What Are Good Sources of Protein? – Protein Quality

Having finished looked at the issue of speed of digestion in What are good sources of protein – Speed of Digestion Part 3, I want to move onto the next topic that I mentioned in the introduction: protein quality.  I’m going to keep this article as brief as possible, for reasons I’ll explain at the end of the article.  If you want or need more, you can pick up The Protein Book which has a detailed discussion of the issue.


What Does Protein Quality Mean?

Quoting directly from The Protein Book:

Protein quality refers, in a general sense, to how well or poorly the body will use a given protein.   More technically, protein quality refers to how well the essential amino acid (EAA) profile of a protein matches the requirements of the body; the digestibility of the protein and bioavailability of the amino acids (AAs) also play a role (1,2).

Essentially, protein quality simply refers to how well or how poorly a given protein is used by the body once it has been digested.  Clearly, any protein that escapes digestion (as discussed in What are good sources of protein? – Digestibility) can’t do anything in the body but that doesn’t mean that all of the protein that is digested automatically works the same in the body.

Repeating myself slightly, protein quality has to do with how well a given dietary protein is used by the body for all of the different purposes that protein is used for.  And the quality of the protein has to do with factors such as the amino acid profile of the protein (amino acids are just the building blocks of individual proteins) along with the speed of digestion issue I discussed in the last series of articles.  I’ll talk about amino acid profile a little bit in the next article of this series.

Recall, for example, that whey protein, because of its rapid digestion, tends to promote amino acid oxidation (burning); obviously amino acids that are burned for energy can’t be used for things like synthesizing muscle tissue or what have you.

With that said, I want to take a brief look at all of the major methods of scoring protein quality.  Again, for a more detailed discussion, please pick up a copy of The Protein Book.

Method of Measuring Protein Quality

Chemical Score:The chemical score of a protein refers simply to its amino acid profile rated to some standard or reference protein, each amino acid is rated on a scale indicating how much of that amino acid is present compared to the reference protein.

For example, let’s say that the reference protein being used contains 100 milligrams of the amino acid leucine.  Let’s say that the protein we’re looking at contains only 75 mg of leucine; that protein would be given a chemical score of 75% for leucine; if instead it contained 125 mg of leucine, it would be given a chemical score of 125% for that amino acid.

Frankly, chemical score isn’t used very much anymore and the whole concept is based on knowing what the ideal protein for human health and function actually is.  Even in that case, chemical score says nothing about digestion or how a given protein is actually used by the body.

Biological Value(BV): BV is one of the more common methods of measuring protein quality and tends to be the one that is seen the most so I’m going to give it the most discussion.  BV is simply a measure of how much of the protein actually entering the bloodstream is retained in the body (e.g. used for proteins synthesis or what have you); that is it takes digestibility into account.  I’d note that some of the protein (again, researchers are actually measuring nitrogen going in vs. out but that’s not important here) that gets into the bloodstream comes back out in the urine.

Since BV is comparing protein in vs. out, the highest possible value for BV would be 100, that would mean that 100% of the protein that got into the bloodstream is being used by the body (note again, some protein won’t be digested in the first place).  No protein has a BV of 100 and claims that whey have a BV of 140 are simply nonsense (they are based on a misreading of a specific paper); this would suggest that for every gram of protein from whey that is eaten, the body somehow stores 1.4 grams of protein.  An impossibility.

BV is measured by feeding subjects a protein free diet for three days and then giving them a measured amount of protein, the amount that comes back out in the urine and poop and skin and such are then estimated and BV is calculated.  This type of study is called a nitrogen balance study and, for a variety of reasons can be very inaccurate.  Again, more detail can be found in The Protein Book.

I’d note that BV is typically tested at very low protein intakes, far below what the average American (and certainly any athlete would eat).  Eating more protein lowers the apparent BV which has led to some humorously bad interpretations of BV. As well overall energy intake drastically affects BV; if you eat more calories, apparent BV goes up, if you eat less, apparent BV goes down.

Because of this, BV has a lot of practical problems.  It’s very accurate under conditions of low protein intake but caloric has to be meticulously controlled. At the types of high protein intakes seen in most modern countries, as well as with athletes, BV doesn’t tend to say very much.

Net Protein Utilization (NPU): NPU is extremely similar to BV.  But while BV is comparing the amount of protein that is actually digested to the amount that is stored in the body, NPU simply compares the amount of protein eaten to the amount stored in the body.  Put differently, BV takes digestion and actual absorption of protein into account; NPU doesn’t.  This doesn’t make NPU very useful.

Protein Efficiency Ratio (PER): PER is a measure of the amount of weight gain (in grams) in rats compared to their protein intake.  It’s always measured in young growing rats and, frankly, has about zero relevance to human physiology.

Protein Digestibility Corrected Amino Acid Score (PDCAAS): The PDCAAS is the newest method of scoring protein quality and is the one most in common use.  Like chemical score it compares the amino acid profile to some reference protein; as well it it takes into account digestion.  Somewhat interestingly, proteins that were scored as low quality (such as soy protein) achieved a much higher score via the PDCAAS.  This is actually in line with research showing that quality soy proteins work just fine for supporting basic human protein needs.

PDCAAS does have a couple of problems, however.  The first is that the highest score possible is set at 1.0, no protein can score above that value regardless of the apparent quality.  Basically, scores higher than 1 are simply rounded back down.

Additionally, part of the premise of the PDCAAS is that the ideal pattern of amino acids for supporting human health (or athletic performance) is actually known.  It’s possible that the ideal protein for supporting basic human health might change with age (for example, amino acid and protein requirements do change with age) or might be different different types of athletes.  The idea that a single amino acid profile can be ideal under all circumstances is tenuous at best.


Does Protein Quality Matter?

Which brings me to my major commentary about the issue of protein quality: I consider it essentially irrelevant.  I noted above that BV, for example, is measured at very low levels of protein intake and this tends to hold true for many of the other methods.   Protein quality is measured under conditions of low intake because the primary application of protein quality has to do with ensuring adequate nutrition for people who don’t have enough food.   Which means that it stops having much relevance at high intakes.

That is to say, small differences in protein quality make an absolutely massive difference if you’re talking about someone in the third world who is eating small amounts of a single source of poor quality protein and doing so in the context of insufficient total caloric intake in the first place.

In that situation, small improvements in protein quality (by adding other foods or even a specific amino acid) may pay massive dividends in terms of improving health or survival of that group.  So would feeding them in general. The World Health Organization (WHO) is very concerned with this issue which is why they make a big deal out of protein quality; it’s relevant to the population that they are worried about.

Anybody reading this article isn’t in that situation for the most part.  If you have the Internet and time to visit my site, odds are that finding food in general, or protein in specific, to eat today isn’t your major concern.  When people are consuming mixed proteins at the levels seen in the general public (typically 2-3 times the RDA level), and especially among athletes (who often eat more than that), protein quality ceases to become an issue.  This is especially true when lots of calories are being eaten.

And while it’s possible that specific proteins might be more or less useful for athletic applications (e.g. providing amino acids specifically needed by those athletes), any athlete eating large amounts of protein will generally be consuming plenty of what they need anyhow.

Athletes tend to get really obsessive about the issue (and of course supplement companies pander to that) but at a protein intake of 1-1.5 grams per pound of lean body mass coming from mixed high quality sources, quality just won’t matter.  There is much more discussion of this in The Protein Book.

One possible exception to this is dieting; when calories are restricted, the way the body uses protein can change and different proteins may be specifically beneficial (the dairy proteins whey, casein or simply milk are valuable in this regards for reasons outside the scope of this article).

I suppose if someone in the modern world was eating small amounts of a single source of poor quality protein, quality would matter as well.  But that would be some weird self-imposed dietary pattern, not the kind that people in many parts of the world follow because that’s all tha is available.

The bottom line is that, for folks in the modern world, eating fairly large amounts of high quality proteins and lots of calories, quality simply isn’t much of an issue in terms of answering the question what are good sources of protein.  Outside of a few weird exceptions, noted above, it simply won’t be relevant.

Next time, I’ll discuss an issue related to this one which is the specific amino acid profile of proteins.

Go to What Are Good Sources of Protein – Amino Acid Profile Part 1



10 thoughts on “What Are Good Sources of Protein? – Protein Quality

  1. So is to early to be inquiring about the Leucine they’ve been pusing on T-nation? Specifically, do you think consuming Leucine with every meal would improve protein synthesis as they suggest, would require less protein to be consumed in general, and/or do you think its completely unnecessary if I’m already getting 1-1.5 g proteing/ lb of body weight from whole food sources?

  2. Tim

    Wait until the next bit, amino acid profile. I’ll touch on this a bit.

    But short answer: of course it’s crap. Typical t-nation, bodybuilding comic book crap.


  3. Interesting post, would like to seek clarification on the PDCAAS as your comments here don’t really match my own understanding.

    My understanding is that the PDCAAS uses essential amino acid requirements for a 2-5 year old as a reference, with 1.0 representing the maximum score as post-digestion this protein provides 100% or more of essential amino requirements and amino acids beyond this point exceed requirements and are excreted or stored as body fat rather than utilised for protein synthesis. So even if the food (whey, egg and some meats would fall into this category) could supply more amino acids in excess of requirement, the score is still 1.0, and rounded down for the reasons stated above.

    The utility of proteins with a “true” PDCAAS score of greater than 1.0 is in complementing other proteins of lesser quality so obviously the PDCAAS score is a reflection of the test protein individually/in isolation, and the true usefulness of a protein in the diet is probably better reflected by a combination of the PDCAAS and the biological value and other measures.

    From what I understand, the amino acid requirement patterns for 2-5 year old are elevated for all essential amino acids when compared to adolesecent and adult populations, and as such a PDCAAS score of 1.0 using 2-5 year olds as a reference pattern indicates that the protein will be equally if not better digested in other populations (see “Energy and Protein Requirements, Report of the Joint FAO/WHO/UNU Expert Consultation” at http://www.fao.org/docrep/003/AA040E/AA040E05.htm#ch5.6 for the table of figures), and that the amino acid profile will be sufficient to meet requirements, though it could easily be in excess and is therefore obviously not exact.

    As the study you mentined hinted at, proteins like soy protein with a PDCAAS of 1.0 if consumed as a sole protein source can fulfill protein requirements when consumed in the requisite amount (daily protein requirement in grams). Obviously this is purely hypothetical as no one protein source is likely to be the sole source in the diet.

    Also, my understanding was that NPU = (True digestibilty X BV) and therefore DOES take into account digestibilty while in fact BV is a measure of the protein that HAS been absorbed and how it is used, but DOES NOT account for protein that has not been absorbed and therefore does not account for absorption/digestibilty (the definitions of BV and NPU you give seem almost in reverse to what I understand).

    I’d be interested to hear if you could expand on the effects of protein feeding freqency, low calorie diets and other factors similar to those mentioned here http://www.fao.org/DOCREP/MEETING/004/M2835E/M2835E00.HTM on protein absorption.

    Also, for endurance athletes with high CHO and protein requirements do you think it possible that protein quality can be an issue as most protein is coming from low quality sources e.g. wheat and large volumes of these foods are almost a necessity, while protein intakes are still up to double that of a normal population.



  4. Just like to clarify that my above post (the long one) is by a different person to the first response, apparently we share the same name though.

  5. Tim (Tim?), now I”m confused.

    Realize that I’m trying to keep this article from turning into my protein book so I left out some details. Assuming the Wiki is remotely accurate, PDCAAS is defined as

    “Using the PDCAAS method, the protein quality rankings are determined by comparing the amino acid profile of the specific food protein against a standard amino acid profile with the highest possible score being a 1.0.”

    To my knowledge, it’s not just the essentials but I”d have to dig into the literature to be sure of that.

    The reason that the AA profile of a 2-5 year old was chosen is because they show the highest overall requirements for essential AA’s (older individuals actually show lower requirements in both absolute terms and amino acid requirements).

    This represents the fact that kids are growing at a fast rate (relatively speaking). So any protein that meets their requirements should meet (in theory anyhow) the needs of adults.

    I’d note that this doesn’t necessarily apply to athletes (none of the methods do, it’s entirely possible that a specific athlete requires more or less of a given amino acid, this is discussed in detail in the Protein Book of course). So rounding to 1.0 based on assumptions about growing kids may not be relevant to folks involved in heavy training. Since that latter group is what I primarily focus on, that’s who I base some of my conclusions on.

    As well, there is some literature suggesting that oxidation of amino acids has its own beneficial effects so rounding b/c oxidation increases above certain levels may be a flawed idea as well.

    My definition of NPU comes from Groff and Gropper’s nutrition text and states “NPU measures retention of food nitrogen consumed rather than retention of food nitrogen absorbed.”

    Meal frequency is detailed (an entire chapter) in the Protein Book although I’ll probably touch on it eventually. You might check the research review on the site about Meal Frequency.

    While endurance athletes often engage in goofy eating patterns, I’d be surprised if any lived on single poor sources of protein like that.

    Thanks for the comment,

  6. Lyle,

    This may be not the place, but I have to ask you what you were referring to when you said: “But short answer: of course it’s crap. Typical t-nation, bodybuilding comic book crap.”

    I respect BOTH you and T-Nation (which is by far the best site on the Internet and has great minds on it) so I am very curious to hear your insights on where you think T-Nation (or perhaps one of their authors) is mistaken.

  7. The idea that adding a bit of overpriced leucine to a meal will do jack squat to a lifter eating lots of protein is typical bodybuilding comic book bullshit. But that’s all T-nation has ever been, a bodybuilding comic book aimed at selling overpriced ineffective supplements to people who still think that there are magic pills to be had.

    I’ll address leucine in the final part of the Amino Acid Profile series but the short-answer is that adding a couple of grams to meals or post-workout shakes is just so much bullshit. Just like 99% of what’s in the sports supplement industry.

    That’s what I meant by it being typical T-nation crap.


  8. T-Nation is a business. A business purpose is to make money. That is T-Nations goal. To make money. Just look at there wayyyy over priced supplements.

  9. Bodyrecomposition.com is also a business and my purpose is also to make a living. But I do it without lying back and forth to people to sell scams and overpriced snake oil to them. So the defense of T-nation that they are just a business is so much crap even if I see people making it all the time.

Leave a Comment