Federica Santoro 0:15

The last two years of pandemic have been a stark reminder of just how tricky risk communication can be. Whether we're trying to convey the risk of dying from an infectious disease or of suffering a serious side effect after vaccination, it's easy to get it wrong. The information is complex and it can be misinterpreted depending on how it's framed and who's listening. Yet people need that knowledge in order to make informed decisions. So, what's the best way to communicate benefits and harms in medicine? My name is Federica Santoro, and this is Drug Safety Matters, a podcast by Uppsala Monitoring Center, where we explore current issues in pharmacovigilance and patient safety. Joining me today is Alexandra Freeman, Executive Director of the Winton Center for Risk and Evidence Communication at the University of Cambridge in the UK. After a long and successful media career at the BBC, where she worked to bring science to the widest possible audience, Alex now helps professionals such as doctors and journalists communicate numbers and uncertainty better. We spoke about biases, narratives, misinformation, and much, much more. Welcome Alex to Drug Safety Matters. It's a pleasure to have you on the show today. How are you? Thank you. I'm great, thanks. Looking forward to chatting. Let's start with the basics, as I often do. We can't really talk about risk communication without talking about risk perception first. And we all know from everyday life, really, that people can have a completely different experience of the same situation. So say some are terrified of flying, others would never set foot on a ski slope, some, we know, are incredibly afraid of catching COVID, others will behave as if the virus will never touch them. Why do people perceive the same risks so differently?

SPEAKER_00 2:27

So I think people trying to communicate risks often try and reduce them down to the likelihood of something happening. It's a probability. But actually a risk is far more than that. It's not just the likelihood of something happening, but it's the impact of what might happen. And that's something that's really different for every single person. I mean, if you just think about the impact of 10% chance of having a headache, say, versus the 10% chance of dying, then of course those two risks feel very different. The same impact can feel very different to two different people as well. If you think about, for example, the chance of catching COVID, and then perhaps you have to self-isolate or be ill for a couple of weeks. For one person whose job doesn't rely on them being at work, they're still going to get paid, they don't have major caring responsibilities, that's one kind of feeling. But for somebody who is caring for an elderly relative or for small children and who might not have a job if they can't work for two weeks, then that's a really different impact as well. So it's completely natural that different people will perceive different risks very differently. And in fact, that feeling can change during the day for a person as well. I certainly know that I feel more vulnerable at some times of the day than others. If I'm really stressed and in the middle of the day and I see an email come in that's really likely to be very difficult to deal with, I'll often not open that email till later when I feel a bit calmer and I feel I can take a bit of bad news now. So our vulnerabilities to particular impacts can be really different, different times of day, different days, different times in our life. I mean, if you think about the risk of death, for example, for a 10-year-old versus for a 90-year-old, those are going to feel very different as well. So risks are not just about probabilities, they're about feelings. And it's really important to bear that in mind.

Federica Santoro 4:26

And there's so many subjective values and experiences that people base their evaluation of risk on. One of our listeners from Brazil, Maria Vitória, has a question about that. She says, How can we explain to people that their beliefs are often based on their own confirmation bias? What do you say?

SPEAKER_00 4:47

Yes, well, confirmation bias is of course something that all of us have. We look at the world around us and we have our own preconceptions and we tend to look and spot evidence that confirms our expectations and what we want to be the truth and what we want to believe. And we all do it all the time. You know, a very good test that I try and do to myself is, you know, if you see quotes from people that you admire, then you just try, what would I think about this quote if it were from somebody who I don't trust? And see how your feelings about that evidence changes depending on the source. On the other hand, these things are really, really difficult to get rid of. Nobody's going to be able to get rid of all their biases. I think all we can do is just be aware of them. I don't know how we can kind of all spread the word a bit more about these things, but I think people are becoming more aware of it. And maybe in schools and in our training, maybe some of the syllabuses are changing in schools, and maybe some of the training courses that we undergo as professionals are bringing these to mind so that we can go through kind of mental checklists of how much of my feelings about this is due to my preconceptions. Definitely.

Federica Santoro 6:02

And that's a good tip about challenging ourselves and trying to keep that bias in check as much as we can. But back to the big problem in risk communication. So, as you said, we tend to evaluate risks based on subjective experiences like our feelings, our personal values, and the like. But specialists like to express risk as objective statistical frequencies. And we've all heard the one in a thousand, the 30%, and so on. There's an added complication that most of us have a hard time grappling with numbers anyway. So, given all that, what's the best way to communicate risk?

SPEAKER_00 6:41

Well, I mean, of course it's going to change depending on your audience and your format and the sorts of risks that you're trying to communicate. But really, I think we have to think in each case, and this was what makes it difficult because there's no kind of easy handbook to risk communication, but in each case we need to think about what does this person need to help make the decision that they're trying to make? So quite often we think of risk communication as just data communication. Here's a load of facts and numbers, and I just want to communicate them. But actually, we need to think, what is the point of communicating them? Why do we think it's important? Why might somebody think that they want to know this information? And usually it's because they're trying to make a decision. Should I take this treatment or another treatment or no treatment? And so to make those sorts of decisions, they need to be weighing up pros and cons, the benefits and the potential harms. And the risks is the potential harms, but actually in the Winton Centre, we do risk and evidence communication. So we don't classify things as necessarily potential harms. They might be potential benefits, they're just as important to communicate. So we need to think how are people going to be weighing up these pros and cons? What might they need to know? So they might need to know, for instance, the quality of the evidence. And they might consider some of the things that we consider as benefits as potential harms. So it's actually really complicated and it's really difficult to do. So I guess my kind of advice would be to think about what decision they're making, how to give them the information in a way that allows them to weigh the evidence up and how to make the numbers easy to see at a glance. Because as you say, none of us are great with numbers. It's a foreign language to all of us. So what we often use is graphics as a way of showing numbers. And we often also try to help people see what they're comparing. So to use those graphics to say, on the one hand, and on the other hand, and that might be on the one hand, these are some benefits of a potential treatment, as in how many people might find a benefit from it, on the other hand, how many people might find a harm from it, or it might be just within one particular category. So if you're talking about a side effect, how many people out of a hundred say do get this side effect and how many people don't? So all of these kind of pros and cons and framing and out-of-numbers, they can all be made easier if you are using graphics and if you're thinking about it from the perspective of your audience who've got a decision to make.

Federica Santoro 9:23

And obviously, some risks are easier to communicate than others. Take cigarettes, where the message is pretty unequivocal. They're bad for you and you should quit smoking. But we deal with risk communication in pharmacovigilance. So when you talk about the safety of medicines and vaccines, it's not that simple and clear-cut. These products obviously carry some risks because there's no such thing as absolute safety, but they also offer great benefits, or we wouldn't be using them in the first place. So the million-dollar question, I guess, is how do we talk about these potential harms of medical products without scaring people away from the drugs they need?

SPEAKER_00 10:09

There's so many interesting things in that question. So, first of all, it's interesting that you used the term the message. So that's a way of thinking about communication that I used to do a lot. My old job was working in television, in documentary making. And usually we would think in terms of a take-home message that we wanted the audience to go away thinking this is good or this is bad, or if you do this, you will reduce your chance of this. In evidence communication, we're not trying to communicate a take-home message. We haven't made a decision on behalf of the audience that we're trying to then communicate kind of our way of thinking and getting to a decision. Instead, we're trying to put the evidence in their hands and help them come to their own conclusions. And that requires the audience to be prepared to do that, so that if they're expecting a take-home message and then you say there is no take-home message, you've got to make up your own mind, then that can be a bit of a challenge. So you have to know how to start the conversation in that I am going to be telling you things in order to help you make a decision, and I'm going to help you understand what you want to understand to make your decision. But it also means that we've got to think about, as you say, the more complex scenarios where it's up to this individual person whether they think the benefits for them outweigh the potential harms for them. And as we said earlier, that's a very personal decision. People value benefits differently, they value harms differently. They may well be more concerned about quality of life than quantity of life. You see that a lot with choices around things like chemotherapy or especially end-of-life care. But those are the sort of the obvious examples, but it goes all the way back. You know, the potential benefits of a drug are not a hundred percent for everybody. When I first started looking at the number needed to treat to get benefit from most of the very common drugs, I was actually really surprised. I don't think a lot of patients realize that for a lot of people they'll get no benefit on average from taking a particular drug. And so seeing the potential benefits and the potential harms in an even-handed way, I think is important for all of us so that we can weigh up those benefits and harms for each of us. And you say without scaring people away from drugs they need. Well, that's really interesting, isn't it? Because that sounds like you are deciding what drugs this person needs. How are you defining that need? Is it because you think this drug will make somebody's life better? Well, I think that's a conversation to be having with the patient about whether they think that this chance for benefit and this degree of benefit is worth this chance of harm or this degree of harm. And maybe, in fact, the sensible thing is to try it for a while and see if it has actual benefit or actual harm for that particular patient because we can never tell. You know, probabilities are not telling the future. So I think this is a really complicated issue. And I know that there are a lot of people concerned about scaring patients and fears that their communication will have led to a patient doing something that they consider is not in the best interests of the patient. But I think quite often we have to be happy to let patients make their own decisions around things. And of course, the patient's decision may be that they want somebody else to take that decision. I mean, it's not at all unreasonable to think that a medical professional, their job is to make a decision on your behalf. So, of course, you know, a lot of people will say, you decide for me, doctor, or if you think this is best for me, I will take it or I will at least try it. In which case, your role as a communicator is slightly different. It's more like a kind of informed consent. You're letting them know the reasons for your own decision on their behalf. But all the time, I think, as you say, we need to be communicating benefits and harms in an even-handed fashion so that people can see the balance between the two. Absolutely.

Federica Santoro 14:29

And another one of our listeners, Josue from Mexico, asks about serious but rare risks. So he wonders, how should we explain to the public that because these risks are in fact rare, we consider them acceptable in terms of overall safety of the medical product?

SPEAKER_00 14:49

Yeah, these are really hard because quite often there's actually a really long list of very rare side effects, some of which, as you say, may be quite serious. But it's it's really difficult to know how much information to give people because it's completely true that we all suffer from information overload. Those long lists of side effects on leaflets just mean that you're not taking in any information. I mean, it looks like somebody's just listed everything that can possibly come out of a medical manual at you. But it is important to be able to give information about what might happen so that people can be prepared for it. I mean, maybe some of these rare risks, you want to be allowing people to look out for symptoms and catching them early if they are about to suffer a rare side effect. So I think it's very difficult to do well. I haven't seen anywhere that does it very well, and I think it's something that we do need to do more research on because I think there's a real trade-off between not overloading people with too much information, but yet giving them information there if they need it. So, for instance, you may consider that out of this list of, I don't know, 50 very rare side effects of a drug, it may be important to pick out three or four key symptoms which maybe patients should be warned about looking for. And if they suffer any of those symptoms, then to contact their healthcare professional, whatever. And to, of course, give them a list of these things, but not expect them to read it, if you see what I mean. So that the information's there for those who want it and for those who've suffered something which they're worried about. But I think we need to really pick out the information that people either need to make the decision or need, after they've taken the decision and they're taking the drug, to be able to protect themselves in the case of a rare side effect.

Federica Santoro 16:51

And it's interesting that you mentioned package leaflets. It's almost as if you read Josue's mind, because he had a second question for you just on that. So the package leaflet is actually one of the most important risk communication tools in pharmacovigilance, and it lists the known side effects of a medical product and how often those side effects occur. The problem, as we probably all know, is that very few people read the leaflets, and if you ever opened one, you know perfectly well why, because they tend to be long and dense. The information can be quite difficult to navigate, and to make matters worse, some countries don't even issue them. What do you think would be a better way to deliver the message in this case?

SPEAKER_00 17:38

It's very hard, but I really think they need to include some graphics to help. Because a long list, well, it's a long list of side effects that you've never heard of in complex medical terms, and they're kind of grouped into these sort of clusters of rare or common, and you might not even really know what that means. I think we need to be really sitting down and redesigning these. Because, as you say, they're one of the most important pieces of risk communication in pharmacovigilance, and the fact that they're basically incomprehensible to all of us is uh a pretty damning indictment of how well we're doing that risk communication. You know, at the moment, it's a bit like the small print on a legal document, and it's not really there for communication. It seems to be just there because people feel better if they've listed it all, or possibly for legal reasons. Whilst I think we need to think about it as what information would be helpful here for the person who is taking this drug or considering taking this drug?

Federica Santoro 18:43

Let's shift gears a little bit. I wanted to ask you about action and inaction and when they carry different risks. And I'll give you an example. When the first cases of unusual blood clots were reported with AstraZeneca's COVID vaccine, several countries in Europe decided to pause vaccinations, and they cited the precautionary principle. So basically they took action to prevent potential harm, even though the evidence was uncertain at that time. But uh critics said that the principle had actually been misused, and that those countries did not consider how many more people would die of COVID without the vaccine. So my question is how do we grapple with this difference, the risks from action versus the risks from inaction?

SPEAKER_00 19:38

It's a really, really important point. And I think it's it's quite an instinctive thing to think that no action bears no consequences. But when you're in the case of an infectious disease, the virus is happening. You it's not a human action. You can't just uh put a pause on the virus whilst you think about your policy. And so, as I was saying, with risks and benefits in general, you have to weigh them up. And you have to think about what are the potential benefits of this vaccine. Well, the benefits depend very much on the prevalence of the disease. And you need to weigh up what is the current prevalence, therefore, how many people are at risk of COVID in this case, if we are doing nothing, and how many people are at risk of harm from the vaccine in this case, if we are vaccinating, and then have a think about it from the point of view of different subgroups of the population as well. So when we were faced with communicating the potential blood clot risk from AstraZeneca, that's exactly what we did. We tried to think about what the decision actually was, and therefore what things would be on the two sides of the scale weighing up. So the blood clots were a serious potential harm. So we put on the other side of the scale, as it were, the potential for ICU admissions from COVID. And we tried to illustrate for different age groups and for different exposures to the virus what that balance between potential harms and potential benefits would look like. We literally, on the one side, had number of people who might be harmed with the blood clots, and on the other side, number of people who might have a similar level of harm from the virus. And we hoped that those would be helpful diagrams to policy level decision makers, because then they could see what age groups the potential harm benefit risk ratio was about the same, and where it might tip in one direction or the other direction. And certainly in the UK, I think that was helpful to people. But it also by putting, we made this graphic for different prevalence levels of the virus. Or exposure levels to the virus, it meant that as in an individual, you could look at that and say, well, I personally, for example, am exposed to quite a lot of this virus because, for instance, I might work in a medical facility or I might work on public transport, and therefore my exposure is quite high. So my potential benefits from the vaccine might be higher, whilst my harms from the vaccine will be about the same. Or the other way around, you might be shielding and therefore your exposure to the virus is much lower than average. So it's important for everybody to think about, as I keep saying, the benefits and harms, but also if you're in the face of an infectious disease, it's a different situation from if you are in control of all the variables. I often think of it as, you know, during the epidemic, decision makers, we were like standing in the middle of a road, a busy road, and you had to decide whether to go left or right. Standing still in the middle of the road was a bad option. You had to make a decision to go either left or right to get out of this road.

Federica Santoro 23:06

We ran an episode on vaccine safety communication a few months ago, and my guests there, two specialists in pharmacovigilance and drug safety, stressed exactly the importance of context, to how it's important when talking about the risks of vaccines, to frame those risks in the context of benefits from the disease. And what I'm hearing from you is also how important context is.

SPEAKER_00 23:33

Absolutely. I mean, a lot of people think about risks and risk communication as being about numbers, but numbers are meaningless. Numbers are an artificial construct. What we're trying to get across is not a number, but a subjective kind of feeling of an expert's idea of what the potential likelihood and the potential impact of something happening is, including the quality of the evidence, about the uncertainties, about what we don't know, so that somebody can know how much weight to put on things in a decision. So, yes, there's a lot to try and get across, which is why I think numbers on their own certainly not good enough, and why graphics can help quite a lot, because graphics are a way of communicating quite a lot of information in a really instinctive way.

Federica Santoro 24:23

Let's talk about stories for a second. Building empathy through storytelling is really important in communications, but certain narratives may be more effective than others, especially when we want to change people's behavior. And let's take climate change as an example. Some within the climate movement have been criticized for painting a really bleak vision of our future on Earth. The sort of doomsday scenario that critics say really paralyzes people with fear and hopelessness rather than spur them into action. What kind of narratives work best then for medical risk communication?

SPEAKER_00 25:08

I'm so glad you've got onto the topic of narratives. It's a topic that absolutely fascinates me because in my previous career in documentary making, narrative was everything. Storytelling is king. You really want to lead the audience, engage their emotions, you want to take them with your music and with your editing, you're going to take them through highs and lows of emotion, lead them to a conclusion. But all of that is about giving people a message or being persuasive and wanting to change their behaviour. Now, most of the time in medical communication, you don't want to be telling people what to do and persuading people to do one thing or another. That goes against informed consent. There are some cases in public health, for example, where you might consider that persuasion is ethically allowed. You talked earlier about smoking, for instance, and helping people to give up smoking. And I think it's really interesting that there are probably topics in which there's a kind of societal consent for persuasion. Maybe drinking alcohol is another one. But there are some areas of health and public health, and certainly all of individual health, where we shouldn't be persuading. It should be informing people so that they can make up their own decisions. And that's a kind of a legal and an ethical obligation. And I think in those areas, narrative can be a really dangerous tool because narrative is persuasive. We're doing some work in the Winton Centre at the moment to see whether there is such a thing as a non-persuasive narrative. Because there's been a lot of work which shows that narrative can be very persuasive. There's some work that shows that narrative can help people remember facts and they can certainly engage people. But is that what you want with a medical communication? Is it possible that some narratives, for instance, patient stories, are actually biasing people because they're empathizing with one particular patient story because maybe that patient resembles them? Are they shutting down their critical faculties because they're being led through a story? Is that what storytelling does when we're transported, when we have that suspension of disbelief? Is that what storytelling does to us? I think these are all really interesting questions. But I think it's key to bear in mind, are you trying to persuade or are you trying to inform? And what are the right tools for doing each of those? Can you just inform with a narrative? Or is a narrative always persuasive? So I would say in medical communication, be really careful about the use of narrative unless you are overtly trying to persuade people. And your example in climate change is where people are openly trying to persuade. They're clearly lobbying or having a, they've got that kind of societal buy-in that this is an area where we as a society agree that we should be doing things to improve climate protection. So there, maybe narrative can be used in a justified way. But I'd be really careful within medicine.

Federica Santoro 28:40

That's fascinating, and I really look forward to learning more about the outcome of your research on narratives. But let me pick your brain on this distinction between informing and persuading a little longer. The Winton Center's mission statement is in fact to present evidence in a transparent way to inform but not persuade. But then don't you also think that organizations like the Winton Center and like ours have some duty to guide?

SPEAKER_00 29:12

Well, what do you mean by to guide? Where do you think you want to guide people to? Do you have a decision about whether this particular person should be taking this drug or not? You might say you think that this drug should be available as an option to people because there is some evidence that it has benefits for some people, and the evidence is that it has harms for fewer people. But it's not up to you to decide whether this individual person should take this drug or not. So I think your role is always to inform and never to persuade in pharmacovigilance. We need to trust the audience. These are people who we are helping to make a decision of their own. And if we communicate well, then we can facilitate that, we can empower people to make their own decisions. I think uh trying to guide people is absolutely fine when they've asked for guidance. As I said, if you've asked your medical professional for their opinion, what would they do in your position, then that's absolutely fine to give an opinion. But you shouldn't be doing it in a really persuasive way, then you should be just stating your opinion and why you've come to that decision. And with policy makers, for instance, talking about whether a vaccine is going to be available to a certain sector of the population, then again, it's a policy-level decision, but you should be able to explain your decision making in an informative way, to allow people to see the evidence that you weighed up and why you weighed it in a particular way. But you shouldn't really be doing that in a persuasive fashion to say, you know, this is the right decision, and I want you to believe me that this is the right decision. You should just say, this was a hard decision to make, this is the evidence that we weighed up, this is what we know, this is what we don't know, this is the quality of the evidence, and this is why we weighed it this particular way.

Federica Santoro 31:09

As if risk communication weren't complex enough, it's uh especially challenging to communicate any piece of information, I'd say, when people are exposed to an endless stream of information, like nowadays, right? And especially when some of that information is more reliable than other. How should we deal with the challenges of social media communication, misinformation, and the infodemic?

SPEAKER_00 31:38

Well, that's a big topic. Uh, I don't have easy answers for sorting out information on social media and misinformation. But I think there are a couple of things that we can bear in mind. And one is helping people to look at information, particularly if you don't know the source of that information, with a critical eye. It's called pre-bunking or even vaccinating against misinformation, where people are taught to look for particular tricks that people who are trying to spread, particularly disinformation, where it's spread with a knowledge that it's false, um, tricks that people use to try and hook you into it. And so these sorts of inoculation techniques allow you to look out for things that are really hooking into your biases, or reasons to make it shareable, you know, things that you might want to look out for. But I think on the converse of that, if you're a communicator, in this world of social media, you have to learn to communicate in a trustworthy fashion, so that those who have learnt to look out for trustworthiness in information can spot your information and perhaps put more credit to it. So this is things like being upfront about your uncertainties, about preempting misunderstandings. So, you know, saying things like this doesn't mean X, and being upfront about potential harms or potential costs as well as benefits. We say that if you're upfront about all of these and you're demonstrating trustworthiness, then you're doing the best that you can for your audience. One really interesting experiment that we've done recently is comparing a piece of text or a message that had none of these, that was persuasive in its intent. So we did an experiment using COVID vaccine information where it was purely about the benefits of vaccination and why you should go and have vaccination, was based on standard government messaging. And we compared that with a piece of information that was upfront that this was a choice, that it was going to have pros and cons, talked about the uncertainties and the unknowns, and preempted some potential misunderstandings or misinformation. And we gave some people one message and some people the other message, and we asked them to rate it on various criteria. And people who were already pro-vaccination, they found both messages were, you know, engaging and positive, and they had no problem with either message. People who were vaccine hesitant or anti-vax were actually turned off by the message that was purely pro-vaccination. But the message that was much more open about uncertainties and framed it as a pros and cons and weighing up and a decision to make, that didn't antagonize them as much. They were much more likely to engage with that sort of message. Neither of these messages made any difference to people's decision making, because actually I don't think anybody's found any messages make differences to people's vaccine decision making. But I think it's really interesting to bear in mind that if you are open about harms and benefits, pros and cons, uncertainties, you're not putting people off, it seems. And you are perhaps engaging with the audience that maybe you want to reach the most, which are those who are initially a bit concerned about this topic. And we've done it again with uh nuclear power as a topic and found similar results. And I think these are really interesting studies to be doing.

Federica Santoro 35:28

It's so fascinating, and there are so many levels to the topic that I feel we could discuss this for hours. To sum everything up, what are your top three pieces of advice for drug safety professionals trying to improve their risk communication?

SPEAKER_00 35:45

Well, I'm sure you won't be surprised. My top three pieces of advice. I'd say first off, listen. Listen to your audience and think from their perspective. And that helps you communicate with them. My second piece of advice would be to be upfront about pros and cons, be upfront about what you do know and what you don't know. And my third is to give people the information they need in a way that helps them make the decision that they're trying to make, that allows them to weigh up the evidence for themselves.

Federica Santoro 36:14

Thank you so much for being here with me today, Alex. I hope you enjoyed our conversation just as much as I did.

SPEAKER_00 36:21

Oh, absolutely. It's been a pleasure. And uh, yes, I could talk about this all day. Thanks again and have a lovely day. Thank you.

Federica Santoro 36:32

That's all for now, but we'll be back soon with more conversations on medicine safety. If you'd like to know more about RISC communication and the Winton Center, check out the episode show notes for useful links. The episode on vaccine safety communication that I mentioned during the interview is available in the archive, along with many other episodes on vaccines and effective communication. If you like our podcast, subscribe to it in your favorite player so you won't miss an episode. And spread the word on social media so other listeners can find us. Apart from these in-depth conversations with experts, we host a series called Uppsala Report's Long Read. A selection of audio stories from UMC's pharmacovigilance magazine. So do check that out too. Uppsala Monitoring Center is on Facebook, LinkedIn, and Twitter. And we'd love to hear from you. Send us comments or suggestions for the show, or send in questions for our guests next time we open up for that. For Drug Safety Matters, I'm Federica Santoro. I'd like to thank Alexandra Freeman for her time and all the people who contributed questions. Listeners Maria Victoria and Joshua, and my colleagues Graham Nadasi, Gerald Ross, and Matthew Barwick. Matthew also helped with post production. Last but not least, thank you for tuning in. Till next time.