Advertisement
Research Article

Exposure-Based, ‘Like-for-Like’ Assessment of Road Safety by Travel Mode Using Routine Health Data

  • Jennifer S. Mindell mail,

    j.mindell@ucl.ac.uk

    Affiliation: Research Department of Epidemiology and Public Health, UCL (University College London), London, United Kingdom

    X
  • Deborah Leslie,

    Affiliation: Centre for Physical Activity and Nutrition Research, Deakin University, Burwood, Australia

    X
  • Malcolm Wardlaw

    Affiliation: Edinburgh, United Kingdom

    X
  • Published: December 05, 2012
  • DOI: 10.1371/journal.pone.0050606

Reader Comments (4)

Post a new comment on this article

Risks of cycling

Posted by ianholloway on 06 Dec 2012 at 10:32 GMT

I very much enjoyed reading the authors' article on like-for-like assessment of road safety by travel mode.

I was interested to read that the results and interpretation, whilst concentrating on the principle outcome measure of fatalities per million hour use, did not draw the reader's attention to the very much higher risk to cyclists according to all other outcome measures. It seems that the risk of hospital admission for instance, is very much higher for cyclists. The outcomes per billion kilometres all showed a much higher risk for cyclists. This measure would be more relevant to commuters, who have a fixed distance to travel to work, and are therefore more interested in this metric, as a measure of their risk for that journey.
As a cyclist recently injured during a commute, I feel that there should have been more emphasis on the negative aspects of cycling compared to driving as demonstrated in Tables 1,2,3,4 and 5.

Competing interests declared: Doctor and cyclist and recent recipient of a non-fatal (!) cycling injury, which could have a negative impact on my career.

RE: Risks of cycling

MWardlaw replied to ianholloway on 10 Dec 2012 at 11:34 GMT

We focused on fatal injuries as these are of the greatest concern, and also indicative of the risk of severe injury. In contrast, risk of admission is a poorer measure. As we point out, the severity of injury (as suggested by duration of stay) varies considerably from one road user type to another. For cyclists that severity is generally low.

Another problem is that pedestrian injuries are substantially under-stated, due to the problem of incomplete coding, whilst cyclist injuries are over-stated, as the location is assumed to be on-highway unless otherwise indicated. This needs to be borne in mind when reviewing the results.

It is true that admission risks per hour were generally higher - and for some groups many times higher - for cyclists than for other modes. In contrast, the results for fatality risk per hour were mixed, all three modes falling in the same range (for males). This tends to bear out the point above that cyclist admissions involve less serious injuries than other modes, even if they are commoner.

For females the results are weaker due to the low number of fatal casualties. The risks for cycling were generally higher (although note wide confidence intervals), although very low as absolute risks.

It is worth bearing in mind that a male alighting from a train to walk home magnifies their risk (fatality per hour) by a factor of between 20 and 50. A tiny risk remains a very small risk even when magnified by what appears to be a large factor.

No competing interests declared.