How dangerous are dogs?

Plenty of people are hospitalised by dog bites every year. Plenty of people are hospitalised by lots of things.

Photo by Celine Sayuri Tagami on Unsplash

People often ask the Office for National Statistics how many injuries are caused by dogs. I assume these people aren’t dog-lovers. Although I also don’t think it’s obvious what a bad number for dogs would be, PR-wise.

In the latest year we know about, NHS England delivered 10,776 Finished Consultant Episodes after a dog “bit or struck” someone. Is that bad? An FCE involves being admitted to hospital, so these dog bites (or dog strikes) must have been at least a little serious.

And the number probably hides some grey areas. I have a friend who broke her arm when a dog she was walking pulled her over. That wasn’t exactly being “struck”, but you’d guess it counts. My son, a confirmed dog-lover, asked whether the data includes injuries to people who had a dog thrown at them. Again I’m not sure. I suspect that would be classified as assault, and it’s probably quite rare.

There are about 57 million people in England and, ten years ago, there were about 12 million dogs. So if you assume, no doubt wrongly, that each bite or strike involves a different person and a different dog, that’s roughly 30 dog-woundings a day, administered by about 0.1% of dogs each year, with a population risk of around 1 in 5,000.

These numbers don’t make me feel endangered. But I don’t own a dog. And I assume that people who do are happy to take the risk.

What’s missing here is a sense of scale, but we can get that by looking for other things that hospitalise people at similar rates. “Fall involving chair”, for example, caused 11,885 FCEs. That’s similar, although I’d guess that chairs are even more popular than dogs, so maybe the risk of sharing your home with one is lower. It really would take a lot of FCEs to make me worry about using chairs.

Then there’s injuries to pedal cyclists: 14,255. That’s a bit higher, but it’s still roughly comparable to dogs, and I think of cycling as fairly dangerous. (These are accidents where a cyclist got hurt, by the way, not the ones where they injured someone else. There were 381 of those.)

Various categories of assault, including neglect and maltreatment, come to 22,869 FCEs between them. So deliberate human violence has around double the impact of dog violence, if you measure it in this reductive way. Assault injuries do seem more serious, as they involve about 2.5 bed days per FCE, compared with 1.3 for dogs.

Another dimension that makes a difference is the age of people involved. Dogs, like bicycles and insects, spread their devastation fairly evenly through the population. Whereas chairs are a particular danger to the very old and the very young, and assault mainly affects younger adults.

Taken together, I don’t know how these numbers should make us feel about dogs. Roughly how we feel about cycling? If they worry you, consider a pet insect.