Wednesday 26 April 2023

The unintended consequences of restricting food displays

Tom Harwood has written a rather good Substack post about the new weight loss drug semaglutide (brand name: Wegovy). Although I am not as convinced as some people that this is a panacea for obesity, it has been interesting to witness the dismay from moralists who are worried that this will 'let junk food companies off the hook'.  

It will also be fun to see people who bandy around fictitious figures of the cost of obesity to Britain, which rise as high as £47 billion and £58 billion, explain why it is not cost effective for the NHS to spend £10 billion on this drug. 
One forecaster from The Swift Centre explains that, on a back of a napkin calculation, the 20% fattest brits taking up semaglutide could save the UK economy billions.

“the cost would be over £9bn a year just to treat 20% of the population. The NHS says obesity costs the organization £6.1bn a year, but £27bn to wider society. Political silliness about paying a single or small number of drug companies such a large amount of money might also cause hesitancy.”

That seems like a decent investment to make.

It isn't really, but to admit that requires admitting that the costs to the NHS of obesity are small or negative and that the costs to 'wider society' overwhelmingly fall on the people who are fat. Obesity is not a public health issue. It is a personal health and aesthetic issue, it's nobody else's business and, in my view, people who want to take Wegovy should pay for it themselves.
Be that as it may, Tom makes the important point that wowsers and anti-capitalists don't really want to solve the problems they whine about. They want to change society and they want people to change their behaviour. Tom, quite rightly, wants to see Wegovy succeed so the government will stop interfering in the food supply. In doing so, he drew my attention to a story I hadn't spotted before with regards to the ban on displaying 'junk food' in prominent positions in shops.

The hilarious thing about all of this is that despite the government now seeing itself taking on a feng shui interior decoration role when it comes to supermarkets - this grand plan doesn’t appear to be working. Retailers are reporting confectionery sales rising, not falling.

He links to this story... 

Spar retailer Julian Taylor-Green, who owns a 2,500 sq ft store in Stafford, said confectionery sales had gone up by 11.5% in the past year, while crisps and snacks rose 14.1%.

“While inflation will play a part in this, I feel this has happened because we’ve now moved HFSS products that were by the tills and on promotional bays all into one place on the confectionery aisle, which has made that shopping experience easier for the customer to access these products,” he said.

This is consistent with other anecdotal evidence from supermarkets. It shouldn't be too difficult to get empirical proof. Thanks to reward cards, supermarkets have a very good idea of what consumers are buying. If they share their data with researchers, we will be able to see whether this is yet another bone-headed 'public health' policy that has backfired.

If it has, can we please stop listening to ignorant 'public health' academics who do not understand the market they are trying to regulate?

Association Between Daily Alcohol Intake and Risk of All-Cause Mortality - a critique

Last month, I wrote about the latest attempt by Tim Stockwell and colleagues to erase the health benefits of moderate alcohol consumption (Zhao et al. 2023). There is now an expert review that is worth reading from the International Scientific Forum on Alcohol Research.

They discuss Stockwell's track record...

The current paper by Zhao J., Stockwell T., Naimi T., Churchill S., Clay J., Sherk A., (2023) is a follow-up paper to two published previously by this group trying to undermine the well-established J-shaped association between moderate alcohol consumption, cardiovascular disease and all-cause mortality. The first paper written by Filmore et al. (Fillmore et al., 2007) re-iterated the old and already falsified hypothesis that the J-curve would be generated by “sick quitters” or badly categorized moderate drinkers. The “sick quitters” hypothesis was duly discredited by reanalysis of the studies (Klatsky and Udaltsova, 2007). Also repeat studies adjusting for the claimed bias still showed a cardioprotective effect for regular light to moderate alcohol consumption (Di Castelnuovo et al., 2006, Ronksley et al., 2011).

The second paper written by Stockwell et al. (2016) was previously critically reviewed by ISFAR. Forum members noted that the authors were very selective in choosing papers to include in their new analyses: they identified 2,575 studies on the subject, analysed 87, but then found some reason to exclude almost all of these studies to reach a conclusion that “ . . . there was no significant protection of all-cause mortality for low-volume drinkers (RR = 1.04, 95% CI [0.95, 1.15]) ” based on what is apparently only six (6!) remaining studies. Their new analysis markedly distorted the accumulated scientific evidence on alcohol and CVD and mortality. As stated by one Forum member, “The biased selection of studies that are included undermines the value of the paper, but more importantly promulgates misinformation in the name of appropriate scientific method. Failure to acknowledge the robust body of knowledge that supports the opposite conclusion, and disqualification of extensive studies that offer plausible biologic explanation of observed benefits, is unconscionable.”

They also notice something in the supplementary material which undermines Stockwell's conclusion, even if you accept his cherry-picking of studies and the vast number of adjustments he makes to the data.
The unadjusted relative risks showed the classical J-shape. Surprisingly, even the fully adjusted relative risk showed a J-shaped curve as well: a significantly reduced relative risk for all-cause mortality for those drinking a ‘low volume’, namely 1.30-24 g/day. The authors, however, state that in their fully adjusted model “daily low or moderate alcohol intake was not significantly associated with all-cause mortality risk” and provide a risk estimate of 0.93 (CI: 0.85-1.01). This discrepancy between figure and text is not addressed by the authors. It is amazing that this discrepancy has not been noted by the reviewers and editors of JAMA Network Open. One wonders if reviewers and editors have read this paper in detail, a first task of reviewers would be to check whether the data support the conclusions. The editors have missed the fact that data and procedures, which were essential to derive at their conclusions have been transferred to the supplemental materials. This makes it impossible to grasp the authors approach from what was written in the original JAMA paper. 

The graph above comes from Stockwell's supplementary material. It shows overall lower mortality risk at up to 24 grams a day. 24 grams is three units. 
The authors have many criticisms of the meta-analysis, but the summary is this:

The key features that clearly demonstrate bias in the present paper are:

1.       This is the third attempt by the same group of researchers to disprove the beneficial relationship between light to moderate alcohol consumption and cardiovascular health. The first attempt (Fillmore et al, in 2006) was discredited by reanalysis of the studies. Reanalysis showed that after adjusting for the claimed bias the cardioprotective effect for regular light to moderate alcohol consumption was still apparent.

2.       Numerous meta-analyses have been undertaken over the past 15 years that have adjusted for this proposed bias and they have consistently shown that there is a cardioprotective effect for regular light to moderate alcohol consumption.

3.       In this new paper, Zhao et al., (2023) have again biased their meta-analysis by ‘cherry picking’ a small number of studies for their meta-analysis – they discarded 3230 studies and analysed only 107. The 107 studies selected did relate consumption to disease and all-cause mortality but the authors carefully avoided hundreds of validated studies that showed reduced disease risk among light to moderate drinkers.

4.       Countless animal and human studies over the past four decades have provided extensive evidence for the biological mechanisms supporting the findings that light to moderate alcohol consumption is cardioprotective. Zhao et al., (2023) seem to have deliberately pretended that they do not exist.


Monday 24 April 2023

Brown bread is the new tobacco

Cross-posted from my Substack.

It’s nearly 20 years since Morgan Spurlock ate nothing but McDonald’s meals for 30 days. Unsuprisingly, he put on weight. He was also found to suffer from liver dysfunction, although that was more likely due to the fact - undisclosed at the time - that he was an alcoholic.

Spurlock filmed his experience and released it as Super-Size Me in 2004. The take-home message was that fast food makes you fat. A science teacher later lost 56 pounds eating nothing but McDonald’s, thereby proving that a calorie is a calorie, although that experiment received less attention.

Dr Chris van Tulleken recently carried out a pound shop version of Spurlock’s experiment on himself. For 30 days, he ate a diet consisting of 80% ‘ultra-processed food’. Today, he wrote about it for the Daily Mail.

His story begins with him opening a box of Coco Pops in front of his three year old daughter. By a stroke of luck, the child has an uncanny knack of speaking and behaving in a way that perfectly suits his narrative.

As I opened the cereal packet, my three-year-old daughter Lyra said: ‘Is it for me?’ No, I told her — she was having porridge.

‘I want the Mickey Mouse cereal!’ Lyra cried, pointing at Coco Monkey.

The packet had clearly been designed with a three-year-old in mind. Lyra had never had Coco Pops before, but Kellogg’s already had her hooked.

Opening a box of Coco Pops in front of a child who has only ever known porridge and then refusing to give her some before eating it yourself seems rather cruel to me, but apparently it is all Kellogg’s fault.

It is interesting that he chooses Coco Pops since this was one of the brands that capitulated to Public Health England’s sugar reduction scheme. In 2018, it was reformulated with barely half as much sugar as the classic recipe, much to the annoyance of loyal customers.

Kellogg’s must now be thinking what many of us are thinking, that it is pointless trying to appease the ‘public health’ lobby.

Again, I told her no, so she collapsed on the floor crying and screaming with rage. Then, while I was seeing to her porridge, she crawled out from under the table, filled her bowl and started to eat great fistfuls of dry Coco Pops, wide-eyed and ecstatic.

I suppose we just have to take his word for this. I’m sure he wouldn’t embellish this story in any way for the purposes of writing an article for the Daily Mail.

Defeated, I poured some milk into her bowl, then read out the UPF ingredients on the packet: glucose syrup, cocoa mass and flavourings.

And barley. And rice. And Niacin, Iron, Vitamin B6, Riboflavin, Thiamin, Folic Acid, Vitamin D and Vitamin B12.

The cereal also contained 20 per cent more salt per gram, I noted, than a typical microwave lasagne.

Why is lasagne suddenly the benchmark against which all salty food is measured? And why is he measuring it by the gram rather than by the serving? A typical microwave lasagne contains 1.8 grams of salt. A serving of Coco Pops contains 0.2 grams of salt. The recommended daily limit is six grams.

Tucking in myself, I found the first spoonful delicious. The texture of the first chocolatey mouthful is extraordinary, some of the ‘pops’ becoming chewy while others remain crisp.

The implication is that he has never had Coco Pops before. How is this posssible? They were around when I was a kid and he’s younger than me.

Just three spoons in, however, the joy was gone: what remained was a brown sludge. Yet Lyra and I continued to crave our next mouthfuls, just like smokers crave the next drag.

This is one of several references to smoking in the article, just in case the reader doesn’t get the hint that food is the new tobacco.

Her eating wasn’t just mindless: it was trance-like. She carried on until her belly was drum-taut. By the time she stopped, she’d consumed two adult servings: most of her day’s-worth of calories.

How has he gone from refusing to give her any Coco Pops to allowing her to eat as many as she wants?

My own consumption was even worse: instead of the recommended one portion, I’d effortlessly hoovered up five.

Five portions? Greedy guts! The whole family seems to have terrible self-control issues. Still, even after this binge - assuming it actually happened - he had only consumed 575 calories and one gram of salt. Not too bad for breakfast.

Sugar, along with salt, is the additive most likely to drive appetite. That’s why you’ll find both additives, in substantial quantities, in nearly all ultra-processed food.

They’re not really additives, are they? The word he’s looking for is ‘ingredient’. And if you try living without one these ingredients, it won’t be long before you go into a coma and die.

And there’s another property in most UPF that drives weight gain. Texture. Take Coco Pops: they’re branded as crunchy, and a few do stay crunchy for a while.

But, as Lyra and I discovered, the milk and Coco Pops quickly form a kind of textured liquid, and each successive mouthful feels more like a slick of wet, starchy globs.

Little does he know that this is a common complaint about the reformulated version of Coco Pops. Sugar provides texture in a way that cocoa powder - which is what replaced it - doesn’t.

The marketing primes us to register an initial crunch of batter, the pop of a puffed rice crisp, the snap of a reformed crisp — but everything yields to the slightest bite. In short, the foods are cleverly textured, which disguises the fact that, within seconds, we’re eating mush.

For CVT, a breakfast cereal getting softer while you eat it is not a predictable consequence of it being soaked in milk, but a red flag that it is ultra-processed food.

Ultra-processed food (UPF) is the latest bogeyman in diet quackery. The concept was devised a few years ago by the Brazilian academic Carlos Monteiro who also happens to be in favour of draconian and wildly impractical regulation of the food supply. What are the chances?!


Laura Thomas has written some good stuff about UPF. The tldr version is that, aside from raw fruit and veg, the vast majority of what we eat is ‘processed’. That’s what cooking is all about. Ultra-processed food involves flavourings, sweeteners, emulsifiers etc. that you wouldn’t generally use at home, often combined with cooking processes such as hydrogenation and hydrolysation that are unavailable in an ordinary kitchen. In short, most packaged food sold in shops is UPF.

Does this mean a cake you bake at home (‘processed’) is less fattening than a cake you buy from Waitrose (‘ultra-processed’)? Probably not, so what is the point of the distinction? This is where the idea breaks down. All the additives used by the food industry are considered safe by regulators. Just because the layman doesn’t know what a certain emulsifier is doesn’t mean it’s bad for you. There is no scientific basis for classifying a vast range of products as unhealthy just because they are made in factories. Indeed, it is positively anti-scientific insofar as it represents an irrational fear of modernity while placing excessive faith in what is considered ‘natural’. There is also an obvious layer of snobbery to the whole thing.

Taken to an absurd but logical conclusion, you could view wholemeal bread as unhealthy so long as it is made in a factory. When I saw that CVT has a book coming out (of course he does) I was struck by the cover. Surely, I thought, he was not going to have a go at brown bread?

Ultra-Processed People: Why Do We All Eat Stuff That Isn’t Food … and Why Can’t We Stop? by [Chris van Tulleken]

But that is exactly what he does.

During my month-long UPF diet, I began to notice this softness most starkly with bread — the majority of which is ultra-processed. (Real bread, from craft bakeries, makes up just 5 per cent of the market…

His definition of ‘real bread’ is quite revealing, is it not?

For years, I’ve bought Hovis Multigrain Seed Sensations. Here are some of its numerous ingredients: salt, granulated sugar, preservative: E282 calcium propionate, emulsifier: E472e (mono- and diacetyltartaric acid esters of mono- and diglycerides of fatty acids), caramelised sugar, ascorbic acid.

Let’s leave aside the question of why he only recently noticed the softness of fake bread if he’s been eating it for years. Instead, let’s look at the ingredients. Like you, I am not familiar with them all, but a quick search shows that E282 calcium propionate is a ‘naturally occurring organic salt formed by a reaction between calcium hydroxide and propionic acid’. It is a preservative.

E472e is an emulsifier which interacts with the hydrophobic parts of gluten, helping its proteins unfold. It adds texture to the bread.

Ascorbic acid is better known as Vitamin C.

Caramelised sugar is just sugar that’s been heated up and is used sparingly in bread; Jamie Oliver puts more sugar in his homemade bread than Hovis does.

Hovis Multigrain Seed Sensations therefore qualifies as UPF but it is far from obvious why it should be regarded as unhealthy. According to CVT, the problem is that it is too easy to eat.

The various processes and treatment agents in my Hovis loaf mean I can eat a slice even more quickly, gram for gram, than I can put away a UPF burger. The bread disintegrates into a bolus of slime that’s easily manipulated down the throat.

Does it?? I’ve never tried this brand but it doesn’t ring true to me. It’s just bread. Either you toast it or you use it for sandwiches. Are there people out there stuffing slice after slice of bread down their throats because it’s so soft?

By contrast, a slice of Dusty Knuckle Potato Sourdough (£5.99) takes well over a minute to eat, and my jaw gets tired.

Far be it from me to tell anyone how to spend their money but, in my opinion, anyone who spends £6 on a loaf of bread is an idiot. Based on his description, the Dusty Knuckle Potato Sourdough is awful anyway. Is that the idea? Is the plan to make eating so jaw-achingly unenjoyable that we do it less? Is the real objection to UPF simply that it tastes nice?

That seems to be about the size of it. CVT mentions a study by Kevin Hall that has become the stuff of legend in the nutritional community. Published in 2019, it already has over 1,000 citations. It was a randomised controlled trial in which one group of people was given an ultra-processed diet and another group was given an unprocessed diet. Importantly, both diets were similar in their overall sugar, fat, protein and salt content. Equally important, the participants were given all the food for free and they could eat as much as wanted.

The people on the ultra-processed diet ended up eating 500 calories more per day than the other group and, after two weeks, had put on nearly a kilogram of weight.

This has been widely trumpeted as killer evidence that ultra-processed food is uniquely dangerous, but I have never understood why. All it seems to show is that if you give people unlimited quantities of tasty food, they will eat more of it than if you give them vegetables. Salted peanuts, for example, are nicer than plain nuts. Of course people will eat more of them.

You can look at the food provided here. The participants were not given ultra-processed versions of the same meals. They were given totally different meals. Here’s what was offered on Day 6 for lunch, for example.

I would be asking more fries but not for more green beans. So what? I’d be asking for more fries even if they were homemade. Why not offer a homemade version of burger and fries to test the UPF hypothesis? Hall didn’t do this because he wasn’t looking at the difference between processed and ultra-processed food, he was comparing ultra-processed food to unprocessed food. The people on the unprocessed diet actually lost weight, but that’s not too surprising when you look at what they were given.

The answer is obviously to not consume too many calories regardless of what kind of food you eat. And, yes, the temptation to overeat is going to be greater if you have a load of Belgian buns on the sideboard than if you have a bowl of celery there. But we kind of knew this, didn’t we? We don’t need to demonise a vast range of foods to make this point, nor do we need to pretend that there is some magical process taking place in factories that makes ultra-processed food fattening per se.

CVT tries to bypass the personal responsibility angle by claiming that UPF is ‘addictive’. He admits that scientists don’t think that food is addictive but he bypasses that by claiming that UPF isn’t really food.

So how do you get round the apparently established fact that there’s no such thing as food addiction? Well, one way is to accept that UPF isn’t real food; instead it’s an addictive edible substance. In other words, it isn’t real food that’s addictive — it’s UPF.

On top of this being transparent sophistry to further an unsustainable argument is that fact that CVT, I assume, stopped eating these foods and went back to porridge after his 30 days were up. If so, there has to be a question mark over how ‘addictive’ they are.

One of the problems with CVT’s argument is that he is talking about food products that nearly everybody is familiar with. Ultra-processed food makes up 51% of the average British diet. Eating breakfast cereals and sliced bread might be a new experience for him, but the vast majority of his readers have consumed such products for years and won’t recognise the sensations he describes.

Sensations like this…

By the fourth week, it had started to have very noticeable physical effects, forcing me to loosen my belt by two notches. And, as I gained weight, so did my family. It was impossible to stop the kids from eating my Coco Pops, slices of pizza, oven chips, ready-made lasagne and chocolate.

I was now consuming a lot more salt, which meant drinking more water and having to get up in the night to pee a lot.

Unable to sleep, I’d go to the kitchen and have a snack, more out of boredom than anything else. I’d also become very constipated because ultra-processed food is low in fibre and water and high in salt. Constipation led to piles — common in people who eat a lot of UPF.

Does this describe the life of a typical Coco Pops consumer, let alone the typical consumer of brown bread? I suspect it doesn’t. Would I recommend the diet CVT subjected himself to? No. No one would. So what’s the point of all this?

Sunday 23 April 2023

Last Orders with Dominic Frisby

A markedly less libertarian episode of Last Orders this week as Dominic Frisby, Tom Slater and I discuss what should be done to people who disrupt snooker tournaments. Also on the agenda: the government’s plan for state-sponsored vaping and the genius of PG Wodehouse.

Friday 21 April 2023

The WHO's guide to journalism

The incompetent and corrupt World Health Organisation has produced a ‘guide for journalists’ to help hacks report on issues related to alcohol accurately. Not entirely unpredictably, it is a catalogue of anti-drinking tropes, half-truths and brazen lies. The very first words are ‘No amount of alcohol is safe to drink’ and it doesn’t get any better thereafter.

The health benefits of moderate alcohol consumption really stick in the craw of the neo-temperance lobby and so that is where the WHO starts:

Isn’t drinking some alcohol good for your health?

No, there is no evidence for the common belief that drinking alcohol in moderate amounts can help people live longer by decreasing their risk of heart disease, diabetes, stroke or other conditions.

No evidence?! Even a casual follower of the science knows that there is at least some evidence. Those who are more familiar with the literature know that there is a huge amount of evidence built up over decades, tested and re-examined from every angle precisely because so many people in ‘public health’ don’t want to believe it.

It is inaccurate to say that “experts are divided” on whether there is no amount of healthy alcohol drinking. The scientific consensus is that any level of alcohol consumption, regardless of the amount, increases risks to health.

This is just a lie. That is not the consensus, and the only reason there isn’t unanimous agreement that moderate drinking is beneficial to health is that anti-alcohol academics such as Tim Stockwell have made it their life’s work to cast doubt on the evidence.

While several past studies did suggest that moderate consumption could, on average, promote health benefits…

Note that this immediately contradicts the claim that there is no evidence.

… newer research (1) shows that those studies used limited methodologies and that many of them were funded by the alcohol industry (2).

The first reference is a short commentary by some WHO staffers which doesn’t discuss methodologies at all. The second reference is a study which found that only 5.4 per cent of research papers in this area were funded by the alcohol industry and concluded that ‘the association between moderate alcohol consumption and different health outcomes does not seem to be related to funding source.’

The WHO must hold journalists in low esteem if it thinks they won’t check up the citations like this.

The dicussion [sic] about possible so-called protective effects of alcohol diverts attention from the bigger picture of alcohol harm; for example, even though it is well established that alcohol can cause cancer, this fact is still not widely known to the public in most countries (3).

The ‘bigger picture’ is overall mortality. When all the risks are taken into account, including the small risks from a few rare cancers, do moderate drinkers live longer than teetotallers? Yes. Yes, they do.

But what about reports that a daily glass of red wine is good for your heart?

There is no good evidence for the pervasive myth that consuming red wine helps prevent heart attacks.

Note that they’ve shifted from ‘no evidence’ to ‘no good evidence’. It seems that no evidence is good unless it shows that drinking at any level is a killer.

The main reason that moderate alcohol consumption reduces overall mortality is that it substantially reduces the risk of cardiovascular disease. The evidence for this is not just ‘good’. It is overwhelming. For example, this meta-analysis of 84 studies concluded that: ‘Light to moderate alcohol consumption is associated with a reduced risk of multiple cardiovascular outcomes.’ The risk of cardiovascular disease mortality was 25 per cent lower than among non-drinkers.

Since red wine is a form of alcohol, it would be very odd if drinking a moderate quantity of it wasn’t good for the heart. All the evidence shows that it is, and there are some reasons to think that it might be even better than other forms of alcohol in this respect.

Why are national drinking guidelines different from WHO recommendations?

Many countries have issued low-risk guidelines, usually recommending no more than 10 standard drinks per week. WHO does not set particular limits because the evidence shows that the ideal situation for health is not to drink alcohol at all.

Again, this is simply a lie.

Doesn’t most alcohol harm come from a minority group of heavy drinkers?

The common perception is that a small fraction of the population causes most of the harm linked to alcohol consumption. But alcohol-related cancers, accidents, injuries and violence are widely distributed across the population (8), including among those who drink moderately. Even though heavy drinkers are undoubtedly at high risk of alcohol-related harm, they contribute only a minority to the total alcohol casualties. In this “prevention paradox”, most alcohol-related harm occurs among low-to-moderate risk drinkers simply because they are more numerous in the population.

This is a reference to one of the ideas of Geoffrey Rose who argued that there are more deaths from a large number of people exposed to a low risk than from a small number of people exposed to a large risk. While this could be true of some risk factors in certain circumstances, it is not a general rule and it is particularly poorly suited to alcohol because the risk of low to moderate consumption is negative, i.e. it is beneficial.

‘Alcohol-related cancers’ are relatively rare and mostly affect heavy drinkers. It is reasonable to assume that most drink-related ‘accidents, injuries and violence’ also involve heavy/binge drinkers. The prevention paradox just doesn’t apply to alcohol and Rose never claimed it did. Some academics have tried to claim otherwise, but they have had to narrow their scope to a limited range of ‘harms’ and portray people who get drunk as moderate drinkers to do so.

The real reason why organisations like the WHO are so keen to pretend that everybody is at risk from alcohol harm and there is no safe level of drinking is that it can be used to justify population-wide policies such as minimum pricing (which the WHO’s new report wrongly describes as ‘effective’). They say this in so many words in the summary:


Alcohol consumption causes considerable harm to millions of people across the world, not just the heaviest users, which is why strong global action that protects the entire population is needed.

Why, you may ask, is the WHO taking such a hard line on this that it is prepared to lie to our faces? Preaching total abstinence is a departure for the WHO. It is not what most of its member states expect of it and it is out of line with its previous messaging.

Part of the reason is that some member states do support total abstinence. You have to remember that alcohol prohibition is still a reality in some countries. Several other countries, such as India and Turkey, allow the sale of alcohol but strongly discourage it. Russia and some of the Baltic states have been clamping down on alcohol in a big way in recent years. The WHO has to represent all these interests.

The bigger and more important reason is that anti-alcohol activists have weaseled their way into the WHO in growing numbers. A lot of WHO reports are written anonymously. This one isn’t, and its list of contributors is very revealing. They include:

  • Phil Cain, European Alcohol Policy Alliance (Eurocare)

  • Florence Berteletti, Eurocare

Phil Cain is a former BBC journalist who has been tweeting about alcohol for years and has become increasingly fanatical. I didn’t know he was now at Eurocare but it makes sense. Eurocare is an EU-funded temperance pressure group formed in 1990 by the Methodist teetotaller Derek Rutherford.

  • Maik Dünnbier, Movendi International

The International Order of Good Templars is a gospel temperance group formed in the mid-19th century. A few years ago they realised that their name sounded a bit too religious/masonic so they changed it to Movendi. Their agenda of total abstinence remains the same. It is now an official partner of the WHO.

  • Øystein Bakke, Global Alcohol Policy Alliance and FORUT

The Global Alcohol Policy Alliance is another temperance group set up by Derek Rutherford. Notice the trend of temperance groups giving themselves bland names so they can slot themselves into the field of ‘public health’. Formed in 2000, Derek Rutherford said its priority was to “make the most of the opportunities provided by the development of the WHO Global Alcohol Strategy and the focus on non-communicable diseases.” It is certainly doing that.

  • Adam Knobel, Foundation for Alcohol Research and Education (FARE)

FARE is a state-funded anti-alcohol organisation set up by the Australian government with a $115 million grant. It says that it spends 50 per cent of its income on ‘policy and advocacy’ and 15 per cent on ‘leading change’. In other words, it is a pressure group.

  • Nason Maani, London School of Hygiene and Tropical Medicine

Maani is an activist-academic who has co-authored many hilariously bad articles with Mark Petticrew about the alcohol industry.

  • David Jernigan, Boston University School of Public Health

Jernigan is also on the board of the Global Alcohol Policy Alliance and speaks at their mad conferences.

Finally, there is Juan Tello. Listed as co-editor of the guide for journalists, he is in charge of the WHO’s Less Alcohol Unit. He recently gave an interview in which he complained about the ‘normalisation’ of alcohol and wibbled on about the ‘commercial determinants of health’. He makes no secret of ‘pursuing a no alcohol environment’.

His interview did contain one bit of good news, however.

Juan Tello: We give countries five recommendations that are cost effective to implement, especially in middle-income countries: increase pricing, particularly through taxation; ban marketing and sponsorships advertising alcohol; decrease the physical availability of alcohol outlets; enforce checks and control of blood alcohol concentration; and treatment. We are well positioned to do this when countries are willing to move forward. 

Think Global Health: What’s the level of interest from countries?  

Juan Tello: The uptake is low. It’s truly low.

Excellent. Let’s hope it stays that way. In the meantime, if journalists are looking for a story about alcohol, may I suggest an article about WHO reports being written by anti-alcohol activists?

Wednesday 19 April 2023

The cost of a healthy diet

Newsnight viewers were treated to a striking statistic from Professor Michael Marmot last week:

There’s a lot of discussion about ‘is it a matter of individual responsibility?’ Let’s look at how much it costs to have a healthy diet. If you’re in the bottom ten per cent of household income, to follow the NHS healthy eating guidance you would have to spend 74 per cent of your income on food. Impossible. So to feed your children junk food is cheaper.

To which the presenter, Victoria Derbyshire, nodded sagely.

The idea that the poor would have to spend three-quarters of their income to eat healthily has been doing the rounds in ‘public health’ circles for a few years, but this is the first time I’ve come across it in the wild. It should strike you as preposterous because
healthy food is generally cheaper than ‘junk food’. ‘Junk food’ is usually tastier and always more convenient, but it is not cheaper than starchy carbohydrates, fruit and vegetables, which are the core components of the government’s Eatwell Guide.

The 74% figure comes from a self-published report from the Food Foundation. It takes an estimate of the cost of a healthy diet from a 2016 study in BMJ Open and compares it with unequivalised household income figures from the Office for National Statistics.

Looking at the BMJ Open study, we find that a healthy diet costs £5.99 per person per day, but there is a problem straight away for the Marmot hypothesis:

The optimised diet has an increase in consumption of ‘potatoes, bread, rice, pasta and other starchy carbohydrates’ (+69%) and ‘fruit and vegetables’ (+54%) and reductions in consumption of ‘beans, pulses, fish, eggs, meat and other proteins’ (−24%), ‘dairy and alternatives’ (−21%) and ‘foods high in fat and sugar’ (−53%). Results within food groups show considerable variety (eg, +90% for beans and pulses, −78% for red meat). The modelled diet would cost £5.99 (£5.93 to £6.05) per adult per day, very similar to the cost of the current diet: £6.02 (£5.96 to £6.08).

So a healthy diet is no more expensive than the average current diet. Awkward.

Are we then to believe that people in the bottom income decile are already spending 74 per cent of their income on food? That can’t be right as they are only spending about 21 per cent of their income on food and alcohol combined. And although food prices have risen since 2020, the share of income spent on food and non-alcoholic drinks by households in the bottom quintile is still below 20 per cent.

So what’s going on?

Based on the figure of £5.99 a day, the Food Foundation assumes that £41.93 is the weekly cost of a healthy diet for an adult and it makes various adjustments to estimate the cost for partners and children. Fine.

It then uses a slightly lower figure of £36.37 a week as the cost of the average diet. I’m not sure why they don’t use the figure from the BMJ Open for this (well, I can guess), but it still isn’t a huge difference: an extra £5.56 a week or 15 per cent more. This isn’t enough to declare that having a healthy diet is “impossible” even at the lower end of the income distribution, as Marmot claims.

What about the income data? Marmot doesn’t mention it, but the Food Foundation figure is based on disposable income. Moreover, it uses a rather expansive definition of disposable income which does not just exclude income tax, national insurance and council tax, as would be normal, but also excludes rent, mortgages, service charges and water bills.

This removes a large part of people’s incomes in one fell swoop. Housing costs alone take up nearly half the income of people in the bottom decile so you can halve the figure Marmot gives straight away. (If you want to get angry about prices, I suggest you focus on housing.)

The Food Foundation also explicitly ignores free school meals which make up a non-trivial proportion of the diet of low income families.

Having done all this, they calculate what proportion of their income a household in each decile would have to spend to achieve a ‘healthy’ diet. Annoyingly, they don’t tell us what figure they are using for an average household nor what they consider an average household to be, although they do give one example: 

For example, we estimate that a household of four (two adults and two children aged 10 and 15) would need to spend £103.17 per week to be able to follow the Eatwell Guide.

In any case, these are their findings…
They say:

On average, the poorest half of households in the UK would need to spend close to 30% of their disposable income to meet the government’s dietary recommendations.

Given their definition of disposable income, most of this looks reasonably plausible. The only thing that looks odd is what is going on at the bottom 10 per cent of the distribution where there seems to be a quantum leap. Is this group really so much poorer than the second-to-bottom 10 per cent?

This is where it gets slightly murky. Economists who work with income data treat very high and very low figures with scepticism. In this instance, the bottom two per cent apparently have no disposable income at all. In fact, their disposable income is negative. In a modern welfare state, this is rather unlikely. It is possible to imagine a situation in which a few people end up in this situation, but they would be so down and out that they are unlikely to come to the attention of the Office for National Statistics. A more likely scenario is that they are mostly wealthy people who are not eligible for benefits and living off savings (retired people under the age of 65, for example).

Since this group makes up a fifth of the bottom income decile, they have a significant effect on the figures. It would be better to exclude them from the analysis. Instead, the Food Foundation counts them as spending 100 per cent of their (non-existent) disposable income on healthy food.

When calculating the proportion of disposable income that the Eatwell cost comprises, two percent of all households in the FRS dataset, comprising nearly 20 percent of households in income decile 1, had a negative disposable income AHC. To enable us to calculate the proportion of disposable income that would be used up by the Eatwell cost, these households were set to 100%.

In fairness, the authors do a sensitivity analysis in the appendix in which these outliers are excluded. They say it brings the figure down from 73.6 per cent to 60.8 per cent, but this must be a typo because the accompanying figure shows it to be around 50 per cent. Presumably it should say 50.8 per cent.

This still seems high when compared to the neighbouring decile, and there are other reasons to be sceptical about figures from the very bottom end of the income distribution, but given the exclusion of free school meals and housing costs, it is not totally implausible.

But even using this questionable methodology, the cost of an average diet already takes up 44 per cent of the bottom decile’s disposable income and if we use the figures from the BMJ Open study, it takes up 51 per cent, i.e. exactly the same as the ‘healthy diet’.

Whichever you slice it, the difference in price between a ‘healthy’ and ‘unhealthy’ diet is either small or non-existent and the idea that the poorest tenth of households would have to have to spend 74 per cent of their income to afford it is patently false. Even if we assumed that a healthy diet was as expensive as the Food Foundation claims, the true figure would certainly be below 40 per cent and probably below 30 per cent.

There are lots of reasons why people on low incomes - or high incomes, for that matter - would prefer ‘junk food’ and takeaways to the food recommended in the Eatwell Guide, but price is rarely one of them.

Tuesday 18 April 2023

Vaping junk science: an overview

Nice video from Reason about the shockingly bad science used by anti-vaping academics. It kicks off with a study I wrote about last year that was so bad it was retracted, not an uncommon event in this field of research.

Just stop Just Stop Oil

The snooker was interrupted last night due to morons on the baize. I have written about it for Spiked.

You never know what match you’re going to see when you buy a ticket to the World Snooker Championship. The two Just Stop Oil activists must have been hoping for something more glamorous than Robert ‘The Milkman’ Milkins vs ‘The Gentleman’ Joe Perry when they booked their seats for last night’s match at the Crucible in Sheffield, but you always take a gamble with these first-round matches. Strictly a red button / Eurosport affair, it wasn’t even broadcast on BBC Four. With all the money they get from the heiress to the Getty Oil fortune, you would think that Just Stop Oil would be able to afford a ticket to the final, or at least a match played with the one table set-up.


Monday 17 April 2023

A swift half with Jacob Grier

The new Swift Half is with the American author and journalist Jacob Grier who many of you will be familiar with from his work on tobacco and vaping. His book The Rediscovery of Tobacco is well worth a read, as is his new collection of essays, The New Prohibition. Look them up on Amazon. I reviewed the former here.

We discuss e-cigarette flavour bans, thirdhand smoke and looming prohibition.

Thursday 13 April 2023

Last Orders with Mark Littlewood

In the new episode of Last Orders we welcome back my boss Mark Littlewood to discuss laughing gas, disposable vapes and Ultra-Low Emissions Zones. Plus, as ever, we answer listeners' questions from our postbag. 

Listen here.

Wednesday 12 April 2023

E-cigarettes on the NHS

The British government is going to start giving e-cigarettes to some smokers, particularly pregnant women. It is also going to start clamping down on underage sales.  

I've been calling for the latter for some time. I've been sceptical of the former but I'm coming round to the idea. I still think that if you can afford to smoke, you can afford a disposable cigarette, but the biggest problem with e-cigarettes at the moment is that people think they are much more dangerous than they are. That has led to fewer smokers switching to them than would be the case if people were well informed.


Consumer ignorance is a major barrier to the consumption of low-risk nicotine products. Myths about vaping causing ‘popcorn lung’ and other diseases have proliferated on social media. Scare stories regularly appear in the press. As a result, a report from Public Health England concluded in 2021 that ‘Perceptions of the harm caused by vaping compared with smoking are increasingly out of line with the evidence’.

Four out of ten smokers in England wrongly believe that nicotine causes cancer and the proportion of smokers who wrongly believe that vaping is as dangerous or more dangerous than smoking rose from 36 per cent in 2014 to 53 per cent in 2020.

In other words, there is a market failure. The blame lies with charlatans in the 'public health' racket, mostly in the USA and at the WHO, often funded by Mike Bloomberg, but it justifies a government response. If the NHS essentially prescribes e-cigarettes, it sends a clear signal to the public that vaping is vastly safer than smoking. As a bonus, vaping advocates in backwards countries like Australia can point to what the UK is doing to illustrate the absurdity of banning e-cigarettes.

The cost of the free vape scheme is not enormous (£45 million). In any case, smokers massively subsidise nonsmokers through tobacco duty so why shouldn't they get a bit back? On balance, it is a good policy.

This was all announced yesterday by the public health minister Neil O'Brien at Policy Exchange (where he used to be the director). You can read his speech here or watch it here. Mostly, it was toe-curlingly awful. I suppose that having to abase yourself in front of the 'public health' lobby by parroting their ridiculous tropes goes with the territory in his job, but I felt embarrassed for him all the same. 

He didn't have much choice other than to repeat the myth that smoking places a net burden on the taxpayer, since the 'evidence' for this came from a Policy Exchange report published while he was in charge of the think tank. He described the report as excellent. As I said at the time, it isn't.
But did he really have to repeat this idiotic canard from ASH? 

Current smokers are 7.5% less likely to be employed compared to never smokers and ex-smokers are 5% more likely to be employed than current smokers.

In places like Birmingham, an additional 6,000 people are out of work because of smoking. Quitting could help to put that right.
Smoking is much more common among lower socio-economic groups these days. Unemployment is always more common among lower socio-economic groups. A correlation between smoking and unemployment is therefore hardly surprising, but the idea that someone will suddenly get a job if they quit smoking is so dumb that O'Brien can't possibly believe it.

He also used the old 'stop hitting yourself, smoker' argument...

As well as the productivity impact, quitting smoking would save the average person around £2,000 a year.

Of that £2,000 approximately £1,800 is tax, and the rate of tax went up steeply last month, so we don't need anyone from the government shedding crocodile tears about the secondary poverty of smokers.
If you can get past this nonsense, what he actually announced was OK. He thanked Javed Khan for his bonkers review before announcing that he would not be going ahead with its main proposal - incremental prohibition - because it would be a gross violation of people's freedom.

Now of course some would go further to stop people to start smoking in the first place. The Khan Review last year advocated the New Zealand approach – a full phase out of smoking, with the age of sale increasing over time to cover all adults.

This would be a major departure from the policy pursued over recent decades which has emphasised personal responsibility and help for people to quit.

The subsequent Q & A is worth watching if you want a reminder of how many parasites are involved in tobacco control. Eric Hoffer once said that every great cause begins as a movement, becomes a business, and eventually degenerates into a racket. Judging by the questions, the room was divided into those who are threatened by vaping and want the state to give them more money (smoking cessation services, Allen Carr) and those who simply want smoking banned (CRUK, ASH). Quite a few of them want smoking banned and to be given more money. Deborah Arnott from ASH was among those panhandling for a tobacco levy despite this being a matter for the Treasury, not the Department of Health (and the Treasury rejected it years ago).

It was interesting to see how aggressive some of the people asking questions were. They are obviously used to bossing ministers about. There were a lot of questions about smoking cessation clinics, which are extremely inefficient and have been made more or less redundant by reduced risk nicotine products. A woman called Penny Steed who works at Whittington NHS Trust and runs Smoke-Free City & Hackney gave a quote for the ages when she told O'Brien:

"We're a great service. We help a lot of people. We really do help them. They tend to go back to smoking and we help them again."

Said without a hint of irony, this was a showstopper. It brought to mind Mark Twain's famous quote about quitting smoking, as well as a more niche reference from Best In Show
She followed this up by calling on the government to make it impossible for people to smoke because "I don't think help is going to cut it." What a ringing endorsement of a service that costs the taxpayer £75 million! Let's pour some more money down that pit, shall we?

Mark Twain quote: Quitting smoking is easy, I've done it ...

Monday 3 April 2023

High status conspiracy theories

I recently wrote about high status conspiracy theories for Quillette, featuring James O'Brien, George Monbiot, Carole Cadwalladr, Philip Hammond, Aseem Malhotra, Peter Jukes, Andrew Adonis, Tom Watson, Jeremy Corbyn and Naomi Klein. 

Who can forget Lord Andrew Adonis, once considered the blandest of centrists, demanding answers from the BBC about what he called “the fake vicar scandal”? Perhaps you have forgotten, so let me remind you. It was a minor conspiracy theory of Adonis’s own invention that stemmed from his mind erupting at seeing a female vicar on a Newsnight panel praising Theresa May’s Brexit deal. With a little digging, Adonis discovered that she had appeared as an extra in a number of films, including a villager in Macbeth and a funfair attendant in Pudsey the Dog: The Movie. Rather than accept that she was a vicar who did a bit of acting, Adonis concluded that she was an actor hired by the BBC to pretend to be a Leave voter.

Replying to one of Adonis’s many tweets about “Vicargate,” Newsnight presenter Emily Maitlis wrote: “To have got to a place where you could chose to believe that enough to write it—is deeply worrying.” There are, as Donald Trump would say, many such cases. If COVID-19 instigated a wave of conspiracy theories on the Right, Boris Johnson’s election victory had the same effect on the Left and Brexit caused many centrists to have a meltdown. Everybody’s had something to break their brains in recent years and some people may never recover.

Many of the classic conspiracy theories arise from a sense of disbelief. A simple car crash seemed to be an inadequate ending to the Princess Diana soap opera. JFK was too important to have been killed by someone as insignificant as Lee Harvey Oswald. 9/11 was too enormous to have been mere terrorism. The logistics of putting a man on the Moon were too mind-blowing to have been achieved in 1969.

COVID-19 and lockdown delivered a psychic shock on a greater scale than any of these events. That fallible politicians were grappling with the extraordinary problem of dealing with a deadly new virus seemed an inadequate explanation when one could instead believe that it was a false flag to sell vaccines and install a world government (or, from the opposite perspective, that the government had a policy of culling the elderly).

The election of Donald Trump and the vote to leave the EU were similarly shocking to those who had previously regarded those developments as unthinkable. As we retreat into our echo chambers, there is a growing inability to even acknowledge the existence of opposing views, let alone consider them valid. When everyone you know agrees with you, it comes as a shock when the votes are counted and you realise that your side is outnumbered by a basket of deplorables. When something seems unbelievable, the temptation is to simply disbelieve it and reach for alternative explanations.


Sunday 2 April 2023

Black market, here we come

I appeared in a short video for the Taxpayers Alliance about tobacco duty which has just risen by an insane amount in the UK - especially on hand-rolled tobacco which is the last stop on the route for smokers before they reach for the black market.