Innovative study shows that cycle tracks and local streets mean fewer injuries for cyclists

I attended a webinar on the "Bicyclists’ Injuries and the Cycling Environment" (BICE) study back in January. The BICE study examined which route types are associated with higher and lower cycling injury rates. The webinar covered a summary of their study's results, which will be available soon online. While some of the study concepts may difficult to understand without an academic research background I thought it was interesting to convey how this study approached the difficult area of injuries and cyclists from a new angle. And from their study we find some interesting results, the most interesting being that they have shown that cycle tracks and local streets that restrict through motor traffic are the two types of routes that reduce the risk of injury for cyclists compared to the typical arterial road with parked cars (exemplified by streets like Dundas, Queen, Bloor and so on).

The innovative BICE study, led by M Anne Harris, was designed with case-crossover design features. In their view, using a case-crossover design is a more reliable way to set up such a study since there are so many variables involved and it would be quite difficult to control for them otherwise. Previous studies faced difficulties in assessing denominators for risk calculations and controlling for confounding variables (variables that correlated with both the dependent and independent variables). The researchers hoped "that the value of this method and the efficiency of the recruitment process will encourage replication in other locations, to expand the range of cycling infrastructure compared and to facilitate evidence-based cycling infrastructure choices that can make cycling safer and more appealing."

Crossover studies according to Wikipedia:

A crossover study has two advantages over a non-crossover longitudinal study. First, the influence of confounding covariates is reduced because each crossover patient serves as his or her own control. In a non-crossover study, even randomized, it is often the case that different treatment-groups are found to be unbalanced on some covariates. In a controlled, randomized crossover designs, such imbalances are implausible (unless covariates were to change systematically during the study).

Second, optimal crossover designs are statistically efficient and so require fewer subjects than do non-crossover designs (even other repeated measures designs).

In their crossover study design they interviewed the participants and took control sites randomly selected from the injury trip. The route type was based on distance ridden on each trip. The sites were then observed by researchers who were blind to the status of the site - whether it was an injury site or control. This prevented observational bias. The crossover design meant that comparisons were made within a person-trip, thus controlling for personal and trip characteristics. Then using statistical modelling they accumulated the comparisons over all the person trips.

Injured cyclists were recruited from the emergency departments of five hospitals in Vancouver and Toronto, Canada. In 18 months, 690 participants were successfully recruited and interviewed. Each participant was interviewed to map the route of their injury trip, identify the injury site and select two control sites at random from the same route. Infrastructural characteristics at each study site were scored by site observers who were blinded as to whether sites were crash or comparison sites. Analyses will compare infrastructural variables between case and control sites with conditional logistic regression.

They found that the typical cyclist profile in the study cycled more than 52 times per year (88%) which might be explained that the more trips one takes the greater the exposure to potential injury. Most wore helmets (59% in Toronto and 76% in Vancouver). I found the Vancouver number surprisingly low considering that there is a helmet law. Could it be that not only do helmet laws discourage some from cycling but that they aren't even all that effective in raising the helmet wearing rates compared to other methods?

Of the injuries investigated (% of 690), 22% involved a motor vehicle, 10% involved a vehicle door, 14% involved streetcar track (and 6% involved both a streetcar track and a motor vehicle where a cyclist avoided a motor vehicle and fell because of streetcar tracks), and 5% involved a cyclist avoiding a collision with a motor vehicle.

The study used 15 route types and compared them. They included more route types and built-environment factors than other similar studies, by also looking at surfaces, intersection type, categorizing streets into minor/major, with/without bike facilities and categorizing off-road into bike paths, cycle tracks and sidewalks.

For the purposes of comparison they presented their statistical analysis as the relative risk of injury as compared to their reference route category, a major street with parked cars and no bike infrastructure such as we see on Bloor Street. The reference category equalled 1 and their other route types either had a fraction of that risk or a multiple.

They found that:

  • All other route categories were lower risk than their reference route type.
  • Cycle track (physically separated bike lanes from traffic) was 1/10 the risk of reference category. This was statistically significant.
  • The reference category with bike lanes didn't provide statistically significant less risk.
  • a bike lane on major street with no parked cars was found to be statistically significant at 1/2 the risk. This may indicate that removing parked cars has a bigger impact than adding a bike lane.
  • A local street with or without bike infrastructure had the same risk factor has bike lanes on major streets with a statistically significant risk of 1/2 that of the reference category.

For the routes fully separated from traffic, they found that:

  • sidewalks had a statistically insignificant and small drop in risk. This is bound to be controversial given that previous studies have found increased risk, though in the case of the well-known study by Wachtel and Lewiston (1994) they found that riding on the sidewalk in the same direction of traffic had almost the same risk but that riding against the flow of traffic on the sidewalk carried 4 times the risk.
  • likewise multiuse paths, whether paved or unpaved, provided only a slight, stastically insignificant drop in risk.

They also looked at non-route features. Of some of the other features studied but were not statistically significant, included intersections with a slightly higher risk and bike signage which provided a slightly lower risk.

The features which did show a statistically significant increased risk, included downhill grade at 2 times the risk of the reference category, routes with streetcar tracks at 3 times the risk, and construction on a route at 2 times the risk. As well they found that uncontrolled intersections were 3 times more risky and the traditional traffic circle (unlike the roundabout) were 8 times more risky. The risk of injury was 10 times for cyclists travelling opposite to motor vehicle traffic (though I'd like to see this broken down into arterial and local streets).

The researchers also looked at risk in relation to cyclist traffic counts. They found that at least at intersections, that an increased number of cyclists actually also increased risk. This runs counter to the safety in numbers research by Dr. Jacobsen, which looked at the correlation on a country-wide scale. What isn't true on any individual intersection may still be true when looking at the cycling rates across a city or country.

They also found that motor vehicle speed affected the risk of injury by cyclists, where motor vehicle speed was less than 30km/h this provided a statistically significant decreased risk.

Preferences
It's important to not just study the safety issues but also the preference by most cyclists. This is because cyclists will still take the routes they prefer over those which are deemed "safe". And also because in study after study it has been shown that the health benefits (physical exercise, less pollution) far outweigh the health costs (in terms of injuries and fatalities). A general policy could arguably be to build route types that best match up lower risk with preferences.

Most cyclists prefer bike paths, multi-use paths, local streets, and cycle tracks. Preference and safety match fairly well. By choose the route types that match both safety and preference would be cycle tracks, local street with motor traffic diverters (to discourage through traffic), and bike only paths.

There were some limitations to the study:

  • most severe and mildest injuries were not included
    • 20 people couldn't be included because their injuries were too severe and they couldn't actively participate
    • the mildest injuries weren't studied since only hospital admittances were focused on
  • the researchers couldn't study different cycle track designs so a comparison could be made between, say, unidirectional and bidirectional cycle tracks
  • there were not enough innovative intersection designs to study

When their study summary is posted online I will update this post with a link or pdf.

herb : 13 years 3 months ago
Tags:

Councillor Wong-Tam trying to stall on Sherbourne separated bike lanes

It was a bit of a shocker to find out that Councillor Kristyn Wong-Tam is getting cold feet on separated bike lanes for Sherbourne, thus potentially putting Toronto's first opportunity for better separation into jeopardy. It has come to my attention that Councillor Wong-Tam has filed a request for Sherbourne's separated bike lanes to be installed only in the North as a pilot for study and further community consultation.

It appears as if the Councillor is trying to stall the project with claims of "needs more community consultation" despite the fact that it has already gone through a completely open process with the community. Most of the residents and businesses who provided comments had said that they supported the bike lanes. There is such a thing as studying a thing to death.

By requesting it only be installed on the North end of Sherbourne, it appears Councillor Wong-Tam, is willing to give up on the improvements for the lower part which were to be coordinated with road repaving. Has Councillor Wong-Tam consulted with Councillor Pam McConnell whose ward covers the other half of Sherbourne? Given that Councillor McConnell supported the separated bike lanes I'd say she hasn't.

In the Winter issue of Dandyhorse, she was asked "How do you feel about having the first separated bike lane in Toronto installed in your ward?"

As excited as I am about having Toronto’s first separated bike lane, I also think that we need to proceed with caution. There are political forces that don’t support biking infrastructure and a planning misstep just gives them a new excuse to declare another fictitious “war on the car.”

At current, there are many new projects coming to Sherbourne that have not been properly addressed (or consulted) in the separated bike lane proposals yet: namely, the many new condo developments, the existing schools, churches and senior home drop-off and delivery areas. The residential community is also wondering what will happen to their local streets when those 159 street parking spaces are removed from Sherbourne. We should study Sherbourne more closely, create a measurable pilot project, invest in a high-quality street design and a comprehensive greening strategy before finalizing a capital infrastructure decision that will be costly to fix if we don’t get it right.

Why would Councillor Wong-Tam want a trial period? Perhaps she wants to defeat it? Either asking for a trial or for yet more consultation is a great way to get rid of a pesky project that you don't agree with. And what's this about a "comprehensive greening strategy"? We're talking about improving the bike lanes not about rebuilding the whole streetscape.

The current unseparated bike lanes have been in place for 16 years. What could we possibly learn with a trial period that we don't already know from all these years of already having bike lanes and from studying similar conversions in Ottawa or Vancouver? A trial period gives those opposed to it a second chance to oppose the separated lanes.

Councillor Wong-Tam feels there needs to be more consultation yet the idea for the lanes first arose at a Toronto Cycling Advisory Committee in January 2010 and has been at public meetings of PWIC over the last 2 years. The concept was reviewed and approved by Council in 2010. There was a public meeting in June 2011 at PWIC where permanent installation of the lanes was approved. Not one person appeared at that meeting and spoke in opposition to separated bicycle lanes on Sherbourne. Prior to the last election Councillor Pam McConnell announced her support of the Sherbourne separated bike lanes and she was re-elected.

The cost of putting in temporary measures would make the project’s final installation much more expensive. And it would make it more difficult to find the money to separate any more bicycle lanes such as Wellesley, about which Councillor Wong-Tam has also been less than supportive.

A few years ago Waterfront Toronto spent a millions dollars on a pilot project of separated bike lanes on Queens Quay. There is still nothing permanent (thought hopefully starting soon). If separated bikes lanes are considered such risky ideas, even on a street with 16 years of on street unseparated bicycle lanes that it needs a test period what are the chances to install any separated lanes in the city on streets that don’t currently have unseparated bicycle lanes like Richmond/Adelaide?

Furthermore the timing is such that the next election would interfere. If the pilot project was done in 2012 the decision to install and installation of the lanes permanently would be made in 2013 just before the next municipal election. And a politician doesn't want to make waves before an election.

Councillor Wong-Tam hasn't been all that enthusiastic about the Jarvis bike lanes for that matter. Although Councillor Wong-Tam is opposed to the reinstallation of the 5th car lane on Jarvis, she seems less interested in maintaining the Jarvis bicycle lanes. She has been quoted as saying that the Jarvis bike lanes were installed with inadequate "consultation".

I ask that the Bike Union take Councillor Wong-Tam to task. She seems to be more interested in what people will think when they will have to park on side streets instead of Sherbourne than about providing safer cycling for the rest of us.

Cyclists lose John, but will we be getting a consolation prize with Beverley, Peter, Simcoe?

At the Public Works and Infrastructure Committee meeting yesterday, the John Street Corridor Improvements Environmental Assessment Study was approved despite the number of people in the cycling community who voiced their displeasure of being ignored. It was a plan that was pushed by commercial interests along John, as well as by Councillor Vaughan, but provided very little for cyclists who comprise up to 1/3 of the traffic along the street.

But surprisingly, there may be a consolation prize for cyclists. A reputable source at the PWIC meeting heard Councillors Minnan-Wong describe how Councillor Vaughan was willing to support separated bike lanes on Beverley, Peter, and Simcoe (and presumably Richmond or Adelaide to connect these streets) if he the John Street report was approved. Since a John Street without bike facilities was approved, I presume that Vaughan is now committed to improving this alternative route, and that Minnan-Wong wanted to make sure this commitment was recorded in the public record.

This is quite a change from a year ago when Councillor Vaughan was sent out a newsletter to his constituents using the nonsense term "barricaded bike lanes" despite the fact that having some separation from motor traffic is exactly what most Torontonians have requested (as in the recent survey by the City). A positive shift away from this view can be seen in his comment about the plan for the Sherbourne separate bike lanes where Vaughan criticized that there is "not a pronounced enough separation.... Unless you make it physically risky to put a car in that spot, you will get cars in that spot. You'll have taxis, you'll have couriers, you'll have vendors."

It's quite commendable that Councllor Vaughan is willing to support this popular separated bike lane network. Despite not getting a better John Street as the most rational link, we can at least count on Vaughan to support the hard work that will be required to make the Beverley - Peter - Simcoe route possible. For instance, there is a major problem with the off-set intersection at Peter and Queen that will be need to be fixed before it can be called truly safe for most cyclists.

It was reported as well that Councillor Minnan-Wong was still championing an expanded network at the meeting (likely seeing that he can still win some political points across the spectrum with his give and take), including extending the Richmond Adelaide bicycle lanes to Eastern Avenue where there are also bike lanes. This would be a great extension to the current EA which is only going to cover a portion of Richmond and Adelaide and ignores the need for connectivity to the other side of the Don.

herb : 13 years 3 months ago
Tags: bike infrastructure, politics, separated bike lanes