Picture credit: Oscar Wong/Getty
Artillery Row

The meaningless models of “public health”

Another brick in the “public health” fortress of unreality

When the government introduced mandatory calorie labelling on menus in 2022, it exempted businesses that employ fewer than 250 people. The original plan had been to cover the entire out-of-home sector, but it was soon pointed out that calculating and publishing exact calorie counts for every meal would be a nightmare for pubs and restaurants that change their menus regularly and do not have standard servings. Reason prevailed and the law was only applied to large chains.

This exemption has always stuck in the craw of “public health” academics who are not interested in making businesses viable nor in the practical implementation of public policy. Although there is no real evidence that calorie labelling has any effect on obesity, they hate to see their ideas watered down. 

The fightback began this week with a study in Lancet Public Health which purports to be an evaluation of the law as it exists. Its headline claim is that it will prevent 730 deaths from heart disease, but that it would prevent even more if it were applied to every cafe, pub and restaurant in the land. The authors stress the latter point no fewer than ten times in the study and talk of little else in the accompanying press release

The authors claim that their study is the “first evaluation of the likely impact of the menu calorie labelling regulation” in England, but it is not an evaluation at all. It contains no information about the real world impact of calorie labelling since it was introduced. No data on food sales or calorie consumption have been collected. The study simply models what the authors reckon might have happened based on experiments conducted long before the law was changed.

The authors assume that menu labelling makes people consume 47 fewer calories per restaurant meal and make various calculations based on that. Have people in England been eating 47 fewer calories per restaurant meal since calorie labelling was introduced? We don’t know because nobody has tried to find out. The figure of 47 calories comes from a meta-analysis of three studies mentioned in a Cochrane Review in 2018. The authors of that review said that the “quality of the evidence for these three studies was rated as low”. In their conclusion they “tentatively suggest that nutritional labelling on menus in restaurants could be used as part of a wider set of measures to tackle obesity” although they cautioned that this was based on “a small body of low-quality evidence”.

They also concluded that “high-quality research in real-world settings is needed to enable more certain conclusions”. You might think that a country with a population of 60 million implementing mandatory calorie labelling offers an opportunity for such research, but apparently not. Instead we have the umpteenth modelling study based on “low-quality evidence” masquerading as an evaluation.

As for the 730 deaths from heart disease that calorie labelling will hypothetically prevent, that would be a small number if it was per year but it is actually over 20 years. The authors’ model runs from 2022 to 2041 and they estimate that there will be between 600,000 and 1,200,000 deaths from heart disease over that period. Their best guess is that there will be 830,000 such deaths and that calorie labelling will reduce that to a mere 829,270. Commenting on the study, Professor David Spiegelhalter made the understatement of the week when he said: “These effects would not be measurable from future data.”

Everything about this study is objectively worthless. Its only purpose is to establish the idea in the academic literature that a policy that has never been evaluated is working well and would work even better if politicians doubled down on it. It is another brick in the fortress of unreality that “public health” academics build around themselves to justify policies that have patently not made any tangible difference to the health problems they were designed to tackle. 

Modelling can be of academic interest by illustrating how an untested policy might work in practice, but it is increasingly being used by “public health” academics to create a world of fantasy into which they can retreat when the policies they spent years campaigning for turn out to be damp squibs. 

I don’t know whether calorie labelling works and I don’t have strong feelings about it as a policy — although people with eating disorders hate it — but I do know that child obesity has risen since the sugar tax was introduced and alcohol-related deaths have hit a 14 year high in Scotland since minimum pricing was introduced. And yet on the spreadsheets of “public health” academics, minimum pricing led to a 13 per cent reduction in alcohol-related deaths and the sugar tax stopped 5,000 girls from becoming obese. In London, which has the highest rate of child obesity in the country, a ban on “junk food” advertising on public transport led to households consuming 1,000 fewer calories a week, according to another ludicrous model. All of this is based on theoretical counterfactuals that are not remotely credible and yet are unfalsifiable by their nature. If challenged, the authors will claim plausible deniability by saying that they acknowledge uncertainty in the text of their study and are only running simulations, but nobody reads the studies and these fairy-tale figures soon become accepted fact. 

Before you know it, you have Sadiq Khan asserting that his advertising ban has prevented 100,000 cases of obesity and Nicola Sturgeon claiming that minimum pricing “is saving lives”. Today the Telegraph has tried to throw a bit of shade on the calorie labelling study with the headline “Calories on menus will prevent only 37 deaths a year, study finds” but even this accepts a false premise. There is no evidence that calorie labelling will prevent, or ever has prevented, a single death. It is all a self-serving figment of activist-academics’ imagination.

Enjoying The Critic online? It's even better in print

Try five issues of Britain’s newest magazine for £10

Critic magazine cover