Transcript - Communicating In a Crisis

Spil

Chemical spills, food scares, terrorist attacks and, of course, pandemics: How can we help manage and reduce the impact of these frightening events with good communication - even when evidence is scant and uncertainty predominates? Joining David to share their experience from the frontlines are: Professor Brooke Rogers OBE - Behavioural scientist at KCL, and SAGE participant advising government on the coronavirus measures and risks to national security Professor Lord John Krebs - former head of the Food Standards Agency where he dealt with BSE in sheep, dioxins and a series of food scares. *** David Spiegelhalter: When confronted with a sudden crisis, a terrorist attack, a chemical spill, a food scare, or indeed an outbreak of a new infectious disease, good communication is crucial to an effective response. Of course we're still navigating our way through the coronavirus pandemic so we've all been on the receiving end of crisis communications quite recently, I will certainly be covering that. But we'll also be considering crisis communications more broadly, trying to pull out principles of best practice that can be applied across very different scenarios. How can governments and other organizations help manage and reduce the impact of these frightening events with good communication? David Spiegelhalter: My guests today both have extensive experience in a range of different crisis situations. Prof. Brooke Rogers, OBE, is a social psychologist whose work focuses on events like chemical, biological, and nuclear accidents or attacks. She advises the UK government on risks to national security and public health and she's currently advising them on the public response to coronavirus measures. Brooke, welcome to Risky Talk. Brooke Rogers: Thank you, David. David Spiegelhalter: Prof. Lord John Krebs originally made his name as a zoologist, but in the year 2002 he became the first chairman of the newly created Food Standards Agency, where he was on the frontline dealing with risks to food safety, including mad cow disease. Welcome, John. John Krebs: Hello, David. Good to be here. David Spiegelhalter: How great to have you both with us. Brooke, I'm going to come to you first to get a sense of the work you do. You're a member of several groups which advise the UK government on issues of national security and public health. Can you explain what kind of situations you plan for and what kind of advice you provide? Brooke Rogers: I have to frame the role that I hold across several different organizations as an independent academic role, they're not paid roles. And I am there to challenge, to question, to try to generate or provide evidence when needed, so all of the comments that I'll make today are very much in my independent academic role from Kings College, London, I not speak on behalf of government. In response to your question, we tend to work with the National Risk Register, which is the public-facing view of the risk landscape that we're considering for the UK. Brooke Rogers: And we very much look across the range of events from fluvial flooding, for example, or chemical releases, terrorist attacks. We would look at pandemic influenza and other types of pandemics, a heat wave, intense cold, drought. We look across the board and say, "Where can we place this in terms of likelihood or plausibility and impact on the public?" And try to put that into a framework that then can enable the experts advising government to provide advice for the government to make decisions about the advice that they then cascade down to the local level. David Spiegelhalter: Where does your experience as a psychologist come into this? Brooke Rogers: I would say that one of the best examples would be through the Behavioral Science experts in the Cabinet Office. And we're a motley crew of behavioral scientists and we look at the risks scenarios that are being used for each risk and threat. And we look at the language that they're using, the way in which they're framing members of the public, the assumptions that they're making about public behavior and we bring evidence to bear on those conversations. Sometimes we challenge the language and we, I would say, have seen quite a change in the way in which they're understanding public reactions. And we've also challenged the measures that they're using and have provided evidence and sent papers up to the government chief scientific advisor and made a case for changing the measurements they were using to understand the impact on the public. David Spiegelhalter: Oh, that is so much for covering there. But let's look at actual SAGE, the Scientific Advisory Group for Emergencies. I was on it back in the days of volcanic ash for heaven sakes, and of course now it's got a lot of work to do, never had such prominence. What's your role? Brooke Rogers: SAGE is not really a membership group, I would say that we're participants and that is underpinned by several other groups. We have SPI-B which is more of the behavioral science approach, we have SPI-M which is our modeling group and we have NERVTAG, which is looking more at the epidemiology and so on. And we bring those lenses together and each group also has many subgroups that work underneath it so there's a layering effect, we can target the questions that we're being asked to consider. Brooke Rogers: And quite often those questions can be broad. We've been working with them to make them a little bit more focused as we move along and in many cases we will suggest where we'll look for the evidence if it exists. Sometimes it won't be directly around pandemic response, we'll have to look at other types of infectious disease and say, "It looks like this will happen." And other times we're very much suggesting principles that should be applied for this communication for understanding behavior moving forward. David Spiegelhalter: Okay. I think it's worth just emphasizing from what you said that the job of SAGE is not to tell the government what to do, it's to provide evidence and projections to help them make decisions. Okay, John, you had an exciting time when you're at the Food Standards Agency, what kinds of things did you have to respond to in your communications? John Krebs: Well, David, I'd divide them into three categories. There are risks that were identified where we knew there was a risk, probably very small and something where you've got to tell people that could be a risk but we're not asking you to do anything different. And a classic example of that was acrylamide. So in about 2001, 2002, some Swedish scientists discovered that if you cook certain kinds of food, particularly starch-containing foods at a high temperature, like making chips or French fries, then you create this chemical called acrylamide which can be carcinogenic. So when you discover that do you say to people, "Stop eating chips?" Do you say to people, "Cook your chips to low temperature so they're not nice and crisp and possibly burn yourself on the fat while you're measuring the temperature?" Or do you say, "This is not a new risk, we think it's very small. Be aware of it but carry on doing what you were doing before." John Krebs: The second kind of risk is one which is a public perception of risk but doesn't actually exist scientifically. And the case where I found that most difficult was GMOs. In the late '90s and early noughties, there was a huge campaign by the green groups against GMOs and many people became fearful that GMOs were bad for their health. And I discovered it's not enough to say the science says that the GMOs that you may be exposed to are completely safe because it's an emotional issue not a scientific issue. And then the third kind of situation was we simply don't know. John Krebs: And the classic example for me was BSE in sheep. We knew that mad cow disease was in cattle but we didn't know whether or not mad cow disease had also infected sheep. And the government of the time was very nervous about us even mentioning that. But I took the view, it's better to treat people as adults and say, "Look, there could be this risk." And we thought a lot about how we'd express this idea of there might be a risk but we don't know, and then what we did there was to say, "There could be a risk, we simply don't know at the moment. We're not advising you to stop eating lamb, but if you want to that's up to you, but we are on the case and we're doing research to find out the answer and we will get back to you very soon." John Krebs: That in a way he was trying to treat the population as an adult group who would understand risk and uncertainty. That in itself was a risky approach but I think it worked. The best answer is, "We simply don't know but we're onto the case." Sometimes you've got to recognize that people's perception of risk diverges from the pure science, GMOs, and sometimes you're communicating, "There could be a risk. It's small but we don't necessarily want you to do anything different about it." David Spiegelhalter: Wow. John, you covered a lot of stuff there and we'll come back to some of it- John Krebs: I'm sorry I- David Spiegelhalter: No, no, that's fine. The acrylamide as well... I think you know way after you left I think the Food Standards Agency had a rather ill-judged campaign- John Krebs: Toast-gate, I think it's called- David Spiegelhalter: Yeah, yeah, the killer toast, which was really, I think taking the wrong approach- John Krebs: Absolutely. David Spiegelhalter: ... To dealing with these risks. Okay, so one of the things we ask everybody in our podcast is to step back a bit. Why are you doing this in the first place? What is your actual aims in doing this? What's it trying to achieve? And I think in the Winton Center we go on about, we're trying to inform rather than persuade. And I'm interested with both of you of where in that spectrum you operate. Brooke, what are you trying to do in your work? Brooke Rogers: I would say that my career, my life has really been about enhancing opportunities for informed decision-making. And it was really as quite a young academic when I started attending... I attended something called a sand pit, and I was still a postdoc and I had the chance to meet practitioners, government practitioners I would have never met before, I hadn't the confidence to meet before. And they locked us in a hotel with people from different disciplines. And we learned about how the challenges that these organizations had and the way in which they were thinking about them. Brooke Rogers: And I felt that quite often the solutions were being drawn primarily from the physical sciences with very little thought about how publics or how people who use those tools would think about them or respond to them. And I also felt that when the public we're considered that Hollywood-esque assumptions were being made about their behavior, their appetite for risk, their perceptions of risk. And so I thought, how can we generate evidence from between members of the public, which is where we started with our work, practitioners more on the frontline and the policy makers. It's just trying to bring more reality to those plans and policies that are being made and the hope that they will trickle down into practice and into the way in which we engage with members of the public and let them engage with our processes as well. David Spiegelhalter: John, Brooke mentioned very strongly that she was interested in informing people so they could make better decisions. How does that fit in with your work at the FSA? John Krebs: Well, the FSA, our watch word was we put the interests of the consumer first. And supposing a problem crops up with food, you've got a range of actions you could take. You could ban it, get it removed from the supermarket shelves, you could give people warning, say, "This is there but it's potentially dangerous." Or you could try to inform people as Brooke was talking about and enable them to make informed choices. I think my objective in communicating risk and uncertainty where we weren't banning, we weren't advising people not to eat something, was to say, "Here is the information we've got, use it in your own way. This is our advice." Our advice might be, "Carry on as normal or if you're worried about it shift to eating something slightly different." John Krebs: But really, I agree with Brooke when she said what's key in this is transparency. Because I think transparency, complete honesty is one of the pathways to building trust, because in the end it's all about do people trust you? Do they think this guy's a shyster? He's just pulling the wool over our eyes or is he somebody who tells it straight? And telling it straight over and over again is one of the ways to help to build that trust. David Spiegelhalter: What about in the virus particularly the period we've gone through, or even going through now where actually there is quite a lot of guidance, instructions, and even laws that are there, to measures to control people's behavior in extent. Brooke, how do you feel the responsibility that you need to provide the evidence to back up those instructions? Brooke Rogers: Certainly. And it's been interesting from the SAGE point of view because the way in which they operated in the past it's always been very rapid. Let's say for Ebola, I think they only had four meetings and so by the time the response was over the information was released but there wasn't a huge public appetite for it and this was part of a much, much larger government communication system. And I would say that all of us are very, very keen on transparency. And it was a bit of a shock but when we saw how long this engagement would carry on for all of us transparency is key, and I believe very strongly that Patrick Vallance and Chris Whitty have been advocating for that. Brooke Rogers: It took a little bit of time behind the scenes to change that structure to enable it but our desire is to have it as real time release of information as close to real time as we can and that redaction would be the exception, if it ever needed to happen it should be because of security implications. I think that science should be a conversation between scientists but also I'm so desperate, I love opening science up to the public. I go into schools and things like that so it should be a conversation with the public as well. Brooke Rogers: And when decisions are made, and we, we do not make the decisions the decisions are political, they have to balance quite a few things so we are one piece of information going in. I do think for any decision across the board it's important to recognize the information that you're drawing upon and to give the rationale for the direction that you're moving in. Because as the science brings in new information we need to be able to consider that for any event and to adapt our response in light of new evidence. So you can have that conversation and say, "We suggested this. We were going with this, new information has come in and therefore we are now changing our advice and these are the implications for you in your daily life." David Spiegelhalter: Okay. If you are telling people to do things for transparency you do need to provide a reason for that and not just an authoritarian diktat. John, have in your work, were you in a situation where you were really telling people, you wanted to tell people to do things? John Krebs: Most of the time if we actually had such a clear view that something was unacceptable, and I gave you an example of that, was a red paint dye called Sudan 1- David Spiegelhalter: Oh, yeah. John Krebs: You remember that that was added to paprika pepper imported from Southeast Asia, because the brighter the pepper the better its market value. When we discovered that this carcinogenic paint dye had been added to paprika we simply said all the products got to come off the shelves. And I can remember at the beginning I was told, "Well, this will be about 50 products." And then an hour later, "Actually when we said 50 it may be 500." And the thing grew and grew, but we felt we had no option. The way I played that was in my mind, supposing we said to people, "Carry on. It's okay that the pizza sauce you were buying or other products contain this carcinogenic red dye but we think it's okay," that would not have played well. John Krebs: There's a case where the instruction was from us to the industry, "Get rid of this product straight away." But in the more nuanced cases... I'll give you another example. During them foot and mouth crisis when there were all these dreadful funeral pyres of sheep and cattle being burned all around the country, it transpired that the dioxins that were coming out of the funeral pyres might be landing on the grass and from the grass into the cattle, from the cattle into the milk, from the milk into butter and yogurt and so on. Should we tell people to stop eating butter and yogurt, to stop drinking milk? Or do we say, "There's this a possible risk we don't really know the answers yet, carry on as normal. If you're worried about it drink low-fat milk because dioxins are fat soluble. If you drink low-fat milk you reduce the risk and we'll tell you when we find out the answer." And the answer was a few weeks later, it wasn't actually getting into the milk for some reason. John Krebs: I think there are different examples of telling people not necessarily strictly what you should and shouldn't do but giving them clear advice and options. And in cases where we had no alternative but to simply cause something to be taken off the supermarket shelves, that's how we handled it. But again, I totally agree with Brooke that openness and transparency is an absolutely essential part of all of this. David Spiegelhalter: Okay. We've had a strong message so far about openness and transparency. Now let's move on to talk about listening. Brooke, you mentioned before the importance of engaging with the audiences, how do you go about doing that? Brooke Rogers: Well, I think that this is where academics and scientists can come in really handy and were very useful with this. Our work, the work I've done in collaboration with several colleagues in the past is focused on the importance of co-production. Even in security spaces we've published around this, looking at protecting crowded places. And I think that for every risk or threat we're considering that there is an opportunity for co-production to take place. David Spiegelhalter: Okay. Just to clarify, co-production means coming up with guidance that's being put together by experts but working with members of the public. Brooke Rogers: Now, when you say that to a government organization who's trying to respond very rapidly to a risk or threat that has a great deal of uncertainty, sometimes I worry that they shut down and I think we just have to get the job done, we just have to give guidance immediately. But I would argue that their guidance will be more effective if they build in that co-production. And my colleagues and I have done quite a bit of thinking around how to enable that in the usual way that we do this with our longer-term research projects but also in real time during an emergency. And I think that right now during this extreme event, during the pandemic and also in the future, we need to really get them used to thinking about and building co-production into the policies and guidance that they're creating. David Spiegelhalter: Oh, that's great. Yep, a message I think we all agree with. But I've got to ask you one thing, the crosses the question you might be asked by journalists, "Do people panic?" Brooke Rogers: The main message is that panic is rare even when there's an extreme event. And there are studies that show that actually when people should be taking something very, very seriously, they overestimate their ability to get out of it and possibly don't realize the danger and don't engage in the protective behaviors so that threat perception isn't activated. So yes, panic is real but I think to always assume panic means that you put yourself on a different footing in terms of the communication that you share, if any, and the way in which you frame it and then in the way in which you talk about risks and threats. Brooke Rogers: And instead we would argue in the evidence that we generate with our view that our public responses need to be understood across a range of behavioral responses from under-response to over-response and everything in between. I'm very keen on understanding the sustainability of those changes in behavior that we see as well when we encourage them or advise them how long can that last. We don't have a lot of evidence around that. So yes, panic is real, it's incredibly rare, we need to understand behavioral response across a scale of range of behaviors and to prepare for that range of behaviors to take place. Speaker 4: I really like the idea of not just categorizing everybody in one group and realizing people do respond to very differently. We've heard from Brooke about the importance of listening to the audiences. In your work, how did you get that feedback? How did you see whether your messages had been successful? John Krebs: Well, first of all it's how do you construct the message? And I'd give you one example. When we were wanting to talk about the possibility of BSE in sheep, we sat around in the office and we said, "Well, actually, what this is, it's a theoretical risk. And we tested that out on focus groups. Do you know what theoretical risk means? And no one had a clue. They thought it was something that people in white lab coats did in laboratories. John Krebs: First thing was get the vocabulary, get the lexicon, right? Having done that and produced our message we did extensive surveys using MORI or other opinion pollers, A, did you hear the message? B, did you understand what it meant? And C, did you respond to it? With the case of BSE in sheep it turned out that about 65% of the population knew that something had been said about BSE in sheep which is quite remarkable actually. They didn't know who said it but that doesn't matter, they knew something had been said. John Krebs: And when we asked people whether they changed their eating habits or results, most people said, "No," a few people were eating less lamb, and a few people are eating more lamb, which is quite interesting. That's how we both constructed the message with feedback from the users and we then surveyed what the message had conveyed and how people responded to it so we could learn a continuous learning experience. David Spiegelhalter: Okay, John, I mean, you've already said how you dealt with the possibility of BEC in sheep. I mean, actually I quote your advice as a set of bullet points. Can you again summarize for the listener, if you had to just bash it out to an audience, how would you deal with some potentially really threatening situation like that? What are the sort of steps you'd go through? John Krebs: Well, I think the steps are first of all, to try and describe in the simplest possible terms, what the risk is, what we know about it. In the case of BSE in sheep the message was it's possible that BSE has got into sheep, we simply don't know, but then to say what you're doing about it. So, "We are on the case, we're trying to find a solution we'll get back to you very soon," and also at the same time to give people some clear guidance. Because simply shrugging your shoulders and saying, "I haven't got a clue, gov," isn't terribly helpful. You're saying, "We don't know, we don't think it's a problem but we're working it out. In the meantime our advice to you is this." I think it's those three elements, being honest about uncertainty, being clear that you're on the case trying to work out what the problem is and what the answer is and to give people choices but with some guidance in the meantime. David Spiegelhalter: Right. Now, let's talk about the coronavirus pandemic. Brooke, you've been in the thick of things advising the government and monitoring the public response to the crisis. Can I ask, what's played out as you might've expected and what's been a surprise to you? Brooke Rogers: I tend to look for positives even when we're murking through all of the negative impacts if we're thinking about coronavirus. And we cannot understand all of those negative impacts without also understanding that we're seeing changes in terms of prosocial behaviors and helping behaviors. And we have a project underway in that area right now so data is being collected, looking at the scale of prosocial helping behaviors, looking at the way in which communities have come together. Brooke Rogers: You're seeing the WhatsApp groups, you're seeing that the different services can't even cope with the number of volunteers that they're receiving. And that there will be certain changes to life that can have some positive benefits for people. How long they will last, I don't know, whether or not these connections that people are building with their communities that hopefully will make them stronger than ever will last beyond the pandemic, we don't know, we're asking those questions. David Spiegelhalter: Were you surprised by the amount of prosocial behavior? Brooke Rogers: In terms of the scale of the prosocial behaviors, yes, I'm surprised, pleasantly surprised. But we haven't really been able to consider the scale, the full scale of a pandemic before we knew it would be bad but we didn't think about this across the entire nation, across the world. In terms of level of adherence that has been really, really interesting as well. I would say that in spite of the media framings and the pictures and the angles that they use with people crowding into spaces, the media rhetoric, the media story wants to be about basically public bashing saying, "Oh, look at all these people misbehaving." But the data shows that a large, large, the majority of people were actually following the advice and doing their best to protect themselves and other people. David Spiegelhalter: Again, that's the impression I get and I get so angry at these zoom images of people on beaches when they're actually properly distanced. Again, were you surprised I think by people and the distancing and the shielding and things like that, how much people embraced that apparently willingly? Brooke Rogers: Perhaps that's surprised than people who don't work in this field or think about this often. I've always been an advocate of the ability of members of the public and the willingness of members of the public to actually listen to information and engage in protective behaviors, protective for themselves and protective for others as well. Again, it's a scale of it's being able to work on the scale and think about the scale and capture data on the scale that is making it more of a surprise to me. But I did expect to see that members of the public would behave in a very rational way. Brooke Rogers: Now we need to think about the length of time that we are asking to them behave. And there's been a lot of finger-pointing, et cetera, about behavioral fatigue from supposedly behavioral scientists. And that did not come from the work of SPI-B but I do think that while we would say it didn't come from us that data needs to be collected on that, that we need to explore that concept and see if evidence can be provided in order to inform that conversation. David Spiegelhalter: Now, John, you've retired from working with the government now so you're in a perfect situation to say how you think they've done in terms of their communications through this crisis. Now, I've got to say I was watching Chris Whitty, the Chief Medical Officer at the start of this and he seemed to be going through the checklist you gave earlier. I thought it was rather good. I thought, "Oh, he's been listening to John same as I do." John Krebs: I'm sorry. David Spiegelhalter: How do you feel it's being dealt with in general? John Krebs: I always go back to Bob May's guidelines for scientific advice and policymaking, and he had a number of points but the three points that I always remember were openness and transparency, acknowledging uncertainty and incorporating the wide range of views. Scientists don't agree, that's just a fact of life. We don't all sit there and say, "Yes, we agree absolutely." There are scientists that have different views about the nature of the disease, how it should be handled. And if you asked me to score on a scale of one to 10, I wouldn't give them a 10, I think it's changed over time but I would give them somewhere around seven to eight. David Spiegelhalter: Okay. Now, here's the tricky bit because I think maybe you feel the same I do, it seems to have changed over time. So it seems to be more political involvement as what was being said at the briefings and so on. Let's just get there some final words on, rather than communicating directly to the public which is what we've been talking about, which is it's got a clear idea about that. What about including the politicians in this discourse, what is your advice on that? John Krebs: I think you've got to tell it straight. As Brooke said, ultimately the politicians are paid to make the decisions. And the prime minister actually has said that last week when he was announcing further release from lockdown measures. He was saying... Chris Whitty and Patrick Vallance were saying, actually this could be risky, it could be dangerous. Johnson quite correctly said, "Well, I'm a politician I'm paid to make these decisions, the scientists are one form of input." I think with the politicians one's got to be A, clear about what the science says and B, not get sucked into supporting a political decision that doesn't have a scientific backing. John Krebs: It's fine to say, "Well, you made the decision because you're worried about unemployment in the catering sector, or you're worried about the Daily Mail headlines, that's your problem. My problem is to tell you what the scientist is saying." And I think sometimes there'd be occasions when Patrick and Chris have been standing next to the politicians. The politicians have said things that go beyond the science and I would have wanted had I been there to grab the microphone and say, "Hang on a minute, this is your political view but it's not the scientific view. Thank you very much, minister." David Spiegelhalter: Do you think that's an argument for actually having really quite separate communications from the scientific community? John Krebs: It's a real difficult one that, David, because Chris and Patrick were pleased to be standing there with the prime minister at the beginning and it showed that they were right in there with Number 10, but it carried with it the risk that they could be sucked in supporting political decisions. And if you asked me what I would have done, I can't honestly say whether they've done it differently other than hoping that I would have the guts when these politicians misrepresent the science or went beyond it, that I could just say, "Hang on a minute," to the press conference, "This is not what the science is telling us this is what the politicians are telling us." But it's a really difficult one and I wouldn't want to dis what Chris and Patrick have done because I think they've been very courageous in a very difficult situation. David Spiegelhalter: Okay, Brooke, can I turn to you on this? You're still working with the government on this stuff at the moment so I'm not going to ask you to evaluate their performance. But can you say something about how that relationship has gone over the crisis, this sort of pipeline between the scientific advisors and the policy makers? Brooke Rogers: We are learning about how to frame our information in terms of length of reports, in terms of fronting up that guidance. I mean, if you look at the SAGE minutes or the SPI-B and it's et cetera, were quite punchy sometimes in terms of really, really highlighting where we're concerned. We work very, very collaboratively and I think that most of us actually feel that that is only some of the most effective work. When we bring we break down those barriers and we create task and finish groups between, let's say the modelers and the behavioral scientists so that we can understand one another. Brooke Rogers: In terms of managing that pipeline between science and the policy makers, I think that you just have to accept that you are there to advise. The advice will be hopefully listened to but they're also taking advice from many other areas and making the decisions, the politically-based decisions on that. If they're publishing our work it's my hope that people will look at those minutes, look at those notes and look at those reports and they'll see what we're really saying and understand that decisions are being made based on other things. So it can be, I guess people would say very frustrating if you put your heart and soul into developing some advice and then you're not always sure where it's going, who it's going to. Brooke Rogers: We do know that other departments are looking at the advice and it is being picked up along the way, but it's balancing it with the information that the politicians are actually receiving from many different groups, that's some of that difficult balance. So I'm there, I'd say for the long game, really trying to work from the inside to bring evidence to bear, to challenge the processes where they don't work to reinforce the processes where they do, and to just ensure that the next time we have an event to deal with. That the things that we have learned are kept in place to enable that conversation and that pipeline between science and policy to be maintained. David Spiegelhalter: Oh, well, that's what I'd like to finish off with. You specialize in a wide range of catastrophes and disasters, which of course haven't happened yet, this is a crisis that we're in the middle of. What is going to be the impact of this? Do you think, or what do you hope will be the impact of this in terms of briefly planning for the inevitable further future catastrophe? Brooke Rogers: I think that this is an opportunity dare I say, to look at the processes, to look at that relationship between science and policy making. To look at the way in which information and data is shared within departments, between departments and what is done with the information that we're bringing and how does that cascade out through the rest of government and how does it make its way up out to the politicians and the policy makers. I think that we are basically rewriting that system and reinforcing, creating a new system. While some of us go and have a little break someday from the very, very full on engagement, that the next event and the continuation of this event will just mean that we're more effective, that we're more efficient and that we are able to be transparent. David Spiegelhalter: Well, that's a nicely optimistic message to draw things to a close. Thank you so much, Brooke and John, it's been a great conversation. I think we've pulled out some valuable lessons about the importance of transparency to build trust, engaging with the public, co-production of guidance and about navigating the tricky relationship between science and politics. It's time say a deep thank you to my guests, Prof. Brooke Rogers and Lord John Krebs, and say goodbye from me and the Risky Talk team.