Sometime in 2020, I began to realise that, despite having trained as a doctor and worked in health for 20 years, I didn’t know anything.
asserted
I → begin → anything
It started with the health and wellness podcasts, whose hosts are mostly men, mostly American, all highly qualified and all very confident about their claims and their conclusions.
uncertain
all → start → claims
In everyday conversation, complicated ideas began to crop up.
asserted
ideas → begin → conversation
Friends and colleagues would show me their microbiome analysis, or the results of genetic tests, or data from the continuous blood-glucose monitor (CGM) they were wearing.
asserted
they → show → monitor
They’d tell me about their supplement stack, including molecules like NMN, glutathione, CoQ-10 and alpha-GPC and show me their fitness data.
asserted
They → tell → data
At this time, I was appearing on BBC TV and radio, and the questions coming in from viewers and listeners reflected this new surge of health and wellness information.
But
asserted
questions → appear → information
I started to investigate the legitimate-sounding advice they were asking me about: Should I take NAD?
asserted
I → start → NAD
Would a CGM help me to lose weight?
asserted
me → help → weight
Will red-light therapy help my joints?
asserted
therapy → help → joints
Should I get tested for microplastics?
asserted
I → test → microplastics
Should I get a blood test to assess my ‘biological age’? – it turned out to be hard to get them definitive answers.
asserted
it → get → answers
This book emerges from my attempts to understand the wellness industry as a health broadcaster, and as a person trying to look after myself.
asserted
book → emerge → myself
Xand van Tulleken is a doctor, broadcaster and presenter who practised in acute medicine in the NHS for a decade.
asserted
who → practise → decade
He trained at Oxford, Harvard and the Liverpool School of Tropical Medicine and is now an Honorary Associate Professor at University College London
I began by asking ‘What works?’ and spent months poring over scientific evidence, with a huge amount of help from colleagues and experts.
asserted
What → train → colleagues
And the answer turns out to be: very little.
asserted
answer → turn → ?
By the end, I hope you’ll have a rule of thumb about what works and what doesn’t.
asserted
what → hope → thumb
I want you to be able to look at a product on a shelf or an Instagram post or an article in the paper and say: ‘Oh, I know why they’re selling that!
asserted
they → want → that
I know what they’re up to!’
asserted
they → know → what
You’ll be able to see who has made you even consider buying the product in the first place, and be aware of all the ways the product might harm you.
uncertain
product → see → you
Let’s start with vitamins.
asserted
’s → start → vitamins
There are only 13 human vitamins, but together they form the foundation of all modern nutrition science, and, most importantly, they’re crucial to how we understand the chemistry of life.
asserted
we → be → life
Vitamin A (found in carrots) is essential for vision in low light; the B vitamins are important for energy production, the nerves and the blood; you don’t need to know any biochemistry to associate vitamin C with good skin and hair, and not catching a cold.
asserted
you → find → cold
It’s completely reasonable to believe that our vitamin intake isn’t optimal today.
asserted
intake → ’ → ?
OK, we don’t see much scurvy any more, but we don’t eat nearly as much fruit and veg as we should: diet-related ill-health is a leading cause of death in the UK and the US.
asserted
health → see → UK
So, a daily vitamin supplement might well prevent a kind of mild scurvy or mild rickets, supercharging our bones, skin, hair and immune systems.
uncertain
supplement → prevent → bones
This is a logical train of thought, helped along by the supplement industry’s suggestions to us.
But there’s a problem.
asserted
This → help → us
Millions have been spent on large research trials in all vitamins and they have almost all come to the same conclusion: multivitamin supplementation – including vitamin C supplementation, which doesn’t actually stop us getting colds − does not prevent chronic diseases such as cancer and cardiovascular disease in the general population.
asserted
us → spend → population
Nor does it decrease your risk of premature death.
asserted
it → decrease → death
And it may even increase your risk of dying sooner.
uncertain
it → increase → risk
In 2007, a team of scientists including Christian Gluud, a professor at Copenhagen University Hospital, published a paper, in a prestigious medical journal, analysing how supplements – including vitamins A, C and E, beta-carotene and trace mineral selenium – affected mortality.
asserted
supplements → include → mortality
The team gathered up data from 68 randomised trials of supplements (collectively including almost 250,000 participants) and combined the data, effectively giving them a huge trial population of their own.
Continuous blood-glucose monitors analyse a person's glucose levels – but the results they give have a significant margin of error
asserted
they → gather → error
At the time, there was such enthusiasm in the medical community for antioxidants that to many it seemed certain that Christian and his colleagues would find a benefit.
asserted
Christian → be → benefit
But the results told a different story.
asserted
results → tell → story
When I called Christian in 2025, he summarised the findings like this: ‘We looked into antioxidant vitamins and found out that they killed people.’
asserted
they → call → people
His team’s study concluded that antioxidant supplements, including vitamin C, do not reduce mortality.
asserted
supplements → conclude → mortality
Even more surprisingly they found that taking vitamin A, beta-carotene, and vitamin E could increase mortality.
uncertain
taking → find → mortality
The study had a huge impact in the scientific world: today it has been cited over 1,500 times.
asserted
it → have → times
Very few papers achieve this level of recognition.
asserted
papers → achieve → recognition
But despite its scale, it wasn’t considered the final answer.
asserted
it → consider → scale
The team went back to their analysis as more data from other trials came in.
asserted
data → go → trials
…and 112 more, not listed.
