Protein: g/kg vs % of Calories — Why Clinicians Prescribe by Weight

Why do guidelines prescribe protein in g/kg?

From the IOM RDA (0.8 g/kg) through ISSN (1.4–2.0), PROT-AGE (1.0–1.5), ESPEN (1.2–1.5), ASPEN and the Academy of Nutrition and Dietetics, every recommendation is expressed in grams per kilogram. Protein requirement tracks lean and body mass and clinical state — a person does not need more protein simply because they ate more carbohydrate that day. This is settled guideline consensus, not a contested claim.

A g/kg target also resolves to a stable number of grams that a meal plan can be built against. That stability is exactly what a prescription wants, and it is why downstream meal-planning targets are anchored to grams rather than to a percentage.

What is the AMDR — and why isn’t it a prescription?

The AMDR for protein is 10–35% of energy. Two facts make it unfit as the primary prescription engine. First, its upper bound is a residual: because carbohydrate (45–65%) and fat (20–35%) plus protein must total 100% of energy, the IOM set protein’s 35% ceiling to complement those ranges rather than deriving it from protein-requirement or protein-toxicity data (IOM/DRI; reaffirmed in the National Academies’ 2024 AMDR review). Second, the band is enormously wide in absolute terms.

• At a 2,500 kcal intake for a 75 kg adult, 10% of energy ≈ 0.8 g/kg and 35% ≈ 2.9 g/kg — roughly a 3.5-fold spread, and the spread widens further at lower calorie intakes.
• A “target” that spans that range is a guardrail, not a number you can prescribe against.
• The band says nothing about the person’s clinical state, training status or energy balance — the very things that set the real requirement.

What does a g/kg target look like as a % of energy?

Because protein contributes 4 kcal/g, any fixed g/kg target lands at a different percentage of energy at every calorie level — the same grams read as a higher share of a smaller diet. The table makes the relationship concrete for a 75 kg adult.

Why does a % target fail on hypocaloric and GLP-1 diets?

A fixed protein percentage moves grams in lockstep with calories. Cut energy and the grams fall — precisely when protein needs to stay high or rise to spare lean mass. A 2025 real-world intake survey of GLP-1 receptor agonist users illustrates the gap: average protein sat at just 17.5% of calories, and most participants fell short of the 1.2–2.0 g/kg gram-based target for their hypocaloric needs. That is descriptive intake data, not a controlled trial, but it shows exactly how a calorie-tracking share of energy under-delivers protein on a steep deficit.

The ISSN’s hypocaloric figure of 2.3–3.1 g/kg of fat-free mass (Helms 2014) — for lean, resistance-trained athletes in an aggressive cut — is expressed per kilogram of tissue, not as a percentage of energy, specifically because a percentage would push the number in the wrong direction. Holding protein grams constant (or raising them) while calories fall is the whole point.

How should you use % of energy? As a cross-check

The AMDR is still useful as a validating band, not an engine. Resolve the prescription in g/kg first, convert those grams to a percentage of the planned energy (protein grams × 4 kcal ÷ total kcal), then read the percentage as a sanity check.

Flag the extremes: a result below ~10% of energy suggests the plan is under-protein for its calories, while a figure above 35% is worth confirming as intentional for a hypocaloric or lean-mass goal rather than an error — recall from the table that a legitimate 2.0 g/kg hits 40% on a 1,500 kcal deficit. Used this way, the percentage catches mistakes without ever becoming a rival prescription method.