Bad teeth are the number one reason kids go to A&E? The number got misread
A British MP said teeth are the top reason children end up in A&E. The actual figure refers to 5-to-18-year-olds hospitalised for tooth decay — not the same thing, and the decay rate is actually drifting down.
The video won't play here. Listen to the audio instead:
The argument · tap a timestamp to hear it
Teeth are the top reason for A&E visits — that claim is wrong
Leila Moran, a Liberal Democrat MP and chair of the Health and Social Care Committee, said on Radio 4's The World at One that the number one reason children go to A&E is teeth. Chris Venatza, a consultant in paediatric dentistry at Newcastle University, flatly rejected it: she very likely conflated two sets of numbers. What actually ranks first is "hospital admission for tooth decay treatment" — the number one reason 5-to-18-year-olds are admitted to hospital. A&E attendance has all sorts of causes, and there is simply no data showing whether dentistry ranks high or low there. Moran's office later confirmed Chris was right, and that she meant the hospitalisation figure, not the A&E figure.
— Chris VenatzaLots of hospital tooth extractions, because the rules changed in 2001
When children are hospitalised for teeth, it is usually for an extraction, not a filling. Extractions are harder on children, so a child who cannot cooperate for an extraction in a dental clinic may need general anaesthetic. There is an institutional turning point here: the rules changed in 2001, and ordinary dental practices were no longer allowed to administer general anaesthetic, so anyone needing GA had to go to hospital. So a large share of the hospitalisation numbers are GA extractions. That explains why "hospitalisation" and "A&E" are two completely different measures — one is a surgical arrangement pushed into hospitals by the rules, the other is unplanned care.
— Chris VenatzaThe hospitalisation rate is actually falling, not a crisis
Chris gave the rate per 100,000 population: about 320 in 2015, down to 270 before the pandemic, and 250 now. Host Tim Harford's reaction was that "this doesn't look like a crisis, the trend is even rather encouraging." Chris conceded that a pre-pandemic decline is a good thing, but stressed that even so, large numbers of children are still having general anaesthetics and hospital dental treatment, which is not good in itself. He added one more thing: for the past decade, dental causes have been the number one reason for hospitalisation, unchanged for ten years — so the "number one" position is stable, but the absolute level is heading down.
— Chris VenatzaTo see the decay rate, look at school surveys, not hospitalisations
Chris said that if you want the real picture of children's tooth decay, there is a better dataset: the government sends dentists into schools to examine five-year-olds. The latest survey found 22% of five-year-olds already have decay reaching the dentine layer — close to a quarter. He said that a quarter of a population having a disease that needs treatment is not a good state of affairs. Over time: about 30.9% in 2008, down to 23% by 2017-2018, then a plateau, with the latest survey slightly down again on the 2022 one. Longer-run data comes from a survey run once a decade since 1973: 72% of five-year-olds had decay in 1973, down to 43% in 2003 — roughly halved.
— Chris Venatza25% of UK medicines come from Israel — off by two orders of magnitude
US ambassador to Israel Mike Huckabee said on the Today programme that 25% of the medicines Britain uses are made in Israel. Lizzie McNeill checked the latest ONS import data: in 2025 the UK imported £10 million of medicines from Israel, against total UK medicine imports of about £28 billion, so Israel accounts for just 0.04% — one twenty-fifth of 1%, not 25%. She also ruled out several possibilities: active pharmaceutical ingredients come mainly from India and China, and the more upstream raw materials mainly from China; and the medicines supplied by Israeli companies (the biggest being Teva) are mostly not made in Israel but in European plants in Hungary, Croatia and elsewhere. Teva estimates it supplies one in eight generic medicine packs in the UK, equivalent to one ninth of all prescriptions, but the other Israeli companies are too small to push the total to 25%.
— Lizzie McNeillBlame Gove for rising NEETs, and Wales dismantles the case
Education secretary Lucy Powell wrote in The Times that a million young people have been left behind by the education system Michael Gove led. Daisy Christodoulou rebutted with a natural experiment: Wales and Scotland took the opposite path to Gove's reforms. Wales's NEET rate for 16-to-24-year-olds rose from 14% in 2019 to 17% in 2025, while England's rose from 11% to 13% over the same period — similar levels, similar trends, with Wales rising slightly more. If Gove is responsible for England's rise, then Wales must be responsible for Wales's rise, yet the two places had completely opposite policies. She said the NEET rate sits on top of economic structure, the minimum wage, tax, regulation and far too many other factors — not just education.
— Daisy ChristodoulouThe teen mental health decline is not unique to England
Against the criticism that Gove's reforms increased academic pressure and damaged mental health, Daisy also looked at the WHO's Health Behaviour in School-aged Children survey, which reports England, Scotland and Wales separately. The results show average life satisfaction among teenagers fell in all three places this century (2000 to 2022), but it also fell across Western Europe and many other regions, so it is not specific to England. Then the share of 15-year-olds who "like school a lot": about 10% in England, about 9% in Scotland, about 7.5% in Wales — all low, but basically the same across the three. The share of 15-year-olds overwhelmed by academic pressure is high in all three, with no big difference.
— Daisy ChristodoulouP-Doom is a subjective probability, and nobody can produce a model
Anthropic researcher Jacob Coxon resigned, warning that Anthropic and OpenAI are irresponsible on AI safety. Anthropic's Evan Hubinger responded on X: he personally thinks the probability that AI kills all of humanity within the next decade is greater than 10%. Alexander Mark Koch, an assistant professor of global risk and resilience at Cambridge, explained that P-Doom is shorthand for "the probability that AI causes human extinction," and that it is almost always one individual's subjective probability estimate. He said the event is extraordinarily complex: it involves regulatory capacity, system resilience, where large-model capabilities are heading, autonomous weapons and biotech tools, whether AI can self-improve into an exponential loop, funding, rules, what China will do, and disaster scenarios ranging from science fiction to "a rogue state using a large model to learn to make bioweapons."
— Alexander Mark KochNobody really believes it, because market prices don't reflect it
Economist Tyler Cowen has repeatedly pointed out that many people talk about human extinction, but nobody can point to any market price — interest rates, share prices — reflecting that anyone truly believes it. Asked whether anyone really believes it, Koch said some people genuinely do, and you cannot say they don't; but people diverge sharply on the probability and on how the catastrophe would happen. Asked about his own P-Doom, Koch said he doesn't have one, because he doesn't know how to assign a number to it: he can conceive of many paths to destruction, and he can also think of ways to mitigate each one, but he doesn't know how to assign a probability to any of them.
— Alexander Mark KochIn their own words · checked verbatim
It's worth just noting, by the way, the number one reason children end up in A&E is because of dentists, dentistry issues. It's because of teeth that are crumbling.
Leila Moran1:01
No, I don't think they are actually, and I suspect that there might have been a bit of confusion with some different figures here.
Chris Venatza2:02
So if we look at the last 10 years of these hospital admission statistics, we see that dental reasons were the number one cause for all of those 10 years. And that hasn't changed.
Chris Venatza5:10
then in 2025, the UK imported £10 million worth of pharmaceuticals from Israel. That doesn't feel like a big number. It's not. Our total pharmaceutical imports stood at around £28 billion. So imports from Israel are only 0.04% of total imports.
Lizzie McNeill9:16
If Michael Gove is responsible for the rise in NEAT numbers in England, then the Welsh education system must be responsible for the rise in Wales, despite the fact that they were following totally different policies.
Daisy Christodoulou16:45
Jacob is correct here. We really do earnestly believe AI could kill all humans. I personally think it is greater than 10% within the next decade.
Evan Hubinger19:59
I don't, because I don't generally know how to put a number on it. It is such a complex thing to predict.
Alexander Mark Koch25:14
Figures
| Number one reason for hospitalisation of 5-to-18-year-olds | tooth decay treatment | 2:02 |
| Child dental hospitalisation rate per 100,000 population | about 320 in 2015, 270 before the pandemic, 250 now | 4:10 |
| Tooth decay rate among five-year-olds (decay reaching dentine) | 22% in the latest survey, 30.9% in 2008, 23% in 2017-2018 | 5:10 |
| Long-run survey of tooth decay among five-year-olds | 72% in 1973, 43% in 2003, 25% in the more recent one | 7:13 |
| Total UK medicine imports | about £28 billion | 10:17 |
| Israel's share of UK imports | 0.04% | 10:17 |
| Share of UK prescription medicines supplied by Teva | one ninth of all prescription medicines | 11:20 |
| NEET rate for 16-to-24-year-olds in Wales | 14% in 2019, 17% in 2025 | 15:41 |
| NEET rate for 16-to-24-year-olds in England | 11% in 2019, 13% in 2025 | 15:41 |
Glossary
- NEET
- Young people who are not in education, employment or training.
- P-Doom
- An individual's subjective estimate of the probability that AI causes human extinction.
- dentine
- The hard tissue layer beneath tooth enamel; decay reaching this layer means the cavity is already fairly deep.
How to listen
Readers following UK public policy, public health and education reform; and tech people who want to see through the "P-Doom" rhetoric in AI safety circles.
The closing Tom Lehrer song and the production credits — skippable.