The Protein Gap – one of the biggest errors in nutritional science?


In the three decades after World War II it became an almost universal belief of nutritional scientists that the world could not produce anything like enough protein to meet human nutritional needs. This deficit in world protein supplies was expected to grow and this so-called protein gap seemed unbridgeable unless alternative large sources of protein could be developed. Primary protein deficiency manifesting as a disease called kwashiorkor was considered to be the number one nutritional problem in the world. These symptomatic cases of supposed protein deficiency were considered to be probably the tip of the iceberg with many millions more impaired to some extent by sub-clinical protein deficiency. Billions of pounds/dollars were spent on a raft of measures and programmes that were intended to increase the availability of protein for human consumption. By the mid1970s this belief in a massive deficit in protein supply was under serious challenge and it gradually became accepted that primary protein deficiency was an unlikely occurrence even in children with diets based upon low protein dietary starchy staples like yams and cassava. Protein deficiency is now seen as normally only likely as a secondary consequence of inadequate total food consumption i.e. starvation. Few nutritionists today think that primary protein deficiency is a substantial world problem. The protein gap had been closed not by increasing protein supplies but at the stroke of a pen by revising sharply downwards estimated human protein needs, particularly the needs of growing children.

Context and aims of this post

There is a substantial overlap between this post and my last one “Do children need more protein than adults”  because one of the main drivers of the protein gap concept was the belief that, on a weight adjusted basis, children needed much more protein than adults and thus a more protein-rich diet than adults. Ideally these two protein posts should be read as a pair.

I have written about the protein gap in newspaper articles, scientific papers and in several editions of my books. I included a discussion of the protein gap in the latest edition of my nutrition textbook (Webb, 2012) even though it could be regarded as a topic primarily of interest in the history of nutrition. I think that there are still important lessons to be learned from an analysis of the protein gap (see below).

  • If the estimates of nutritional requirements used to set dietary standards (e.g. RDA) are overstated then this can result in the mistaken classification of many adequately nourished people as deficient and the initiation of measures to correct this illusory problem. Past exaggeration of the protein requirements of children resulted in the false perception that primary protein deficiency was very prevalent amongst the world’s children.
  • Experiments with animals and non-human species have played a key role in most advances in human biology and medicine e.g. almost all Nobel Prize winners in physiology or medicine have made use of non-human species. Nevertheless, animal experiments do have the potential to mislead human biologists if the results are uncritically extrapolated to humans. Baby rats double their birthweight in six days and consume rat milk which is very rich in protein. Did experiments with laboratory rats contribute to the belief that human babies have a very high protein requirement even though they take months to double their birthweight and consume human milk which only has about a quarter of the protein content of rat milk?
  • The belief that protein deficiency was a major world problem and high protein intake was a key characteristic of a good diet has had a pervasive and persistent effect upon attitudes in human nutrition. Scientists rarely challenged this general scientific faith even though there were clear deficits in the supporting evidence. Dairy products and meat were once viewed very positively as important high protein components of human diets but are now more often seen as major sources of undesirable saturated fat. Vegetarians in particular are still seen by many as vulnerable to protein deficiency even though their diets almost certainly contain ample levels of protein.

Past belief in a protein gap and major past initiatives to close this gap

The term protein gap refers to the shortfall of protein supply that was apparent when estimates of protein availability for human consumption were compared to the inflated estimates of human (especially children’s) protein needs. There emerged a widespread and persistent belief that protein deficiency was:

The most serious and widespread dietary deficiency in the world” (Waterlow et al 1960)

And, that:

In many parts of the world the majority of young children suffered some protein malnutrition” (Trowell, 1954).

From the 1940s through to the early 1970s estimates of human protein needs seemed to be much greater than the protein available for human consumption; unequal world distribution of protein supplies was seen by some as an important exacerbating factor. This protein gap was predicted to increase as the world population grew and seemed to be unbridgeable unless alternative and previously untapped novel sources of protein could be found. This idea of protein deficiency as a major cause of worldwide malnutrition persisted for well over 2 decades; as late as 1972 the following statement was made by the chairman in his opening address to an international scientific conference:

Every doctor, nutritionist or political leader concerned with the problem of world hunger has now concluded that the major problem is one of protein malnutrition……The calorie supply [of developing countries] tends to be more or less satisfactory but what is lacking is protein, and especially protein containing the essential amino acids. Gounelle de Pontanel (1972)

In the 1950s, 1960s and 1970s, there were many very expensive projects aimed at increasing protein supplies from alternative sources to try and bridge the growing gap between the human requirement for protein and the worldwide supply. Billions of pounds at today’s prices were invested in these projects and as they were aimed primarily at solving an illusory problem (i.e. closing the protein gap) so most of this money and effort was wasted. These measures and their cost were discussed very thoroughly in a book by Kenneth Carpenter (1994) entitled Protein and Energy and they include:

  • Projects to develop cheap and palatable high protein foods that could be produced from locally available products like fishmeal, soybeans, peanuts, coconut, sesame and cotton seeds.
  • The development of single cell protein (SCP) from yeasts, algae and bacteria and this led to products like Quorn
  • Companies like BP initiated major research and development programmes to produce SCP from by-products of the oil refining process
  • The breeding of genetically enhanced plants with higher and better quality protein content; these generally gave lower yields and required more fertiliser inputs and so none of those developed were ever widely used
  • Addition of synthetic amino acids to plant proteins like wheat to increase their quality.

An agency of the United Nations, The Protein Advisory Group, was set up in 1955 to “Advise on the safety and suitability of new protein-rich food preparations”. My own PhD (1970-74), although not directly about protein, was connected to the development of alternative rich sources of protein for Third World children. This programme eventually led to the development of Quorn which is now marketed as an alternative to meat for affluent vegetarians rather than an alternative source of protein for malnourished Third World children.

The concept of a protein gap loses credibility

This belief in a protein gap was under serious challenge by the mid1970s and it gradually became accepted that this was an incorrect belief. Detailed analyses of the diets of children in the Third World indicated that primary deficiency of protein was uncommon and that any protein deficiency was usually a secondary consequence of low total food intake (see Waterlow and Payne, 1975). This downgrading of the importance of protein deficiency was recognised by Miller and Payne as early as 1969 when they concluded that almost all dietary staple foods contain sufficient protein to meet human needs and that even those diets that are based upon very low protein staples (like yams and cassava) are unlikely to be specifically protein deficient.  A paper published by Donald McLaren in the Lancet in 1974 seems to have been the watershed moment when the protein gap was openly challenged so forcefully and persuasively that it came to be disregarded. McLaren’s short paper is a very persuasive repudiation of this theory that had had so much influence upon nutrition research and education and upon aid programmes; he used the title “The great protein fiasco” for this paper. In a 2011 interview Professor McLaren  said that:

“In my opinion, the belief in the “Protein Gap” is one of the greatest errors committed in the name of nutrition science in the past half-century”

My personal enlightenment

I only became aware that this protein gap theory had been abandoned by most nutritionists when in 1986 I went to King’s College, London on a year’s study leave. I was shocked by the discovery that this theory that I had been taught as an undergraduate, and had acknowledged as a partial justification for my PhD research had been largely abandoned. This enlightenment not only triggered my researching and writing about the protein gap but also encouraged me to question other accepted scientific beliefs and to become very aware of the fallibility of scientists and current scientific practice.  My study of this topic was a major trigger for my subsequent writings about error and more recently fraud in biological and medical science. I believe that a critical analysis and discussion of past errors can help us to learn from these mistakes of the past.

What caused the protein gap mistake?

I have concluded from my analysis of the scientific literature that three major factors led to this false belief in a crisis of world protein supply.

  • The general belief that growing children needed a much higher protein concentration in their diets than adults. As discussed in my previous post, it was widely believed that children needed five times more protein for each kilogramme of body weight than adults. Requirements were further increased if only low quality proteins of vegetable origin were consumed.
  • It was assumed that a nutrition-related disease called kwashiorkor was due to primary protein deficiency i.e. it occurred when their diet contained enough calories but lacked sufficient protein.
  • Kwashiorkor was assumed to be the most prevalent form of malnutrition worldwide and these clinical cases represented only the tip of the iceberg with many more children impaired to some extent by lack of protein in their diet.

Exaggerated estimates of the protein needs of children

As discussed in my previous post, there was a massive overestimation of the protein needs of growing children in the early dietary standards. In the first (1943) edition of these standards, the estimated daily protein need of a 2 year old child was 40g/day but in the 10th edition (1989) the estimate was just 18g/day (see Webb, 1989). To get 40g/day of protein a child would need to get about 13% of their calories in the form of protein whereas to get 18g/day they only needed about 5% of their energy to come from protein. Similar calculations suggest that the protein needs of all adults should be met when 8% of dietary calories are in the form of protein. It would also have been assumed in the 1940s and 1950s that some of children’s protein needed to be first class protein from animal products or good quality vegetable proteins like those in peas and beans. Using the 1943 figure then primary protein deficiency looked highly likely and in some Third world countries where only low quality protein from cereals and root vegetables was eaten it would have seemed almost inevitable. If the 1989 figure is accepted then protein deficiency seems unlikely unless a child is starved of food generally. The emphasis on protein quality also diminished because people do not normally eat just a single dietary protein and when you mix dietary proteins of low quality, the overall quality of the mixture is higher than each of its constituents (referred to as mutual supplementation of proteins).

It was difficult to directly measure the protein needs of children in ethical experiments so much protein research was conducted using laboratory animals. You can limit the protein intake of young rats and test directly the point at which their growth is impaired but one could not ethically do the same to human babies. All of the evidence from laboratory and agricultural species suggested that weight for weight rapidly growing young animals required much more protein than adult animals (5 times as much in a rat). According to Hegsted (1959) cross species generalisations played an important part in setting the then very high estimates of children’s protein requirements. Yet most animals grow much faster than human babies and their milk is also much richer in protein than human milk so such cross species generalisations would not seem to be very appropriate in this case. Table 1 shows the growth rates of a number of laboratory and agricultural species and also the protein content of their milk. In common laboratory and agricultural species, protein makes up 19-27% of the total energy in the milk. In human and chimpanzee milk, the protein content is less than a third of this i.e. primate milk is very low in protein compared to other species.

Table 1            Growth rates and milk composition of seven species.

Species            Days to double weight    Milk protein (g/100g)          %calories as protein

Man                    120-180                                 1.1                                                     6

Cow                      47                                          3.2                                                    19

Sheep                   10                                          6.5                                                   24

Cat                         7                                           10.6                                                27

Rat                         6                                            8.1                                                    25

Mouse                  5                                             9.0                                                    21

Chimp                 100                                         1.2                                                    7

Simplified from: Webb (1989)


Kwashiorkor, due to primary protein deficiency, is the dominant form of worldwide malnutrition?

 The other factors that encouraged acceptance of a worldwide protein crisis was the belief that kwashiorkor was caused by eating a diet that was specifically deficient in protein and that this disease was the most common worldwide manifestation of malnutrition in children. Neither of these assumptions is now believed to be true and there was never much critical evaluation of the evidence used to support them. Once again Hegsted had prophetically warned in 1959 that there was a paucity of evidence for a specific causative role for protein deficiency in kwashiorkor. Kwashiorkor has a number of symptoms that could be theoretically explained by lack of protein or lack of the amino acids of which proteins are composed such as those listed below.

  • Oedema. There is swelling of the tissues due to an increased accumulation of fluid in the tissues which was traditionally said to be the result of depletion of the proteins in plasma. Plasma proteins normally exert and osmotic pressure which keeps fluid in blood and prevents it accumulating in the tissues.
  • Retention of some body fat stores. The fact that some body fat often remains in children with kwashiorkor suggested that it was not caused by a simple lack of energy.
  • Flaking of the skin with a loss of pigmentation (flaky paint syndrome). These symptoms could be explained by a lack of protein to replace skin and a lack of the amino acids necessary for production of the pigment melanin.
  • The liver may be enlarged with the accumulation of fat. Fat that is synthesised in the liver is normally transported away from the liver by being attached to protein to make it soluble in blood. Lack of protein to transport the fat could be used to explain this symptom.
  • Apathy and lack of appetite. Many nerve transmitters are synthesised from amino acids so perhaps lack of amino acids for transmitter synthesis is responsible for these psychological changes.

There was no convincing direct evidence that children with kwashiorkor were eating less protein-rich diets than children suffering from the other major form of childhood malnutrition called marasmus and when this was eventually tested it was found not to be the case (see Miller and Payne, 1975).

The assumption that kwashiorkor was the world’s most common manifestation of severe malnutrition in children was largely based upon 1952 survey results from parts of rural Africa where kwashiorkor was common. This limited survey was assumed to be typical of the whole world but that was certainly not the case in Asia for example where most of the world’s malnourished children were then located.

Lasting impact of the protein gap myth

As noted earlier, huge amounts of money and resources were wasted in trying to increase availability of protein in order to close the growing but illusory shortfall in world protein supply. If this money, resources and effort had been more usefully directed e.g. the provision of clean water and more calorie-rich food, then how much more might have been achieved?

This emphasis on protein probably increased the prestige of meat and dairy products which are now viewed as major sources of undesirable saturated fat. This positive emphasis on protein may have made it even more difficult for later health promoters to persuade us to eat less meat and dairy produce in order to moderate our saturated fat consumption. This belief in the importance of a plentiful supply of good quality protein in human diets persists to this day despite primary protein deficiency having been shown to be improbable even on Third World diets based on low protein dietary staples. The following quotes from a 1991 edition of a respected American nutrition text seem to sum up the ambivalent attitude of modern nutritionists to the importance of protein in human diets.

Protein is at the heart of a good diet. Menu planners build their meals around the RDA for protein.

Most people in the United States and Canada would find it almost impossible not to meet their protein requirements. (Hamilton et al, 1991)

Even though the second quote sums up the current consensus view of most orthodox nutritionists, there is still a widespread belief that high protein content is a highly desirable characteristic of a human diet especially for children. Affluent vegetarians are still warned or concerned about their protein adequacy even though most consume high protein foods like pulses, milk or eggs. Even without these high protein foods, protein is an unlikely limiting nutrient in mixed vegetarian diets.

Whilst scanning the internet for references to the protein gap, I came across a number of sources that were reporting the results of a survey of the diets of Indians which suggested that nine out of ten Indians lack proper protein intake including a report in The Times of India . As a British nutritionist this seems an unlikely proposition even allowing for my lack of any practical experience of diets in India. It is not clear from the accounts that I have seen how this survey was conducted or who commissioned it. It was conducted by an international market research company IMRB based in India and involved 1,260 Indians in several cities apparently using an interview rather than a direct assessment of food consumption.  In several of the commentaries a rule of thumb that adults need 1g per kg body weight per day is suggested; this seems high by UK standards (UK RNI is 0.75g/kg/day) even allowing for differences caused by protein quality differences. Note that the RNI is set well above average requirements because it is intended to indicate the requirement of those adults with the highest protein need. People put on low or very low protein diets seem to adapt and use protein more efficiently. Are Indians being classified as protein deficient because of exaggeration of their likely requirements? Some foods, particularly breakfast foods, are referred to as low protein (e.g. oats) but are not particularly low in protein (by my calculations over 12% of the energy in oats comes from protein). The general tone of these reports is very reminiscent of some of the material that I have read from the protein gap era in Britain and the USA and indeed the term protein gap actually occurs in some of these reports.  In table 2 the percentage of calories in various foods that are in the form of protein are listed.


Table 2 – The protein content of some foods expressed in absolute terms and as percentage of the energy yield (values are generally for raw food)


Food                            g protein/100g food                            %age of calories as protein


Whole milk                              3.3                                                       20.3

Skimmed milk                         3.4                                                       41.2

Human milk                            1.3                                                       7.5

Natural yoghurt                      5.0                                                       38.5

Egg                                              12.3                                                     33.5

Chicken (meat)                        20.5                                                     67.8

Cod (white fish)                      17.4                                                     91.6

Salmon (oily fish)                 18.4                                                     40.4

Quorn                                        11                                                        51.8

Wholemeal bread                  8.8                                                       16.3

White bread                            7.8                                                       13.4

Chapatti flour (white)           9.8                                                       11.8

Poppadum (raw)                     20.6                                                     30.3

Basmati rice (raw)                   7.4                                                       8.2

Oats (raw)                                  12.4                                                     12.4

Peas                                             5.8                                                       34.6

Broad beans                              4.1                                                       34.2

Lentils                                        23.8                                                     31.3

Canned baked beans               5.1                                                       31.0

Chickpeas                                   20.2                                                     25.3

Potatoes                                      2.1                                                       9.7

Mushrooms                                1.8                                                       55.4

Broccoli                                       3.3                                                       57.4

Cauliflower                                1.9                                                       58.5

Onions                                        0.9                                                       15.7

Carrots                                        0.7                                                       12.2

Okra                                            2.0                                                       47.1

Apple                                           0.3                                                       2.6

Tomato                                        0.9                                                       25.7

Banana                                        1.1                                                       5.6

Orange                                         0.8                                                       3.2

Peanuts (groundnuts)             24.3                                                     17.1

Almonds                                      16.9                                                     12.0

Walnuts                                        10.6                                                     8.1

Coconut                                         3.2                                                       3.6

To put these numbers in table 2 into context, the energy yield of the maximum adult protein requirement (RNI/RDA) is about 8% of the estimated average energy requirement. These values in the last column of table represent something akin to the protein concentration of foods. My above calculation suggests that a diet with a protein concentration of around 8% should be adequate for adults (note that similar calculations in my earlier post suggest a figure of nearer 5% for children). Some discussion points from this table:


  • The high percentage values for some vegetables like broccoli, cauliflower, okra and even tomatoes – this is because even though there is a relatively small absolute amount of protein it represents a high proportion of the energy yield because there are not many calories from other sources i.e. essentially no fat or starch and just a little sugar.
  • The relatively low percentage levels for nuts – this is because nuts contain relatively large amounts of high energy fats which may provide 80-90% of their calories.
  • The relatively low percentage levels of rice and potatoes – this because they are rich in starch which provides most of their calories.
  • Only a few foods on this list have less than 8% of their calories as protein; human milk, the ideal food for rapidly growing babies, is one such food!
  • Oils, fats and sugar are essentially protein-free so if one adds lots of these to a food then it will add calories but no protein and so reduce the percentage of the energy yield from protein i.e. it will tend to dilute the protein. It will also more generally reduce the nutrient density of the diet i.e. the amount of any nutrient per calorie in the diet.


Sources used – listed because most do not have internet links

Carpenter, KJ (1994) Protein and energy: a study of changing ideas in nutrition. Cambridge: Cambridge University Press.

Gounelle de Pontanel, H (1972) Chairman’s opening address. In Proteins from hydrocarbons.  The proceedings of the 1972 symposium at Aix-en-Provence. London: Academic Press pp.1-2.

Hamilton, EMN, Whitney, EN, and Sizer, FS (1991) Nutrition. Concepts and controversies. 5th edn. St. Paul, MN: West Publishing Company.

Hegsted, DM (1959) Protein requirement in man. Federation Proceedings 18, 1130-36.

McLaren, DS (1974) The great protein fiasco. Lancet ii, 93-96.

McLaren, DS (2011) The legacy of a life (interview by Jonathan Steffen). Sight and Life 25(3), 50-53.

Miller, DS and Payne, PR (1969) Assessment of protein requirements by nitrogen balance. Proceedings of the Nutrition Society 28, 225-234.

Times of India (June 4th 2015) Nine out of ten Indians lack proper protein intake.

Trowell, HC (1954) Kwashiorkor. Scientific American 191(6), 46-50.

Waterlow, JC, Cravioto, J and Stephen, JML (1960) Protein malnutrition in man.  Advances in Protein Chemistry 15, 131-238.

Waterlow, JC and Payne, PR (1975) The protein gap. Nature 258, 113-117.

Webb, GP (1989) The significance of protein in human nutrition.  Journal of Biological Education 23(2), 119-124.

Webb, G.P. (2012) Nutrition: maintaining and improving health. 4th edition. Oxford: Taylor and Francis.



Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s