Skip to 0 minutes and 8 seconds (Fiona Morgan) Pharmaceutical industry sponsored trials tend to show more positive results than trials that are not sponsored by the industry. What happens is that the more negative results don’t get published. And so you get an overly positive idea about how effective a particular treatment is. (Dr Andy Williams) Because essentially, we’re only getting part of the picture in the published research. (Fiona Morgan) Exactly. So the idea of the AllTrials campaign, which was setup by Ben Goldacre, the campaigning doctor and journalist. Fiona Godlee, who’s editor of the British Medical Journal. And what they are arguing, and arguing very successfully is that if people conduct clinical trials they should publish the results, and make the results of those trials available.
Skip to 1 minute and 16 seconds It’s not actually ethical to do anything else. (Dr Andy Williams) Has the pharmaceutical industry followed suit? (Fiona Morgan) Some companies have. I believe my former employer, GlaxoSmithKline, again declaration of conflict of interest, has signed up and said, it will make all its trial results available. But I think it’s quite a slow process. (Dr Andy Williams) And the assumption is that they’d be less likely to want to sign up because it might hit their bottom line because it might reduce the perceived effectiveness of their drugs. Yeah. OK. (Keren Williamson) And is it going to be difficult for people to read through all of that, to come to a conclusion? How will people be able to understand all of that literature?
Skip to 1 minute and 59 seconds (Fiona Morgan) Particularly when it’s published in trial reports.
Skip to 2 minutes and 5 seconds That’s why it’s quite helpful to look for guidance on the topic which is based on reviews of the evidence. This is an area I work in where I produce systematic reviews, rigorous reviews of the available evidence. So it’s looking at those kind of summaries of research, and the guidance that flows from it to make those decisions. But it has to be based on all the evidence, and not just the published evidence. (Dr Andy Williams) So a big take home point for people who might be watching this now is, if you’re looking at a report based around the evidence, around any particular drug to make sure that that is a systematic review.
Skip to 2 minutes and 52 seconds Or a review of all the available evidence, and not just the positive findings which are released by the pharmaceutical industry. Or which are perhaps more likely to be published in scientific journals. (Fiona Morgan) Yes. Absolutely. Scientific journals in the same way as the news media like good news, like positive, important stories. They like the stories that say, this does something. The fact that something doesn’t work is much less interesting, and it tends to be less published. (Dr Andy Williams) Isn’t that interesting? I’ve been talking about the way that the news values of the journalists might impact on the accurate presentation of the evidence.
Skip to 3 minutes and 38 seconds But you’re suggesting that the, the scientific journals where we as academics publish our work also have a set of news values. And that might skew the evidence that’s available to people as well. (Keren Williamson) You rarely put your work in for publication if it’s come out that it doesn’t really make any difference because people are not willing to read that. They knew that already. (Fiona Morgan) But that’s important to know. (Keren Williamson) But it’s very important when you’re making your decision, particularly with health. It’s very important to hear fully both sides, if it doesn’t make a difference, or if it does make a difference, or if it isn’t any better, then you really need to know this.
Skip to 4 minutes and 14 seconds (Dr Andy Williams) So what AllTrials is going to do is going to help people like you who do these reviews to make the reviews more comprehensive, to make the reviews more accurate, to make sure they’re less partial than they have been up to this point. (Fiona Morgan) Absolutely. Yeah.
Discussion: publication bias
In the second part of our weekly discussion, the team look at the thorny issue of bias in the reporting of health research.
Does this make you more sceptical?