Want to keep learning?

This content is taken from the Cardiff University's online course, Making Sense of Health Evidence: The Informed Consumer . Join the course to learn more.

Skip to 0 minutes and 8 seconds [Fiona Morgan] Hello, and welcome to the final week of the Informed Health Consumer and our final discussion, which this week is looking at the topic of risk in health research. Once again, thank you to everyone who’s added new evidence-based resources to Flipboard. We’re very grateful for the resources and your comments on them. Joining me this week are my co-educators Andy Williams, Mark Kelson, and Keren Williamson. So this week we’ve been talking about risk and how that risk is presented in the media. Andy, I think, again, going back to the media, this is very much your area of expertise. Yeah. Risk is what we call a “key news value,” in the coverage of science and health and research, to be honest.

Skip to 1 minute and 7 seconds And what that means is that the research that we read most about in the news media, that we hear most about on broadcast news, tends to be the unusual stuff– the stuff which presents an unusually high risk or an unusually gruesome risk or something which isn’t run of the mill. And this kind of penchant for the risky, for the dangerous, in the news can really give people the wrong impression about how dangerous the world that they live in actually is. That’s the big problem with risk in the media, generally speaking, from the research base in my field. Yeah. Mark, you were talking about individual versus population risk. Yeah.

Skip to 1 minute and 48 seconds I mean, I think it’s fair to say that probably all of the low-hanging fruit has been plucked, in terms of medical research. So the real smoking-gun causes have been identified, and we’re slowly chipping away at the smaller risks now. And a lot of time what an epidemiologist would consider a success or a large risk for an individual might translate into a very small difference, indeed. So I think one of the key things is to interrogate any news report that you see, or a research study, and not just looking for relative risks but for looking for absolute risks. So you know exactly what sort of change an intervention might make for you.

Skip to 2 minutes and 29 seconds I think it’s important that we recognise that we do analyse in groups and then try to apply that to the individual. And a lot can get lost in the translation. So it’s of little consolation to somebody who has a very low risk of developing cancer if they go ahead and develop that cancer. Their risk doesn’t mean anything to them, once they have that illness. So I think that’s a very important distinction to make. That’s really relevant, to read in the news-media reports about risk, as well. Because journalists, because they’re looking for the riskier elements of any story, often report relative rather than absolute risk, for example.

Skip to 3 minutes and 10 seconds So– just a hypothetical– if a study has found hot cross buns double your risk of getting bowel cancer, for example, that’s a relative risk, isn’t it? Yeah? If you eat hot cross buns, you’ve got double the chance of getting bowel cancer. This is hypothetical, by the way. Hot cross buns are, as far as I know, safe. Other types of buns are available. [LAUGHTER] But without reporting the absolute risk– without reporting what chance there was in the first place, that people who don’t eat hot cross buns in the first place– then that is meaningless. It might be that there was only a 0.01% chance, in the first instance, of anybody getting that particular cancer.

Skip to 3 minutes and 53 seconds So eating hot cross buns which doubles that risk is such a marginal increase in risk that it makes no difference. But the headline will say “Hot Cross Buns Double Risk of Cancer.” Yeah? Is that the way– is that the kind of thing you’re talking about? Absolutely. I mean, you can double your risk of winning the lottery by playing twice. But that doesn’t actually mean you’re any more likely– in theory, really– to win the lottery. You’ve got a two in 500 billion chance of winning the lottery. So the base level of risk is really crucial, and that often does get missed. Because we like simple summaries. We like a simple story, and that’s very appealing to people.

Skip to 4 minutes and 27 seconds And when it comes to cancer, sometimes you have celebrity influence to raise awareness of risk. And talking about Angelina Jolie and her very brave declaration of what her risk of breast cancer was and what her quite profound decisions she took to address that risk were, I think that raises awareness within the population. It can also raise worry and concern. Because, as you say, it’s not always the true element of that risk that people take on board. And there could be– when you’re talking about Angelina Jolie’s case, for example, I mean, it has to be proven that you have that particular mutation– a gene that would cause you to increase your risk– if it’s in your family.

Skip to 5 minutes and 14 seconds Then it’s raising awareness for people to go and possibly explore that. But then you’ve got the consequences of living with the risk, as well. So there are very different ways of looking at how people will take that risk on board and deal with that. With other– sorry. Sorry, but at least there’s a kind of positive outcome from that case study. But it doesn’t mean that we should trust what celebrities say about science in the news media, more generally, because sometimes they make completely cuckoo comments which aren’t based in evidence and which can leave people with the wrong impression about things like vaccinations, for example. Lots of celebs speak out about that. It’s just an opinion, isn’t it? Yeah.

Skip to 5 minutes and 50 seconds You also have the issue that comes– that we explore in this, which is that the kind of research that establishes risk is called “observational” research. It’s by looking at groups of people. And what you can’t prove from that is causation. You can say it’s “likely,” it’s “very possible.” And “the overwhelming balance of evidence,” I think, as in the case of smoking and lung cancer. What you can’t do is talk about something causing cancer, in this context. Essentially, observational research identifies associations and correlations. Simply put, something could be doing this. But there could be other things that interfere with it. We talked about confounders and other explanations for why something happened.

Skip to 7 minutes and 1 second And you need to have a lot of research if you’re going to try and establish a causal link. Yeah. Well, one of the big– the key– research studies that took decades to convince everybody was the research undertaken by Sir Richard Doll, when he was looking at doctors. I mean, initially, he became quite frustrated because him and his colleague, Hill– they did a study which showed that there was probably a link between the raising incidence of cancer– lung cancer– in smokers, postwar. And people still didn’t really believe it, at that time. The tobacco industry was very powerful. So he wanted to know how he could really reinforce that message.

Skip to 7 minutes and 47 seconds So he chose to get huge numbers and follow them over a long time to try to look at a prospective– look to the future– to see if he could prove his point. And over the 50 years that his study actually ran, I don’t think anybody can doubt that there is now a causal link between smoking and lung cancer. But you also have spurious correlations, which I think we’ve identified. And there are wonderful ones like the number of Nicolas Cage movies, compared with suicides or death. I can’t remember exactly, but we do point you towards those during this week’s materials. And they’re quite fun, those.

Skip to 8 minutes and 34 seconds The tobacco industry clearly had a massive influence on the amount of time it took for the link between smoking and lung cancer to be accepted. We know that they spent vast amounts of money. And that’s happening now. And you think, well, nobody’s going to be arguing this after all this time. But it’s still happening. They did it every step of the way. Every time there was a concerted effort by policymakers and public-health experts to regulate tobacco in the public interest, there was an equally– an equal or, if not greater, push by the tobacco industry, usually in the form of communications and public-relations kind of tactics, to trash the science– to junk the science.

Skip to 9 minutes and 24 seconds And oftentimes they didn’t do it by funding other science, by taking the scientists on on their own terrain by publishing rival stuff in journals and by engaging in open and honest scientific debate. Very often, they used front groups and astroturf groups like FOREST and many, many others to trash the scientists’ reputations, to play the man not the ball, to frame it more in terms of rights and personal freedoms rather than health risks. And each step was a delaying tactic, and each step delayed action to effectively regulate this in public-health terms. We’ve talked quite a lot about news representations of risk.

Skip to 10 minutes and 9 seconds And we have a little segment where we look at how a particular paper published in 2011 suggested there might be a possible link somewhere between mobile phones and brain tumours. Now this was presented in some papers as a possible association. But by the time it got to Britain’s number one tabloid newspaper, it was a front-page story saying “Mobile Phones Cause Brain Cancer.” Yeah. And, I mean, the reason that they go for this is because it’s an unusual risk. Because it’s something that’s not necessarily been thought of by many people. And also it’s sometimes an opportunity for news organisations to campaign on something. That’s a really, really valuable thing for news organisations to do. It brings them money.

Skip to 11 minutes and 5 seconds It brings them readers. If you’ve got an ongoing campaign, especially where you can prove at the end that you’ve made some kind of difference, then it’s bankable, in commercial terms. So you’ve got commercial realities of the news industry butting up, yet again, against the clear and accurate communication of evidence to populations about health risk. And a quantification of it would be useful. So if you were lying on the floor, bleeding to death, should you not pick up your mobile phone because it’ll give you cancer to ring an ambulance? Then no. You know? I mean, I think you need to know what is the risk of on the phone, compared to what you’d be missing out on.

Skip to 11 minutes and 45 seconds So it’s, again, a heads-up to anyone who might be taking this course to not just look at one source of information about these things. If it is a new story, look at other media stories about this. Look outside the mainstream media, to the experts in the blogosphere. Go to the source. Try and train yourself to be able to read the abstracts of papers, to look at papers. Go to the myth-busting places that we’ve been talking about. It’s just– yeah, relying on one source like that can be really, you know, quite the thing to do. And in the UK we have several really good myth-busting sources which are available openly, freely available, to anyone, anywhere in the world.

Skip to 12 minutes and 29 seconds If it’s cancer-related stories you’re looking at and cancer risk, there’s the Cancer Research Science Blog. There’s also the NHS Choices behind the Headlines service, which we mentioned on several occasions. Both of these provide really good, well-balanced summaries that are very readable. That’s on the science. And if you’re looking for a bit of information about a group you see quoted in the news, like FOREST or like any of these front groups or astroturf groups, there are sites like Powerbase or like SourceWatch, in the US, that you can go to, which I’ve highlighted in some of my resources on the course, so far. And you can just type these groups in.

Skip to 13 minutes and 17 seconds And, more often than not, they will have been a university researcher or an investigative journalist or a member of the public who has, in a rigorous and evidence-based way, checked out and followed the money– to come back to something we were talking about a while ago– to see if there are any undisclosed conflicts of interest here. To see if this group’s funded by the cigarette industry, for example. And then, once you’ve got that information, you can make a much more informed decision about whether to trust what these people are saying. That’s great. Thank you very much.

Discussion: risk in health research

Share this video:

This video is from the free online course:

Making Sense of Health Evidence: The Informed Consumer

Cardiff University