How can we improve policymakers’ access to and use of high-quality statistical models for decision-making? R4D collaborates with research partners in Burkina Faso, India, Kenya, and Nigeria to understand critical lessons for this knowledge translation work.
The Challenge
Decision-makers in the health sector often face complex choices and trade-offs. Modeled evidence – evidence generated using mathematical models that simulate different health scenarios – can be a valuable tool to help inform policy- and practice-level decisions, with 95% of surveyed modelers and decision-makers in one World Health Organization (WHO) survey agreeing that modeled evidence should be used to inform guidance for public health recommendations, particularly to determine the relative effectiveness and cost-effectiveness of various interventions (Norris et al., 2018). However, decision-makers do not always use modeled evidence for reasons including a lack of policy-relevant models, the perception that models are too complex to understand or based on too many assumptions, and a lack of communication between decision-makers and modelers (Knight, G. M., 2016; Campbell et al., 2009; Innvær et al., 2002; Oliver et al., 2014). The inability to ensure that decisions are informed by the best available data can result in losses in efficiency, effectiveness, and impact, which affect the end users of the health system.
The Opportunity
R4D’s study, conducted between October 2021 and June 2022, aims to understand how to bridge the gap between modeled evidence and policy/program decision-making by:
- Identifying the factors and approaches that facilitate or inhibit exchange between decision-makers and modelers
- Evaluating current practices in forums where translation work is already occurring
- Proposing changes in funding approaches, organizational structures, and country or global policies to enable success
Our Work
R4D is partnering with a health research firm or consultant in each country of focus (Burkina Faso, India, Kenya, and Nigeria) to implement this study and facilitate discussions with stakeholders about how findings and recommendations can be translated into policy. The study uses a mixed-methods approach including a desk review and surveys and in-depth interviews with over 200 decision-makers, modelers and brokers of modeled evidence across the five countries.
The R4D-led Translating Modeled Evidence for Decision-Making Working Group represents stakeholder groups in the research countries and provides a forum for members to provide input to the research process, learn from the experiences of other members, and share feedback with donors and other global partners about good practices for enhancing access to and use of high-quality modeled evidence for country-level decision-making.
Key preliminary findings from this work include:
- COVID-19 has led to a surge in interest in modeling and new models for coordination and communications structures between modelers and decision-makers.
- There is a need for long-term, sustained capacity development for both modelers and decision-makers to understand how best to communicate and interpret models. Long-term capacity development ensures that a strong foundational capacity for modeling exists when emergencies like COVID-19 arise.
- Models are most useful when modelers and decision-makers work together throughout the modeling process with common objectives in mind.
- Trust and utility of models for decision-making is greater when models are locally developed, context-specific, and based on high-quality data.
- Formal coordination mechanisms, like task forces or working groups, that provide a platform for discussion of available modeled evidence between modelers and decision-makers can help make models more transparent, easier to interpret, and relevant for policymakers.
