Nutrition Science
Why the Healthiest Eaters Are Often the Most Confused
The people who read the most about nutrition are frequently eating no better than those who've never thought about it once.
The Idea
Nutrition science has a problem it rarely admits out loud: it is structurally bad at producing reliable knowledge. Most of what reaches the public — the superfoods, the banned ingredients, the miraculous macronutrient ratios — originates from epidemiological studies that can identify correlations but cannot establish cause. When researchers find that people who eat blueberries report better cognitive function, they have not found that blueberries sharpen thinking. They may have found that people with the time and money to eat blueberries also sleep better, exercise more, and experience less chronic stress. Untangling those threads is extraordinarily difficult. This is compounded by the media cycle. A single study — often small, often preliminary, almost always hedged in its actual conclusions — gets translated into a headline that strips every caveat. 'Coffee linked to reduced liver cancer risk' becomes 'Coffee protects your liver'. The original researchers may wince, but the claim is already circulating. What makes this particularly insidious is that the misinformation doesn't feel like misinformation. It arrives wearing the clothes of science: graphs, citations, university affiliations, confident spokespeople with credentials. Our pattern-recognition instincts, which are very good at separating real threats from false alarms in most contexts, are poorly calibrated for evaluating scientific claims. We tend to accept findings that confirm what we already half-believe and dismiss those that challenge it — which means the noise accumulates and the signal gets buried.
In the World
In 2015, a Danish researcher named Mathilde Touborg Sørensen was part of a team that published a paper on saturated fat and cardiovascular disease. The findings were nuanced — different saturated fats appeared to carry different risks, and the picture was considerably more complicated than the prevailing 'fat is bad' consensus suggested. A version of this research made it into the press as evidence that butter had been unfairly maligned. Butter sales rose across several markets. The original researchers subsequently gave interviews clarifying what they had and hadn't claimed. This is not an unusual story. Journalist Gary Taubes spent years documenting how the low-fat dietary guidelines that dominated from the 1980s onward were built on weaker evidence than the public was led to believe — and how the sugar industry had quietly funded research that redirected blame away from sucrose and toward fat. The result was a generation of 'low-fat' products that replaced fat with refined sugar, potentially making things worse while appearing virtuous. The cycle continues today, just with different protagonists. Seed oils, red meat, gluten, intermittent fasting — each generates a wave of confident claims, counter-claims, and eventually a muddled public that has been told opposite things by sources that all sounded authoritative. The confusion is not accidental. It is, in part, a product of how nutrition research is funded, published, and consumed.
Why It Matters
There's a real cost to navigating nutrition in this environment — not just the wrong food choices, but the anxiety, the decision fatigue, and the erosion of trust that comes from being confidently misled more than once. People cycle through conviction and betrayal: gluten-free, then keto, then Mediterranean, then carnivore. Each new regime arrives with a compelling story and a set of before-and-after testimonials. Each eventually gets complicated by subsequent research or simply stops working. The more useful shift isn't finding the next correct framework — it's developing a more sophisticated relationship with nutritional claims in general. That means asking: is this based on randomised controlled trials, or observational data? Who funded the study? Has it been replicated? Is the effect size large enough to be meaningful in a real life, or just statistically detectable in a lab? None of this requires becoming a scientist. It requires becoming a slightly more sceptical reader — one who can hold provisional conclusions lightly and resist the pull of a clean, confident answer in a domain that rarely produces them.
A Question to Ponder
What's one food belief you hold with more certainty than you can actually justify — and what would it take to genuinely change your mind about it?
Get a new one of these every morning.
Start learning with Thinkable