PDCAAS vs DIAAS: What Protein Quality Scores Can and Cannot Tell You

PDCAAS and DIAAS are protein-quality scores, not another way of stating grams of protein. PDCAAS uses an indispensable-amino-acid pattern and a single overall measure of protein digestibility. DIAAS uses digestibility estimates for individual indispensable amino acids at the end of the small intestine. A DIAAS figure can also exceed 100 for an individual ingredient, whereas conventional PDCAAS is capped at 100. This distinction is useful in technical nutrition work, but a higher score does not automatically make an entire food, meal or dietary pattern healthier.

Mealisa already has a general introduction to protein digestibility and quality. This guide is deliberately narrower and more technical: it shows the arithmetic behind a score, explains why the reference age group changes the result, and separates the FAO recommendation from the actual rules used on food labels. All figures in the worked example describe a fictional test protein, not soy, milk, wheat or any marketed product.

PDCAAS versus DIAAS: the methodological differences

The FAO expert consultation report, Food and Nutrition Paper 92 recommended DIAAS as a replacement for PDCAAS in protein-quality evaluation. That recommendation is not the same as a universal legal change: some regulatory systems still use PDCAAS, and ordinary EU protein-content claims operate on a separate basis. A 2024 scientific review also notes continuing gaps in reference data and measurement methods.

What is the limiting amino acid?

Humans need a set of indispensable amino acids from their diet. A scoring pattern specifies how many milligrams of each are needed per gram of reference protein for a defined group. To compare a food protein with that pattern, divide its corresponding amino-acid content by the reference value. The lowest ratio is the limiting amino-acid ratio. The calculation is not a judgement about taste, price or whole-food quality.

For a simplified PDCAAS calculation, multiply that lowest uncorrected ratio by overall true faecal protein digestibility, then cap the reported score at 1.00. DIAAS first adjusts each amino acid by its own true ileal digestibility, compares each adjusted amount with the reference, and uses the lowest adjusted ratio. In both cases it is important to know precisely which requirement pattern and analytical method are being used.

If a cereal protein has relatively little lysine, lysine may be limiting. Some legume proteins have different limiting amino-acid patterns. Those broad observations do not supply a numerical PDCAAS or DIAAS for a particular bread, stew or supplement. A real score needs appropriate measurements for the actual test material. For another explanation of protein composition, see our complete-versus-incomplete protein guide.

Worked example: a fictional protein, nine amino-acid groups

Consider a hypothetical protein powder called Sample X. The values below were invented solely to demonstrate the calculation. They are not measured facts about any food. The reference column uses the FAO pattern for people older than three years; the nine indispensable amino-acid groups include the paired sulfur-containing and aromatic groups. Every content value is expressed in milligrams per gram of test protein, not per 100 g of food.

Step 1: calculate the uncorrected amino-acid score. Lysine has the lowest gross ratio: 50 ÷ 48 = 1.0417, or 104.2%. Leucine is 65 ÷ 61 = 1.0656; all other gross ratios in this particular imaginary dataset are higher. Thus lysine is the limiting amino acid before applying digestibility.

Step 2: apply the fictional overall faecal digestibility of 90%. The traditional PDCAAS calculation becomes 1.0417 × 0.90 = 0.9375. Expressed as a percentage, that is 93.8%, rounded to one decimal. Since the result is already below 100%, truncation does not change it. No separate adjustment is made for lysine's relatively low digestibility in this method.

Step 3: apply each fictional ileal coefficient. Digestible lysine is 50 × 0.78 = 39 mg per gram of test protein. Compared with the older-than-three reference, 39 ÷ 48 × 100 = 81.3%. The other eight adjusted ratios appear in the final column, and none is smaller. Sample X therefore has an illustrative DIAAS of 81.3% for this age pattern. The gap between 93.8% and 81.3% comes from the two different digestibility approaches in this constructed example—not evidence of a difference for any real food.

Change the reference group and the answer changes

The FAO report provides age-specific scoring patterns. For children aged roughly six months to three years, the recommended lysine reference is 57 mg/g rather than the 48 mg/g used in the pattern for older children and adults. Holding this fictional sample's digestible lysine fixed at 39 mg/g, its lysine ratio becomes 39 ÷ 57 × 100 = 68.4%. In the same illustrative table, none of the other eight amino-acid ratios falls below that, so the illustrative DIAAS becomes 68.4% with the younger-child pattern.

These are teaching calculations, not published scores or regulatory certifications. FAO also discusses a distinct reference-pattern choice for regulatory applications, including the younger-child pattern for many foods. Do not compare scores from different publications without checking their age pattern and methodology. An apparently precise difference between two numbers may disappear or reverse when the reference groups differ.

Why one number can hide useful information

A DIAAS is the minimum across indispensable-amino-acid ratios, not their average. Sample X has 81.3% for lysine and 140.6% for histidine in the older-than-three example. The single score highlights lysine as the bottleneck; it does not mean every amino acid is only 81.3% available. That distinction matters when discussing protein blends: another ingredient could complement a relatively weak amino-acid profile, but you cannot calculate the blend's DIAAS simply by averaging two published scores.

Conversely, a high individual-ingredient DIAAS does not say how much protein the serving contains. A food providing 3 g protein and one providing 25 g may have similar quality scores but very different contributions to a meal. The protein in a mixed meal comes from each ingredient's quantity and amino-acid contribution, not from multiplying total protein grams by a single internet score.

For example, if a food tracker records 20 g protein, mechanically multiplying by our fictional DIAAS of 81.3% to declare 16.3 g of “real protein” is not a valid nutrition-label conversion. The score is a relative indicator of a limiting digestible amino acid under a specific reference pattern. It is not a simple correction factor for all proteins or for a complete day's food. For practical food choices, our high-protein foods guide focuses on quantities and portions rather than laboratory scores.

PDCAAS 100 and DIAAS above 100 are not contradictions

A separate limitation concerns truncation. If an imaginary protein had an untruncated PDCAAS of 1.14, its conventional reported PDCAAS would be 1.00 (100%). An individual-ingredient DIAAS, if supported by separate measurements, may instead remain above 100. The two values need not come from the same underlying measurements. The figure above 100 means the limiting digestible indispensable amino acid exceeds the selected reference amount per gram of protein; it emphatically does not mean the food is more than 100% protein or that the body absorbs more than all of it.

The FAO framework makes an important distinction between single ingredients and calculations for mixed diets or sole-source foods, where scores above 100 are treated differently. Processing can also alter lysine availability, especially when heating causes reactions between amino acids and sugars. A score for a raw ingredient should not be assumed identical to the score of a heavily processed finished product.

Where published DIAAS numbers can mislead

Some published digestibility values rely on growing pigs or other experimental models because direct human ileal measurements are difficult. Availability of human-specific data remains limited, and processing, antinutritional factors, food structure and analytical techniques can change the result. These uncertainties are discussed in the 2024 methodological review and a review of limitations for plant-based diets. They do not make protein science useless, but they do mean a ranked table of scores from incompatible studies is less reliable than it looks.

When reading a claim such as “protein A scores 110 and protein B scores 75,” check the actual material, preparation state, reference age group, test model, year and whether the score was truncated. Separate isolate results from whole foods. A soy isolate, cooked beans, bread and a complete dinner are not interchangeable test units. Avoid describing a plant-based diet as inadequate on the basis of a single ingredient's score; well-planned mixtures can provide adequate indispensable amino acids.

Food-label rules: FAO recommendations versus US and EU requirements

FAO recommendations are methodological guidance, not automatic legislation. In the United States, FDA guidance on Nutrition Facts points to 21 CFR 101.9(c)(7)(ii), which uses PDCAAS when calculating protein percent Daily Value in the specified labeling context. This does not mean every product is required to print a DIAAS or PDCAAS number alongside grams of protein.

In the European Union, the annex to Regulation (EC) No 1924/2006 defines the claims “source of protein” and “high protein” by the proportion of the food's energy supplied by protein: at least 12% and 20%, respectively. These thresholds are not DIAAS cutoffs. If a hypothetical product provides 15 g protein in 400 kcal, its protein contributes approximately 15 × 4 = 60 kcal, or 15% of total energy. On that arithmetic alone it meets the energy threshold for a “source of protein” claim but not the 20% threshold for “high protein”; all other applicable labeling conditions still matter.

Do not mistake a proposed quality-score category in a research paper for a currently authorized EU claim. Nor should “DIAAS 110” on marketing material be accepted without the underlying amino-acid data and digestibility method. Claims and scientific measurements answer different questions.

How to use these scores without overinterpreting them

Protein quality becomes particularly important when studying diets with limited protein quantity, formulating sole-source foods, or comparing ingredients whose indispensable-amino-acid profiles and digestibility differ. It can be useful to food scientists and professionals, but everyday meal choices should also consider total protein, dietary variety, calories, fiber, allergies, budget and preferences. Our guide to high-protein vegan foods gives practical alternatives rather than implying that one isolated score defines a whole diet.

Use this short reading sequence for any published score: identify whether it is PDCAAS or DIAAS; confirm the reference pattern and age group; confirm whether the result is truncated; check if the food is an ingredient or prepared product; and ask for the actual data behind the limiting amino acid. If any of these are missing, a neat-looking percentage may not be comparable to another one.

Sources

Primary basis: FAO Expert Consultation, 2013; 2024 ten-year review; FDA regulatory labeling guidance; EU nutrition-claims regulation. Every Sample X number is an explicit educational assumption and is not an estimate of a real food's protein quality.

• review of limitations for plant-based diets