What is the actual evidence behind 1.6 g/kg, and does it still apply when I am in a 40% deficit?

1.6 g/kg is quoted as though it were a physical constant. I would like to know what study it came from, what population, and whether the conditions of that study resemble mine, because I suspect they do not.

My situation: 86 kg male, 45, on retatrutide in a research context, running what I estimate is a 35-40% energy deficit because appetite is essentially gone, resistance training four times a week. If 1.6 g/kg was derived in people eating at maintenance and trying to grow, it is not obvious to me that the same number is the right one for someone eating at 60% of maintenance and trying not to shrink.

Is there a deficit-specific number, and is there any trial that actually randomised protein intake during a large deficit and measured body composition rather than nitrogen balance?

Your suspicion is correct on both counts: 1.6 g/kg came from energy-balance hypertrophy data, and the deficit-specific evidence points higher.

Where 1.6 comes from

The figure is the breakpoint from a dose-response meta-analysis of resistance-training studies, where the regression of lean-mass gain against protein intake flattened at approximately 1.62 g/kg/day [1]. Two features of that derivation matter for you:

• The included studies were overwhelmingly conducted at or near energy balance. Nobody in that dataset was in a 40% deficit.
• The outcome was gain in lean mass, and the breakpoint is the point past which additional protein stopped buying additional gain. A breakpoint for maximising gain in energy balance is not the same quantity as the intake required to minimise loss in a large deficit.

The confidence interval around 1.62 was also wide, with the upper bound above 2.2 g/kg, which is routinely dropped when the number is quoted.

The deficit-specific randomised evidence

The trial you are asking for exists: young men on a 40% energy deficit for four weeks with intense resistance and interval training, randomised to 1.2 versus 2.4 g/kg/day, with body composition measured by DXA [2]. Result: the higher-protein group gained roughly 1.2 kg of lean mass while losing about 4.8 kg of fat; the lower-protein group was roughly lean-mass neutral at about +0.1 kg and lost about 3.5 kg of fat. Both groups trained identically and ate the same energy.

Notice that 1.2 g/kg was not catastrophic, it was merely worse. And notice that the group at 2.4 g/kg gained lean tissue in a 40% deficit, which is the strongest available demonstration that the deficit itself is not the binding constraint if protein and loading are handled.

Two caveats before you extrapolate: the participants were young, they were previously untrained or lightly trained, and the trial ran four weeks. Untrained young men gaining lean mass under any stimulus is the easiest result in exercise science. At 45, trained, over months, you should expect retention rather than gain. But the direction of the protein effect is well supported.

What number to use

Your body weight:                86 kg
1.6 g/kg (energy-balance figure): 1.6 x 86 = 138 g
2.0 g/kg:                         2.0 x 86 = 172 g
2.4 g/kg (deficit trial arm):     2.4 x 86 = 206 g
Per-meal at 3 eating events, 2.0 g/kg: 172 / 3 = 57 g
Per-meal at 4 eating events, 2.0 g/kg: 172 / 4 = 43 g

For a large deficit with serious resistance training, the defensible range is 1.8-2.4 g/kg of body weight, or equivalently about 2.3-3.1 g/kg of fat-free mass, which is the range recommended for lean athletes dieting hard [3]. Treat 1.6 as your floor, not your target.

Two important qualifications for your specific case. First, if you are carrying meaningful fat mass, scale to fat-free mass rather than body weight or the body-weight figure will over-shoot. Second, and more importantly: at 206 g of protein you would be at 824 kcal from protein alone. If your total intake is genuinely 60% of maintenance, the higher end of that range may be arithmetically impossible without protein displacing all other food. In that situation the right move is to raise total energy toward a 20-25% deficit rather than push protein to the exclusion of everything else. The Longland design worked because the participants ate the prescribed energy; a 40% deficit with adequate protein is a very different thing from a 40% deficit produced by pharmacological aversion where you eat whatever you can manage.

Kidney function is worth checking at the top of that range if you have any reason to suspect impairment; in people with normal renal function, high protein intakes in this range have not been shown to cause harm, but "in people with normal renal function" is doing real work in that sentence and it is a blood test, not an assumption.

edited 13 Mar 2026 by swirl_dont_shake — tightened the wording; no substantive change

A counter-argument in the interest of balance, because the thread is heading toward "more protein is always better" and the evidence for the top of that range is thinner than the enthusiasm.

Points against pushing to 2.4 g/kg for most people in this situation:

• Displacement. On a constrained energy budget, every extra gram of protein displaces fat or carbohydrate. Below a certain carbohydrate intake, training quality falls, and training quality is a larger determinant of lean retention than the marginal 30 g of protein. Trading the second for the first is a net loss.
• Diminishing returns are real. The gap between 1.2 and 2.4 g/kg in the deficit trial was about 1.1 kg of lean mass over four weeks in untrained young men under near-perfect conditions. The gap between 1.8 and 2.4 has never been isolated and is presumably much smaller.
• Adherence. A target you hit 90% of the time at 1.8 beats a target you hit 55% of the time at 2.4. Under appetite suppression the achievable target is the correct target.
• Satiety. Protein is the most satiating macronutrient. On a drug whose problem is that you already cannot eat, loading protein makes the total-intake problem worse. This is the opposite of the usual dieting situation and it inverts the usual advice.

My reading of the same evidence: 1.8-2.0 g/kg of body weight is the sweet spot for a person in a substantial deficit with resistance training, and effort beyond that is better spent on total energy, sleep and progressive overload. The last of those three has by far the largest effect size and costs nothing.

To the sub-question about nitrogen balance versus body composition, since it is the reason this literature is so muddled: most of the older protein-requirement work, including the basis for the official recommended intakes, used nitrogen balance. Nitrogen balance answers "what intake prevents net loss of body nitrogen in a person at energy balance", which is roughly a starvation-avoidance question. It is not the same as "what intake preserves contractile tissue in a person in a deficit who is loading their muscles".

That gap is why the official recommendation of 0.8 g/kg and the sports-science figure of 1.6-2.4 g/kg can coexist without either being fraudulent. They answer different questions. Anyone who cites 0.8 g/kg at you as evidence that higher intakes are unnecessary is comparing a floor for preventing deficiency against a target for optimising an outcome.

The methodological upgrade that resolved a lot of this was the indicator amino acid oxidation approach, which produced requirement estimates roughly 40-50% higher than nitrogen balance for the same populations. When you see a study quoted at you, checking which of the three methods it used tells you most of what you need to know about whether its number is comparable to the one you are using.