Someone on Twitter send me the following link. “Safety effects of permanent running lights for bicycles: A controlled experiment“:
Making the use of daytime running lights mandatory for motor vehicles is generally documented to have had a positive impact upon traffic safety. Improving traffic safety for bicyclists is a focal point in the road traffic safety work in Denmark. In 2004 and 2005 a controlled experiment including 3845 cyclists was carried out in Odense, Denmark in order to examine, if permanent running lights mounted to bicycles would improve traffic safety for cyclists. The permanent running lights were mounted to 1845 bicycles and the accident rate was recorded through 12 months for this treatment group and 2000 other bicyclists, the latter serving as a control group without bicycle running lights. The safety effect of the running lights is analysed by comparing incidence rates – number of bicycle accidents recorded per man-month – for the treatment group and the control group. The incidence rate, including all recorded bicycle accidents with personal injury to the participating cyclist, is 19% lower for cyclists with permanent running lights mounted; indicating that the permanent bicycle running light significantly improves traffic safety for cyclists. The study shows that use of permanent bicycle running lights reduces the occurrence of multiparty accidents involving cyclists significantly. In the study the bicycle accidents were recorded trough self-reporting on the Internet. Possible shortcomings and problems related to this accident recording are discussed and analysed.
This was a controlled trial of putting permanent (daytime) running lights on bicycles. It was cluster randomized. And, over the course of a year, the incidence of accidents for those with the running lights was 19% lower. The incidence of multiparty accidents with injury to the participating cyclist was 47% lower.
How do I not know about this?
Policy sometimes baffles me. This was a randomized controlled trial of almost 4000 participants. They found a real reduction in an important outcome. There are no randomized controlled trials for bike helmets, because people argue that it would be unethical to do so. In fact, I think most of the research is case-control. Don’t get me wrong. I understand why that is, and I’m not recommending that people stop wearing their bike helmets. The odds ratios for preventing head injury are compelling.
But we have an RCT here, with a real reduction in accidents. Why aren’t we acting on that?!?!?!