Nutrition advice keeps changing because most of what reaches you is tactics, and tactics have a short shelf life. The underlying idea has been far more stable. In 2018, David Ludwig and Cara Ebbeling laid it out in JAMA Internal Medicine: the type of food you eat, not only the amount, may influence the hormonal signals that drive fat storage and hunger. That argument has been debated hard for years and it has not been swept away. The eight-week diet trend you saw last spring did not survive the summer. Notice the difference in durability.
If you are tired of being told something new every year, you are reacting to something real. But the fix is not to give up on nutrition. It is to learn which layer of advice you are looking at.
Why does nutrition advice keep changing?
Because there are three layers of it, and they move at wildly different speeds.
- Principles. How your body responds to food. These change slowly, over decades, as evidence accumulates.
- Applications. Eating patterns built on those principles, like eating protein and vegetables before starch, or narrowing your eating window. These get refined as research fills in details.
- Hacks and headlines. The apple cider vinegar shot, the celery juice, the one supplement. These churn constantly because they are content, not science.
Most of the whiplash you feel comes from layer three being reported with the confidence of layer one. A single small study gets a headline, the headline gets contradicted six months later, and it feels like scientists cannot make up their minds. They usually have not changed their minds at all. The coverage did.
What did Ludwig and Ebbeling actually argue?
Their 2018 paper in JAMA Internal Medicine pushed back on the idea that body weight is explained by energy in and energy out alone. They set out the carbohydrate-insulin model, which proposes that highly processed, rapidly digested carbohydrates raise insulin, that elevated insulin promotes storage of fuel in fat tissue, and that this can leave fewer usable fuels circulating, which may increase hunger and lower energy output. In that framing, overeating is partly a consequence of the hormonal response, not only its cause.
This model is contested. Other researchers argue the effects are smaller than proposed, and the debate is genuinely unresolved. That is worth saying plainly. But notice what almost nobody in that debate disputes: a bowl of lentils and a can of soda do not do the same thing inside you. The quality of what you eat shapes your hormonal response. The argument is about how much it matters, not whether it matters at all.
If eating fat does not make you fat, why did we hear the opposite for decades?
This is the objection that stops most people, and it deserves a straight answer. Fat was targeted partly because it carries more energy per bite than protein or carbohydrate, so reducing it looked like the obvious lever. What happened next is the part worth remembering: as fat came out of packaged foods, refined starch and sugar frequently went in to replace it, because low-fat products taste like cardboard without them.
The carbohydrate-insulin argument suggests that swap may have been a poor trade, because refined starch and sugar tend to provoke a larger insulin response than fat does. Olive oil, avocado, nuts and oily fish provoke very little insulin response by comparison, and fat slows gastric emptying, which can blunt the glucose rise from a meal. That does not mean fat is a free pass or that you should pour cream on everything. It means the old rule was aimed at the wrong target.
How do I tell a durable principle from a passing hack?
Three questions I use before I take any new nutrition claim seriously:
- Does it have a mechanism? Can someone explain what is physically happening in your body, not just what a study found? Fibre slowing glucose absorption has a mechanism. A superfood berry usually does not.
- Has it survived disagreement? The carbohydrate-insulin model has been attacked repeatedly by serious researchers since 2018 and is still standing as a live scientific position. That is a much stronger signal than a claim nobody has bothered to test.
- Does it require a purchase? Principles are usually free. Products are usually the hack.
Apply those three and roughly ninety percent of what crosses your feed drops away, which is the point.
What stays the same no matter which study lands next?
Here is the short list that has not needed revising, whatever happens to the finer points of the debate:
- Whole foods tend to provoke a gentler blood sugar and insulin response than their processed versions. Whole oats behave differently from instant oats. An orange behaves differently from orange juice.
- Protein and fibre slow digestion, which tends to flatten the glucose curve of a meal.
- Liquid sugar hits fast, because there is no chewing, no fibre matrix and almost no delay.
- Eating the vegetables and protein on your plate before the starch tends to reduce the glucose spike from the same food.
- Walking for ten to fifteen minutes after a meal gives working muscle somewhere to put the glucose.
None of that is new. None of it will be overturned by next year's headline. It is boring in the best way.
What does this look like on my plate tomorrow?
Pick one meal you already eat and change its shape, not its existence. Breakfast is usually the easiest. If it is currently cereal and juice, that is two fast-releasing carbohydrate sources with almost nothing to slow them down. Eggs with spinach and half an avocado, or plain Greek yoghurt with berries and walnuts, delivers protein, fibre and fat that tend to keep the glucose curve flatter.
Then leave it alone for two weeks. Most people fail at nutrition not because they chose wrong but because they change again before the last change had a chance to prove itself. The churn in the headlines becomes churn on your plate.
Where does an insulin score fit into this?
The Insulin Fix score exists so you do not have to relitigate the science every time you pick up a package. It runs from 0 to 100, and a higher score means a lower insulin load. It weighs glycemic load, protein, fibre, how processed the food is, the quality of its fat, whether the energy arrives as liquid, and the order you eat things in. Those inputs are the durable principles, translated into one number.
It is a shortcut for the layer of advice that does not change. That is the whole design intent.
Frequently asked questions
Is the carbohydrate-insulin model proven?
No. It is a well-argued and actively debated scientific model, published by Ludwig and Ebbeling in JAMA Internal Medicine in 2018 and contested by other researchers since. What is far less controversial is the narrower claim underneath it: different foods provoke different hormonal responses, so food quality matters alongside quantity.
So does that mean energy balance is wrong?
Not wrong, incomplete. Energy balance describes what happens. The carbohydrate-insulin argument is about what drives it, including why hunger and energy output are not fixed but respond to what you eat. You can accept the physics and still ask what is steering it.
Can eating more fat really help with fat loss?
Research suggests that replacing refined starch and sugar with protein, fibre and quality fats may reduce the insulin response to a meal and improve satiety, which can make eating less feel easier rather than harder. That is different from adding fat on top of what you already eat.
How long before I should expect to notice anything?
Most people report changes in energy and hunger, particularly the afternoon slump, within one to two weeks of changing the shape of their meals. Body composition moves more slowly. Judge the first fortnight on how you feel, not on the scale.