Protein Quality Explained: PDCAAS, DIAAS, and Leucine

Quality means two things at once

Protein quality scores combine two separate questions into a single number. The first is compositional: does this protein contain the nine essential amino acids in proportions that match human requirements? The second is practical: how much of it do you actually absorb?

The first question is what the complete-versus-incomplete distinction captures. A protein short of one essential amino acid is limited by it, because that amino acid caps how much of the rest can be used for building tissue. Grains are typically limited by lysine, legumes by the sulphur amino acids methionine and cysteine.

The second question is why two complete proteins can still score differently. Some proteins are digested more completely than others, and some plant foods carry compounds - fibre, phytates, tannins - that reduce how much of their amino acid content is absorbed. Cooking and processing shift this too, sometimes substantially.

A quality score is the product of those two: how good the profile is, discounted by how much of it gets through.

• Composition: are all nine essential amino acids there in adequate proportion?
• Digestibility: how much of it is actually absorbed?
• A score multiplies the two - a great profile poorly absorbed scores badly.

PDCAAS: the long-standing standard

PDCAAS stands for Protein Digestibility-Corrected Amino Acid Score. Adopted as the reference method in 1993, it takes the limiting amino acid's ratio against a reference requirement pattern and multiplies it by the protein's overall digestibility.

It has one famous quirk: scores are truncated at 1.0. Any protein that meets or exceeds requirements is recorded as 1.0, no matter how far past the threshold it goes. That is why whey, casein, egg white, milk protein and soy protein isolate all sit at the maximum and look identical on the scale, even though they are not.

The truncation was a deliberate choice - the logic being that excess amino acids beyond requirement are not useful as protein - but it flattens exactly the comparisons people most want to make. PDCAAS also uses faecal digestibility measured across the whole gut, which can overestimate how much was really absorbed, because gut bacteria in the large intestine alter the picture after absorption has already finished.

DIAAS: the newer method

DIAAS - Digestible Indispensable Amino Acid Score - was proposed by an FAO expert consultation in 2013 to address both weaknesses. It measures digestibility of each individual amino acid at the end of the small intestine, where absorption actually happens, rather than inferring it from what leaves the body. And it does not truncate.

The result is a scale that separates foods PDCAAS lumped together. Dairy proteins come out highest, above the ceiling PDCAAS would have capped them at. Egg protein scores high. Soy sits below dairy but clearly above most other plant foods. Wheat protein scores poorly, which is unsurprising given its lysine limitation. Pulses fall in between and vary a good deal by preparation.

DIAAS is a better instrument, but it is not yet a universally applied one - measuring amino acid digestibility at the ileum is demanding work, so published values cover far fewer foods than PDCAAS does. In practice you will meet both scores, sometimes on the same package.

• PDCAAS: 1993 standard, truncated at 1.0, whole-gut digestibility.
• DIAAS: 2013 proposal, no truncation, measured at the end of the small intestine.
• DIAAS separates dairy, egg and soy proteins that PDCAAS scores identically.
• Neither score says anything about how much protein a serving contains.

Why leucine gets singled out

Leucine is one of the nine essential amino acids and one of the three branched-chain amino acids, but it also has a signalling role that the others do not share to the same degree: it is a trigger for the cellular machinery that initiates muscle protein synthesis. Amino acids are the building material; leucine is part of the start signal.

This is where the idea of a leucine threshold comes from - the proposal that a meal needs a certain minimum of leucine before that response is fully switched on. Whey is unusually leucine-rich, which is a large part of why it performs so well in short-term studies, and dairy proteins generally sit high.

Two caveats keep this in proportion. First, leucine on its own is a signal without materials; supplementing leucine while eating too little total protein does not achieve much. Second, the threshold idea comes mostly from acute measurements taken over a few hours, and the relationship between those measurements and long-term outcomes is looser than it is often presented to be.

For someone eating ordinary meals containing 25-35 g of mixed protein, the leucine content takes care of itself. It becomes a live consideration mainly at very small doses, or for people building meals from a narrow set of lower-leucine plant sources.

How much should quality change what you eat?

Less than the amount of attention it receives. Quality scores are most useful in two situations: comparing isolated protein products against each other, and evaluating diets that are narrow enough for one limiting amino acid to genuinely bite.

For a mixed diet, the scoring largely resolves itself. Different foods have different limiting amino acids, and eating several of them across a day means each covers the others' gaps. This is the same complementation principle that makes rice and beans work, applied across a whole diet rather than a single plate.

The place quality really does matter is at the margins: a fully plant-based diet built narrowly on grains and vegetables, a very low total protein intake where every gram has to count, or a product marketed as a protein source that scores badly. Collagen is the standard cautionary example - high in total protein per scoop, no tryptophan at all, and therefore not a substitute for a complete source no matter what the label says.

The blunt version: quantity first, variety second, quality scores third. Getting 30 g of protein from a decent mixed meal contributes more to a day than 15 g from a theoretically superior source.

• Comparing two isolates? Quality scores are genuinely useful.
• Eating a varied mixed diet? It mostly takes care of itself.
• Narrow plant-based diet or very low intake? Worth paying attention.
• Never let a quality score override the per-serving amount.