Age composition of all cycling road fatalities, 1991-2013, Australia (source data: Bonfous & Olivier)

A new paper on road fatalities in Australia shows that the number of cyclists who died in crashes declined over 1991-2013 at an average rate of 1.9% per year.

Although the downward trend is not as fast as it is for car and pedestrian fatalities, it’s nevertheless a distinct improvement and goes against the popular wisdom (e.g. see Is the risk of getting killed while cycling on roads increasing?).

It’s all the more remarkable because exposure – particularly the number of cyclists – increased over the period.

Recent trends in cyclist fatalities in Australia is written by Soufiane Boufous and Jake Olivier from the University of NSW. It’s published in the July issue of the journal, Injury Prevention.

They counted 959 reported cycling fatalities over 1991-2013. Most (88%) were the result of a cyclist colliding with a vehicle, such as a car or truck.

Deaths from this cause fell at the rate of 2.9% p.a. over the period. The average age of those who died increased substantially; from 26 years in 1991 to 39 years in 2013.

But fatalities where no vehicle was involved – such as death resulting from a fall – went against the overall trend; they increased at the rate of 5.8% p.a.

These crashes accounted for only 12% of all cyclist deaths over the period but increased their share e.g. they went from 3% of all bicycle fatalities in 1991 to 22% in 2013.

The average age of those who died in non-vehicle crashes also increased substantially; from an average of 35 years in 1991 to 55 years in 2013.

The non-vehicle group had a markedly higher age profile. Over the 1991-2013 period, 63% of riders who died were aged over 40 years and 28% were 60+; the corresponding figures for fatalities involving a vehicle are 24% and 17%. (1)

The reason for the fall in overall fatalities isn’t clear.

The authors speculate that the increase in age at death in both categories might reflect “the rise in popularity of cycling among middle and older age groups coupled with a drop in children and youths”.

They think the higher average age of non-vehicle fatalities might be due to differences in exposure arising largely from improvements in cycling infrastructure.

Older riders (are) more likely to use routes with maximum separation from motorised traffic than younger riders. Previous research indicates that safety concerns were a key deterrent to older adults’ cycling and most would only consider riding on cycle paths or roads with little motor vehicle traffic.

Older riders are more vulnerable in a crash because they’re frailer; they tend to sustain more serious injuries than younger riders. This is also true for older occupants of vehicles involved in a crash (e.g. see What’s the problem with elderly drivers?).

The researchers don’t offer a specific hypothesis for why fatalities involving a vehicle fell, but one possibility is of course better infrastructure.

That seems an incomplete explanation because it implies cyclists are travelling fewer absolute kilometres in unprotected conditions even though their total numbers are growing.

There might be other explanations for the overall fall in fatalities.

Perhaps the improvement is due in part to slower traffic speeds resulting from higher congestion, lower speed limits, or possibly even stronger enforcement. Or it could be that the “safety in numbers” effect got stronger with the growing popularity of cycling.

A new report by the Bureau of Infrastructure, Transport and Regional Economics, Pedestrians and road safety, provides another possible explanation.

It shows the number of pedestrian traffic fatalities in Australia is also trending down. However the number of cases of pedestrians suffering injuries serious enough to require hospitalisation is relatively steady.

This difference suggests the possibility that improvements in medical access and/or care might be part of the explanation for the fall in pedestrian fatalities. If this difference also holds true for other modes, it might help explain the fall in cycling fatalities.

The numbers could also be affected by the way “near misses” are treated. Some deaths counted as not involving a vehicle may in fact have resulted from cyclists having to take emergency action to avoid a collision with a motorist (this cyclist was very lucky).

Whatever the reason it’s good news that cycling road deaths are falling overall. It would be even better if it’s matched by a decline in serious injuries.

The increase in fatalities where no vehicle is involved is worrying and is bound to attract attention; but as the authors hypothesise it might primarily reflect the greater numbers that now have the option of cycling in protected conditions, as well as the composition of those riders e.g. by age. (2)

All riders can cite examples of hazardous conditions for cycling on roads and in bike lanes that can lead to a fall or worse. These need to be addressed with the same vigour that governments have tackled unsafe conditions for motorists.

Cycling is getting safer in terms of the risk of being killed, but as I’ve noted before it’s still a lot riskier than driving (see Is cycling more dangerous than driving?). There’s still a pressing need to reform road laws to support cycling as well as provide better infrastructure.


  1. These figures refer to the age composition for the period 1991-2013, but as a point of comparison, note that 46% of Australians were aged 40 or more years at the 2011 Census and 20% were aged 60 or more years
  2. There’s good evidence that sport cyclists are markedly over-represented in cycling-related hospitalisations (e.g. see Should cyclists need a licence to ride on public roads?).