The pandemic is a global healthcare emergency that requires a series of restrictive measures and rules to contain the spread of the virus. But how can you make decisions during a context of uncertainty? How can you reconcile political decisions with scientific data? How can you communicate them to citizens? We discussed this with Giacomo Galletti, a behavioural science researcher and expert working at the Toscana Regional Healthcare Agency, and Enrico Girardi, the chief scientific director of the Lazzaro Spallanzani National Institute of Infectious Diseases.
During the Covid-19 pandemic we found ourselves in a context of uncertainty because the data available were- and still are- limited. A systematic review that was recently published on the British Medical Journal [1] showed that the amount of robust data to support the public health interventions that have been implemented to tackle the pandemic is scarce. In such a context what should be kept into consideration when decisions and rules are made?
Enrico Girardi. When the emergency started the first operation that had to be done was to assess how much of what was known and what had been planned to tackle potential future pandemics could be applied to the current pandemic. I have to say that part of the interventions, such as the use of facemasks or social distancing, proved to be effective. At the time the urgency to make decisions meant that policy-makers could not have waited for a formal confirmation that these measures were effective so they had to start by looking at the best data that were available then.
Giacomo Galletti. I was impressed by the fact that in this systematic review the researchers examined an initial list of 36,729 studies and excluded 36,079 of them that were deemed “irrelevant”! Basically at the end of the selection process the authors focused on 72 studies, half of which assesses the individual measures implemented to tackle Covid-19 while the other half examines groups of interventions. The behavioural scientist in me was immediately drawn to the last column of the summary table on the characteristics of the studies that were examined, the one that reads as “Risk of bias”. How many studies were eventually deemed to have a low risk of bias? Five! Therefore out of roughly 37,000 studies on the efficacy of the interventions you can only truly pick five of them with a certain degree of confidence around the “solidity” of the results. The fact is that, despite the large amount of research studies, the application of rigid criteria that can guarantee the solidity of the scientific method highlights the complexity of such operation, which is often long and laborious, especially when you are working within a complex and uncertain context such as the midst of a pandemic, where it is essential to share the outcomes of an investigation in a timely manner.
Therefore what should be kept into consideration when the rules are made? The authors of the article hope that the public health department will make efforts to plan interventions with a focus both on specific health aspects of the community and the sociocultural needs of the targeted population groups. It is necessary to make more in-depth and structured research studies in order to better understand the efficacy of the measures that were adopted to tackle Covid-19, particularly with regard to the vaccination. If it were up to me I would start with the assumption that the issues in question are extremely complex and that the only common factor for every single aspect of the pandemic is uncertainty. Here, being aware of uncertainty can be a good starting point as it helps you to contextualise the policymaking decisions made around new rules. Besides, if scientists too were to start by acknowledging uncertainty they would be able to ask more in-depth questions on how to adopt and propose engaging methods that can generate useful knowledge- with its own limits- to inform those that would have to make decisions either way.
Do you think that a greater investment in research- resulting in better data- would promote a more informed decision-making process?
Giacomo Galletti. You don’t have many other options if you still value the scientific method today. You need every type of data- epidemiological, economic, social and- especially- the scientific ones. With regard to the investment, let’s think about how much time and resources Galileo Galilei invested into building the telescope, without which he would not have been able to prove his scientific hypotheses. We cannot comprehend the complexity of the universe if we are not equipped with the right lenses to explore it in depth. But lenses have to be built somehow and that takes time, resources and multidisciplinary work.
Enrico Girardi. With regard to generating data there are different challenges concerning funding. First of all there is a theoretical issue around how to study the public health interventions, especially during the course of epidemics. During the last twenty years methodological research has been specialising more on assessing treatment interventions, both in terms of generating primary data and conducting secondary analysis of such data. However, the same thing cannot be said about public health interventions. Conducting a controlled study can be very challenging in several areas of intervention either because of the nature of the interventions themselves or because of how quickly these assessments are needed. Therefore one initial aspect that should be developed is about how to generate and assess the data from these contexts. A second aspect is about maximising the chances of putting together data that were obtained from different geographical contexts and this would require more shared planning of the research work. Obviously this means investing, but we should do this especially in “times of peace” because the framework to conduct assessments and generate data should not be organised and created during an emergency. It should also be highlighted that the context represents a significant factor- perhaps even more so than treatment and diagnostic interventions- for the evaluation of public health interventions. The notion that the same one-size-fits-all measures can be implemented in very different social, cultural and political contexts without any adaptations is problematic.
Suitability and the economic and social impact represent fundamental factors to define the rules. – Enrico Girardi
When it comes to making the rules what are the factors at play, other than having better data available?
Enrico Girardi. In such context many of the factors at play concern the practical feasibility of some interventions. You can deem some measures effective because they have the potential to reduce the incidence or the seriousness of a condition, but these types of interventions cannot be considered abstractly. Suitability and the economic and social impact represent fundamental factors to define the rules. Therefore you cannot delegate the final decision-making onto the public health expert or the epidemiologist because making an assessment is a complex process that requires the participation of all stakeholders, meaning the community and its different components.
Giacomo Galletti. I would like to tell you an anecdotal story because the communication experts say it works… During the spring of 2005 David Foster, the late American writer, made a speech at Kenyon College, in Ohio. The speech started with a story on two young fishes swimming beside each other in a pond. They come across an older fish that greets them and says, “Hello folks. How is the water?” The two young fishes greet the elder fish and keep on swimming in the opposite direction. After a while one of them turns to the other looking at bit lost and asks, “Excuse me but… what the heck is water?” That’s it: what is water? In what context can you establish the rules? As a behavioural scientist my answer is a partial but obvious one: behaviours.
In 2019, when we started to organise the yearly Nudge Day- the days dedicated to behavioural science- at the Toscana Regional Healthcare Agency, we adopted this motto: “There’s no progress without improvement, there’s no improvement without change and there’s no change without adopting new behaviours…” Basically the point is this: rules are as effective as their ability to align people’s behaviours to the socially desired ones. Yet it’s not enough to impose sanctions or to propose incentives. Two behavioural science experts that work for important European institutes, Chiara Varazzani from the Organisation for the Cooperation and Economic Development and Marianna Baggio from the Joint Research Centre of the European Commission, have highlighted that an approach focused on understanding the factors that impact the adoption or avoidance of certain behaviours is essential to understand how to guide institutional policymaking. Overall, even though there are multiple factors at play, change ultimately takes place through (new) behaviours. And, to complete the previously mentioned motto, there is no adopting new behaviours without putting people in the condition to do this as comfortably as possible.
The rules established on the basis of the same data differed from country to country and even varied within the same nation. How can the knowledge based on scientific and epidemiological data- as well as the know-hows acquired during past epidemics- be reconciled with the social, economic and political needs of the country and with the expectations of the citizens?
Giacomo Galletti. Roberta Villa, who you know well at Forward, recently wrote that, “comparing the epidemiological situations of different countries to identify what works, what doesn’t and the best strategy to follow by copying others is an approach that will inevitably take you off the path”. I really don’t know how this reconciliation can take place. My economist demon- that wakes up here and there when it comes to simplifying extremely complex issues- would turn to utopian notions from classical economy straight away and it would whisper in my ear that if all the agents of a market were to have free and equal access to information and exchange this would eventually self-regulate and establish quantities and prices.
If we were to transfer this notion from a utopian context to a contingency one, one of the possible keys to facilitate the progressive reconciliation between rules and actions would be to deliver high quality information and communication that is widespread and discusses data in a timely manner, which would result in guiding the political decision-making as well as the expectations and sensitivities of citizens. On the contrary, I think that the reconciliation between countries- and at a local level- can be achieved following individual “experimental” initiatives: for example you observe the outcome of the Israeli vaccination program and then you decide whether to follow that approach or not; you look at the Italian Green Pass system and then you adopt it elsewhere if appropriate. The issue is that during an emergency the pandemic cost of not doing something can lead to significant repercussions.
Enrico Girardi. It is a political act to reconcile them because the utmost expression of politics is indeed to make decisions while keeping into consideration all these aspects. This partly explains why different countries made more or less open and flexible decisions based on a prediction of how society would respond. The practical application of any public health measure would be deficient without the involvement of the community so you cannot be tied to theoretical and abstract models but should assess how to implement them starting from the knowledge you already have.
Rules are as effective as their ability to align people’s behaviours to the socially desired ones. – Giacomo Galletti
How can a balance be found between political and evidence-based decisions in order to make rules that can tackle the pandemic effectively?
Enrico Girardi. Political decisions should be evidence-based but they are not a mechanical application of the knowledge that is generated through research. Let me clarify what I mean: we experts can provide models predicting the possible scenarios that would take place after individual interventions. However, such models don’t tell you what you should do but provide a series of possibilities. Then society and politics are responsible to decide the most harmonious way to move forward.
Nevertheless I believe that politics has made a mistake when it presented some choices as certain and inevitable. When you are communicating with society you should imagine addressing individuals that are able to understand so you should discuss any uncertainties and any possible evidence-based alternative. Some talked about “dictatorship of the experts” but there should not be- and there is not- anything of the sort. Experts should provide you with the basis to make the best decisions around what rules to apply.
Giacomo Galletti. “I’ve seen attack ships on fire off the shoulder of Orion” and groups of multidisciplinary professionals- epidemiologists, statisticians, economists, scientists, politicians, public and scientific communication experts and policymakers- reason methodically on how to adopt policies and rules that would tackle the pandemic effectively. They started by analysing behaviours, sharing strategies that favour the adoption of new behaviours and defining the rules and the criteria to communicate them accordingly. Then I’ve also “watched C-beams glitter in the dark near the Tannhäuser Gate”… Jokes aside, the matter is quite complex. Creating a task force that operates well is no easy task.
However, I believe this issue should be discussed, especially when people start to talk about a syndemic rather than a pandemic. It should also be acknowledged that the decisions made in a specific context have more or less direct repercussions on many other contexts. Furthermore, how do you assess such effects? Let’s imagine a scenario where there is a statistically significant relationship between the GDP decimals and the number of Covid-19 hospitalisations in the ICU and decisions are being made around the characteristics of a new lockdown. Is the nature of the problem clear? Are the weight and limitations of reasoning on individual data clear? What about the weight and limitations of political decisions that are based on social pressures and the expectations of the citizens? If you start from the premise that the role of scientists- and experts in general- should be to provide evidence and that politicians are responsible to make decisions- ideally evidence-based ones- then it is not easy at all to figure out which tools should be employed to facilitate the relationship between these two contexts. However, I can say this much: in light of the National Recovery and Resilience Plan whereby some “control rooms” are being reinstated, it would be interesting to make a detailed assessment of what worked and what didn’t within the Covid-19 task forces, how political and clinical/epidemiological factors weighed in on the decision-making process around rules and, last but not least, the efficacy itself of the rules that were created in these contexts. Once all that has been achieved you can talk about balance.
One of the possible keys to facilitate the progressive reconciliation between rules and actions would be to deliver high quality information. – Giacomo Galletti
The decision to introduce new measures that are more or less restrictive should be made in advance, when the epidemic curve is not growing yet, but we always seem to wait for it to be growing. That would mean acting late according to scientists and epidemiologists but not to decision-makers and politicians who have to deal with other priorities as well. What is the best time to make the rules and how can the right compromise be found?
Giacomo Galletti. I think that the professionals with solid expertise in statistics and epidemiology are the ones who can help to answer this question, by reasoning together with the experts of the other sectors. That’s because- going back to talking about the syndemic and the complexity of the situation- the timeframe of an intervention in a specific context necessarily impacts other contexts, so before making a decision you need to be aware of the domino effect you are going to trigger. The same multidisciplinary group we mentioned earlier should then agree, together with the policy-maker, on what the most appropriate time is.
Enrico Girardi. There is no such thing as the best time. There are different possible times and each one comes with its own pros and cons. I don’t see a contraposition between scientists and politicians. They’re all doing their job. Let’s think about the closure of schools for example: this was done for a period of time and it might have also been based on the epidemiological context, but it has enormous collateral repercussions and it is appropriate to take these into consideration together with the evolution of the epidemiological situation.
The practical application of any public health measure would be deficient without the involvement of the community. – Enrico Girardi
What is the correct way to inform citizens about rules that were made during an emergency context so that they are accepted and applied for the good of the individual and the collective?
Enrico Girardi. Society should already be involved at an earlier stage, when rules are being discussed, whenever possible. Another aspect that should be kept into consideration is the transparency of the decision-making process: it would be appropriate to explain how and on what basis you created certain guidelines. Then you also need to be honest and open to discuss the level of scientific solidity of the rules; society needs to know that some of these regulations are based on ever-evolving knowledge. In conclusion, transparency, honesty and involvement are the main factors at play when you want to communicate the rules effectively and create the favourable conditions for them to be followed.
Giacomo Galletti. I noticed that communication tends to refer to risk as an abstract notion. Operationally speaking, risk is analysed in terms of the probability of the event and nature of the damage happening, but these two concepts don’t seem to emerge much. A more effective way of communicating the need to impose a rule to tackle a specific risk would be to specify the probability of a certain damage occurring in a scenario where you follow the rule and in one where you don’t. It would also be helpful to discuss any “collateral damages”, their nature and how likely it is that they will take place. Communication experts would probably say, “Look, this thing of comparing risks of different things happening is not very useful when it comes to communication” but I still believe that this attitude could have long-term positive effects on educating people on how to reason when you are assessing risk.
With regard to communication, I will just say that I would have liked to hear a message along the lines of, “Dear citizens, here it’s a mess! We’re trying to figure out how to end this nightmare and- based on what we know so far- this initiative has good chances of tackling the problem and curb further damages. There are some contraindications but they are much less likely to take place and we will keep watching out for them. I understand that for many of us this can be a bitter pill to take but, if we care about protecting everyone’s health, it is the most effective pill there is right now”. Would that work? I have no evidence to prove that but I think that learning how to communicate uncertainty and risk well usually turns out to be useful! Once again, let’s hope there will be more thorough communication, one that stems from rational and liberal debate and is based on evidence as much as non-evidence, one that treasures opportunities to listen and aims to value building a trust-based relationship between those who observe, the decision-makers and those who are called on to adapt their behaviours. And let’s hope that works…
[1]Talic S, Shah S, Wild H, et al. Effectiveness of public health measures in reducing the incidence of covid-19, sars-cov-2 transmission, and covid-19 mortality: systematic review and meta-analysis. BMJ 2021;375:e068302.
© 