform endorsement of prophylactic-dose anticoagulation in 2020 to recommendations supporting therapeutic-dose anticoagulation in 2022, with variation in certainty across guidelines. Individual participant data were obtained after completion of data transfer in April 2024 and included 7 RCTs comprising 6,362 patients. Therapeutic-dose anticoagulation reduced the odds of organ support or death (12.9% vs 16.2%; adjusted OR: 0.81; 95% CI: 0.67-0.97). Major bleeding was rare (0.8% vs 0.5%).
CONCLUSIONS: The COVID-19 anticoagulation experience highlights how delays in coordination, data sharing, and synthesis can slow the translation of evidence into practice. Therapeutic-dose anticoagulation was associated with reduced odds of organ support or death and, in this context, serves as a case study of how evidence emerges and is acted upon under conditions of uncertainty. Strengthening coordinated research networks, platform trial infrastructures, prioritized funding, and near-real-time cross-trial synthesis may improve the timeliness, reliability, and responsiveness of evidence systems during future health emergencies.
PMID:42523015 | DOI:10.1016/j.jacc.2026.05.028