More research is needed

We researchers love to say ‘more research is needed’. Sometimes, it seems to mean ‘more research is needed, we can’t possibly do anything in the meantime’. It can be a barrier to ambitious policies, because they’ve never been done – and researched – before. So I like to qualify this. In a recent co-authored paper, we write:

‘Although more research is needed, it is important not to wait for this, and we can still act to ensure the preferences and needs of under-represented groups are foregrounded in policy and planning.’

But when it comes to road safety, near miss research is definitely something we need to do more of. Right now, we seem a long way from taking near misses seriously. We seem to be struggling to act even when people are dying. This morning on the radio I heard a story of a ‘shocking’, ‘most sickening’ crime. This referred to a man having stolen a purse from an 84-year-old woman who had been run over by a lorry and left dying. The story illustrated the extent to which death on the roads has become normalised. That’s almost expected: an accident, certainly wouldn’t make the national news. Violating the property of the recently dead is taboo, but we are apparently unable to prevent their violent death in the first place.

This wasn’t meant to become a rant, although it is hard to avoid getting angry when critics of safer streets express horror at threatened ‘economic disbenefits of £200 million’ while death is taken for granted. (These disbenefits are largely comprised of summing lots of possibly slightly longer trips for motorists over a thirty-year period. Not something that horrifies me, personally, as much as a twenty-five year old woman being killed by an HGV last night – but perhaps I’m biased).

Rather, my point is that reactive is not enough. Acting when one, or two, or four people die is not enough. We need immediate action to stop further loss of life, and research can help. Specifically, looking at where near misses occur can help identify and measure ‘early warnings’ before anyone is killed or seriously hurt. (I think it’s also very important to look at near miss type incidents to identify situations that negatively impact the cycling experience without being lethal, but given where we are, I guess we’d better start with preventing death and serious injury).

My suggestion is that local authorities first look at several datasets that can provide information on cycling near misses in their area (some of the data will also be useful for pinpointing problem locations for pedestrians, either because of conflict with cyclists, or with motorists). I can soon make local mapped data available from the Near Miss Project; Collideoscope may have further data, and perhaps the CTC might provide information on locations where bad driving has been reported through their Road Justice project. In London, there’s also Roadsafe.

An initial go at mapping Near Miss Project incident locations in one borough.

An initial go at mapping Near Miss Project incident locations in one borough.

Where a near miss or similar incident might incidate an infrastructural problem – for example a pinch point where cyclists are driven at, or a junction where near left hooks are reported – the authority should send people out, at rush hour, to observe and record any systematic threats to vulnerable road users. This can be done in a pretty low-tech way, but will then provide numbers of incidents that can be used to derive a rating of how problematic the location is. (This can be complemented by the use of new audit tools such as TfL’s Junction Assessment Tool).

Optimistically, I would like to believe that new data and new tools might play a small part in helping us do better at preventing deaths. I take inspiration from Jan Gehl’s work counting pedestrian activity and mapping pedestrian desire lines, which has done so much to help create more pedestrian-friendly public spaces.

Bookmark the permalink.

8 Responses to More research is needed

  1. You are so right about that purse theft thing, I nearly spat out my cornflakes upon hearing that the copper had never seen anything worse…either his career has been very short or sheltered from reality! It’s bad to steal of course. The case is arguable either way whether it’s worse from a living/dying/dead person. But surely it’s far far worse to kill someone violently, that’s why the sentences are stiffer…unless of course you kill by motor vehicle, in which case you will get off almost scot-free every time.

  2. Can I suggest mention of my LTT article http://rdrf.org.uk/2013/11/15/if-we-want-safer-roads-for-cycling-we-have-to-change-how-we-measure-road-safety/ on how to measure road danger, in particular to cyclists?

    As with your work, it goes in to how we have to do something other than just look at where people are reported as hurt or killed when cycling.

    • admin says:

      Sure 🙂 I agree that measuring risk caused is important – I’ve previously argued for a ‘Vehicle Risk Unit’ measure instead of PCUs, where we’re talking about recommended limits on motor vehicle flow on key cycle routes.

  3. Andy R. says:

    Unfortunately this seems very similar to Spicer’s Conflict Study approach, where conflict (degree of braking or avoidance manoeuvre) was a proxy for collisions and could be used to predict where collisions might occur. That only seems to be known by academics and those of us who’ve been on road safety courses where the lecturers have had time to fill.
    To my knowledge it is not, nor ever has been, used by any Local Authorities, even in times of almost decent budgets; they all act retroactively in producing Collision Investigation and Prevention reports.
    Unless this approach to identifying danger is widely publicised throughout the highway/traffic engineering profession, and there’s significant follow-up pressure from campaigners, I feel even if it is ‘low tech’ it will be seen as an additional burden on hard pressed LAs.

    • admin says:

      Hi Andy, agree with your points re: professional education/culture.

      I believe the Conflict Study approach is used in Scandinavian countries – albeit more high-tech with video data collection and analysis. I agree the principle is similar to what I’m suggesting although I would want to use slightly different definitions that go beyond avoidance manouevres – such a big chunk of the recorded near miss data is about problems such as tailgating/close passing where a cyclist doesn’t necessarily swerve or brake – perhaps can’t. Which raises the question: how close should a vehicle be for a manouevre to count as tailgating or a close pass, at different speeds? Another useful piece of research, I think.

      • admin says:

        Of course, thinking about it, we DO use the Conflict Study approach in this country. We use it to establish whether the presence/behaviour of cyclists is posing unacceptable risk/annoyance to pedestrians – e.g. TfL’s studies of bus stop bypasses, numerous studies of parks, shared paths, etc. Nothing wrong with that, particularly where we’re using / trialling new forms of infrastructure. But isn’t it a bit weird, that we don’t also use that kind of approach to protect pedestrians and cyclists from the much greater threat posed by motor vehicles?

      • Jitensha Oni says:

        A possible definition of a close pass or tailgating is: if the rider were to fall over toward the centre of the road just before the close pass or with the tailgating occurring, he/she would be damaged, if not driven over, by the vehicle, then it’s an unacceptably close pass or tailgating. This takes care of the speed aspect since a driver should be able to stop to prevent damaging the rider under any circumstance. Stopping distances are in the HIghway Code.

  4. Andy R. says:

    Well I do think the conflict study approach was ahead of its time (1970s?). It needed low cost but relatively high quality video monitoring – something we now have.
    In fact the only time I’ve used this technique it made use of an existing CCTV installation at an MSA. With the ubiquity of CCTV coverage in UK cities (even if we reduce it to that owned and monitored by LAs), it should be the technique whose time has come.

Leave a Reply to Andy R. Cancel reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.