Although an ever-increasing volume of evidence exists on what works in global development, participants identified a need to ensure that this evidence is accessible to both evidence producers and users. Making evidence widely accessible holds a range of advantages; it can help to secure the foundations of the existing knowledge base and identify where gaps exist, reduce the unnecessary duplication of efforts, and allow decision makers to tap into existing findings on the impact and cost-effectiveness of specific interventions. Meanwhile, synthesising evidence can help identify where a body of high-quality evidence paints a clear picture of the effectiveness of policy solutions and enable policymakers who may have limited time or statistical capacity to integrate key learnings into their decisions.
“Although policymakers need to make decisions based on evidence, they often don’t have the time or inclination to go back to original studies.”
Integrating existing evidence repositories. Participants highlighted the existence of discrete evidence markets and libraries that allow researchers and policymakers to search for existing evidence on the impact of specific interventions. However, existing evidence repositories are not consistently integrated with one another, while assessments of quality often vary. The formalisation of links between existing repositories was identified as a priority, while the introduction of universal quality standards was suggested as a way to ensure that high-quality evidence is identified and promoted. Working in partnership with other evidence generators was positioned as vital to allow for linkages to be made across existing evidence libraries and for gaps in the evidence base to be identified.
AI and evidence synthesis. Synthesising evidence was seen as a particularly powerful use case for artificial intelligence (AI). Although participants flagged instances of individual organisations piloting the use of AI in evidence synthesis, it was noted that much of this work is being undertaken in isolation. A call was made for organisations to work more closely with one another to share lessons in this area, and to collaborate on identifying additional use cases for AI (eg. for country-specific evidence chatbots to also connect to international repositories such as ImpactAI).
Enhancing evidence accessibility. While evidence synthesis has a range of benefits, contributors emphasised the need to look beyond synthesis and make evidence accessible to a wider range of audiences. Borrowing good practice from academia represents a potentially fruitful way forward, with suggestions including the utilisation of a system similar to Digital Object Identifiers (DOIs) for evidence reports, forming better linkage mechanisms between papers by particular authors or covering a particular theme, and geo-tagging to allow for evidence centred on particular geographic contexts to be easily identified. More fundamentally, participants recognised a need to ensure that evidence users are not prevented from accessing relevant findings, and advocated for changes in the current system of publication to incentivise institutions to allow open access to the evidence they produce. Participants also highlighted an ethical requirement to ensure that locally-produced research is freely accessible to local evidence users, and noted the potential for this to contribute to driving forward the locally-led development agenda.
Supporting evidence adoption. Beyond the free availability of evidence, participants highlighted the value of services to interpret and summarize studies for key target groups such as program managers. Examples include rapid evidence assessments and evidence hotlines as employed by various countries.