Is cycling on roads getting safer or more dangerous?

There were 32 cycling fatalities on Australian roads in 2015 according to BITRE’s Road Deaths Database. That’s a lot fewer than the 45 deaths in 2014 and 50 in 2013. It’s the lowest number since 2009.

And 32 very likely overstates the dangers to cyclists from other road users. Nine of those fatalities didn’t involve another vehicle; it’s probable that most of this subset were the consequence of health issues like heart attack from exertion.

You can see a chart of historical cycling fatalities in this article I wrote in September last year (luckily, my prediction for 2015 was spot on). It shows cycling fatalities in Australia declined slightly over the last 20 years.

Index of injuries in Victoria requiring hospitalisation, by mode (source data: TAC Road Trauma Database)

However, as the exhibit indicates, the number of serious injuries suffered by cyclists on roads (i.e. that required hospitalisation) increased significantly over the 14 years from 2000 to 2014. The data is for Victoria, but I expect the general trend applies nationally.

In Victoria, 386 cyclists were admitted to hospital in 2014 as the result of road trauma. Fortunately, only 9% of them required a stay longer than 14 days (compared to 13% for drivers and 27% for pedestrians). Ten cyclists died on roads in Victoria in 2014 (and 45 nationally).

Cycling hospitalisations grew a lot faster than those for motorists and pedestrians, consistent with the intuition that the number of cyclists on Victoria’s (and the nation’s) roads grew significantly over the period.

That, in turn, confirms that cycling fatalities on roads fell significantly in “real” terms over the last 10 to 20 years, i.e. fatalities declined while the level of cycling increased.

Why have cycling fatalities apparently fallen so much faster than serious injuries?

If it were because, say, the composition of the population of cyclists changed (e.g. newer cohorts of riders are more careful), or because of improvements in cycling infrastructure (e.g. more segregated on-road paths), fatalities and serious injuries should at least be going in the same direction.

Another possible explanation is better emergency medical care; perhaps a few of the 36 cyclists who were hospitalised in Victoria in 2014 for more than 14 days might have died in past years?

What we don’t know for sure, because it’s very difficult to estimate, is whether or not the level of cycling on roads is rising at a faster or slower rate than the increase in serious injuries.

If the factor(s) leading to lower fatalities also apply, to some extent, to hospitalisations, then serious injuries might also be declining in “real” terms, albeit probably at a much slower rate than fatalities.

While welcome, an improvement in “real” road trauma outcomes for cyclists doesn’t reduce the need for significant infrastructure improvements. Deaths and injuries are still unacceptably high and cycling is still more dangerous than driving.

A critical issue is the perception of danger from drivers discourages cycling and limits its potential contribution as a key mode of transport.