Research, Evidence and Learning – The Need for a Global Infrastructure
Covid-19 has put the lives of millions of people at risk, creating uncertainties and heightening existing fragilities, particularly where social inequities and inequalities are most pronounced. Global health requires equitable, inclusive responses, informed by research, data and evidence. Existing global health research infrastructure is afflicted by weak institutional mechanisms and perpetuation of evidence hierarchies and silos and excludes and devalues different knowledges and lived experience. Major challenges include unevenness of financial support to global health research, evidence generation, and learning, policy engagement with too narrow a range of evidence; and insufficient investment in infrastructure for promoting international learning and exchange of health-related knowledge, evidence and data. Recommendations for action by G7 members to address these challenges include: (1) a jointly negotiated quota of 0.5% of G7 members’ national GDP for R&D funds administered through multilateral channels; (2) establishing a centralized health research clearing-house with joint governance for communication and action; (3) establishment of Pandemic Centres of Excellence in all world regions providing collaborative, regional mechanisms for medical research, social science research relevant to health, and vaccine production, distribution and delivery; (4) support to collaborative research networks that represent different forms of knowledge and experience, and use a diversity of research approaches and methodologies; and (5) investment in adaptive, agile national and regional systems for monitoring, early warning, and crisis preparedness, underpinned by open data and digital utilities. The paper also offers practical suggestions for implementing these recommendations in the short, mid and longer term, including G7 members working jointly with the UN, G20 and other international actors to join a global call for a 2023-2032 UN Decade for Health Research.