 The way we're going to do this is we're going to ask, to guess to present some slides, showing examples of visualisation, to talk about briefly about ten or fifteen minutes each. And then we're going to begin a conversation which we do hope you will join in. Kick around a few thoughts from the beginning, but after that we'd very much welcome your questions and contributions and comments. So, Sally, can we begin with you? Would you like to show us what you've been thinking about in relation to visualisation of health? The Chief Medical Officer post started as an adviser to the Privy Council. It's a post that started in 1855. A statutory post as an independent adviser to government on all medical matters. The great fun of it is that I get to speak truth to power, that I have to speak without fear or expectation of favour to ministers, government and also talk to the public about public health matters. I'm the 16th, and if this works properly, you'll see that I'm not only the sixth most powerful woman, but I'm the first woman CMO. In this role I'm head of the public health profession, as well as leading research for the NHS, because I come from a scientific background. Now, one of the ways I get to speak independently as CMO is by publishing an annual report, which is totally independent. It's a very public way of sharing the public health agenda. I speak to the country from this position, and I'm often called the nation's doctor. I can speak about the annual report and the issues there, or actually about the science. Many of you have heard me talking about mitochondrial transfer and the so-called three-parent families, and trying to explain the science on pinning that. I usually front the annual stop smoking campaigns, and this gives me the opportunity to speak about many issues to the public, to health professionals, academics, funders, commissioners, policy makers, and indeed now I'm on the executive board of the World Health Organization to all of them as well. Today, for example, I've spoken to the local government association, I've done some filming for the BBC, and here I am. I'm quoted in the medical press, I write for the medical press, I'm quoted elsewhere, and I've even, on a subject that I am working hard on, antimicrobial resistance, done a TEDx talk, The Drugs Don't Work, and I clearly missed an opportunity because I have a book I could have brought and signed. But you can get it on Amazon, it's called The Drugs Don't Work, and it's about antimicrobial resistance published by Penguin, and I don't take the royalties, so I can tell you it's worth a read. My colleagues take the royalties. I do all of this, and I can see that my team gave you a particularly serious picture, top left, and I try and help people understand the issues, not just through using words, but also as I'm going to show you through diagrams and pictures, because, for me, visualisation is very important. I've got all the public health priorities that you would expect, obesity, physical activity, and my most recent call to the nation was to take on board the fact that average weight, and I'll come back to this with an infographic, is now unhealthy weight that we've normalised over weight. We've normalised a sedentary life in this country, and I'm working on alcohol, antimicrobial resistance. So I have the public health priorities, but I also have some that arise out of the annual reports. I've mentioned antimicrobial resistance, and I have to, on behalf of all of us, respond to sudden issues, and the obvious one is horse meat, and many of you will remember that I had to explain that even if you had eaten horse meat, it wasn't going to harm you. I've eaten horse meat knowingly, it's perfectly reasonable, and if it was contaminated with phenyl butazone, you wouldn't need to eat five or six hundred big horse burgers to get a single human dose. Fergus Walsh of the BBC did this wonderful picture of the five, six hundred burgers, and it just is so revolting, you quite understand, you could never eat enough to get a human dose. An issue that's arising that I'm very concerned about are e-cigarettes and vaping, and we could talk about how marketing is being used by tobacco companies and e-cigarette companies for that. But let me go back to the annual reports because that's where I'm using visuals. We've got the CMO annual reports going back to 1877, so a long time, and there's been one every year except during the Second World War when there was just one report for the whole of the Second World War, and in 2010 when I was appointed for a year as interim and not as a substantive CMO. I've chosen to do my report quite differently. I'm doing two volumes each year. One volume is about the health of the nation and its surveillance, and I'm going to show you some of that and the example in the exhibition comes from one of these. One of the important things about this is collecting a lot of data actually for public health people to use but of interest to all of us. How do you show it so that it matters to people and it means something, and then making sure that behind it all that data is available by local authority so that people can use it themselves and reuse it because I think that's very important, that transparency. So a surveillance volume each year, and then each year I've chosen to do a deep dive, as I call it, an in-depth review of the subject, getting in the scientific experts who write the chapters, tell me what this means, and then I write with the chief editor the policy implications of that, and I have to find different ways of doing it and it was that first one on infection, which I thought was going to be totally non-controversial, I should tell you, but went from the NHS to the community and impacts on all of us when all these experts said that antimicrobial resistance is a problem, and I realised the experts hadn't got a voice, and as someone said to me at an international meeting recently, I changed from being chief medical officer to being chief marketing officer around antimicrobial resistance, and one of the problems I want to show you in this diagram, we've coped with antimicrobial resistance, antibiotic resistance very easily in the past, it's natural selection, it happens very easily because we had a golden age of discovery, but what you see in this, and this comes from Davos, the World Economic Forum, is that we were making new antibiotics right through there till 1987, and since then, we've had no new antibiotic families. Now, I think this is a useful infographic, but it didn't, to me, have the strength I needed to get this issue over, so we redid it in this way to try and make it a prominent issue. I have to say that the Prime Minister called me in a month ago to discuss it, and he's now got it on his agenda as we'll begin to be clear. So, this was me taking graphics from other people and working with my team to try and make the real concern, because this is like climate change, if we don't get new antibiotics, we'll die of antibiotic resistance before climate change hits us and we're doing it to ourselves. So, I thought that that was better to make it clear, and in that report, we used some very nice infographics, and this was one that actually, the British Medical Journal used a year ago, March 2013, as a picture of the week, and what is important here is to show you this section of 36% that are infections in the blood of something called E. coli, and there's a lot more than we knew until we put it together in this way, and the important thing is that 30% of E. coli set to senior patients die if you have antibiotic resistance, and only 15% if you don't that models out at 5,000 people dying in this country every year of antibiotic resistance of their E. coli infections, and that's much bigger than the deaths of MRSA, C. difficile, but by putting it on this circle in that way, there's probably a proper name for this, we drew it to people's attention and made much more of it. I've got another example to show you here out of the same volume, our chief editor Tom Fowler is here, and he pulled it all together, and this is about liver disease, and it got a lot of attention because the black line, this line is England, and these are mortality, so deaths rising due to liver disease. Now, liver disease causing deaths caused either by alcohol, or obesity, or on diagnosed viral infections, almost entirely. You can get Wilson's disease and things, but they're very few, so they're all preventable, yet look at that going up, and the red line is the EU 15, the countries you'd want to compare us with, coming down. This really demonstrated to everyone that we were going the wrong way compared with everyone else. So, by using this, we were able to demonstrate we have a problem with liver disease, predominantly actually alcohol. It clearly takes time to sort, and I wanted to use a lot of different ways of looking at things in my report. One of the things we did use was not just, I clearly got to click it again, standard maps, which I've learnt to call chloropleths, did I say that right? Chloropleths, where the size of the map relates to the geographical size, but we've used, with input from Danny Dawling, the social geographer up in Sheffield, who's fantastic, he provided the files to help us do this, cartograms, and here, what you see is the size relates to the population. So you're getting on this, not just geography, but the issue and the population size. I think they're terrifically powerful. And if we go to the next one, which is one of the ones that we used in that particular annual report, what it actually shows is good news that people who live the longest spend the shortest amount of time living with long-term disability or illness, and so it really is a myth that increasing longevity means people will spend more of their life in ill health, but inequality is evident. Now, Tom Fowler, who's the chief editor here in the audience, added to these cartograms and he added two things. He added this histogram so you can see the distribution of the issue and the problem, and we made sure that dark was always bad and pale good. And then he added red edges to local authority areas, I'm losing this, like this, where there's statistical significance. So dark surrounded by red, and it's really bad news. Have I got that right, Tom? Good. They have to train me well. And by using these different ways of showing things, we hope to draw people in and to get them thinking about the issues. I think the great thing about these cartograms is that they allow you to build hypotheses that you can then explore and think about and debate. I think, I do think that they are fantastic the way they are. This one is an obesity one showing you the prevalence among reception year children. I think for me the interest is putting that next door to children going into secondary school and seeing they go in with 9% overweight or obese and they move into secondary school to double that overweight. What you see here is obesity differs with geographic location. We could show you ones about gender and age group deprivation ethnicity. So, you know, we can begin to tease things out. It was only when I got the whole report and looked at it that I realised how badly we were doing for the public in the north west, the northeast and Cornwall. I mean, there are bits of London too, but it really brings it home to you by using these. And here we've got the proportion of population on the left aged over 16 who report abstaining from alcohol and then on the right the proportion of drinking population who engage in higher risk drinking. So you can compare are the same places abstaining and my geography is very bad so I'm not even going to guess where that is. But look, a high proportion of abstaining and over drinking and that leads you to wonder whether they're directly related. But you can see other areas where they're not the same. That area there there's much less over drinking and more abstaining. So you can begin to think what's going on. It makes you think about it so you can test it by looking at the data or asking questions. I use them for graphics this is one from my infections report and it's actually quite complex when they did it for me I had to spend quite a bit of time looking at it but it begins to tease out which of the important infections where they come from are they imported, do they relate to travel history or not, things like that quite important at the moment malaria but what with Ebola going on here's another hemorrhagic fever dengue I was horrified to know that in that year we'd had 294 cases of dengue coming in and you can see some of these other issues there we're all very concerned about TB and I should know where it is on there but I've lost it but TB is on there and is an issue so it begins to show you things around hospital bed days used there's some 1% of hospital bed days for infection but if we don't sort out the antibiotic resistance that'll go up and age of death will come down and to explore the issue about foreign travel and staying abroad the final one I want to show you is our latest report and the chief editor for this experimented with more user friendly images for the layman and the left hand one is around obesity of every 100 adults in England two are pathologically underweight look at the number that are obese and overweight and it is therefore a minority who are a reasonable healthy weight and this side is around what's the most you drank in one night last week and isn't it a good thing that 40% I don't know how they managed it didn't drink last week at all since I know the literature I drink much less than I ever did and I'm within the CMO guidelines on alcohol consumption I want you to know but drinking nothing in a week I can't remember when I did that oh dear, I probably shouldn't have said that about a third drink within the recommended limit and there's that third that I worry about which is more than the recommended limit and we've got some groups at the moment looking at the evidence base for what is safe to drink and reasonable to drink and they're working now with other people including David's going to have a look at it what's the best way of communicating this to people because it's one thing me harangig people and saying you shouldn't over drink but we all know that will have no impact how do we get people to understand what's in their interests how do we make a difference for them so that they don't damage their health but we don't stop them enjoying life and those are the kind of questions that we need to ask and I think visualisation of the data helps us move that forwards so I do want to thank the British Library I thought it was an absolutely lovely exhibition I really enjoyed going round it I'm going to go round again before it closes I learnt a lot from it and I hope you've all enjoyed it too thank you Sally just before we hand over to David you said at one point that it made this point about being the chief marketeer are you selling an opinion or are you presenting data ooo interesting I'm presenting data but I am selling to governments around the world that if they don't take action in my opinion they are heading for a doomsday scenario and with the report audience is it the public is it politicians is it doctors who are you primarily thinking of when you construct those graphics they're aimed at the doctors be they public health community or the people in the NHS and the politicians so they have to be intelligible to politicians or I'm not doing my job which is advising government and politicians so my most recent one was on children and looking at the return on investment for taking early action to prevent ill health or problems and I was so shocked by the data that we got from the scientists that I called it our children deserve better so I do have an advocacy role and I am allowed an opinion but I try very hard to make sure my opinion is based on fact and scientific evidence and that there's a few scientists who'd stand up and say thank you she's right we gave her that data I'm sure we'll come back to this David do you want to show us how you think it looks from the point of view of the citizen the consumer of information the public so I've got a slightly different perspective on this which I hope if Sally hasn't messed up my stuff completely I can make work I'm interested in the stories that are told on the basis the kind of public health data that Sally has at her fingertips so I'm interested in what it means to you how it's communicated to you by the media or how by anybody else so this is by starting point of the sort of stuff that the daily express will feature so brace yourselves for the daily express this sort of daily boost cancer risk by 20% so what this says is showing that consumption of processed meat 50g per day increases your cancer risk by 20% in fact the epic group in Europe have just released a new paper showing again reinforcing this that in fact it looks like a daily 50g of processed meat that's about three rashes of bacon or a large floppy sausage that's about that increases your annual risk of mortality by about 18% what's known as the hazard ratio the processed meat at least is fairly strong evidence now how is that communicated now that's what's called a relative risk communication it just tells you but what does it mean do I worry, do I care and so there's a lot of research has been shown that this relative risk communication actually exaggerates the magnitude of the effect of something and it's generally considered that the way in which these things should be communicated for a transparent communication an honest appraisal of the evidence by individuals should be in terms of absolute risks what does it mean to perhaps 100 people like you so we could ask what does it mean for me just giving people we'll come back to that to the example in a moment just giving people info is unlikely to have much impact on behaviour as Sally said just telling them what's going to happen is known is not going to change people's behaviour very much but I do believe it's actually necessary if only ethically to provide that information it's necessary but not a sufficient condition we'll get back to behaviour change later I'm sure relative risks are misleading or at least they exaggerate the importance and the other way to communicate this which is the classic public health way to say oh you know 10,000 people a year will die of X or 5,000 people in the country will die of X is actually I think completely inappropriate and irrelevant to communicating to an individual extremely relevant to communicating to a profession and to policy makers but for me what do I care I care about what does it mean for me okay so the standard way in which it's recommended to communicate this kind of thing now which is being taken up by many organisations is to say what does it mean to 100 people we saw that in Sally's similar sort of icon arrays so for this 400 people I should say this is particularly this study of pancreatic cancer which is an extremely nasty cancer and very horrible indeed but fortunately only affects one in 80 of the population so let's say 400 people I'm sure like you who sit down to your nice healthy smug middle class breakfast of muesli and nut, fruit sadly still, 400 at 5 will get pancreatic cancer during their lifetime so let's compare it with 400 complete slobs who sit down every morning to a great big greasy three rasher bacon sandwich that's how many will get pancreatic cancer according to this study so did you notice the difference that's the difference because that's the 20% increase over one in 80 chance one in 400 400 people are going to stuff that down their gob every day of their lives in order to get one extra case of pancreatic cancer put like this it doesn't look so impressive at all and it's well known that it doesn't look that impressive however this is now considered the sort of standard way I'd like to show some examples of where this is being put into practice and the first example I'd like to show is the new breast cancer screening leaflets I was on the committee that drew these up and these are staggeringly innovative it's the first I think in the world to take the approach that's being taken it's basically based on trying to take a non paternalistic approach it says to a woman consider the offer of having breast screening here is the information here is the evidence about the pros and cons and it's explained in a particular way I'll show it in a moment and it doesn't make a recommendation same for cervix and bowel cancer the new leaflets do not recommend that you have screening this is an amazing innovation and we still don't know quite what the effect will be the breast cancer attendance rates have dropped slightly over the last few years but this is before the leaflets come in very slightly so it's not known what the effect of this much more transparent open effect might be and the information is based on an evidence review commissioned by the Mike Richards the cancer czar and various methods of communication were tested I was one of the presenters to a citizen's jury very good engagement with the public about what they understood and they've been taken up by for example Breakthrough Breast Cancer UK which we recently won an award for the communication about breast screening on their website I recommend going to it website where you say would you like to know more would you like to know more and so for example there infographic for explaining breast screening is this one I'll show a different one in a moment this is one way of doing it so this is 200 women who don't attend breast screening over 20 years they say 12 will be diagnosed with breast cancer 8 will be treated and survive and 4 sadly will die from their breast cancer so that's 4 out of 200 between 50 and 80 for 200 women going for screening 15 will be diagnosed with breast cancer 3 more 12 will be treated and survive a very good survival rate 80% survival rate 3 sadly will die early from their breast cancer so what's the difference of the experience of those 200 women from that side or that side in this group one more has died in this group the group went for screening 3 have been over treated they've been diagnosed and treated for cancer that would not have affected them they wouldn't have even known they had it if they hadn't gone for breast screening what that corresponds to in terms of numbers is that over the year in the UK 1300 women essentially have their lives saved or early deaths prevented from the screening program at the cost of 4000 women being treated totally unnecessarily so that's the facts and that's what's gone into the new screening leaflet the 3 and the 1 being the balance of the harms and benefits here's an alternative graphic instead of an icon array there's a frequency tree saying exactly the same thing this is 200 women who are 10 screening 15 will develop breast cancer 12 treated and survive 3 die from breast cancer 200 women not going for screening again here we have the 15 all develop breast cancer but 3 are unaffected 4 die from breast cancer 5 are unitive graphics the citizens of the jury got presented with both of these and actually they understood them they actually like the table as well just for the numbers and they like the frequency trees I like frequency trees quite like you don't see the full 200 you don't quite get the impression this is actually very low numbers of people who either benefit or are harmed but it's there now one of the tragedies is that neither of these graphics ended up going into the leaflet to which I shall now bang my head against the wall they were in we were diddering around for ages and they were in it to almost the last draft and they got taken out because they tested it on some women and they said oh I don't understand that because you look at that if you're on your own and you're not very numerate and you look at that, that's quite difficult to understand people find it difficult to understand so it's something I'd like to I believe though there is a numeracy paradox in this kind of communication what happens is that the leaflets are designed for people with a reading age of 11 and a numeracy of you are probably even less however, those people there's increasing evidence that these are the people who do not particularly want to engage with shared care informed choice reading the leaflets the leaflets are designed for people who don't want to read the leaflets that's the case with health information at the moment I think it's a real paradox I think we need to go into multiple levels of explanation I was pleading, let's have two pages of the back and a little print so people who want it to know more so I think websites again can do this multi-level the crucial thing that's been shown in psychological research and communication is that one size does not fit all you can have something attractive but for people who really engage with it it needs to be geared up to their levels and engage for interest and numeracy okay, next example chronic risks, this sort of stuff I don't know about you, this is my favourite this is my idea of a meal out I love this sort of stuff I couldn't care less what people do people can eat and drink and smoke themselves to an early grave couldn't care less, that's not part of my job but I do believe they should sort of know the consequences another moment, they're badly served so for example Daily Express again fancy that it says eat more, less meet more veg it's a secret of a longer life I'm trying to eat less meat and more veg but they say things like if people cut down the amount of red meat they eat to less than half a serving a day 10% of all deaths could be avoided whoa, isn't that amazing why don't you have that in your graphic that would really convince people isn't statistics wonderful so this is not great and it's not very honest what epidemiologists show is that changing your behaviour or different behaviours lead to different annual risks of death and what they're always referring to is what's known as the hazard curve this is also known as the force of mortality this is a great graphic so Ashley what this shows this is the chance of not living to the next birthday so for me I'm 60 it's about 1% not living to the next birthday on average so 1 out of 160 year olds does not reach 61 you get up to 1 in 10 by the time you're about 83 it's about 1 in 1,000 for 30 year olds and 1 in 10,000 for 7 year olds nobody has ever been safer ever in the whole history of humanity than a 7 year old at the moment in this country 1 in 10,000 will not make the 8th birthday this sadly is the bulge of risk taking youth it's a really really sad very serious so how can we use this how can we sort of communicate something with this let's see we've got an animation of this so for example that's men and that's women women are lower than men at every single age lower risk but we cannot the reason why this is so fundamental is that the behaviours move this up or down is what's known as the hazard ratio that's what epidemiologists calculate a 550g of processed meat is hazard ratio 1.8 it moves this up by 20% what's the interesting fact is that one I really like is the fact that gone puts in 1825 you must have got down like this the curve didn't look that much different 200 years ago except it was way up but it's still in the same basic shape it's a straight line apart from idiotic youth there's something about our bodies that between 7 and 90 makes them die at the same increased rate 9% per year every year we get older it's a 9% increased risk of dying it's like compound interest great rate of interest fantastic it doubles every 8 years so every 8 years your chance of dying before your next birthday doubles amazing in compound interest 9% men women 9% so every year we get older it's associated with a hazard ratio of 1.09 so another way we can think about this a metaphor we can use is that this bacon sandwich with a hazard ratio of 1.18 it's like being 2 years older if you stuff your gob with 50g of processed meat every day you're making yourself 2 years older affecting your age and a couple of cigarettes that's a year older you have a couple of extra drinks you're making yourself another year older two hours of sedentary behaviour of a day down to now be Christmas special another year older you are by that so just be warned just be warned so that's now taken up by this idea of heart age which I'd like to come on to something to do with premature ageing you're making yourself getting older faster than you need be and who wants to be older perhaps teenage girls people do not want to be older than they are it's not being used as a form of communication telling people how old their heart is how old their lungs are being shown to be help giving up smoking you can also show survival curves this is the proportion of people women who will be alive at a particular age this is assuming given that everything carries on the same of course it doesn't carry on the same because if we look at life expectancy it was 82 in 2006 it's been going up at 3 months a year not just for the last 25 years about the last 50 years slowing down I believe it's still going on so every year we only age 9 months we go back again all that public health stuff all those drains that's what it was a few years ago nothing drains so much so 3 months every year so how can we then change our life expectancy so another way of looking at it is to go on to behaviours hang on ah ah ah ah oh I know this is my screen it's not working how does changing your behaviour change your survival hang on this is not working I think I'm going to have to give this one up it's going to full screens it's going to work no ok sorry it doesn't come over on the thing basically what this does is what I was going to show is that if you smoke and everything like this this goes up or down but these ideas just like to finish off by saying that these ideas have been now implemented in a calculator which is now generally available to anyone who wants to use it and the JBS3 calculator and I want to push this slightly because again my team created this calculator it's now available online and a little plug hopefully with the help of NHS England it might go into every general practice it could anyway anyone it's just available online and it's going to join British society Cardiovascular society all these societies and basically the idea is that you can put in your factors based on huge public health studies and find out about your risks of Cardiovascular disease so let's say 74 I'm going to put myself in the mail he's 40 or so and he smokes and his cholesterol's a bit high oh let's put his blood pressure up goodness me yeah yeah bring it on okay so let's see let's see what state he's 40 okay so let's go he hasn't got any family history or anything like that so he says his heart age is about 48 so his heart is about 10 years older than he is because of his behaviour what this means is that he's got the same risks, annual risks of a heart attack as someone who is 48 who's got better habits so this is to do with premature ageing with ageing yourself using that metaphor there's other metaphors you could use this is the metaphor that says how long do you want to see your grandchildren how long are you going to go before you have your first heart attack or stroke if you expect to have your first heart attack or stroke on the thwonto what's the chance of getting to the next 10 years without a heart attack or with a heart attack or stroke and then we're we're tentative about this this is your chance of surviving without a heart attack or stroke at the moment you're here and it's plummeting down like that but look look if you were a good boy and maybe got brought your blood pressure down what have you took your statin even though you're only 48 brought and did that you might get your HDL up as well but of course what you damn well should do is stop bloody well smoking because you go wham you're real bang for your buck there but even if you don't want to do stoking you can do quite a lot with behaviour and drugs so that's a very strong visual image that we designed by clinicians like it and the people tested out on like it as well it's actually mathematically true what's nice is that this area here really is your increased number of years before you expect to have a heart attack or stroke and it's mathematically correct and visually powerful but we in the spirit of one size does not fit all we then possibly slightly over the top put in every damn bit of communication device we could think of you like little smiley faces you can have little smiley faces so you can show that if you want to this is 100 people like you by age 50 one will be dead or something else a heart attack or stroke but four people like you would have had a heart attack or stroke but they won't have now because you've taken your interventions but if you don't stop smoking then that goes back and if you don't stop smoking a much more increased chance you'll die of something else anyway in the meantime so all that comes up there some people like the smileys I quite like icon arrays I like looking at 100 people like you this is just like the screening this is if you carry on as usual this is if you start being a good boy this is what's going to happen to those people in 10 years time let's make it 20 years let's make it 20 years so this is how many will be living without a heart attack or stroke this will have had a heart attack or stroke and these will have died before something else but this is the reduction if you intervene oh you haven't stopped smoking come on stop smoking so now you see but that's maybe a bit difficult to compare so if you take the difference though you can then see very powerfully on a difference diagram that these people who would have been either had a heart attack or stroke or died are now going to be bouncing around like young fawns so that's 100 people like you so these images some of the various representations have been tested out not by psychologists in randomized trials to show the impact they have on people being strongly suggested that people with low numeracy and good explanation can grasp these this tool is not particularly supposed to be used directly by the patients although they can it's supposed to be a three way device between the doctor and the patient so that's the kind of thing I think is really quite exciting at the moment so just for a final metaphor ok here's this is storytelling these risks don't really exist this is stories different images the story of living longer might not be very attractive because who wants to have another year being old and dribly but the story of aging faster might be more gripping because we don't want to get old quicker so for example I went to my GP I've got about 12% chance of a heart attack or stroke at the moment there's one in a million chance I will have it in the next hour so then got some doctors Sally can you remember what to do I'm relying on you so I could take provided I take the staff reduced 33% so the standard icon array for that would be 100 people like me and in 10 years time 12 of them will have a heart attack or stroke but if they all take their statins like good people for the whole 10 years and they don't mind about any side effects or they change their statins to reduce the side effects that many four will have their heart attack or stroke prevented so actually that means a number need to treat it 25 people like me will have to take a tablet every day for 10 years to prevent one heart attack or stroke ok well maybe I do maybe I don't want to take it there is another metaphor because it's 100 people like me and it's been shown that some people can actually distance themselves from one of those there aren't 100 people like me so another way to think about it is 100 futures for me 100 ways things might turn out for me not 100 people other people I don't know what's going to happen in my future I think all these futures are disappearing off in how many of them will I have a heart attack or stroke so of course the right image for that is to really personalise this is really personalise medicine you wait you wait and if you can use your web you can use your webcam you can be an integral part of your risk communication so there's 100 people like me and this is what happens it's even better if you scatter them around it gets a much better impression of the fact that my future is essentially equivalent to throwing a dart over my shoulder and which one I hit is the luck or the draw it's been shown again in randomised trials that scattering the icons increases the feeling of unpredictability makes it more difficult to count much better at communicating the essential unpredictability about the future and with that I shall stop thank you David there's a terrific temptation to suggest that aggregated risks which are really averages apply to you personally or apply to an individual personally so that when we say that sitting down and watching down to Naby every day is two hours or whatever off your life or something like that half an hour per day apart from the two hours wasted watching it apart from the hours wasted wasted but it isn't that's an aggregate effect for an average population not for an individual it's simply a probability which might apply or might not to an individual and you'll never know and you'll never know if you're that individual where you'll never know the effect of watching down to Naby just as actually strictly speaking you don't even know the effect of smoking possibly I've got lung cancer anyway pretty unlikely so is there still scope for confusion there that people will think this is a literal read-off for my own prospects and this is the problem with what honesty I mean all these are associations as well generally we're not sure apart from giving up smoking very good evidence changing diet or something like that well the actual evidence of your health of changing diet as an intervention is quite limited these are generally population associations we've got what half a million people in the epic study or something like that they haven't done had an intervention yes we can compare groups of people that's why in the screening one these are 200 other people they aren't the same people who went there but of course that nuance which I try to keep to does get lost and that people are interested in what will be the effect on me my next drink my next drink of this cigarette and that of course is actually impossible for a start it is impossible to know it's not even that it even exists as a phenomenon because risk doesn't exist I don't think it's something you can go out in the world and you can't even theoretically measure it I don't think what the effect would be so you have to use these analogies and metaphors about large groups that are essentially making an analogy it's as if you were doing something that did have this but it's not saying it did have this I know and that gets lost for the hard ones we do know because the effect on reducing blood pressure and cholesterol are shown in randomised trials of statins and blood pressure reducing drugs so those hard ones are and the effect of stopping smoking are evidence based this should be the effect on you on your risk of doing this Sally how do you deal with the problem of all the caveats and the qualifications and the sort that David has just been describing because you know you're communicating to the public there might be this argument that we do need the numeracy age of 8 in order for this thing to reach everybody but what does that mean I mean clearly you can't put all the confidence intervals and all the rest of it around your data into the main report well in the main report in our main report the clinical scientists do put all that nuance in the policy chapter I don't because I can't I'm talking to politicians and I'm talking to about what needs to be done policy wise and around the visualisations and you have to simplify I mean as a doctor with patients I had to simplify if you take this treatment the odds of you recovering are this or the likelihood 80% do respond to this treatment 20% die despite taking this treatment you have to find ways where people can relate to it in some form and it's simple of course when it's about individual treatment you often end up with what would you advise doctor and then you have a different discussion because that brings in what are their attitudes to life and the quality of life and things but I think on population health when we're advising it is you try and simplify these randomised control trials and get people to live as healthily as possible but I still remember as a young doctor I was told off by my consultant saying to a patient who was dying of lung cancer you're not supposed to smoke but it's not going to do you any more harm so that's where you could go and do it my consultant said you're supposed to stop the model smoking I said well what could just one more general question and then we'll open it out because there's a question of trust here some people might respond by saying well you've not been quite full and frank you've not told me the whole truth here and a related point really that visualisation because it's so compelling it's almost visceral you can get to the point where well as Tim Harford has said misinformation is beautiful too or that the problem with the most accomplished information design can be the most convincing lie so this is a slight danger that by simplifying and by the colour and the energy and the vividness of all this kind of thing there's both the potential for people to be more mischievous and the potential for people to feel they're being slightly sold something on a complete basis there's some very interesting stuff that's starting to be written about the ethics of information design the fact that of people wanting to arouse a certain amount of emotion but not too much and certainly not to do it in a misleading way so I think there's very strongly indeed about what are you trying to do this is supposed to be transparent communication to allow people to better exercise their feelings and their judgements and their attitudes in a way that isn't manipulative and where no, I can make a number look big I can make it look small just by changing the colour you can do all sorts of things putting things on a log scale just change everything so I think this is strongly where ethics come very strongly into it and your attitude to risk so if you take the breast cancer one and you said to me you've got a small lump on mammography you know we might find if we don't treat you that it's nothing and so you would have had treatment but if it is something you're more likely to survive I would opt for the treatment on the grounds that better to A to have the treatment in case but also I would take that treatment in the hope of saving other women if that happens as well as a solidarity issue that there are two things that play there one is it might save my life but even if it doesn't save my life it may save another woman's life so there are value judgements personal to you about whether you're going to interpret this that's very interesting because I tend to usually if someone picks me up from the audience when I'm talking about this stuff saying you've been completely individualistic what about vaccines where you're making a decision not just on your half but on behalf of all your child behalf because of herd immunity and so on because you can be a free rider if you want to be so again this is very different I was communicating about vaccine risk I would say it's quite reasonable not to treat it just as an individual level but to treat it as a societal issue do either of you think that we're in any danger of making the same old statistical mistakes just prettier of course you can and people can manipulate but I think the sort of stuff that we've been talking about is a major step forward because there is it's to do with transparency it's to do with in a way telling stories and trying not to be misleading and it's to do with it's using in a way it's also using metaphors and images that are tested on that can be empirically tested to see what their impact is and how they're and how they're seen and so for example in terms of the framing the idea of looking at the whole experience of 100 people tries to get away from the idea of just concentrating on the bad or just concentrating on the good so for example the framing in the US they always talk about mortality rates for heart surgery which are 2% in the UK we say our surgeons have got 98% survival rates which is much better of course our surgeons are hugely better and that is done that's the different framing that's done in the two different countries and of course what one really should really be doing to somebody saying well out of 100 people like you two will die and 98 will survive even the order in which you put them can have an effect and certainly the colour you use and the graphics can have an effect but essentially that is a story you should be trying to communicate the uniform reporting of harms and benefits is the sort of mantra I keep on buzzing around my head whenever I see anything is this doing uniform reporting of harms and benefits? so I think we need transparency and I like this and it's what we should do but I fully recognise that as we give people more transparency there's less paternalism in the system maybe maternalism and that we may have less of a take up of either screening or something else but that's the risk we have to take I was just going to say that's something that will be empirically tested the other thing is that I think it's very reasonable and something I hear again from my clinical colleagues is repeated that when people are presented with transparent information a very common response is thank you so much I'm really glad you showed me that now what do you think I should do? or what would you do with your wife or your family and I think it's completely that's how I respond as well I think it's completely reasonable because it's something which is honest and fair but the clinician is showing trustworthiness and then the patient then says we would like to have their advice I don't think that's paternalistic that's being helpful that's what they think they're not being paternalistic does anybody want to have a question down on the front here I should start off by saying I'm chair of the transport statistics groups so when it comes to statistics I'm not too sure about that but I've actually got two comments the first is are people not necessarily about medical actually becoming cynical about statistics simply because it is used as a weapon by groups look at the immigration debate look at just about anything my other point is you talked about statins which is an interesting thing now let's say I take 75mg of aspirin to reduce my risk of heart attack that's why I'm also told that it might increase my risk of internal bleeding I'm told there are side effects from taking statins I mean I've really heard inside health well in the series my heart is always going on about these statistics I think there are so many factors when it comes to taking your bacon or cholesterol actually isn't so much the category that I've thought it was the risk of cholesterol and everything like that and it seems to me that you do something there are so many ways it can affect you some possibly, some negatively are you ever going to be able to explain it to me because at the end of the day if I go with the doctors I'll probably say a bit about statistics you're the doctor what do you recommend I think that's a very sensible approach the point is that trying to identify the effect of one bit of behaviour one sort of thing really is very tricky and actually is doomed you can't say because our behaviours are hugely correlated and trying to identify the effect of one single thing is really tricky indeed the point is that when you do all this quantification you end up with amazingly new insights like don't smoke don't drink too much take some exercise and have a healthy diet wow you can ask a doctor but that's what he will say because it's what works and you can try to look at the relative contribution to these and smoking it dominates very much but basically that's the cluster of behaviours that's going to help the health of you personally and the nation so I think in a way statistics being used as a rhetorical device I don't like that because I'm very happy to do it myself and you could say that that's in fact what a lot of this to some extent is because in the end there's some fairly basic rules of behaviour that we know it would be a good thing to do and the statistics and that's why actually I believe that as I said at the beginning the statistics by themselves are not shouldn't be the main tool of trying to change behaviour because they're ineffective anyway because people say another statistic however they communicate in a transparent way and consistently from different people and accompany crucially by public health interventions that actually change what's in the trade known as the choice architecture the whole nudge, non-conscious influences on people just make it easier to do a better thing especially influencing social norms trying to influence what people around you are doing then you can change behaviour because behaviour is changing it does change the ultimate objective isn't it Sally to change behaviour I'd like people to lead healthy lives but is it to change behaviour or is that up to them it is up to them only they can decide to do it but clearly if we can nudge the population into healthier behaviour they will benefit and we will spend less in the NHS which is another advantage I think it's very difficult why take physical exercise the evidence is absolutely clear that sedentary lifestyles are bad for you and physical exercise is good for you but we have a population that's getting more and more sedentary and we've found it very difficult to find our way through how do you increase physical activity in children let alone in adults Do you share some of the skepticism we've heard about the possibility that visualisation is going to be an important part of changing people's behaviour Can it do it? I think it's a minor part I think we're into real behavioural interventions and social marketing but you need all of this to want to pin that so you know that you're doing something that's evidence based you're trying to get people to something that they will find pleasant and good that will help them This is actually a question to Sally Davis all day today we have been discussing ageing here in this very conference hall and the way we culturally deal with ageing populations in the United Kingdom and I was so surprised that you came out with the increased longevity doesn't mean spending more on healthcare for the elderly because they're not necessarily getting more ill health My question is then when are we going to convey this message is this a new finding or when are we going to convey this message to the many journalists the many people and politicians in this country who demonise the elderly every day the way we talk about elderly people in this country, in this culture is the burden on the NHS the financial burden of the elderly population the burden of the state pension I come from a culture where we fear the elderly but also 31 years at the bedside of patients elderly patients working every day with them in the NHS as an immigrant of 31 years I don't recognise these people who are such a burden on this economy and this society Thank you very much Just on the aisle there are a couple of rows back It's now possible via NHS choices although it's not very accessible to find out individual data on consultants and surgical outcomes and deaths but only for some specialties I was wondering do you think data like this is a good idea to show to the public is the data we shouldn't show to the public and who gets to decide David do you want to take the second one first the data on individual surgical performance I've been involved because that's political innovation to put the individualised thing I like transparency of data all sorts of surgical data has been available for some time if you can find success rates for kidney transplant rates for which the professions are run for cardiac surgery they're all available so I like transparency it depends again on the packaging for individual surgeon it's not the issue it's often the unit post surgical care it's not just the surgeon the name surgeon so I think that can be very misleading the other thing is that it's absolutely crucial to avoid any sort of league tabling because people can be different but they're not actually significantly different there's always some variances people have runs of good or bad luck but on the whole it doesn't make a difference adult cardiac surgery in this country is absolutely staggering there used to be quite a lot of variability in practice but then once the profession started really monitoring success rates risk adjustment success rates everybody just came in got rid of those it doesn't seem to make any difference where you have your bypass graft done in the country isn't that a staggering achievement absolutely amazing and that was brought about through sharing data and a degree of transparency for it so I think there is a really good incentive at least for the profession knowing that somebody could see these things so I'm one in favour of transparency but it's got to be done in an appropriate way avoiding league tables i media attention and using good transparency which can be done I've seen the photographs of particular surgeons in the paper with their failure rates as they've been described and it's pretty hard some of them might have only done six operations or something like that one fatality it's shocking unless it's done in a statistically competent way and there are ways to do it to avoid league tabling, to avoid ranking is outrageous and should never be done Sally could you tell us about the life expectancy and healthy life expectancy because it was fascinating to hear you say it well that particular bit of data I'd have to go back to Tom probably to inquire where it came from but we picked it out because we thought it was interesting does show that as people get older we are not the worry was that as people get older their ill health would start at the same age and we'd have years more ill health and that's not actually what we're seeing we're seeing for many people the ill health starting later and so it's not a bad thing to be living older for the NHS costs I share your concern that we have a culture that doesn't respect the experience and everything of our elders and I hope we'll find ways to get there but I think there are two separate things this is really important because I said I made the rather crass joke about another year being old and dribly but that's is a popular perception there's a lovely quote by Kingsley Ames he said I'm not going to give anything up for the sake of another year in a geriatric home in western Superman because he was assuming that the age where he went into the home was fixed and it would just be another year in the home but if it means delaying by a year I'm sure he might have given it up and actually if you look at the data with the statins and anti high blood pressure treatments and everything we can put off strokes, we can put off cardiovascular events and people are healthier for longer it is worthwhile Does it make you wonder how attitudes could change if we got the data presentation right? That's a really interesting question I haven't seen a presentation of that you'd have to have healthy life expectancy in life showing that as you got older the margin of ill health stayed the same width would be incredibly important We'll give you the data to do it for Final thoughts on transparency again this conundrum about whether people are really going to believe you I'm not sure it's about whether they'll believe me the issue is that we should have transparency so all the data I use and those diagrams they're all on the web for people to find them but then how do you put your message over and you start with so who are you talking to and what is the message that will work with them I do simplify, I think we all do if we're trying to get messages over but I make sure that the data is there for people who want to see the rest of it because it's very important So whatever the politics surrounding the message you have to comes back to the hard data in the end as you said earlier I think it underpins I think it's a very important point that you do have to simplify stuff and that's why I do think that again for ethical reasons when needs are multi levels of explanation where people can always go to more if they want to I'd like to know more screening leaflet, endless discussions because there's three and the one they're somewhere around there and in the end they got labelled some people want to put intervals on them and that's too much in the end I think they're given qualitative labels as being the best current judgments so three levels of qualification they're provisional, they're judgments and they're as good as we can do they're only as good as we can do and that's what they're labelled as which I thought was after three hours of discussion it wasn't a bad place to get to just being honest OK, we're going to have to finish there thank you all so much for coming along and for your questions and thank you again David Spiegelhalter and Sally Davis thank you