Nov 22, 2012

Does cycling infrastructure reduce serious accidents?

A new Canadian study confirms what cyclists intuitively know: providing quality cycling infrastructure reduces the risk of accidents and serious injuries

Alan Davies — Editor of The Urbanist

Alan Davies

Editor of The Urbanist

Bicycle route preference vs route safety (source: Teschke et al)

In the latest issue of the American Journal of Public Health, a team led by Professor Kay Teschke from the School of Population and Public Health at the University of British Columbia, reports on a study examining injury risk associated with 14 types of cycling route.

They interviewed 690 adult cyclists in Vancouver and Toronto who’d been injured seriously enough while riding to be admitted to hospital within 24 hours (the severity of the injuries is not reported). Two thirds of those eligible for the study agreed to participate.

The 14 route types were categorised in four groups: major streets with parked cars; major streets without parked cars; local streets; and off-street routes.

The three on-street groups were further subdivided to account for various types of lane markings and traffic calming. The off-street routes were refined to account for variables like sharing with pedestrians and type of pavement.

The researchers found a third of accidents involve a direct collision with a motor vehicle (although another 14% of injuries are sustained in avoiding a vehicle). Most of those injured are regular cyclists, male, younger than 40 and well educated.

Accidents are distributed more or less evenly across the four main road types. However more than half of all injuries are sustained on streets with no cycling infrastructure (i.e. no works, no signs, no on-street markings).

Compared to major streets with parked cars and no bicycle infrastructure (the reference type), segregated on-road bicycle tracks have about one ninth the risk of serious injury. Marked bike lanes on major streets with no parked cars have nearly half the risk, as do dedicated off-street bike paths.

However off-road paths shared with pedestrians have about three quarters of the risk of the reference type. The authors don’t say but I expect this reflects conflict with other users.

The study also found that aside from cycling infrastructure, other route characteristics are also associated with risk reductions: quiet streets (i.e. local streets); and no car parking on major streets.

On the other hand, downhill slopes, tram tracks and construction work are much riskier than the reference route type (more than three times in the case of tram tracks).

The study also compared the risk associated with the route types against those route characteristics cyclist’s in general say they prefer. As the exhibit shows, there’s a high correspondence between these two variables. Cyclists intuitively know what’s safer.

Many route types with positive preference ratings were also among the safest: cycle tracks; local streets; bike only paths; and major streets with bikes lanes and no parked cars. These provide a range of options with potential to both lower injury rates and increase cycling. This in turn may create a positive feedback cycle because increased ridership has been associated with increased safety.

Teschke et al’s findings are consistent with those of Buehler and Pucher’s study of 90 US cities. As I’ve discussed before, they found the supply of bike lanes and paths per capita is a statistically significant predictor of bike commuting.

Cities with a greater supply of bike paths and lanes have significantly higher bike commute rates. They found this to be true even after controlling for land use, climate, socioeconomic factors, gasoline prices, public transport supply, and cycling safety.

The link between safety – whether real or perceived – and cycling for transport purposes is compelling. Better infrastructure encourages more cycling and lowers injury costs.

(Visited 1 times, 1 visits today)


Leave a comment

Your email address will not be published. Required fields are marked *

17 thoughts on “Does cycling infrastructure reduce serious accidents?

  1. Russ

    Alan, Dylan, your discussion raises an interesting problem with this type of research actually. When I had a student look at infrastructure and cyclists, the proportion of (probably) risk averse cyclists (non-lycra wearing/women) on bike lanes is significantly higher than on other streets. If an aversion to risk reduces accident rates (reasonable assumption), some proportion of the actual risk of an accident is attached to the type of cyclist using the route. A lot of the infrastructure itself might be much of a muchness in terms of safety, but the perception of safety adds risk-averse cyclists to a route, which significantly reduces accidents, and increases awareness of cyclists in motor vehicles. A virtuous cycle, even if there is no net benefit in safety all things being equal.

  2. Dylan Nicholson

    Sure, I’m not particularly risk averse, but I’d say most of my falls don’t happen during what I’d think of as high-risk activities – they’ve happened because of something unusual that I’m not used to yet (or maybe because I mentally switch off a little when I think I’m safe). And my partner generally *is* quite risk averse, but because she doesn’t ride very much any more (she used to in Japan), she lacks confidence/experience that would avoided the falls she had.
    Either way, I don’t particularly think infrastructure planning should be based around avoiding bruises and scrapes 🙂

  3. Alan Davies

    Dylan Nicholson #14:

    No, but then neither was/is your partner. Getting used to cleats would only account for the first year. I suspect you’re not as risk-averse as I am.

  4. Dylan Nicholson

    Were you riding 300km a week?
    I’d say it would be a rare 12 months I’d go without having some sort of minor fall or bump. Especially if you’re just getting used to riding with cleats!

  5. Alan Davies

    Dylan Nicholson #12:

    Who said anything about driving? I haven’t had a cycling accident of any sort in 15 years even though I ride regularly (on roads) and commuted every day by bike for three years.

  6. Dylan Nicholson

    Well, yes, it is rather a lot compared to the # of minor accidents you’d normally have driving, but the health (and other) benefits of cycling are worth a few bumps and scrapes in my book.

  7. Alan Davies

    Dylan Nicholson #10:

    That too.

    Only two accidents (fortunately with minor consequences) in the last year each!?

  8. Dylan Nicholson

    Alan, actually you’d think the ‘skew’ would be that the people who end up in hospital with injuries are more likely to people who cycle a lot, and don’t necessarily represent the general population – but having said that, I cycle almost 100 times as much as my partner (by distance, not time), yet in the last 12 months we’ve both only had two (minor) accidents each. FWIW two of them almost certainly wouldn’t have occurred with better infrastructure (one was a dooring, and another was because my partner had ridden onto the footpath because the road felt unsafe, only to fall over as she tried to avoid some pedestrians).

  9. Russ

    hk, the point about separating travel paths is an interesting one, because at some point travel paths will intersect, and it is how that intersection is dealt with that matters. In this study, for example, major streets with no bike lane but no parked cars are slightly safer (albeit within the margin of error) than major streets with bike lanes and parked cars. Interestingly, they didn’t actually find intersections to increase risk, which seems counter-intuitive. Other studies I’ve seen have also indicated that the treatment of bike lanes across intersections matters a lot: ie. that having a bike lane that just ends at the point where travel paths intersect is more dangerous than none at all, because of visibility and the perception of space and rights.

  10. Alan Davies

    hk #7:

    Good points. Off road paths have a higher accident rate than segregated and marked on-road paths but the severity of accidents would be lower because there’s no exposure to cars.

    The other thing to keep in mind is this sample of cyclists wouldn’t be representative of all cyclists. These are the ones who ended up in hospital – we should expect risk-takers to be over-represented.

  11. hk

    Reducing the exposure time of the probability of the number of cycling collisions between vehicles is one issue. The other main issue is reducing the severity of the injury from collisions with motorised vehicles.
    The first is best handled by separating travel paths. The second is best managed by enforced speed reduction of BOTH cycling and vehicle traffic.

  12. IkaInk

    In regards to the presence of bike lanes and tram tracks in Toronto, nowadays a fair chunk of the streets with trams running along them do have bike lanes, however they’re usually little more than a painted lane between parked cars and traffic as is standard in Melbourne.

    Here’s a cycle map of Toronto if you’re interested. Note that only the red lines, the red/blue and the purple lines are anything that actually resembles cycling infrastructure. The yellow lines are nothing more than the odd bike painted in the middle of the road, to remind drivers bikes are allowed to use the full lane (which of course they’re allowed to do on any road…). The blue lines are nothing but signs on the side of the road saying that bikes use the street, although this at least is usually reserved for quiet streets. The dotted yellow lines are nothing at all and shouldn’t be there. Toronto’s bike infrastructure really seems no better to me then Melbourne’s and in many cases is much worse.

  13. Alan Davies

    Dudley Horscroft #4:

    69% were wearing helmets at the time of their accident. The city-specific rates were 79% Vancouver and 59% Toronto.

    Can’t comment on your tram lines question.

    Bike counts for each of the four groups (median hourly count): Major street with parked cars (108); Major street no parked cars (150); Local street (84); off-street (234). In the three street categories, the most used sub-category is marked bike lanes.

  14. Dudley Horscroft

    Two questions arise: out of the 650 people admitted to hospital, how many had head injuries as a result of not wearing a helmet, and how many were actually wearing a helmet? And, whereas Vancouver has no trams and Toronto does, how is it that tram tracks are rated three times as dangerous as streets with parked cars and no bike markings? To the best of my remembrance, all Toronto streets with tram tracks also have parked cars, and when I was there, no markings for bikes.

    In addition, there does not appear to be any indication of the actual distances trvelled on the various types of road, which surely must affect the results?

  15. Cyclesnail

    What I find interesting that this type of study needed to get done, and some of the background is in Peter G. Furth article found in John Pucher’s “City Cycling”: The reasons are hidden in the AASHTO 1999 Guide (American Association of State Highway and Transportation Official’s Guide for the Development of Bicycle Infrastructure).

    This guide was heavily influenced by John Forrester, a traffic engineer who since the 1970’s was active in the formulation of safety regulations surrounding cycling. He was active in bicycle racing clubs and saw the separation of bicycles from cars as a threat to cyclists who wanted to use the road. His basic tenet was that cyclists fare best when they act as, and are treated as, operators of vehicles. This puts into perspective the comment by Teschke: “Many, though not all, of the previously reviewed studies found higher risks on off-street route types, but this was not the case in the present study.”

    We hear similar arguments advanced by vocal people in the Perth cycling community, and they might be interested in taking a really close look at this newest piece of research. The BTA has always called for the separation of traffic based on speed and volume, as there is little doubt that most people on bicycles feel safer when they do not have to share the roads with cars travelling over 30kmh.

  16. Last name First name

    Parker Alan •OAM
    the residential speed limit in Canada is 40 Km/hr on some roads and 50 kms/hr on others. In School Zones zones the speed limit is 30 km/hr or 40 km/hr.In most US states the residential speed limit is 25 Miles per hour.
    The above speeds apply to far many more miles of local roads than main roads and is therefore the most important form bicycle infra structure

  17. Dylan Nicholson

    Would be interesting to make some sort of approximation of the economic value of building bike paths, making use of this information. It wouldn’t surprise me in the last that you could show that a well-placed bike path would easily save millions more than the cost of building it within a few years.

Share this article with a friend

Just fill out the fields below and we'll send your friend a link to this article along with a message from you.

Your details

Your friend's details