Evidence synthesis in pandemic times to inform global guidance: WHO HQ Rapid Review Group experiences

Article type
Authors
Askie L1, Ballesteros Silva M1, Grimmer K2, Kothari K1, López Alcalde J3, Movsisyan A4, Turner L1
1World Health Organization, Geneva, Switzerland
2World Health Organization, Geneva, Switzerland; Faculty of Medicine and Health Sciences, Stellenbosch University, Stellenbosch, South Africa
3World Health Organization, Geneva, Switzerland; Instituto Ramón y Cajal de Investigación Sanitaria (IRYCIS), Unidad de bioestadística clínica - Hospital Universitario Ramón y Cajal (CIBERESP), Madrid, Spain
4World Health Organization, Geneva, Switzerland; Institute for Medical Information Processing, Biometry and Epidemiology (IBE), Chair of Public Health and Health Services Research, Faculty of Medicine, LMU Munich, Munich, Germany
Abstract
Background:
The World Health Organization (WHO) does not have a dedicated within-house evidence-synthesis service. The Rapid Review Group (RRG) was formed within the WHO Science Division in March 2020 to support WHO Technical Units responding to the COVID-19 pandemic. We report on challenges, innovations, and lessons learned by this group.

Methods:
The RRG comprised experienced methodologists from around the world who were available immediately, prepared to work virtually and out-of-hours, on short-term contracts. The WHO COVID-19 Research Database was the main evidence source for reviews, and RRG worked closely with an information specialist. RRG tasks included collaborating with technical units to clarify questions, structuring protocols, and systematic, comprehensive searches, efficiently identifying relevant literature from the exponentially increasing database, producing synthesized evidence reports, and advising end users on evidence relevance, interpretation, and use.

Results:
The RRG worked 16,780 hours over 46 months. They produced 205 evidence synthesis products underpinned by comprehensive search strategies (69 evidence inventories, 44 rapid reviews, 20 scoping reviews, 16 stand-alone search strategies, 8 systematic reviews, 3 protocols, and 1 major living review) and provided methods consultation and/or peer-review for 44 additional projects. Few research questions the RRG were asked to address were answerable by experimental studies. Most review findings required significant contextual framing because the available evidence was often too broad to inform immediate real-world situation needs. Standard steps for rapid evidence production in nonpandemic times were rarely achievable, given the complexity and volume of requests, the urgency of producing answers, and time and resource constraints. Nonnegotiable components of systematic, high-quality rapid reviews were debated within the RRG, and steps to ensure optimal review rigor within time and resource constraints were determined on a case-by-case basis.

Conclusion:
The RRG provided WHO with consistently rigorous, transparent evidence synthesis products for use in a broad range of emergency response questions. The lessons learned from producing evidence synthesis quickly on a global scale by such a small team highlighted not only how rapid review steps can be adapted, as required, to produce trustworthy evidence within time and resource constraints but also the enormous value to WHO of an in-house evidence synthesis unit.