Why nutrition is so hard to prove
Ed, our resident nutritionist, gets asked constantly about the latest diet headline doing the rounds. His usual answer: be sceptical, and here's why. Almost everything we "know" about diet comes from weaker forms of evidence than people assume — mostly observation, rarely proof beyond doubt. If we insisted on absolute proof for everything, we'd struggle to even confirm smoking causes cancer.
The gold standard, a randomised controlled trial comparing two groups over time, is genuinely difficult to run well for nutrition. Results have to be adjusted for confounding factors like smoking, activity levels and body weight. Trials comparing one diet against another are often designed by people who already favour one side, which tends to shape the result before it's even run. And to prove, beyond argument, that one diet is best across a lifetime, you'd effectively need to follow people from the womb to old age, controlling everything from breastfeeding onward. It's not a small ask, which is why it's never actually been done.
What we get instead
What we get in its place are shorter trials measuring narrower markers, even though many chronic conditions take decades to develop. Layer the media's appetite for a dramatic new diet every week on top of that, and you get a lot of confident-sounding claims built on very little.
A few flags worth knowing
- A book that opens by telling you to forget everything you've heard about nutrition is not how nutrition science actually works, and is worth treating with caution
- Being well known is not the same as being qualified — plenty of persuasive diet advice comes from people with no relevant expertise at all
- If a diet book cites studies, it's always worth checking whether those studies actually say what's claimed
None of this means nutrition advice is worthless, only that it deserves the same scepticism you'd apply anywhere else someone's trying to sell you certainty.
