Public science communication and the PURE diet study
The Vitality ObeCity Index 2017 (pdf), released in October, “analysed data from Vitality members living in six cities across South Africa” in order to present an overview of South African eating trends and health statistics. The report describes some positive trends, while highlighting that “we still have a lot to do to combat obesity”.
I contributed three opinion pieces on science communication, the food industry and marketing to children, and the importance of consumers making informed choices. Here’s the first of the three, with the others to follow in the coming days.
We are bombarded with claims made in the name of science. Headlines tell us — often in language of absolute certainty — what is safe or unsafe to eat. The onslaught is even worse on the internet and on social media, where hyperbole seems to have completely replaced considered evaluation of complex issues.
The problem is that we have to take our guidance from somewhere, given that few of us have the knowledge to make fully-informed choices about what to eat. So when having to make choices under these conditions of uncertainty, many of us might simply choose to eat what we always have, or what our parents fed us.
Others will be persuaded by a headline, by a charismatic and persuasive health practitioner, or perhaps even by their personal experiences of what makes them feel healthier. None of these can guarantee that you’re making the best choices. More importantly, there are no guarantees that scientific knowledge won’t later emerge that tells you you’re making a mistake.
This is because science doesn’t work like that. It doesn’t advance one press release at a time, or change based on subjective experiences. There are also no guarantees.
This doesn’t mean that a well-justified consensus is impossible. But it is exactly because of this uncertainty that we should commit ourselves to the idea that evidence-based nutrition is our best strategy for having longer and healthier lives.
This is true regardless of the fact that scientific consensus changes, because it is far more rare that the scientific consensus is dangerously wrong than it is for an extremely contrary view to be correct (even though they are of course sometimes proven to be so).
So what should we do, and where should we get our information from? One thing we should do is remember that much of science involves a slow triangulation on the truth, and that it usually avoids the language of certainty. So, the more certain and the more hyperbolic the claim being made is, the more sceptical of it you should be.
We should also keep in mind that we live in filter-bubbles, often getting our information from like-minded sources. This can create a false impression of the credibility of views that are not evidence-based. For example, celebrities like Gwyneth Paltrow can spawn a large enough community, and enough content, that an unwary reader could get the impression that their advice is mainstream, even though it might be completely contrary to the scientific evidence.
Then, there is our confirmation bias: the tendency to over-emphasise information that confirms something we already believe, while downplaying the importance of contrary evidence. By way of example, even though there is no scientific evidence that sugar causes hyperactivity in children, every children’s party “confirms” the myth that is does (whereas in fact, groups of children at festive occasions should always be expected to be hyperactive, whether you feed them sweets or sashimi).
Recent headlines discussing the 18-country PURE study of diet demonstrate the contrast between what science says and what headlines say, and how we can so easily be misled via our filter bubbles and biases.
Two headlines told us the following: “Low-fat diet could kill you, major study shows” (The Telegraph); and “The Low-Fat vs. Low-Carb Diet Debate Has a New Answer” (TIME). Unfortunately, the PURE study itself doesn’t justify those claims at all. But a tiny fraction of people will read the study, and will instead rely on the summaries from an ever-shrinking pool of competent science journalists.
The headlines are misleading first because PURE was an observational study, where we see how a population responds to an intervention without ever having full control over that intervention. So, while it might provide clues about cause and effect, it can’t prove anything. The data were also obtained through self-reporting, and people aren’t good at giving accurate information regarding what they eat over time.
Second, the effects of a low-carbohydrate diet that PURE showed were far more modest that the headlines would suggest. Dr. Andrew Mente (McMaster University), one of the authors of PURE, said they amount to “effects in the neighborhood of a 20% reduction in relative risk. So if the annual [absolute] risk of mortality is 1%, it would be reduced to 0.8%. At the individual level it is tiny.”
These examples of newspaper coverage of PURE demonstrate how many of us can emerge with a false impression of what the science says, unless we are very careful to choose our sources wisely.
Besides learning how to identify warning signs of misrepresentation, we can also remind ourselves that subject experts have a place, because they are informed and competent to assess far more of the evidence, more competently, than we are. Listen to them, rather than to enthusiastic wellness bloggers or lifestyle gurus.
Originally published at Synapses.