Select Page

The Background

If you are a health conscious consumer, you would have heard of, how A2 milk from humped cattle like Gir, Ongole, Sahilwal is better than the A1 milk from breeds like Holstein Friesian (HF). The A2 milk movement has gained a lot of steam in India in the last few years with milk being sold at extremely high prices like Rs.150 per litre and ghee at a whopping 4000 per kg in some cases. Being a sustainable farmer, natural product store owner and a health enthusiast i had the urge to verify this claim in an unbiased manner. So I have been extensively reading all possible scientific research on the subject for the last few years. In this article I will bring out the hypothesis of A1 and A2 milk followed by the scientific research and present my conclusions on the topic that could help you make better choices for your health.

Hypothesis of A2 milk

The classification of milk as A1 and A2 is based on the type of beta-casein present in the milk. I have been reading this line for quite some time without understanding much. So I took the help of my brother in law Bharat to create this infographic below which would make your life easier!

Basically A1 and A2 type is decided based on the type of beta-casein and the primary difference being the 67th position of 209 amino acid chain. Histidine at 67 is A1 and Proline at 67 is A2. Now the major culprit as per the pro A2 milk brigade is the BCM7 which is a 7 amino acid long chain produced during the digestion of A1 beta-casein. Various studies have been published linking the role of BCM7 in disease.

Scientific Evidence

The first arguments in favour of A2 milk have been made basis two studies which correlated the average per capita consumption of A1 beta-casein and the incidence of certain diseases in various countries. One of the studies was published by Laugesen and Elliott and the other by MacLachlan (its interesting that MacLachlan eventually starts the A2 Milk Corporation company which holds patents for A1/A2 gene testing of herds). The graphs showing the correlation in these studies are shown below:

There are two major problems with this kind of studies. Firstly ‘Correlation’ and ‘Causation’ are two different things. Secondly both the diseases discussed above namely diabetes and heart disease are multifactoral diseases.

Correlation and Causation are not the same!

Correlation studies can be often misleading. For example two parameters that have very good correlation can be absolutely unrelated. A classic example of a bad correlation study is the diet heart hypothesis by Ancel Keys which linked dietary cholesterol to heart disease, and resulted in Indians considering ghee as a risk factor, only to realise later that dietary cholesterol has no bearing on the heart disease, as indicated by the latest2015 USDA guidelines. And correlation theories can always be countered by anomalies. For example in Switzerland, Type1 diabetes has increased three-fold since 1990 but milk protein consumption has not changed (Crawford et al, 2003). Just to make this more easier, Jessiein Chauspe shows how extremely high correlation could mean nothing.

In this case if we just go by the correlation between internet usage in a country and breast cancer death rate per 10000 women, we will end up with a conclusion that internet usage causes breast cancer!!

Mutifactoral Diseases

Both Diabetes and Heart disease are multifactoral diseases which simply means that they have multiple cause factors. For example its highly likely that heart disease is caused by smoking, alcohol or stress as compared to BCM7. So, to say that incidence of heart disease in certain countries is higher due to BCM7 from A1 milk is not very scientific. Similarly type1 diabetes incidence could be genetic or due to viruses.

Further studies were published by research made on rabbits by feeding them A1 protein. This study was done on 60 rabbits for a period of 6 weeks with 10 subgroups. It essentially means that the number of rabbits tested for A1 protein induced disease were 6. Also rabbits generally eat leaves and milk protein would be an artificial diet for them. So this study would not be of great consequence.

Major Blow to the A2 movement

As the A2 movement gained momentum the European Food Safety Authority (EFSA) did an analysis of the claims and published a report. The EFSA’s report, issued in 2009, rejected most of the evidence. It found that in “most if not all” animal studies prior to its publication, scientists had injected BCM-7 directly into their animal test subjects, rather than administering it orally, which in the EFSA’s view rendered the results irrelevant to human consumption. Subsequent to the EFSA report the A2 industry tilted their positioning from healthy heart and anti-diabetic towards easy digestion and tolerance.

More recent studies

However there is one animal study published in 2014 by researchers from National Dairy Research Institute, Karnal in India which related BCM7 to inflammatory response in mice.

Finally two studies involving humans were published in 2014 and 2016. The 2014 study one of whose authors is Dr.Woodford K (former stakeholder of A2 milk corporation till 2007) had observed positive association between A1 milk and abdominal pain. However the authors themselves concluded that there is a need for larger confirmation of participants as the sample size was only 41 of which only 10 reported intolerance to commercial cow milk upon administration of 750 ml a day.

The 2016 study studied the effect of A1 and A2 beta casein proteins on gastrointestinal physiology, symptoms of discomfort. However even this study was done only on 45 subjects and was based on association. The authors were establishing a link between discomfort and A1 protein. However it is to be noted that lactose intolerance had been a major study area for gastrointestinal discomfort and protein type has no bearing on the lactose intolerance as it due to deficiency of lactose digesting hormone lactase.

On the whole it is interesting to see that there is no single human study which establishes the causative risk of BCM7 on human health. This is remarkable given the fact that it has been almost 25 years since the debate started. Further most research that is pro A2 has been supported by the A2 milk corporation which is a private entity holding patents on A2 genetic testing.


  1. The biggest take away for me is that there is absolutely no reason for us to pay premium for ghee from A2 milk (assuming all other factors like feed, water, grass remain the same) as the whole debate is about milk protein and ghee has NEGLIGIBLE protein in it!
  2. The A2 milk phenomenon is not limited to only desi Indian cows. A lot of foreign breeds like Guernesy, Jersey from Channel area and Asian, African cattle have a predominant A2 gene frequency. So demonising certain breeds of cows that underwent a genetic mutation 7000 years ago when humans were not as greedy as they are today, is not very logical.
  3. Luckily in India a lot of milk comes from the water buffalo (Bubalus bubalis) and it has been proven by genetic studies that all buffalo breeds of India produce only A2 milk. So feel free to consume buffalo milk.
  4. When we consume animal products we eat what the animal eats. So most A1 milk comes from factory dairy farms where corn and soy is used as feed resulting in increased milk yields. I feel this is more important than the genetic mutation. So always prefer milk from animals that are free range and are grass fed. Further all the medicinal benefits of cow milk/ghee came from the fact that they grazed on forest lands which had a high biodiversity of medicinal properties. So it does not matter much if the cow is A1/A2 if all it is eating is corn and soy!
  5. One of the major problems with the milk is the number of antibiotic and hormone injections the cattle is given. So this needs to be considered while choosing the source of the milk.
  6. A lot of modern entrepreneurs who promote A2 milk do abstain from use of injections and industry feed. This would tilt my balance a little more in their favour. However if someone is doing the same with a buffalo or a jersey cow, then its fine to support them too!
  7. Most nutrients and beneficial bacteria in milk are lost due to processes like homogenisation and pasteurisation. So its better to buy unprocessed whole fat milk from a local farmer who gives real milk