Implementation of the multicountry WHO COVID-19 pregnancy cohort study: challenges and lessons learned during the pandemic

Introduction: A generic research protocol was developed for a prospective cohort study to allow systematic, harmonized data collection of the impact of SARS-CoV-2 infection and vaccination during pregnancy on maternal, obstetric, and neonatal outcomes across different settings. This article describes the study conception, development, implementation, challenges, and key lessons learned within study sites across the world. Methods: The protocol was implemented in 43 facilities in 10 countries during the pandemic, involving consecutive recruitment of over 16,000 pregnant or postpartum women. We evaluated selection of study sites, ethical approvals, staff recruitment and training, recruitment and follow-up, and incorporation of new elements over the course of the pandemic across the study sites. Results: Study implementation in multiple LMIC settings was feasible; however, major challenges included delays in study implementation due to ethical approval procedures and availability of testing for exposure assessment. Implementation of research during a constantly evolving pandemic context led to the need for amended protocols, adjusted sample sizes, new outcomes and variables, repeated review by the Ethical Committees and adapted laboratory protocols. For example, the first COVID-19 vaccines became available after the study had started, with the need to modify the data collection forms and serologic testing algorithm to allow incorporation of this information in the study structure and analysis. Conclusion: Study implementation during a pandemic in different countries and periods was challenging but is not only expected to provide important information on the effects of SARS-CoV-2 infection and vaccination on pregnancy, but also on conducting research during future outbreaks. More streamlined ethics reviews during pandemics, availability of generic protocols in advance, and sites in LMICs ready to activate in an outbreak, as opposed to triggering processes during a crisis, would be highly beneficial.