Remember Devi Sridhar? She is that American public health academic based in Scotland who was always on TV during the pandemic explaining that if she were in charge she would simply eliminate Covid-19. It’s probably very unfair but I always imagined her as a sort of Rasputin figure in the court of Nicola Sturgeon whispering to the leader about Zero Covid. She later backed away from the Zero Covid strategy, such as it was, and wrote an article titled “Don’t blame scientists for what went wrong with Covid — ministers were the ones calling the shots” which some people might think was self-serving.

In her latest article, Devi is in the mood for some debunking, but what she actually does is highlight the awesome power of confirmation bias. The target of her scepticism is a report showing that the UK and Ireland have a larger proportion of the population consuming five or more portions of fruit and vegetables a day than any other OECD country. A third of British and Irish adults say they get their five-a-day whereas the OECD average is just 15 per cent. This has displeased Devi.

At the dawn of the new year, newspaper headlines reported that the UK and Ireland were ranked “the world’s best at eating fruit and vegetables”. And this seemed surprising to me, as a public health expert. Just on economic grounds, the UK imports 50% of its vegetables, and 84% of its fruit – largely from Europe, Africa, and the Americas.

I am at a loss to understand why Devi thinks this is relevant information, let alone why she thinks it supports her argument. There is more demand for fruit and vegetables than British farmers can meet and so we have to import lots of them. Doesn’t that imply high levels of consumption?

These foods are relatively expensive in the UK, costing on average £11.79 per 1,000 kcal, versus £5.82 for 1,000kcal of processed foods. 

This is a ridiculous way to compare the cost of food. Vegetables tend to be low in calories and that is one reason why they are viewed as healthy. When preparing a meal, you don’t consciously attempt to reach a calorie threshold. You put something together that is going to fill you up. You can get a kilogram of carrots for 65p and a kilogram of potatoes for even less. If you want to compare the cost of different diets, you need to look at the cost per 1,000 grams rather than the cost per 1,000 calories. On that measure, the likes of peas, potatoes and bananas are very cheap. 

Devi seems to be suggesting that fruit and vegetables are more expensive relative to other countries but she provides no evidence for this and the reality is that food is unusually cheap in Britain.

“Ask a senior researcher what two parts of a study they read first, and it’s methodology followed by financial support — in short, how did the authors find the answers and how was it paid for?”

This is a telling comment, firstly because I doubt it is true — surely you look at the conclusion first? — and secondly because I get the impression that if the study had been funded by the food industry, Devi would have dismissed it out of hand. As it happens, the study is from the OECD and uses data from the European Union so she has to press on.

Devi’s objection to the study is that it is based on a survey which asks people how many portions of fruit and vegetables they eat each day. She rightly points out that people might forget or lie or misunderstand the question. This is always the risk with self-reported data, but it is not confined to this particular survey. Virtually all nutritional epidemiology is based on asking people what they eat. Devi didn’t have a problem with this last year when she cited survey data to claim that 75 per cent of the calories in school lunches come from “ultra-processed food”, nor when she claimed that ultra-processed food causes cancer.

Devi doesn’t specifically mention social desirability bias or recall bias, but these are legitimate problems with survey-based data and they help explain why nutritional epidemiology is such a train wreck. But you can’t pick and choose your moments to be sceptical. Is there any reason to believe that the British and Irish are more forgetful or more dishonest when answering the question about fruit and vegetables than other nationalities? Perhaps there is, but I can’t think of one and Devi doesn’t offer one.

Instead, she simply says that 26 per cent of UK adults are obese as if this self-evidently disproves the theory that a third of us are eating our greens. She says that rates of overweight and obesity have “increased over the past decade” which, as a matter of fact, they haven’t, and then claims that British children are shrinking. 

Data shows that children in the UK are becoming shorter compared to other countries, and it’s been asserted that the average height of a five-year-old in the UK is likely to have decreased because of rising child poverty and Conservative austerity policies.

This is the meme of the month in public health and it needs to be very carefully worded to give the intended misleading impression without actually lying. Devi gets it right the first time when she says kids are “becoming shorter compared to other countries” but gets it wrong the second time when she claims that average height has decreased. The fact is that British children have never been taller. They have grown by about a centimetre on average since 2010. It is true that they have been overtaken in the international rankings by some other countries, but it seems likely that this is at least partly due to the large number of babies being born to parents of Asian heritage.

Although Devi does not explicitly endorse the claim that children are shrinking due to Conservative policies, she notes that others have “asserted” this. She does not mention that the assertion came from the Labour Party. It seems strange to give the nod to wild speculation from an obviously biased source in an article about the need for empirical rigour, but Devi also insists that “Brexit increased overall food prices by 6%” based on a highly implausible piece of research so she is capable of believing all sorts of things if she wants them to be true.

Maybe the survey is a load of nonsense and everyone is lying. Who knows? Frankly, who cares? The world of “public health” is awash with junk science and ludicrous claims, so why is Devi Sridhar so bothered about this particular statistic? The answer, surely, is that it represents good news. 

A few years ago, some (anti-)alcohol researchers wrote an article titled “The problem with good news: how should public health actors respond when alcohol consumption declines?” The title was revealing. Alcohol consumption fell by 18 per cent in the UK between 2004 and 2016 despite the government failing to implement the “evidence-based” policies favoured by public health academics and despite the licensing laws being relaxed in 2005. This was a problem for “public health actors” not only because it made them look like a waste of space but because it made it difficult to claim that Britain was in the grips of a spiralling epidemic of drunkenness that required more taxes and bans.

A sense of panic is essential to the nanny state project. If the British people understood that they were living in a pretty average country, they would be less interested in the radical surgery proposed by the health zealots. Hence the need to make silly claims about children shrinking and the desperation of people like Devi Sridhar to dismiss any evidence that the UK might not be so bad after all.