Helen Pearson: How Evidence Shows What Really Works

What does it really mean to ask, “Where is the evidence?”

In this episode of The Science Intersection, I’m joined by Helen Pearson, award-winning science journalist, editor at Nature, and author of Beyond Belief: How Evidence Shows What Really Works.

We talk about how evidence-based thinking transformed medicine and began influencing other fields, including education, social policy, conservation and everyday decision-making. Helen explains why evidence can be powerful, but also why it rarely gives simple answers on its own.

We also discuss storytelling in science writing, why politicians may struggle to act on evidence, how to assess claims more critically, and what misinformation, influencers, AI and social media mean for public trust in science.

Helen’s book Beyond Belief is available now.

To find out more:

Helen Pearson’s website:
https://helenpearson.info/

Beyond Belief: How Evidence Shows What Really Works:
https://helenpearson.info/book/what-to-believe/

The Life Project:
https://helenpearson.info/book/the-life-project/

Nature overview/interview about Beyond Belief:
https://www.nature.com/articles/d41586-026-01359-1

Transcript

SPEAKER_01 0:10

So on this episode of the Science Intersection, I'm joined by Helen Pearson. Helen is an award-winning science journalist and editor at Nature, a TED speaker, and the author of Beyond Belief: How Evidence Shows What Really Works. She was named European Science Journalist of the Year in 2025 and is an honorary professor of practice at UCL, where she teaches science writing. Her previous book, The Life Project, explored Britain's birth cohort studies, remarkable long-running studies that have followed thousands of people across their lives and helped us understand why childhood, education, health, poverty, and social conditions shape human outcomes. In Beyond Belief, Hearn explores the rise of evidence-based decision making, how medicine moved away from relying heavily on expert opinion and convention, and how the idea of asking, what does the evidence show is now influencing fields from government and education to conservation, business and parenting. So I'm currently in the process of reading the book, and I'm finding it really insightful. And I wanted to explore in this conversation a bit about your background, your writing process, why evidence matters, why it's not always listened to, and how we can get better asking what really works.

SPEAKER_00 1:31

Yes, I'm I'm happy to do that. And thanks for talking to me today and for your interest in the book. So I'm I'm gonna read two paragraphs right from the very introduction because it sort of explains why I wrote the book and a little bit about its flavour. Okay. I first became interested in writing a book about evidence in May 2013 when I visited London's impressive Royal Society of Medicine. I was there to interview Ian Chalmers, a doctor and researcher, as part of work on my previous book. Chalmers is incredibly self-deprecating, despite his work having helped probably millions of people around the world. He told me I'd recognise him by looking for someone who was short, fat, and balding. As a science journalist, I was already steeped in the world of research and medicine, but my conversation with Chalmers would change how I thought about both. In the 1960s, Chalmers realized that much of what he'd been taught in medical school was wrong. Most decisions about treatment were based on someone's opinion, do this because I think it's right, or conventional wisdom, do it this way because that's how it's always been done. This meant that two doctors with differing opinions might give a patient different advice for the same predicament. Chalmers was troubled by this. What was he supposed to believe? He later realized that some children in his care probably died because he handed out wrong advice that he'd been taught. So those are the two paragraphs I wanted to read to you.

SPEAKER_01 2:49

So you you interviewed Ian Chalmers. Was there anything kind of surprising about the conversation with him and how did it kind of shape the book?

SPEAKER_00 2:57

I interviewed him because he sort of played a role in these big birth cohort studies, which I'd written about for my previous book. But those studies actually kind of helped him realize that sort of observational studies or or these type of longitudinal studies that we were talking about, where you track people over time, can only ever produce associations, right, between you know childhood upbringing and poor health in the future. And it and it but and he he was he actually became quite frustrated with their inability to show him causation in in science. He was working in obstetrics and gynecology as well, and told me, you know, this story of how, which I just alluded to in the excerpt of how um he would see doctors handing out different advice to women when they were pregnant. So, like one doctor might say, you know, prenatal vitamins or supplements are really important, or bed rest is really important, and another doctor would say no such thing. And so he was frustrated by this. And he felt like, you know, that didn't provide him with the information he needed. And in fact, he told me that this was so difficult that when he was woken in the night sometimes to care for a woman, his first question would be what's wrong? And his second question would be, which doctor is treating her. He went on and came across the work of Archie Cochrane, who's a kind of towering figure in in sort of medicine, who and and through this became very enamoured with the idea of testing whether things work in randomized trials, which is this kind of powerful technique in science where you randomly allocate people to have a treatment or to not, and then you see if the treatment group improves um faster than the or better than the control group. And from this, he he went on and uh carried out this kind of really uh pioneering exercise in which he decided to bring together all of the clinical trials which had been done to test treatments in pregnancy and childbirth. And and he not only brought them together, but he he synthesised the evidence, which is just like really kind of big theme in the book, is is like how do we make sense of bodies of knowledge. So he synthesised the evidence um on what really works in pregnancy and childbirth, published these two seminal books in the 1980s, which showed that many practices, in fact, in pregnancy and childbirth didn't really have evidence to support them, things such as epesiotomies, for example, which is this highly invasive um technique, and and then through this process of synthesizing the evidence, managed to sort of start to change practices for for women. So had this really sort of pioneering influence on medicine.

SPEAKER_01 5:21

I'm I'm curious in terms of communicating the information to the public. So talking about communicating evidence-based medicine and things such as the Cochrane collaboration, how do you get the story element out there? Stories make things more memorable. So how do you kind of balance?

SPEAKER_00 5:44

Yeah, well, I I think that's I mean, that's such a great question, and that was that was the challenge, in a way, because Chalmers has this really interesting personal story. In a way, his story of like realizing that that so many practices in this field are based on like conventional wisdom, anecdote, that's the way it's always been done. That plays out across my entire book. So in in every single field, it was like somebody who had the same realization, whether it's conservation or management or policing. It's like, yeah, what why are we doing things the way that we're doing them? And should we not test whether things work? And so so I found that story fascinating. And the reason I kind of wanted to write the book is these terms, like evidence-based medicine, almost anyone who goes to the doctor is is being helped, assuming they are helped by the doctor, is has been, you know, touched in some way by evidence-based medicine, but most people have never heard of it. And and so I felt there was this fascinating story around how medicine and all these other fields were sort of challenging conventional wisdom and turning to evidence, which which people, you know, most lay people don't know. And that's and that's kind of why I wanted to write it. And the and the other reason, I mean, I'm getting to your question about the story in a minute, but but the other reason was each of these fields has kind of individually gone on this journey towards evidence. And I wanted to bring it all together and show how it was part of this, what I call an evidence revolution, right? Because all of them have sort of been been doing this at the same time. So show how it was part of this bigger movement. But the the question how to tell the story, I mean, absolutely, you know, in general, I think it is true to say that that stories resonate with people, right? I mean, we like hearing, reading about people and and and just you know, stories that unfold, whereas terms like evidence-based medicine and oh my god, evidence-based policy and decision-making are really dry. And so the way that I approached that with the help of my editor was basically to kind of you know go go hard on these personal stories because because as people, as people individually like started to challenge how things were done and realized themselves that their field should be based on evidence, then through that I could show the kind of the the the how these fields were changing. So it so it was all about in in the end, these I I I sort of centred each chapter on people's stories, or often people who had this realization and and became enamoured by evidence and then went about changing their field.

SPEAKER_01 8:04

A lot of people seem to see evidence as more problematic in the social sciences. They might feel that the information is more subjective or the questions being asked are more shaped by values, assumptions, or political debates than in the field of say medicine. Is that fair? And are there particular challenges in social policy or education where the outcomes might be more contested or harder to measure?

SPEAKER_00 8:32

It's a very sort of quantitative or scientific sounding field, is medicine. Yes, yeah. Well, in a way, I mean, looking at all these different fields, you know, because I do look at social policy, and I mean, you know, I I think maybe it's a misperception to think there is an enormous amount of really rigorous science in in social science and and and other disciplines. So I I wonder whether many scientists have these biases about other fields, about you know, social sciences, which are not necessarily justified. And it's sort of surprising. I mean, it's surprising to people who work in medicine to discover that there were this sort of enormous body of evidence, including many randomized controlled trials, which test, for example, whether policies work or whether educational techniques work. I mean, there's this huge movement in evidence-based education with this big, you know, build-up of trials which test what really works to help children learn. There is a lot of evidence in all of these fields. I mean, I do, I do agree that there are more trials in in medicine and that other fields have sometimes have to be a little bit more pragmatic about the evidence they're able to collect because things might be more difficult, for example, to do a to do a randomized trial. Each field is working within these sort of different boundaries. And you know, for example, in conservation, it it is very difficult to do randomized trials because you know you can't randomly allocate birds to to treatment and control groups, and you can't randomly allocate oceans or or forests. But nevertheless, that you know, obviously that this these fields find different ways to try and do the most rigorous experiments they they can to work out what really works to say species and ecosystems.

SPEAKER_01 10:14

I was thinking, for example, that in medicine you've got a kind of very obvious thing to measure, so like changes in blood pressure, changes in physical health, whereas in social science maybe you're looking more at outcomes in terms of psychological health, which is potentially more subjective and a bit harder to kind of say, all this really works.

SPEAKER_00 10:41

Yeah, no, I I think that's a good point, is that the outcomes that is a real problem in other fields. I mean, in medicine, people agree on the outcome as well. It's like, well, we want to see whatever, a reduction in blood pressure, or we want to see more people survive. Whereas the outcomes can be more difficult to define. I don't know, let's let's think of an example. So, for example, I looked at some pioneering experiments that tested in massive randomized trials in the 80s whether welfare to work schemes could help people move more quickly from unemployment into employment. And welfare to work means you ask people to, you know, engage in work or certain activities in order to receive their welfare payments. And these trials, which were in the US, suggested that, yes, on some measures, you know, it seemed to have an effect. It did sort of slightly shorten the time, I believe, that it took people to find employment and it saved US states some money. But on other measures about outcome, for example, it didn't help more families move out of poverty. So perhaps there's more of a debate about you know what outcomes matter. And I also think you can have more kind of contested outcomes or or people with different ideas of what the outcome should be. So, for example, in conservation, obviously a conservationist really wants to see more species survive, and that's their priority. But there are other priorities that the road developer might think that that's saving money is is the the outcome that that matters the most, for example, you know, and you're in and or be willing to sacrifice some species in, I mean, not whole species, but you know, some organisms, let's say, in order to save money. So that so there are these other things which have to be factored into decisions, and that's kind of true across any any field really.

SPEAKER_01 12:22

How do you communicate that evidence is only one factor in decision making alongside politics, costs, values, and public opinion?

SPEAKER_00 12:32

I mean, I suppose one way I did it was, I put this in some of my talks, but there's this pie chart. So a policymaker who was working in Congress for had worked in Congress for years in the in the US and was really, you know, passionate about getting evidence to be part of the conversation when decisions were being made. And then after after years of experience, sketched this pie chart out, so you need to imagine that in your mind, it was a visual way of showing all the factors which influenced a decision in producing legislation in Congress. And this pie chart, I mean, basically 99% of it is not evidence, so it's it's it shows costs. And you know, a politician has to consider politics, of course, and whether they're going to be elected, and what do their local constituents think, and and lobbying, and and so there were all these other influences coming in on a decision, and and he sketched, and then he sketched that the influence of research evidence as being 1% of the um the influence, which I thought was kind of like really telling. And I felt like it was it was just this fantastic way of also showing what a lot of the people who who are campaigning for evidence want, which is just to kind of grow that part of the pie. They what they want it to be part of the conversation, they're realistic that evidence alone doesn't provide the answer or or make the decision for you, but they want it to be on the table when the big decisions are being made and perhaps to have a little bit more influence than just one percent.

SPEAKER_01 13:51

Yeah. I am I'm kind of wondering if part of the issue is thinking about things long term. So if people are sort of trying to get elected, they might kind of be quite short-termist, and maybe some of the things that are evidence-based require quite like long-term strategies, but also some of the evidence-based is sort of directed at things which would improve things, but you don't see the results straight away, as it were.

SPEAKER_00 14:18

So it's such a smart point. I completely agree with you. So politicians need to have results quickly, right? Because they want to get elected in the next four or five year cycle. But the evidence that a policy is having an effect, whether it's work to welfare payments or or whatever it might be, or an action to save an ecosystem, let's say, I mean, that might not become apparent for a decade, or in the case of conservation, you know, decades, because of course ecosystems change really slowly. And so that is absolutely a problem. Not only might it mean that policymakers might decide that they don't really care what the evidence the evidence shows that it works, because they just want to be seen to be taking action, right? But the other problem with that is it disincentivizes policymakers from collecting evidence on whether their own policies work. Because you're going to be gone in five years. So what why should you invest in evaluating whether this policy works? So this creates this kind of vicious cycle where policymaker would really like to have evidence of whether what they're about to do works, but previous policymakers and even they themselves might not be incentivized to collect that evidence. How would one get around that? How does one get around that? I mean, in some cases, people have just ended up sort of championing, becoming so passionate about collecting evidence that that has got things going. I wrote in the book about another kind of pioneering randomized trial that was done in Mexico to test this poverty alleviation program. It's this famous program introduced in the 80s called Progressor. This is a famous randomized trial that took place, an early randomized trial of a social policy in Mexico. There was a kind of poverty crisis, there was a need for a new policy. The policymakers devised this kind of radical new approach, which is called a conditional cash transfer, where poor families are given a payment on condition that they do certain things. For example, send their children to school or attend health clinics. And so the question was, you know, is this helpful to people? And the the policymakers realized, I think, that um because it it was often the case in Mexico that when a new government came in, they often just cancelled the policies from the last government, that they wanted to collect rigorous evidence in order to basically try and ensure that didn't happen. And they felt that having convincing evidence from a randomized trial would would be hopefully enough to kind of support the program in the long term. And sure enough, it you know, it did suggest that it was beneficial to families, and it did actually manage to help support that program for several decades, although eventually it was um cancelled, but not before this idea of conditional cash transfers had also become quite popular and and had been implemented in other countries around the world. So that trial ended up kind of being really influential.

SPEAKER_01 17:04

We've talked about social sort of social science and we talk about medicine. I'm kind of interested in how people can kind of use the evidence in everyday life. So if we kind of think about medicine, how can people kind of I suppose do their own research and think about what treatment is right for them, but still sort of open up space to have a sort of conversation with their GP, which kind of maybe takes into account evidence-based medicine?

SPEAKER_00 17:34

Yeah, really good question. I'll say two things about that. What one of the kind of core concepts of evidence-based medicine is that evidence alone is not enough to decide what to do. And so the idea of evidence-based medicine is that yes, you you look at evidence which suggests what works, hopefully, based on clinical trials across large populations, but it also factors in the expertise of the clinician and the values and preferences of the patient. It is supposed to be a conversation. That's how you know it has to work because the trials might have been done on slightly different populations. So you're kind of trying to work out okay, does this work for this particular person in this particular situation? Now, of course, whether that really happens in the kind of time-pressured appointment that you've got with the doctor, that's going to vary from place to place. Now, in terms of helping people kind of do their own research, and I at the end of the book, um, I put these evidence hacks. I mean, I found reporting the book to be incredibly empowering myself in terms of helping me kind of understand evidence on a personal level, and I kind of wanted to help people start to be able to do that. Now, I'm not saying it's easy. I mean, one very basic thing that I encourage people to do is to actually ask for the evidence, right? When you encounter claims that you're not sure of that or that sound dubious, and to say, okay, is there evidence behind them? Then to think about whether if if there is a study which supports a particular claim, let's say a claim on social media by some you know influencer, is it good evidence? And and the most basic indicator of that might be is the study peer-reviewed and published in an academic journal? Now, there's plenty of poor quality evidence, which is also gets peer-reviewed and is published nevertheless, but it's a kind of crude heuristic for quality. Another thing that I would suggest, which I do much more myself now, is to look for what's called evidence synthesis. So the thing about looking at individual studies, which many science journalists do and plenty of you know scientists do too, is we're sort of chasing after these individual studies, but individual studies risk being wrong just because of the way the study was done. It could be by chance, it could be that it was a poorly conducted study. So it's much better, and it's standard practice in medicine to do what's called a systematic review, which is a way of synthesising all of the trials together and seeing, okay, what you know what happens when we put pool all this data, like what where's the signal and the noise? And so looking for something like a systematic review might give a better overview of evidence in a particular area if if that's something that you're looking for. Another brief hack I would mention, I mean, there's all kinds of perils, right, in using AI to provide evidence for us. It can hallucinate, it might get it wrong. But there are now some kind of tools, some AI tools, which, for example, things like there's one called illicit, there's one called consensus, they have free versions. And I don't want to be a huge advocate for individual tools, but the difference of those is you can ask them natural language questions and they will only draw on published research rather than just kind of synthesizing any old information off the internet. So that can be quite you know a useful way to get a again an overview of a field or you know to pull out systematic reviews if that's something that you're interested in.

SPEAKER_01 20:41

I think I should actually check that one out. I didn't I haven't not heard about it before. Talking about looking at evidence. So do you think that science journalism is kind of becoming a bit more difficult in an age where there's misinformation, polarization, and social media?

SPEAKER_00 21:00

Yes, for all of those reasons. Not just science journalism, but communicating science in general, you know, by scientists is of of course becoming more difficult. I mean, I'm actually reporting a piece at the moment about trust in science, and you know, most sort of surveys show in a crude sense that actually trust in science is quite high. Scientists tend to be a really trusted profession in polls compared to, let's say, journalists like me or politicians. But that doesn't mean there aren't problems, and and there is a polarization. Some polls suggest quite a sort of political polarisation around trust in science in the US in particular, where Republicans or Republican leaning groups have shown a drop in trust in science over the last kind of five to ten years, whereas Democrats' trust in science has made a stayed quite high. So there's there's kind of that going on, and some people feel like, you know, politicians at the moment in the US are kind of using this, I mean, not just fueling it, but kind of using it to say, okay, well, we're going to attack these scientific institutions because they're not trusting. The other kind of big problem is around kind of scientists losing influence online because, of course, now there's there's just many, many more voices coming from from influencers and from other places. It becomes much harder to know what to trust or what's valid, right? Experts I talk to feel very concerned that valid evidence-based information is losing influence, is failing to stand out now within this much more kind of complicated, fractured kind of media ecosystem that we've got.

SPEAKER_01 22:34

I suppose teaching critical thinking whilst someone is young, so maybe teaching it in schools. Do you talk about, for example, how to not only thinking about if something's been peer-reviewed that someone's talking about or or it's sort of there's been a trial, do you ever kind of talk to people about how you can recognise language that shows that maybe something on social media isn't quite what it seems?

SPEAKER_00 23:05

Yeah, I mean, that's such a good question. I'm actually reporting a piece for nature at the moment. Well, I'm not sure what it's going to turn into, which is sort of around how do you teach critical thinking in an age of AI? Because I do agree that I think that that's part of part of the solution is arming young people with the ability to kind of spot misinformation or or to understand a claim. I write about a specific effort to teach critical thinking when it comes to claims in in the book, where the the researchers developed um sort of concepts which children need to understand to know, for example, it was specifically around health claims, like if a claim is like, oh, this magic pill, you know, is going to make you younger, or then how can young people approach that and know whether it's reliable or not? And and they came up with sort of concepts, for example, you know, in order to know whether something works or has had an effect, to have been tested on a group and compared to a group that it isn't taking that supplement or pill, right? So there has to be a control group. And and those types of sort of basic concepts, they they found a way to kind of teach these and actually, and then they went off, which of course is fantastic, and tested whether this teaching approach worked in a in a randomized trial, which it it appeared to. So there are some really, I think, exciting ways which scientists are beginning to tackle this problem. I I don't think teaching concepts in school is the answer alone, because of course people forget things, but I think it has to be part of it.

SPEAKER_01 24:29

So were there any chapters that were taking difficult to write, or is there anything you you left out that you weren't sure whether you should include? The first draft of the book was almost double the length of the final book.

SPEAKER_00 24:40

So, in a way, I think I think a friend who read it said it was sort of like a an evidence synthesis in itself of just this kind of huge collection of information. Every chapter is hard. It's really, really hard to write a book, or at least I found it hard. I was trying to be quite ambitious. It went into fields I really didn't know anything about. I mean, I didn't know much about evidence-based policing or evidence-based education before I went in. So each of them was a kind of you know exploration for me that involved a lot of research. And in terms of what what eventually got cut out on the advice of my editor, whether it was the right thing to do, I don't know. I've just did a massive edit on it myself and really try to get it down to these stories of people and use the stories of the people to kind of help explain these concepts and how the evidence revolution had unfolded in these different fields. But I I did take out quite a lot of maybe the problems around evidence. I mean, I include a lot of the problems around evidence, but I would say there were two chapters which were really kind of getting at that a bit more. So more kind of details of, I guess, you know, like the reproducibility crisis in science and how a lot of science ends up being difficult to reproduce. And then something I found is I've been going out to give talks about the book, at least if I'm speaking to people who already really invested in evidence, is a lot of people want to know how to get evidence used. This is a really common problem across fields now, is that there's a lot of them have like built up loads of research, and it's like, okay, but how do we get teachers to use the evidence showing what works in schools, or police to show what really works to cut crime, or policymakers to use the evidence that exists on what policies are most effective. And that's a really common problem, which I'm not saying there are simple answers to and I didn't solve in in the bits I cut out of the book, but I did kind of cut quite a lot of that because the problem, potentially in policy or in academia. The reason we maybe we cut it was because it it it feels quite in the weeds for maybe the more general reader.

SPEAKER_01 26:33

What would you recommend for someone who wants to go a bit more beyond the general reading? And what kind of steps would you recommend to people in terms of thinking about evidence? I know there's a campaign fairly recently called Ask for Evidence, and that maybe goes some way towards kind of making the public think about how evidence is important. If you want to talk a bit about that campaign or any other kind of campaigns.

SPEAKER_00 27:01

The Ask for Evidence campaign, in a way, is where I got inspiration from for one of the evidence hacks at the at the end. The Ask for Evidence was a campaign which was launched by a group called Sense About Science in the UK, which is a really huge campaigner, a successful campaigner, to kind of empower people to be able to find evidence from science when they have important decisions to make or they feel that they need it. And it's exactly what I had sort of said before around, you know, just encouraging people to be skeptical towards claims and say, okay, is there any evidence behind this? And oh, in terms of finding sort of extra reading, well, I mean, there's just this, I mean, in the book itself, there's like a huge list of references which point towards some of these, obviously, a lot of the references which I used, but there are a lot of just really important organisations, such as the Cochrane Collaboration, sort of become famous for producing gold standard syntheses of evidence. And there are other groups like that which act almost as like sort of evidence banks, I suppose, such as the Campbell Collaboration. I mean, there's one called 3IE in policy and international development. So I think once people sort of start delving in, then they inevitably can find those types of groups if they look.

SPEAKER_01 28:09

How should decision makers act when evidence is incomplete? Doing nothing also has consequences?

SPEAKER_00 28:17

Oh, well, yeah, I mean, this is a really common situation of, I mean, often evidence is incomplete, and certainly in the case of policy, it it's I mean, that's almost the the norm, I suppose. I think most of the evidence champions in the book, and I would probably say that nevertheless it is worth looking at the evidence, right? I mean, very rarely does evidence provide these perfect answers. Um, you know, often it is flawed or incomplete or uncertain, like we saw within COVID with COVID, right? I mean, people didn't know much about the virus at the beginning, and yet decisions need to be made. I think the answer in that situation is to try and synthesize the best evidence that we have. And often I think it's not, you know, not letting perfect be the enemy of the good. So there may be very useful evidence that can be found. And and one brief story which kind of exemplifies this, which is this like fantastic evidence synthesis that I wrote about in the book, was around a kind of high-stakes conservation decision of whether to implement a captive breeding program for this endangered caribou population in Dasper National Park in Canada. And the question was, will this program work? Right? You've got one shot at doing this program. And the fact is, you you there is no simple answer to that from the evidence because it's a unique ecosystem. Nobody's tested captive breeding on caribou before in a careful trial. But they did this really amazing approach, which was like, let's break this question of will this work into some assumptions and look for the evidence behind each assumption. And so they did this huge evidence gathering exercise. For example, have all the other threats to the caribou been mitigated, such as wolves. Will the caribou survive being captured? So that was something that could sort of be answered for the evidence. So they broke it down into assumptions, that gathered all the evidence, got all these groups as well, like indigenous communities, reindeer experts from Finland, you know, all these different groups who who know something about caribou to come together and look at the evidence as well and discuss it, and then reach this conclusion on the basis of that, that actually, yes, most of the assumptions seemed to be true, and therefore it was worth proceeding, and therefore they, you know, the the national park did proceed. And that was just a really kind of unusual and rigorous way of gathering evidence and kind of it would it's it's a form of sort of decision making in which you you know you you break it into assumptions and look at each one. But I thought that was really kind of impressive and also quite unusual, and a way of showing how you can take a bunch of incomplete evidence and use it in the best possible way to answer a really difficult question.

SPEAKER_01 30:49

I do wonder about advice for science writers in terms of how they can present stuff and think about stories to the public and their use of evidence to actually improve science communication.

SPEAKER_00 31:04

Yeah, I mean, I you know, that's in a way exactly what most science communicators are trying to do, right? Is to convey evidence uh in an accurate way and in a compelling way, because you you need your audience to be engaged as well as to be informed accurately. So that's that's at the heart of what we're all trying to do. I mean, I I can say one way that I've changed how I approach journalism as a result of writing the book, which is I I feel like before myself, and and also I see this in a lot of science journalism and communication, is it's sort of chasing after the single study. You know, studies come out, let's write, you know, quick news stories saying, oh, we found this result. And and that's why you get you know these kind of flip-flopping ideas of you know, red wine is good for you one week and then next week it's bad. And and and really what we ought to be doing is is trying to communicate or put findings in context and look at the body of evidence as a whole. And that that's now, I mean, I'm in a fortunate position where I get to do that because I often write features which try and look at kind of complicated or contentious bodies of evidence. It but in in very practical terms, it means I'm much more likely now, when I'm going into a new field, to sort of be looking for systematic reviews and and thorough overviews of the evidence rather than saying this study shows X and this study shows Y.

SPEAKER_01 32:23

And do you think that people kind of flip-flopping does kind of impact people's trust in science and then also don't really grasp the scientific process and maybe lose a certain amount of faith in science?

SPEAKER_00 32:38

Yeah, yes, definitely. Yeah, because if you do see things pointing in different directions, like on red wine or latest nutrition advice is, then it you you do become sceptical, right? And and questioning, rightly so. So that's why the context is important. We have to remember that you know, you this is also I mean, even if you've got high-quality science journalism coming out, which I'm trying to do, then then it's taking place in this now really complicated media ecosystem where you're surrounded by influencers or companies which are trying to promote their message, and there's all these sort of commercial influences as well behind the scenes, which are potentially sort of working against people who are trying to put out accurate information. So that's that's a kind of huge challenge.

SPEAKER_01 33:25

I I do think quite a bit of the actual problem is the fact that people are trying to sell things, they have to put out these things that sound very they sound sort of very certain, quite dramatic. But obviously, in terms of social media, you get you do better if you you work with the algorithm, which basically means you need to present things in a way that is like attention-grabbing.

SPEAKER_00 33:53

Yeah, I mean I absolutely I think it's really important. I mean, I I spoke last week with with a health science journalist called Deb Cohen, who's written a brilliant book called Bad Influence about all of the kind of commercial forces which are working behind the scenes and it which are you know changing what messages are put out and off influencers, I mean, almost always have got some type of commercial interest going on. And from my reporting on trust in science, you know, that the concern is very much around what you said, which is the algorithms on social media reward posts which evoke outrage, right, and emotion and have really strong messages to them. They don't necessarily reward posts which give this sort of steady overview of a bunch of dry facts and data. Scientists have got that working against them, and I don't think anyone has got an answer to that problem at the moment. I mean, we can use stories, of course, we should be using stories and telling those stories where data and and science have improved the situation. But that there's there's a there's a tension there, right, between wanting to give a a sort of thorough overview and a balanced view of what the evidence says versus how you get that message to spread with an algorithm that that rewards, as I said, outrage and emotion and extreme views.

SPEAKER_01 35:10

Thank you for listening to this episode of the Science Intersection, and thank you so much to Helen Pearson for such a thoughtful conversation. We talked about evidence-based medicine, why stories matter in science writing, how evidence travels into policy, what happens when evidence is incomplete, and how all of us can get better at asking whether a claim is actually supported by good evidence. Helen's book is Beyond Belief, How Evidence Shows What Really Works, and I'll include links to the book and Helen's work in the episode notes. If you found this episode interesting, please follow or subscribe to Science Intersection, leave a review if you can, and share it with someone who might enjoy a conversation about evidence, science, policy, and how we decide what really works. Thanks again for listening.