One meal a day or two: what changes

Eating once a day is not difficult to survive, and the trials below put adults on a single evening meal for eight weeks without incident. The question worth answering is what has to move when the eating window shrinks: the protein dose in one sitting, the fat load a stomach will pass at a sensible rate, and the morning blood glucose that follows. The evidence on those three is better than most diet arguments, and it does not all point one way.

The 30 g ceiling was an artefact of the measurement window

Muscle protein synthesis was long said to top out at 20 to 30 g of protein per meal. That number came from tracer studies that sampled for two or three hours and then stopped looking. Trommelen and colleagues at Maastricht ran the longer version: a quadruple isotope tracer feeding study, 100 g of protein in a single bolus against 25 g, sampled over twelve hours [1].

The 100 g dose produced a greater and more prolonged anabolic response, running past twelve hours, with dietary amino acids still arriving in muscle protein at the end of it. Whole-body protein breakdown did not rise to compensate, and amino acid oxidation barely moved, so the surplus was not being burned off as the older model assumed. Their conclusion is that the anabolic response to a protein feed has no demonstrated upper limit in magnitude or duration, and that it had been underestimated.

For someone eating once a day this removes the strongest objection to the pattern. A single meal of beef and eggs can carry 100 g of protein or more in one sitting and put most of it to work, rather than sending three quarters of it to the liver as the 30 g rule implied.

Fat, not protein, is what limits a single meal

Emptying rate is set by calories per millilitre, and the relationship is close to linear. Calbet and MacLean gave six healthy subjects 600 ml test solutions at 0.1, 0.2 and 0.7 kcal/ml and measured half-emptying times of 9.4 minutes for the dilute glucose solution, 16.3 and 17.2 minutes for the two peptide hydrolysates, and 26.4 minutes for the milk protein solution [2]. Caloric density against half-time gave a correlation of 0.96.

The numbers are from liquid test meals, so do not read them as a stopwatch on a steak. The direction is what matters. The same calories delivered more densely sit in the stomach for longer, and fat is the densest thing on the plate, at roughly twice the calories per gram of protein and more than twice that of cooked meat by volume.

That is the arithmetic behind the pattern people actually report: as the window narrows there is less room for lean volume, so the fat share rises even when the protein target stays fixed. Two meals leave space for a large salad or a bowl of vegetables. One meal at the same calories rarely does. The practical limit on a single sitting is not appetite in the abstract, it is how much fat the stomach will release per hour before the meal sits there.

What eight weeks of one meal does to the measurements

Stote and colleagues ran a randomised crossover with two eight-week periods. The same volunteers ate all their weight-maintenance calories as either three meals a day or one meal a day, with the single meal eaten in the evening. They held their weight within 2 kg across the six months of the study. Hunger rose on the one-meal arm. Fat mass fell. Blood pressure rose, total, LDL and HDL cholesterol all rose, and cortisol fell [3].

The glucose data from the same cohort is the part worth keeping. Carlson and colleagues ran morning oral glucose tolerance tests after each period. On one meal a day, fasting plasma glucose was higher, the glucose rise during the test was greater and more sustained, and the insulin response was delayed. Ghrelin rose in response to the single-meal regimen. The impairment reversed when the volunteers went back to three meals [4].

These were normal-weight, middle-aged adults eating to weight maintenance, so the trial isolates meal pattern from weight change. It says nothing about what one meal a day does during a deliberate fat loss phase, and it should not be read as a verdict on weight.

Shrinking the window did not make people eat less

The TREAT trial is the largest clean test of the assumption underneath most intermittent fasting advice. One hundred and sixteen adults with a BMI of 27 to 43 were randomised to eat ad libitum between noon and 8pm, or to three structured meals a day, for twelve weeks on a study app [5].

Weight fell 0.94 kg on time-restricted eating and 0.68 kg on three meals. The between-group difference was 0.26 kg, with a 95% confidence interval from -1.30 to 0.78, which is a result compatible with no difference at all. Estimated energy intake did not differ between the groups. Appendicular lean mass index fell more on the restricted arm, by 0.16 kg/m2.

If the mechanism was supposed to be that a shorter window means fewer calories, twelve weeks of it produced no measurable difference in intake. Whatever the 16:8 pattern does for some people, it is not that.

Meal frequency and body composition

Schoenfeld and colleagues pooled fifteen experimental studies on meal frequency and body composition. Feeding frequency came out positively associated with fat loss, body fat percentage and fat-free mass, which is the result the small-frequent-meals camp has been quoting for years. Their sensitivity analysis then showed that the positive finding was driven by one study, which leaves the association without support [6].

What I would do with it

Two meals a day is the easier default, and the trial data above is why. The morning glucose shift on a single evening meal is a measured change rather than a feeling, and it is the one you can check for yourself with a meter and a fasting reading, which is a better use of a fortnight than guessing at satiety.

One meal a day is defensible for people who prefer it, given the protein data, but it needs two deliberate choices. Put the protein in first, because the fat that fills the rest of the plate does not carry any. And expect the first ten days to feel heavy after eating rather than hungry before it, which is gastric emptying doing what the Calbet numbers describe.

If you take insulin or a sulfonylurea, changing when you eat changes when a hypo can land, and the timing follows the meal rather than the clock. That is a conversation with your GP or diabetes team before the first week, not a self-experiment with a meter afterwards.

Stote's volunteers held their weight within 2 kg on both patterns for six months, and their LDL cholesterol still rose on the one-meal arm. Frequency is a tool, not a free variable.