Authors: T. Christopher Bond, PhD1; Aaron Galaznik, MD2; Ann McMahon, MD, MS, FISPE3; Sabina Nduaguba, PhD4; Montse Soriano Gabarró, MD, MSc, FISPE5
Acknowledgments: Natasha Pratt (chapter review), Daniel Prieto-Alhambra, Patrick Ryan, George Hripcsak (OHDSI), Noelle Cocoros (Sentinel), Sebastian Bacon (OpenSAFELY), Miriam Sturkenboom (VAC4EU, EU PE&PV), Olaf Klungel (EU PE&PV), Aniketh Talwai (COVID-19 Research Database), Laurence Torcel-Pagnon, Kaat Bollaerts (COVIDRIVE)
1Bristol Myers Squibb
3Office of Pediatric Therapeutics, US Food and Drug Administration
4West Virginia University School of Pharmacy and Cancer Institute
5Bayer AG, Germany
The COVID-19 pandemic laid bare the limited interoperability capacity of the existing health care data infrastructure for assembling data to quickly address critical questions about a novel disease. Several initiatives have emerged to address this shortfall using real-world data (RWD) for COVID-19 observational research. This chapter features a selection of multi-stakeholder COVID-19 RWD initiatives that have contributed to our understanding of the COVID-19 pandemic and/or are structured to continue to provide opportunities for observational research about emerging issues related to the disease. The authors spoke to and/or emailed with representatives from each of the initiatives for the writing of this chapter.
RWD and real-world evidence (RWE) have never been needed as widely and urgently as in the COVID-19 era. Some initial analyses from RWD were needed for decision-making in real time, despite the known limits of the underlying data. Other questions continue to be explored and refined, and new questions emerge over time. COVID-19 is an ongoing pandemic and its effects will stay with us longer than we had hoped. The methods and resources that have emerged to address COVID-19 will ideally endure and evolve into the future. In turn, the way we think about and perform RWE research using RWD will change.
Perhaps the height of public awareness on the value of RWD sources came in January of 2020 with the Johns Hopkins COVID-19 Dashboard1 (more formally the COVID-19 Data Repository by the Center for Systems Science and Engineering at Johns Hopkins University), which uses publicly available epidemiological data from aggregated government, national, and regional data sources.2 Much has been written about the dashboard—how it came to be, how it was relied on, and how the COVID-19 data were interpreted. Spurred by a global emergency, the Johns Hopkins COVID-19 Dashboard supplied a constantly updating source of aggregated data to track COVID-19 cases, deaths, vaccinations, and more.3 Those raw numbers were and continue to be widely used. The real-time aspect of the dashboard and the public availability of the compiled data allowed both researchers and the general public to monitor the evolution of the pandemic as it occurred, and continues to do so.
Patient-level COVID-19 RWD sources have been used since early 2020 in different parts of the world to further characterize COVID-19 and to evaluate the impact of medical and public health interventions on the disease. Many of these data sources have been made available or established as part of multi-stakeholder consortia and networks, maximizing the use and value of RWD and methods as part of the pandemic response. Some of these have been established with long term goals and with enhanced data infrastructures, software, and tools.
Multi-stakeholder COVID-19 RWD initiatives with the potential to expand the depth of real-world observational studies have been compiled by the Duke Margolis Center for Health Policy4,5 and on a compilation of tools to support COVID-19 research produced by the Food and Drug Administration (FDA).6 Additional COVID-19 RWD multi-stakeholder initiatives have been added after conducting a targeted review of recent COVID-19 data sources and consultation with other researchers.
Some of the existing RWD initiatives leverage data platforms or distributed data networks to conduct COVID-19 research, while others are efforts newly launched during the pandemic. The initiatives that existed prior to the pandemic are mostly based on the commonly used health care RWD—claims and/or electronic health record (EHR) data. New initiatives evolved during the pandemic expand the types of RWD used to include surveillance data, consumer data, and epidemiological survey data. While most of the initiatives are closed systems with restricted access to their participating organizations/stakeholders, others were set up as open-source platforms or data aggregators, with the goal of making data rapidly available to a broader research community or to the general public in the COVID-19 era.
Six COVID-19 RWD multi-stakeholder initiatives are covered in depth in this chapter, as illustrative examples that may help us understand the future of COVID-19 research—and the future of RWD and RWE. Four of these were initiatives adapted from existing infrastructures, while the other 2 emerged during the COVID-19 pandemic. In our selection of initiatives for detailed description, we considered the need to cover different geographic locations (international, United States [US], Europe) and stakeholders (open science collaborative, government, health care-focused private stakeholders, technology-focused private stakeholders, private-public partnerships). The initiatives are:
- Observational Health Data Science and Informatics (OHDSI): an initiative leveraging existing international distributed health care data in an interdisciplinary collaborative that facilitates open-source analyses to conduct observational studies on COVID-19 disease characterization, treatment, and care
- FDA Sentinel: an FDA system leveraging a distributed data network with a common data model as well as other standalone data sources to conduct COVID-19-related studies
- OpenSAFELY: an initiative that enables the access of multiple United Kingdom (UK) government data sources in a reliable and protected platform to address COVID-19 research needs
- Vaccine Monitoring Collaboration for Europe (VAC4EU): an international non-profit association set up with the aim to conduct collaborative real-world analysis on vaccines. The entity was founded as a result of the Innovative Medicines Initiative-funded ADVANCE project that was initiated after the H1N1 pandemic
- COVID-19 Research Database: a cross-industry, cross-sector initiative composed of institutions that donate technology services, health care expertise, and de-identified data in the US for COVID-19 observational research. The data repository contains integrated, linked data sets from multiple sources, from the more traditional RWD (claims, EHR) to life insurance claims, consumer data, and mortality records
- COVIDRIVE: a public-private partnership leveraging the existing vaccine effectiveness platform in Europe to comprehensively examine COVID-19 vaccine effectiveness across a range of products and assist vaccine companies in fulfilling their regulatory obligations
As we continue through the COVID-19 era, questions to consider include:
- What other epidemiological RWD sources have emerged, expanded, changed, or become more accessible for COVID-19 observational research purposes?
- Which ones will continue to serve as methods to further understand COVID-19 disease and the effects of preventive and therapeutic interventions?
- In what ways has this global health emergency altered the way we approach RWD to generate RWE?