Eat vegetables and protein first, carbohydrate last, with roughly ten minutes between. Identical meal, identical calories — but the post-meal glucose peak comes out substantially lower. It was measured in people with prediabetes by researchers at Weill Cornell, and later tested as a standalone behavioural intervention in a randomised pilot.
The experiment that makes the point
The reason this isn't just another internet tip is that someone controlled it properly. Shukla and colleagues gave people the same meal on separate occasions and changed only the order it was eaten in — carbohydrate first versus carbohydrate last — then measured what happened to glucose and insulin afterwards.1
Same food. Same quantity. Same person. The only variable was sequence. And the carbohydrate-last condition produced a markedly lower glucose excursion.
What makes this more than a curiosity is the follow-up: the same group ran a randomised controlled pilot testing food order as an actual behavioural intervention in prediabetes — that is, asking whether real people can adopt it and whether it holds up outside a single test meal.2 That's the step most nutrition tips never take.
Why the order matters at all
Three mechanisms, working together.
1. Gastric emptying slows down
The biggest one. Your stomach doesn't release its contents all at once — it meters them into the small intestine, where absorption happens. Fat, protein and fibre arriving first slow that release rate considerably.
So when the carbohydrate finally shows up, it's queued behind everything else and trickles through rather than arriving in a bolus. Same total glucose, delivered over a longer window — and a slower delivery is a lower peak. That's most of the effect right there.
2. The fibre gets there first
Soluble fibre from vegetables forms a viscous layer in the gut that physically slows how fast glucose reaches the intestinal wall. Eating it first means the barrier is already in place before the carbohydrate arrives, rather than being mixed in with it.
3. Gut hormones get a head start
Protein and fat entering the small intestine trigger release of incretin hormones — GLP-1 among them — which prime the pancreas to respond and further slow gastric emptying. Eating them first means that machinery is already running when the glucose load lands.
Interestingly, the sweet-tasting part of a meal seems to play its own role in gastric emptying independent of calories, which is one more reason the ordering of what hits your tongue isn't a trivial variable.3
How to actually do it
The whole method:
- Vegetables and salad first. Non-starchy ones — leaves, broccoli, peppers, courgette. Not potatoes, not corn.
- Protein and fat second. Meat, fish, eggs, tofu, cheese, olive oil.
- Carbohydrate last. Rice, pasta, bread, potato, dessert.
- Leave a gap if you can. Roughly ten minutes between the first course and the last is what the research protocols used. Even a short pause is better than none.
| Meal | Usual order | Resequenced |
|---|---|---|
| Sandwich lunch | Everything at once | Side salad first, then the filling, then the bread. Or open it up and eat the contents first. |
| Pasta dinner | Pasta with sauce, salad on the side, all mixed | Salad, then the meat or sauce, then the pasta. |
| Curry and rice | Spooned together | Vegetables and curry first, rice at the end. |
| Restaurant meal | Bread basket while you wait | Skip the bread until the main arrives — the bread basket is the single worst-sequenced item in Western dining. |
| Breakfast | Cereal or toast alone | The hardest one to fix by sequencing. Better solved by adding protein and fat, not reordering. |
Honest limits
I'd rather you got this from me than found it out later.
It doesn't work on a carbohydrate-only meal. If your lunch is a bowl of pasta and nothing else, there's nothing to put first. Sequencing needs something to sequence — which means this technique quietly requires you to eat vegetables and protein in the first place.
The effect size varies a lot between people. Study averages are averages. When I tested this on myself with a monitor, some meals changed dramatically and others barely moved. Yours will differ from mine.
It doesn't reduce total calories. You're changing the shape of the curve, not the amount of food. This is a glucose intervention, not a weight-loss one, though the reduced post-meal crash tends to make people snack less — which is a real second-order effect and not nothing.
It's not a treatment. If you have diagnosed diabetes, food order is a useful habit alongside your prescribed management, not instead of it. And if you take glucose-lowering medication, changes to how you eat can affect your numbers — worth a word with your prescriber.
Stack it with the other free lever
Food order changes how fast glucose arrives. A short walk changes how fast it leaves — muscle contraction pulls glucose out of the blood through a pathway that doesn't require insulin at all. That mechanism is explained here.
Doing both is the highest-value ten minutes in this entire subject: vegetables and protein first, carbohydrate last, then a walk while the curve is rising. Free, immediate, and better supported by evidence than anything sold in a bottle.
And if you're still curious about supplements
Fine — but do this first, because it's free and it works. When you do look at products, look at the label before the sales page. My Gluco6 review takes the most-advertised formula in the category apart ingredient by ingredient.
Disclosure: that review contains affiliate links, and it still scores the product 3.4 out of 5.
References
- Shukla AP, et al. The impact of food order on postprandial glycaemic excursions in prediabetes. Diabetes Obes Metab. 2019. PMID 30101510
- Shukla AP, et al. A Randomized Controlled Pilot Study of the Food Order Behavioral Intervention in Prediabetes. Nutrients. 2023. PMID 37892527
- Kashima H, et al. Suppression of sweet sensing with glucose, but not aspartame, delays gastric emptying and glycemic response. Nutr Res. 2019. PMID 31421394
Medical disclaimer. General information written by a layperson, not medical advice. Nothing here diagnoses, treats, cures or prevents any disease. If you take glucose-lowering medication, talk to your prescriber before making significant changes to how you eat.