So straight up: there are risks of consuming raw milk. Lets not pretend otherwise. Raw milk may contain pathogens which can cause pretty unpleasant infections including Salmonella and Listeria. Milk is an excellent medium for growing bacteria, and consequently there are outbreaks of food poisoning caused by raw milk and cheeses. The point of this post is to carefully weigh up those risks, putting them in context with other foodborne risks, so that, hopefully, we can make a more informed decision, and if we are so inclined, more effectively present the case for consumer choice to the regulatory authorities.
Some data on the prevalence of the more common and serious pathogens in European milk samples is shown below:
These figures indicate that pathogens are in the milk supply. But it is not clear if they tested raw milk outlets, or raw milk destined for pasteurisation. Either way, the figures do not seem very reliable. Belgium could not muster any numerical data apparently, and Europe as a whole seems to have a disproportionately high share of the world’s milk related infections. Do Europeans drink more milk than than other countries? or perhaps they are just better at reporting outbreaks. In their 2013 review of risks from milk-borne pathogens Claeys et al highlight the potential severity of these risks:
The consequences of a milk-borne infection may be limited to the common symptoms of diarrhea, vomiting, nausea, fever, abdominal cramps, etc., but a certain percentage of persons can develop more severe clinical symptoms such as Guillain-Barré syndrome (Campylobacter spp.) and hemolytic uremic syndrome (HUS) (E. coli O157:H7), or long-term and sometimes chronic complications, e.g. reactive arthritis, or even death.
Those at most risk are the immunocompromised, young children, the elderly and pregnant women.
Because of these potential risks most scientists, regulators and doctors advise against drinking raw milk. Their position is that pasteurisation is effective, makes milk safer and does not affect its nutritional qualities significantly – so why take the risk? On the other hand, the views of raw-milk advocates tend be the polar opposite: they claim the risks from raw milk are overblown, whilst pasteurisation damages the health-giving properties of milk, for example by denaturing enzymes and proteins. In England in the last 12 months, there has been a public consultation where both sides of the argument have been aired. Indeed I hosted a well attended meeting with the Food Standards Agency at Rosemary Cottage as part of the consultation (see previous post).
Are raw milk risks overblown?
So how do the claims on both sides stack up? Firstly lets look at the risks. Some of the most recent data on foodborne infections, albeit from the USA rather than Europe or Britain, come from the excitingly titled “Outbreak Alert! 2008” by the Centre for Science in the Public Interest. The following graph shows the number of cases of outbreaks (bars) and illnesses (points) caused by different foodstuffs in the USA between 1990 and 2006
The first thing to note is that the dairy data includes infections caused by both pasteurised and unpasteurised dairy products all rolled in together. We will try to tease them apart later. What is interesting, however, is that there are far more cases of infection from seafood and produce, poultry, beef, eggs and pork than dairy. Again, we can’t tell from this data whether these infections arose from raw or cooked items, but it is notable that each of these categories includes foods which are often bought, sold, stored, handled and prepared raw before being cooked. Produce (fruit and veg) and sea foods are also often eaten raw (think salads, fruit, sushi and oysters). Even cooked foods such as bread and bakery goods were a significant source of infection. Where cooked foods cause infections it is usually due to post-cooking contamination – a reminder that heat treatment, such as pasteurisation, does not guarantee safety at the point of consumption.
What this graph does not tell us is the severity of the illnesses associated with each category. Whilst the CSPI reports that approximately 5000 people die every year from foodborne infections in the USA, these are mostly from fruits, nuts, eggs, meat, poultry, fish and shellfish. Dairy related infections are usually mild, with only a ‘handful’ of deaths from pasteurised products, and not a single death from raw milk products since the 1980’s, despite an estimated 10 million Americans consuming raw milk regularly.
Despite this apparent vindication for the relative safety of raw milk, the Centre for Disease Control (CDC) recently published a study claiming “Majority of dairy-related disease outbreaks linked to raw milk “. They made the shocking claim that raw milk was 150 times more likely to lead to infection and 13 times as likely to lead to hospitalization than pasteurised milk. Worrying stuff! But is it justified?
Thankfully Chris Kresser has done a great job at deconstructing these headline provoking statistics. What he shows (and I won’t repeat it all here) is that:
- CDC included a large number of cases caused by Mexican-style Queso Fresco ‘bathtub cheese’ made illegally at home.
- They used an old estimate of 1% raw milk consumption, whereas the most recent survey (2007) placed the figure closer to 3% (strangely, this was carried out by the CDC too!)
The long and short of it: Chris Kresser reckons that the relative risk of raw compared to pasteurised milk is closer to ten to one. But before that puts you off raw milk, consider the absolute risks associated with milk consumption:
Annual chance of falling ill from milk consumption
Pasteurised milk: 0.00011 %
Raw milk: 0.00106 %
In other words the risk of illness from raw milk consumption is very small. furthermore, the estimated risk of serious illness (hospitalization) is even lower, at 1 in 6 million per year. For the purpose of comparison, you are three times more likely to die (not just be hospitalised) in a plane crash than by drinking unpasteurised milk!
Comparing risks from raw milk consumption to other foods
One of the problems with the graph above is that it does not take into account the quantity of each food that is consumed. When you consider that most people consume dairy most days, whereas seafood may only be consumed weekly, the relative risks look quite different. In fact dairy was only responsible for 1.3% of the foodborne illness each year in this report. Helpfully, the good people at The Centre for Science in the Public Interest have created a graph (below) that shows the risk corrected for consumption. Not only that but they have compared them to dairy-borne illnesses. Very convenient:
Of course the dairy data is still made up of pasteurised and unpasteurised milk combined. If we separated them the pasteurised dairy risk would drop, and the unpasteurised risk would increase. Exactly how much is a bit tricky to say, but based on the 10:1 risk difference, contributed by 3% raw milk consumers I estimate the ‘mixed’ risk is made up of 78% pasturised and 28% raw drinkers. That would make the bar for pasteurised milk drop to 0.28 and the bar for raw milk just under 8, placing the risk close to that of pork and below beef, eggs poultry and sea foods.
Furthermore, if we had the data (which we don’t) to split each of the other bars above into raw vs cooked we would likely see similar patterns with the cooked foods having a lower risk and the raw ones having a higher risk than the mixed data presented above. Chances are that raw milk would end up the least risky of all the raw food categories above, quite possibly it would be less risky than raw produce such as fruit and salads, including bean sprouts, that we all eat at least every week without a second thought.
Should it be ‘fresh’ milk or ‘raw’ milk?
Why have food regulators focused so heavily on raw milk when it is clear from the considerations above that all common food groups have a risk associated with the raw products?
It appears raw milk is being unfairly singled out and demonized in a way that other raw food ingredients in our shopping baskets have not. At the same time, advocates of raw milk can make extreme claims of the benefits of raw milk that lack good evidence. I’d like to suggest a way forward that I think provides a better fit with the facts. I think we need to see it not as raw, but as fresh.
Why do we call it ‘raw’ milk in the first place? We don’t call salad ‘raw’ lettuce. We don’t call a fresh cut of meat ‘raw’ meat. We call them fresh. Our fruit does not have to carry a label saying “This product is unpasteurised and may contain pathogens harmful to health”, yet there are far more deaths each year from infected fruit (e.g. melons) than from raw milk. And for all the health benefits we claim for raw milk, the moment we use it in our teas or coffees, or add it to recipes we cook it anyway.
Before you bite my head off, I’m not suggesting raw milk isn’t desirable. It is equally desirable in a healthy diet as are fresh vegetables and fresh cuts of meat – whether we cook them or not we still want them fresh. We do not buy the bulk of our meat and veg tinned, sterilised or irradiated, and many of us prefer salad that has not been washed with highly chlorinated water whether or not we eat these foods cooked or raw. The issues are substantially similar whether it is milk or carrots, beef steak or rocket. We do not want any of them pasteurised, homogenised or processed.
What does processing do to milk?
This table is remarkable as it summarises changes to the milk structure and quantity at each stage of processing. There is no doubt that oxidation of lipids, and denaturing of proteins and loss of some vitamins takes place. However, several of these steps (mechanical milking, cold storage) apply to raw milk purchased at the farm gate also.
However, the extensive review paper Does homogenization affect the human health properties of cow’s milk? (Michalski, 2006) from which this table comes, finds little evidence that any of these changes significantly reduce the health benefits of milk.
Imagine if we could replicate this kind of analysis for other foods. For example, we all know that tinned, frozen and fresh carrots taste different. How do the nutrients in them change when they were picked, cleaned, chilled, packed ‘in a protective atmosphere’, diced, tinned. I’m sure there would be drops in vitamins and changes to lipids and proteins too. One wonders why milk has had so much research compared to other foods.
Scrutinising health claims for raw milk
The officials who will decide for us whether we will be allowed to continue to purchase raw milk are often faced with a barrage of rather unsubstantiated claims from raw milk advocates, which does not necessarily help their cause.
So lets look at some of the health claims made for raw vs pasteurised milk.
1. Raw milk tastes better
This is somewhat subjective, but as a raw milk drinker I would say this is certainly true. There is evidence that almost every stage in milk processing affects the flavour of the milk. Many artisan cheese makers insist on using raw milk as pasteurised milk does not elicit the subtle flavour-notes that distinguish their products.
2. Raw milk is better tolerated or digested
Many anecdotal reports, but little hard data. In fact a Weston A Price Foundation funded study by Stamford University found no difference between raw and pasteurised milk on lactose intolerant families.
3. Raw milk contains more vitamins
True, B1, B6 , B12 & C are reduced by heat treatment. However, they are also reduced by cold storage so raw milk can still suffer.
4. The proteins and enzymes in raw milk are intact
True, but this is true for any food comparing raw to cooked. The point is whether this is important or not. Most proteins and enzymes are denatured just as much by changes in pH as they are by temperature. Stomach acid and digestive enzymes are going to hydrolyse dietary proteins and denature enzymes anyway. There is good evidence that many proteins are more bioavailable once cooked (see Richard Wrangham’s book ‘Catching Fire’).
5. Raw milk reduces allergies
There is quite strong evidence that raw/farm milk consumption in childhood is associated with a reduction in asthma and atopy (asthma, eczema, hay fever). The reasons are not yet understood. However, there is no evidence for a protective effect of raw milk on adult allergies. Unfortunately, trials are hampered by the obstacle of the ethics committees. As there is a known risk with raw milk, and children are one of the most at risk groups for milk-borne diseases, it is currently impossible to carry our randomised controlled trials to test this theory properly!
6. Live bacteria are probiotic
Whilst raw milk does contain probiotic bacteria, it is generally considered that they are far too few in number to provide a meaningful probiotic effect. However, many traditional societies ferment their milk before consumption and almost certainly increases the beneficial bacteria to useful levels. Remember, however, that milk contains many prebiotic components (lactose, milk fat globule membranes) that stimulate the growth of beneficial bacteria already present in the gu, but this effect is present even in pasteurised milk.
7. Raw milk contains more n3 PUFAs and CLA
Pasteurisataion does not destroy the essential fatty acids ALA, DHA or CLA. The quantity of these in milk is generally very low. However, organic and grassfed milk tend to contain considerably more of these health giving components, regardless of pasteurisation, and thank goodness for the fact that most of the dairy cows in Britain are grass fed for most of the year.
8. Pasteurised milk is at a greater risk of foodbourne pathogens
The risk of infection from raw milk has been shown time and again to be higher from raw milk. However, the common idea that pasteurised milk is sterile is incorrect. The most common procedure for pasteurisation is currently High Temp Short Time pasteurisation, where milk is heated to 73° for 15 seconds. This process kills virtually all free-living bacteria. However, it does not destroy pathogenic spores (E. coli, Paenibacillus spp etc). These can potentially grow rapidly if conditions are right. Some of these spore forming bacteria are psychrotrophic, meaning they can grow in refridgerator temperatures. In practice, shelf life of pasteurised milk is short enough to prevent them becoming a health risk for the most part.
More important is post-pasteurisation contamination. Listeria monocytogenes is present in raw milk, but is held in check by antimicrobial components and commensal (friendly) bacteria. Although Listeria is killed by pasteurisation, if it gains re-entry following pasteurisation it can grow rapidly as the inhibiting factors have been destroyed by the heat treatment. Evidence of this is that raising pasteurising temperatures increases the growth of bacteria during subsequent refrigeration storage, and increases the risk of listeria infections and deaths.
9. Raw milk is safer because its small scale, local and sustainable.
As UK law stands, you can only purchase raw milk direct from the farmer. However, internet sales mean they no longer need to be local. Either way, they have an interest in making sure the milk they supply is safe. Firstly, because if they make you ill you will not come back again; secondly, because by selling direct they make a bigger profit margin. Milk intended for wholesale is collected from multiple farms, mixed, pasteurised and homogenised by large dairy operatives and provides a very slim profit. Individual farmers do not need to be as careful about hygeine when producing milk for this system nor take the same pride in the way they run their dairy.
a) the milk may only be sold direct to consumers by registered milk production holdings (at the farm gate or in a farmhouse catering operation) or through milk roundsmen. Sales through other outlets have been banned since 1985 (although sales by the farmer at farmers markets are allowed);
b) the supplying animals must be from a herd that is officially tuberculosis free, and either brucellosis free or officially brucellosis free;
c) the production holding, milking premises and dairy, must comply with hygiene rules;
d) the milk must bear the appropriate health warning;
e) compliance with a) to d) above is monitored by inspections twice a year; and
f) the milk is sampled and tested quarterly under the control of the Agency to monitor compliance with standards for total bacterial count and coliforms.
This arrangement seem sensible and helps maintain the relationship between farmer and consumer. The FSAs current consultation on raw milk suggests these rules are likely to remain unchanged. Scotland, however has had a full ban on raw milk for some years now. The food standards agency claims:
The Scottish ban on sales of raw cows’ milk and cream was introduced in 1983 following a number of milk-related illnesses and 12 potentially associated deaths. The introduction of the ban resulted in a marked decline in milk-related illness, which has been maintained in subsequent years.
However, what they don’t say is whether the English system outlined above, has led to an equivalent decline in milk related illness also.
Take home message
- The benefits of raw milk are modest for most of us, but may be significant for some
- Any benefits of raw milk are questionable once used in tea/coffee/cooking
- The unprocessed nature of raw milk lends it the same desirability as fresh meat, fresh vegetables, fresh fruit & fresh fish
- Under the current regulations raw milk probably has similar risks as other fresh / raw meats, seafood, fruit and vegetabes.
- Problems may depend on milking practise and herd health: so know your supplier!