When I see a medical statistician on a bicycle…

…I do not despair for the future of the human race.

In my day job I work for scientific and medical journals, a million miles from transport and planning policy.  Except that this week I was pleasantly surprised to find that one of our papers was on all the bike blogs.  (I had nothing to do with the paper, and didn’t even notice it until it was published.)  In BMC Public Health, Andrei Morgan and colleagues have done what we at AWWTM love: taken the best methods that we have for evaluating evidence and applied them to the topic of London cycling; specifically they have described the data on cyclists killed in London traffic.

The interesting factoids are:

  • While annual death rates remained relatively static over time, when considered against a background growth in estimated cycling kilometres from 0.85% to 1.48% of the total estimated traffic kms in London, they have fallen considerably in terms of deaths per estimated cyclist km.
  • Three quarters of the killings were on or at junctions with main roads.
  • Women were more likely to be killed in inner-London and during daylight; Men got killed day and night throughout London.  Men accounted for more of the total deaths, but the authors did not normalise any of these data, so we can’t say whether it’s because men are more incompetent or women more safety conscious, whether drivers behave differently around them (as Ian Walker previously found), whether men are more likely to be using the main roads where crashes happen, or whether there are simply more men cycling, especially in the outer boroughs and at night.
  • 40% of the killings were in the outer boroughs, despite there being much lower levels of cycling there.  The two are probably not unconnected: the outer boroughs have bigger faster roads and fewer cycling facilities.  The inner boroughs have slower speeds and “safety in numbers”.  People who think that London’s main roads are dangerous places are not entirely stupid.
  • There were five reported incidents in which “only the cyclist was involved”.  Presumably people riding in lampposts, or “just falling off“?
  • And of course, the most timely finding: over 40% of incidents involved freight vehicles, half of those on left-turns.

The authors — themselves London bicycle users at the School of Hygiene & Tropical Medicine in Bloomsbury — conclude that trucks should be banned from central London.

The paper is published coincidentally (it was written over a year ago, before peer review) in the same week that Dennis Putz was sentenced for killing Catriona Patel with a truck at Oval last year, and that Boris Johnson promoted his own ideas about banning HGVs from central London (despite delaying the LEZ which might have helped a bit), amongst many other events that have highlighted the problem of trucks in central London this autumn.  So to an extent the paper only adds more weight to a conclusion that most of us had already reached, through previous studies and through our own amateur observation and experience, and for other reasons additional to safety issues: that trucks do not belong in city centres.

Rather, the important message that I got from the paper was to highlight just how poor the evidence-base for cycling safety policy is.  The authors repeatedly had to acknowledge and apologise for the limitations of their work — in the places where I, in my lowly blog post, can speculate wildly about possible explanations for the authors’ observations, the authors themselves must stay silent because there simply isn’t good enough data on things like the characteristics of London bicycle users and their bicycle journeys.  It’s an issue that we keep coming across, in arguments over bicycle helmets, segregated infrastructure, and every other policy, intervention, and initiative: the documented evidence rarely approaches the quality necessary for making important decisions on important policies.

We at AWWTM are very much of the (evidence-based) opinion that a policy or intervention isn’t worth pursuing unless it is informed by evidence of the way that the world works, and how it might affect the way that the world works.  And it depresses us that this even needs saying, but it seems that many politicians and planners are happy to dogmatically follow policies that have been shown to fail, and to implement new ones without doing anything to check that they are working.

More of that later.

at junctionsThe important o
Advertisements

4 responses to “When I see a medical statistician on a bicycle…

  1. Pingback: Tweets that mention When I see a medical statistician on a bicycle… | At War With The Motorist -- Topsy.com

  2. Interesting about the gender distribution. It wasn’t that long ago that there were a lot of handwringing articles about women being killed by lorries on their bikes which seemed to just be an excuse for the authors (and the commenters) to air their prejudice about women not being assertive enough on bikes / too vain to wear helmets / not fast enough / or just too air-headeadly blonde and busy doing their make up to see the great big lorry about to run them over. It will be interesting to see what the reason is now we find out it’s actually the men who are at risk…

  3. Pingback: What’s wrong with this number? | At War With The Motorist

  4. Pingback: When I see a medical statistician on a bicycle… | Joe D

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s