Cattle Attack Statistics: an alternative view.

The trampling season?

In this opinion piece, one of our COWS members, David, challenges the official UK statistics, and offers his own risk analysis.


If you have found this website and opened this article, you may well have recently had an encounter with cattle and are trying to find out more.

You are not alone. I experienced a very close and personal encounter with a herd in July 2014, from which I was lucky to survive, and I too spent time during my convalescence researching the issue to find out what I could have done to avoid the attack.

There is no shortage of official guidance – from the Health and Safety Executive (HSE), from the National Farmers’ Union (NFU), and from the Ramblers. These organisations warn of the risks inherent in cattle, and recommend safety precautions for both farmers and walkers. But, after serious and fatal attacks on members of the public, reported statements from HSE and NFU are full of platitudes, and downplay attacks to the extent of denying that cattle pose any risk to the public.

On the other hand, on my discharge from intensive care, one of the consultants gave me the best advice I have come across, which was to stay out of fields with cows and calves because “it was trampling season”.

Killer cattle -every cow is a potential killer

So, why is there a disconnect? Why isn’t the common knowledge of the medical profession – that cows are dangerous – shared by those organisations which should be closely involved in understanding the risks? As a general summary of the issues I will discuss in this article, this one aspect stood out from my research.

It seems to me that official organisations control statistics (such as they are) about cattle attacks and misuse the statistics which are available. The HSE and NFU cite the size of the cattle herd (9.8 million), and 300,000 farms, the predicted number of countryside visits by the public (2.93 billion) and the actual fatalities (between 1 and 2 members of the public each year, not forgetting 5 farm workers). Their inference is that the statistical likelihood of being subject to any cattle attack appears to be so negligible as to be discounted, and so preventative measures are a waste of effort.

In comparison, attention is drawn to 3 people killed by lightning each year and the 400 pedestrian deaths. Lightning strikes can be thought of as an Act of God, beyond even the scope of the HSE to control. As for the pedestrians killed by cars, I found a blog from Oil & Gas Safety which cites that 400 figure for pedestrian deaths (as well as 24,000 injuries caused by cars) of which 316 pedestrians were killed when crossing the road (assessed at 4 times a day). From this the risk of injury was calculated at 200,000 to 1. Needless to say, a lot of road safety engineering and car restrictions have been put in place to reduce the risks to that level. Remember that risk assessment (200,000 to 1) when it comes to considering cattle risks.

Given the figures cited by the HSE and NFU on the risk of cattle attacks, what is the problem? It can indeed be fairly claimed that the risk to the public appears remote in the extreme. But, if this is a non-existent issue, why were cattle attacks discussed in Parliament, why did a recent Coroner’s Inquest in Wiltshire call for a public debate, and why did my consultant talk about the “trampling season”?

It is worth digging into those figures to see if they reflect the real position.

Is the official and broad-brush comparison of numbers a valid approach or is it a matter of “lies, damned lies and statistics”?

This article considers the statistics by putting forward an alternative assessment for discussion. One purpose of this website is to set up a reporting system so more specific statics are available and so better judgements can be made about the risks.

Note: This article is primarily from the standpoint of a member of the public and not the farming industry, whose members are also at risk but who should be in greater control of the hazards.


By way of introduction, I will refer to some of the official pronouncements after the publicity generated when David Blunkett was injured by a cow.

HSE statement after cow attack

Tony Mitchell, of the Heath and Safety Executive, says that Cattle are classed as a non-dangerous species and by and large are generally docile. Their inquisitive nature is often mistaken for aggression… when maternal instincts are aroused, then they may behave in a threatening manner. This stock statement seems to be on HSE autocue. It is issued in almost identical form by other HSE spokesman after later attacks.

NFU statements after cow attack

An unnamed spokesman from the National Farmers’ Union gives the advice that if felt threatened “just to carry on as normal…. and remember to close the gate”.   Another NFU person, Ed Rees, thinks it sufficient to just take a walking stick …be bold and walk straight through them” and another suggests using a walking pole like a sword and stabbing the cow’s nose like a matador.

The passive and neutral terms used and advice given (and from my personal experience entirely optimistic and unpractical suggestions) are interesting and deliberate. Why not say what is actually the truth; that cattle can be dangerous.

The Animals Act, 1971, defines “dangerous animals” as “an animal not normally domesticated which if unrestrained could cause serious injury“. This definition appears to exclude cattle. But, the NFU is wrong in saying “Cattle are classed as a non-dangerous species.” Cattle are not classified as “non-dangerous”. They are just not classified as “dangerous animals” which is different in emphasis. A so-called non-dangerous animal is just as capable of inflicting serious injury.

When I was attacked…

When I was the victim of an unprovoked and coordinated herd attack by a group of cows, I was pleased to learn from the HSE that I was just mistaking the cows’ natural inquisitiveness as a deliberate attack. I was also told that the trampling was just a casual threat on the part of the cattle, because if the herd had intended to carry out their threat, then I really would have been in trouble!

Referring back to the NFU spokesman’s advice after the Blunkett attack: just to carry on as normal…. and remember to close the gate.” I clearly did not follow his advice. Acting normally when being charged by a herd was not an option, and the furthest thing from my mind while being airlifted away from the field was to worry if I had left a gate open.

Neither did I “just take a walking stick …be bold and walk straight through them.” The walking stick I was carrying did not survive the first hit of the skirmish and although fencing at school may have helped me to take out one beast, the other 24 would have been beyond my skills.

The HSE and NFU spokesmen either knew nothing about the real nature of cattle attacks (in which case, why were they acting as spokesmen?) or their statements were deliberately misleading.


Of the claimed 2.93 billion annual visits to enjoy the countryside, how many of those people actually encounter any cattle? What is the basis for that figure?

The 2.93 billion figure comes from one of the annual Natural England MENE studies into public visits to a natural environment, and has nothing to do with visits down to the farm. Natural environments include urban parks, the seaside, woods and similar, none of which is typically classed as cattle grazing land. For instance, visits to the coast include sitting in a car looking at the sea view or eating at a promenade restaurant, so I think from these caveats that juxtaposing a statistic including those activities against cattle attacks stretches credulity.

The closest most people will get to cattle will be sitting in the beer garden of The Bull public house.

But the MENE study has some points of interest. Of the total visits, it is assessed some 28% were to urban parks and 17% were on footpaths or bridleways which may be where cattle could be encountered. But within that percentage many footpath walks will also be in towns or on formal paths, forest or lake walks, rather than across farms. Even so, many farms are specialist arable or sheep farms, rather than for cattle. Cattle are not uniformly distributed across the UK. The predominantly arable areas of the south and east side of the country have lower cattle density than Wales, northern areas of England and south and east Scotland.

So, to use the figure 2.93 billion and imply that represents countryside visits is misleading. The estimated figure for public encounters with cattle is not a figure I can find.


There are 9.8 million head of cattle, and it is accepted that not all cattle are rampaging through the countryside causing havoc, and indeed most may normally be docile. But that is not really the issue. The risk posed by each cattle herd should be individually assessed, based against overall figures and similar cattle herds.

No insurance actuary assesses the accident risk of teenage drivers based on accident rates for 50-year-old women drivers. They look at peer-group statistics.

Using DEFRA (Department of Environment, Food, and Rural affairs) figures, it is estimated that 20% of cattle are permanently kept indoors, and will not encounter the public. Indeed, cattle are housed over 4 to 5 winter months and so any potential cattle encountering annual country visits should be reduced to reflect seasonal variation where cattle encounters are precluded.

The herd total includes a mix between dairy and beef cattle, bulls, breeding stock, young stock and calves. Just as for car drivers, each group presents different characteristics and risks. There are also different cattle breeds, farm management and handling methods which may incline cattle towards the docile or aggressive ends of the spectrum.

So a simple headline total is misleading for a complex mix of reasons and figures. The hazard and risk a cattle herd presents should be based on comparable herds, not a general total figure. You don’t equate and expect the same performance from a Porsche and a Ford Fiesta just because they are both cars.

Limousin suckler cows are dangerous


What seems to be anecdotally accepted (although at this time statistics do not seem available to clearly evidence the point) is that within the overall numbers, there is a group of cattle which seem likely to behave far more aggressively than the common herd.

What are grouped as “continental” breed cattle, as opposed to traditional British breeds, are regarded as more unpredictable and less calm and docile in behaviour. Unfortunately, these more highly strung cattle also have a greater body mass and capacity to cause injury than British breeds.

These continental breeds, including Limousin, Charolais and Simmental, were introduced in the UK in the 1970’s and now make up much of the beef herd, although full figures are difficult to obtain. The British Limousin Cattle Society estimates that breed accounts for 32% of the beef herd and the Limousin gene is present in 75%. Interestingly the Limousin Society has an article setting out the efforts of stockmen to breed more docile cattle, which presupposes there is a recognised issue with the lack of docility in the breed.

Continental breed beef cows kept as suckler herds (cows with calves at foot) are regarded by the farming community itself as posing a particular risk of aggression. It is recognised that individual suckler cows that may otherwise be docile can change character because of maternal instincts with their calves, leading to protective and aggressive behaviour. As was given in expert evidence at the 2015 inquest into the death of Mr Porter “A farmer who puts a suckler herd in a field with a footpath is aware there is a high risk of attack to the public”.

That situation is made worse in those suckler herds that are extensively farmed and left to themselves. According to Defra, some 60% of such herds are on remote moorland areas where public encounters are rare. With little farmer and public contact, familiarisation with humans is limited.


According to data from the AHDB (Agriculture and Horticulture Development Board), of the total 9.8 million head of cattle, only some 1.57 million are breeding beef cows, of which I suspect most are Continental-breed suckler cows. Of these, 60% are kept in remote areas away from regular human contact, giving an estimated number of 942,000 cows who could be considered to pose the highest risk to humans.

So, by looking a bit further into the figures, and applying some simple risk analysis, the “official” headline risk of a public death of some 9,800,000 to 1 can be reduced to some 942,000 to 1.

But even that analysis is incomplete. The simple number assessment considering cows as individuals, and not as a co-ordinated herd, is wrong. Cows are kept in herds.

The majority of stories one reads about are whole herd attacks and not a single rogue cow. There may be one particularly aggressive, hormonal leader cow, with the rest just following to attack out of group cooperation and to protect the herd as a whole. But the lead cows may change as different individual cows go through different phases in the maternal calving cycle. Although it seems spring and summer predominates for calving, in fact calving occurs all year round.

With different cows each month feeling particularly maternal, and taking the aggressive lead, the whole herd is permanently on a war footing.

According to AHDB, the average size of a suckler herd is 27. If we take the 942,000 cows that pose the greatest risk (the remote-breeding, individual suckler cows), we can divide them into attack groups of 27. This gives us 35,000 potentially dangerous herds that can put an individual member of the public at risk.

In other words, the risk of a member of the public being killed by an individual herd is not 9,800,000 to 1, but more like 35,000 to 1. That risk can be referred back to the 200,000 to 1 risk for pedestrian fatalities.


The official figures almost universally concentrate on fatalities. Death is indeed an inconvenience to the deceased, more so to family and dependents left behind. Less often discussed is the number of injured victims, which can range from potentially fatal but recoverable injuries, permanent disabilities, broken bones and simple bruising, down to mental trauma from situations where escape was made.

Those situations are less often discussed for the simple reason that it is difficult to get a grasp of the number in those categories.

Whatever the victim suffered, it can be assured that their future walking habits, and enjoyment of the countryside, will have been permanently altered and constrained.

One issue that official bodies seem not to understand or deliberately under-play, as exampled in the neutral and passive language cited above, is that all cattle attacks and incidents are potentially fatal. Being biffed by ¾ tonne of beef from a single cow is no trivial matter, let along being trampled multiple times by a herd. Once the herd shows “inquisitiveness” towards an interloper into their domain, the severity of injury suffered, or fortune in escaping unscathed, is simply a matter of luck.

Cattle do not think that they will just gently “threaten” one interloper to teach a lesson, but go harder on another. The herd intention is to eliminate a threat.

So, more numbers…

I have seen a Scoping Review about the risks of cattle to the public, based on media report searches over 20 years up to 2013. A figure of 54 attacks was found, of which about a quarter were fatal (13). So that gives some 41 media-reported injury incidents, or roughly 2 a year. That number seriously understates the issue because it is known from HSE that there were more injuries.

A BBC news report in 2009 (after the David Blunkett incident) gave a figure obtained from the HSE for the previous 8 years of 18 deaths (2 a year) and 481 serious injuries (60 a year). The 18 deaths is the figure usually reported for public deaths over the 15 year period up to 2013 (so clearly not the same statistic as in the Scoping Review above). I assume the 481 injuries just refers to members of the public, and not to farm workers. It is not known how serious those injuries were, but for the HSE to have been involved and recorded the injury number it can be assumed the severity was at the higher end of the scale.

The HSE concede that there is likely to be an under-reporting of injuries due to lack of awareness of the need for farmers to report a cattle attack injury as a workplace injury under RIDDOR.

Another BBC News publication from 2014 (Perils of the English Countryside) is aimed at the public and gives various figures for different animal and other hazards. I do not know the source of the figures but in the years between 2001 and 2011 the deaths from bulls is stated as 15, major injuries as 77 and injuries requiring treatment as 82. Equivalent figures for cows are 17, 362 and 328 respectively. So cows attack more, but maybe that’s because there are more of them.

These BBC figures give us a total of nearly 900 incidents. The total average annual figures are roughly 3 deaths, 40 serious injuries and 37 lesser injuries, but all of these 80 annual incidents had the potential to be fatal. Those figures are similar in magnitude to the HSE figures cited previously.

David cows 1


But many incidents involving cattle and the public cause no specific injury. The member of the public escapes the field, or does not even enter, or with the benefit of a walking stick walks boldly through to scatter the submissive herd, closing the gate behind.

The numbers of those incidents is even more difficult to obtain. Subjectively, just as there is a multiplier between deaths and the greater number of injuries (x 25), so it is to be expected that the ratio of injury to escape would escalate.

After cattle attacks reported in the media, typically many comments are submitted by members of the public relating to similar personal incidents where they had a lucky escape.

Ramblers have a Pathwatch reporting line for footpath problems and Ramblers have provided me with figures for 2015. The public logged 469 incidents where a path was “inconvenient” to use because of the presence of cows or bulls, and 114 incidents where the path was “unusable” because of cows or bulls.

These are not injury reports, just nearly 600 ramblers or countryside visitors who in one single year were sufficiently disgruntled to go to the trouble of reporting concerns because cattle in some way reduced or prevented their use of a public footpath.


So make of these figures what you will. I am not a statistician and fault may well be found with my logic and methods, but I suggest fault can also be found in the logic and methods of official organisations which collate the statistics and partially represent conclusions drawn from those figures.

It is clear that the issue of cattle attack is not limited to a simple 1 in 9.7 million chance, and from such headlines as “Killer Cows” it seems the media and public are becoming increasingly aware that cattle incidents are relatively frequent and apparently becoming more frequent.

There are published figures available which give a glimpse of the overall extent of the hazard of cattle, which indeed merits a public or parliamentary debate. What is also needed is greater research and refinement of the figures.

You can help by reporting any incidents on this web site.

However, none of this further research should delay action being taken to reduce what is clearly a known and significant problem.

In later papers what improved action that can be taken will be discussed.

If you have an opinion you want to share, do let us know.

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

Powered by

Up ↑

%d bloggers like this: