Hands holding a sanitary mask

Opinion piece by Maxime Stauffer, based on the presentation given at the Geneva Health Forum on 3 May 2022

Building pandemic preparedness requires knowledge and action. But building knowledge and triggering actions are often separate processes. And whether they inform each other depends on how we design the science-policy interface for pandemic preparedness.

We are currently facing an opportunity. With COVID-19, political attention has turned to pandemics. This alignment of attention is an opportunity for change, as evidenced by negotiations on a pandemic treaty or the creation of the European Health Emergency Response Authority. Such an opportunity does not happen frequently and should be leveraged, as it also allows to change institutions and practices differently. One way to act differently is to avoid previous mistakes when designing a science-policy interface.

Here are six examples of the challenges that governments, international organisations and academics face when attempting evidence-based policymaking.

Governance using evidence tends to still happen in silos. The actions that we implement to mitigate the consequences of a shock can be detrimental to other policy issues. For example, responses to the 2008 economic crisis, such as austerity measures, have led to disinvestment from health systems, which likely reduced resilience to pandemics and thus COVID-19. Therefore, we might know how to react to a shock. But the reaction to one shock can prove unpredictable as to its effects on future shocks.

Budget allocations to pandemic governance are heavily skewed towards mitigation rather than prevention. For instance, the United States and the European Union have deployed recovery plans amounting to hundreds of billions of dollars and euros in mitigating the consequences of COVID-19. But investments in preventing future pandemics, despite their expected severity, amount to millions of dollars only. This discrepancy between mitigation and prevention remains while there is evidence of the financial and ethical benefits of prevention.

Evidence-based policymaking processes are cumbersome and often not fit for purpose. Decisions based on systematic reviews require considerable amounts of time to search, filter and synthesise evidence. This process is too slow to react fast and adequately. As an example, the World Health Organisation was forced to adapt and speed up its synthesis processes to be able to recommend evidence-based actions.

Evidence-based policies in the form of hard law can constrain mitigation and adaptation. It is helpful to have International Health Regulations based on evidence, but their fixed legal structure constrains their adaptability to new, fast-changing contexts. A legal perspective on evidence-based policy begs the question: how to leverage more soft law, norm-based measures for pandemic preparedness?

Image of the Swiss Parliament room from above, only one third of seats are occupied.
On 2 May 2022, IPCC experts gave a briefing to the Swiss parliament – a session boycotted by two-thirds of parliamentarians. Decidedly, information and expertise are not enough. (Image: RTS)

Evidence-based advisory structures are designed to inform rather than shape decision making. The Intergovernmental Panel on Climate Change (IPCC) provides an excellent synthesis of climate science. But, in most cases, information is not enough to trigger action.

Last but not least, putting scientists into policy roles has limitations. During COVID-19, we have seen several countries whose key decision-makers were scientists (e.g. Sweden). While hiring scientists to make policies is a natural way to foster evidence-based policy, it has limitations. A scientist is trained in producing rigorous information. A policy professional is trained (or should be trained) in handling uncertainty, regardless of the state of the evidence. Having scientists manage uncertainty can lead to myopia on evidence rigour, lack of speed, and lack of ability to navigate trade-offs productively.

The above examples illuminate two aspects of evidence-based policymaking. First, governments and multilateral organisations have gone a long way in attempting to use evidence in policy. It is not a new field. Second, past attempts have faced difficulties and led to failures. This gloomy but realistic picture of the challenges at the science-policy interface is the basis we can use to do better.

So, what can we do to better design the science-policy interface for pandemic preparedness?

The best solutions will be contextual. Therefore, delineating specific recommendations is out of scope here. Instead, there are ingredients for success that, if followed, can help the design of context-specific proposals. The following three ingredients are not new, but they are worth repeating. Overall, if you need to remember one sentence: we need to go beyond information.

First, we need to align research with decision realities.

Preventing and reacting to pandemics require evidence on the interconnections between pandemics and other social systems, such as economic, social, institutional, and security. How does preventing pandemic trade-off with other policy issues? Why? And what are the co-benefits?

Good decisions need timely information. If knowledge is not available on time, decisions still need to be made. Therefore, there is a need to build infrastructure to produce knowledge rapidly and relevantly.

It is essential to adapt information needs as a function of the risks associated with decisions. Typically, we do not need the same level of evidence to advise wearing masks as to recommend vaccines. The latter is riskier and needs more rigorous information. The former is safer and can be recommended as implementing it will also help generate relevant additional evidence.

The above requirements – addressing interconnections, building fast infrastructure, and adapting information to decision risks – can shed light on important research questions and strategies that may increase the impact of science.

Note that many actors push back on aligning research priorities with policy needs, claiming that it is at the expense of scientists’ objectivity. But, while, perhaps, only about 10-30% of current research is policy-relevant, it would be impactful to increase this percentage to 30-50%. This would still preserve the independent and exploratory nature of scientific research.

Second, we need to build relationships and trust between scientists and policy actors. Imagine we succeeded at aligning research with decision realities. Will this research be used? Not necessarily.

The uptake of this research depends on whether policy actors are aware of it, trust it and get access to it. There are many ways to do that. For instance, there is an emergence of digital tools to facilitate knowledge translation and access, such as interactive data visualisations. But the primary mechanism is relationship and trust. Relationships can adapt as a function of context. Trust ensures long-term mutual understanding and co-benefits between scientists and policy actors.

We need to involve policy actors and scientists in science-policy mechanisms and nurture ongoing relationships. If a pandemic occurs, the relationships will be in place and can be more easily leveraged.

Third, we need to train decision-making under uncertainty. Adapting research as a function of decision realities and fostering relationships and trust are strategies to ensure that relevant information is produced and used. But, at the end of the day, what we want are better decisions. And while scientific knowledge helps, making better decisions is yet another skill.

We need policy actors to learn how to acknowledge and handle both uncertainty and complexity. They need to know how to make decisions that can be adapted over time by preserving instead of hindering future decision options. They need to identify trade-offs and avoid paralysis in the face of compromises.

The capacity to make decisions under uncertainty is critical. Science will never reduce uncertainty entirely. Therefore, policy actors will have to make decisions in the face of known and unknown unknowns.

Simulated settings such as digital or role-play simulations can help policy actors to learn by making decisions and receiving feedback loops on their effects. Additionally, simulations help them experience uncertainty and complexity, which also trains their communication skills, stress-management, and analogical thinking in the face of future threats.

The above points should ensure we do not commit the same mistakes in the face of future pandemics. We can likely find new and more effective solutions by pushing the boundaries in how we conceive of policy-relevant research and by integrating decision-making in our conception of the science-policy interface. The current attention dedicated to pandemics as a policy issue is not only an opportunity for change. It is an opportunity to think fresh and build an interface that is better fit for achieving its purpose.