OBJECTIVES: The European Health Data and Evidence Network (EHDEN) is a federated network of European partners, with real-world data standardised to the Observational Medical Outcomes Partnership Common Data Model (OMOP CDM). We demonstrate how EHDEN can be used to generate real-world evidence (RWE) in the context of COVID-19, and highlight how RWE may generate questions on how to interpret it when it conflicts with randomized controlled trial (RCT) data.

METHODS: Evidence on the effectiveness of remdesivir and tocilizumab in people aged 18 years or older hospitalized with COVID-19 was collected from: 1) RCT data collected from a systematic literature review 2) RWE from a historical cohort study run using Institut Municipal Assistència Sanitària Information System (IMASIS, Spain) data. RCT data was synthesized in a Bayesian network meta-analysis (NMA). RWD were analysed using propensity score matching, logistic regression and cox proportional hazards. As the RWE is currently from a single data partner, this was compared qualitatively with the RCT data.

RESULTS: The RCT data (n = 15,246) did not show an effect on all-cause mortality for remdesivir vs tocilizumab (odds ratio: 1.10, 95% confidence interval 0.72-1.40). RWE evidence (n = 475) showed lower odds of mortality for people treated with remdesivir compared with those treated with tocilizumab (odds ratio: 0.38, 95% confidence interval 0.20-0.69, P value 0.002).

CONCLUSIONS: Federated data networks like EHDEN can be used to generate RWE. However, with the data analysed thus far, the RWE results are not consistent with RCT data. This difference may be attributable to remdesivir and tocilizumab being given at different stages of the disease course in people hospitalized with severe COVID-19. How health technology assessment (HTA) and regulatory agencies make decisions when presented with conflicting RCT and RWE evidence warrants special consideration.