Like this site? Help us to make it better.


DfT casualty statistics rank driving, cycling, walking and motorcycling by risk

Motorcyclists are most vulnerable users, but pedestrians and cyclists also exposed to vastly greater risk than car drivers

Figures released last week by the Department for Transport in its annual report on reported road casulaties for 2011 reveal the risk of death or serious injury to different classes of road users by distance travelled and highlight the vulnerability of pedestrians, cyclists and motorcycle riders compared to drivers.

For each billion miles travelled by the separate modes of transport, more than ten times as many pedestrians and cyclists than car drivers lost their lives on Great Britain’s roads in 2011. For motorcyclists, the rate was around 40 times higher.

The gap is even wider when it comes to looking at the risk of death or serious injury by billion vehicle miles. More than 20 times as many pedestrians and around 40 times as many cyclists were killed or suffered a serious injury for each car driver. Once again, however, it was motorcyclists who were exposed to the greatest risk – they were 75 times more likely to be killed or seriously injured than a car driver. 

Relative risk of different forms of transport, Great Britain: 2011    
Casualty rate per billion vehicle miles  

                Killed   Killed or seriously injured  

Car driver           3               26  
Pedestrian *        42              542  
Pedal cyclist       35            1,035  
Motorcycle rider   122            1,868  

* 2010 National Travel Survey data were used to calculate 2011 pedestrian rates    

Source: Table RAS30070, DfT Reported road accidents and casualties, Great Britain

It’s unsurprising that car drivers are at a reduced risk of being killed or seriously injured in a road traffic incident than pedestrians, cyclists or motorcyclists; the protection the vehicle affords means a cyclist or pedestrian will always come off worse than the motorist when they are hit by a car, and manufacturers invest heavily in safety features such as airbags.It’s true that in absolute terms, more car occupants were killed on Britain’s roads in 2011 than other classes of road user – 883, compared to 453 pedestrians, 362 motorbike riders and 107 cyclists.

Moreover, when it comes to road casualty statistics, there are difficulties in directly comparing data, because for example even those who walk and cycle a lot cover a lot fewer miles over the course of a year than someone spending the same amount of time driving.

And while casualty rates per billion miles travelled is the closest we have to being able to compare different types of road user, there are problems involved in collating reliable data regarding distance travelled for cyclists and pedestrians.

Even so, the data do suggest that more needs to be done to protect the vulnerable.

Existing road infrastructure is designed around the car, often exposing cyclists and pedestrians in particular to unnecessary risk – negotiating a busy urban junction on a bicycle, for instance, or trying to access an edge of town retail park on foot, can be very different experiences to those enjoyed by motorists cocooned in their cars and for whom the road layout has been designed.

Simon joined as news editor in 2009 and is now the site’s community editor, acting as a link between the team producing the content and our readers. A law and languages graduate, published translator and former retail analyst, he has reported on issues as diverse as cycling-related court cases, anti-doping investigations, the latest developments in the bike industry and the sport’s biggest races. Now back in London full-time after 15 years living in Oxford and Cambridge, he loves cycling along the Thames but misses having his former riding buddy, Elodie the miniature schnauzer, in the basket in front of him.

Add new comment


qwerky | 11 years ago

Just to add some scope; if the accident rate for air travel were to be in the same league as road travel then a route from London to Sydney (assuming 400 passengers) would have 1 death per return flight.

10573 miles x 2 x 400 = 8458400 passenger miles
1 death / 8458400 miles = 118 deaths / billion vehicle miles

How many people would take a flight if they knew that statistically, one passenger wouldn't make it back?

doc | 11 years ago

The statistics are a very basic - even crude - summary. THe method of data gathering is not specified. But it's probably at best inaccurate to a great degree, traffic counts are fine for motor vehicles but cannot properly account for human powered movement in any meaningful way. There are dangerous assumptions of average speeds beyond the motor powered sector, and often mileages are hugely underestimated.
Therefore any comments around hours and risk is also likely to be invalid as the base data from which the extrapolation is made is probably inherently flawed in the first instance.
In a nutshell, it's all related to that well known political name, Mr E Balls.
The old adage, GIGO (garbage in, garbage out) has to be considered.
There's lies, damn lies, and statistics!

sparrow_h | 11 years ago

Interesting. I wonder what the margin of error on the estimated numbers of vehicle miles per year is (it must be massive!), I don't know how they can possibly estimate those anywhere near accurately.

kie7077 replied to sparrow_h | 11 years ago

I wonder what the margin of error on the estimated numbers of vehicle miles per year is (it must be massive!), I don't know how they can possibly estimate those anywhere near accurately.

Amount of fuel sold, average mpg of vehicles on roads would be one way, I'd bet that statistic is measured several ways and would easily be the most accurate of those mentioned.

Jimmy Ray Will | 11 years ago

Ok, just did some loose maths, to convert miles to hours. Assumed all car journeys averaged 35mph, bike journeys 12mph, pedestrians 3mph, and motorcyclists 40mph.

The figures aren't too inspiring.

Deaths per billion hours are as follows

Cars - 105
Pedestrians - 126
Cyclists - 420
Motorcyclists - 4880

So the major takeaway is this... Motorcycling is bloody lethal compared to everything else. Cycling is more than ten times safer, but that is still four times more deadly than driving a car.

KirinChris | 11 years ago

Looking on the positive side, it helps discourage the perception that cycling is an inherently dangerous activity.

If a million people cycled 1000 miles each year then the individual's risk of death or serious injury is around one thousandth of one per cent. That's not so bad.

SpamSpamSpam | 11 years ago

Hmmm. A "per hour" measure isn't great. If I cycle ten miles to get to work, or choose to take a car, then the distance is the same. I want a comparison of how much more/less safe each mode of transport is for this journey. So the per mile measure is probably better.

hoski | 11 years ago

Presumably if you multiply the deaths per billion miles by the average speed of the mode of transport you end up with deaths per billion hours?

SpinnakerMagic replied to hoski | 11 years ago

..I'm not sure how accurate that could be though - knowing how long journeys take with enough precision to produce meaningful data might not be possible.
(assuming you live for 80 years, that's 7.0128×10^-4 billions of hours - i fully intend to live forever, but still..)

dave atkinson replied to hoski | 11 years ago
hoski wrote:

Presumably if you multiply the deaths per billion miles by the average speed of the mode of transport you end up with deaths per billion hours?

speeds vary, and the rate of incidents also varies, depending on where you are. in london average bike speed is broadly comparable to average car speed; that's why you always get people moaning on daily mail stories that they're constantly having to overtake the same cyclist. it's because they're not going any faster.

out in the country, for example on rural A and B roads, the average speed of a cyclist is far below that of a car. and motorway travel, which is fast and relatively safe, pushes up the averag speed of cars. We don't have (as far as i know) any statistics from the DfT that breaks down journeys by these criteria, so just taking an average will be some help, but it certainly isn't the whole picture.

SpinnakerMagic | 11 years ago

It would be nice to see 'per-journey' statistics - also somewhat flawed, but i think would give a better indication of the real risks - perhaps, oddly showing that driving is perhaps more dangerous than you'd think, and cycling rather safer.

JohnS | 11 years ago

First, these figures should be per billion hours, not miles, to reflect the differing speeds of different modes of travel.

Second, why are these figures for victims of KSI and not perpetrators of KSI? That would tell us which road users need to be given most attention by the authorities.

HKCambridge replied to JohnS | 11 years ago
JohnS wrote:

Second, why are these figures for victims of KSI and not perpetrators of KSI?

One can be established rather more reliably than the other, surely?

Also in some situations there may not be a second or more vehicle involved.

The DfT link in the article isn't working for me: I get a page not found. Anyone else?

OldRidgeback | 11 years ago

It is no wonder given the questions over flaws in the data gathered by bodies such as the DfT then that the EC's road safety policies are often so wide of the mark and have such little effect when implemented. Much of the safety policy is geared towards the victim rather than the instigator, reactive instead of proactive in other words.

Since car, truck, bus and van drivers are the cause of most of the accidents, surely the most intelligent policy will be to tackle bad driving on the part of those road users?

A V Lowe | 11 years ago

Yet again a failure of DfT number crunchers, quite patently measuring incidents by miles travelled does not really equate to the time of exposure to the hazards for the differing modes. Walking quite clearly takes a lot longer to rack up a large mileage, and the insidious nature of pedestrian traffic means that counting, especially using the same cordon lines as a carriageway count is always going to deliver a questionable figure.

Those using DfT cycle counts are also noting wide variation and failure to correlate to project-specific counting. For example standard DfT counting 08.00 to 18.00 and 07.00-19.00 on the A6 in the Manchester area missed a whole group of cyclists who were riding because of the lack of buses for early shift work or setting off early to get in to the city because of distance and traffic considerations. When Spokes in Edinburgh had cyclists counting cyclists their figures were frequently double those of the Local Council's surveys at the same junctions, counting cyclists as part of a general traffic count.

Sums & DfT a hot topic of the moment!

Matt_S | 11 years ago

Is there any breakdown in the figures relating to the cause of the accidents? E.g. how many of the cycling accidents were related to riding too fast down a country lane and leaving the road on a tight corner. Or how many of the car driver deaths were due to a cyclist rear-ending them. etc.

Latest Comments