Rapid Responses to:

SPORT: Donald A Redelmeier and J Ari Greenwald Competing risks of mortality with marathons: retrospective analysis BMJ 2007; 335: 1275-1277 [Abstract] [Full text]

Rapid Responses: Submit a response to this article

Rapid Responses published:

Are marathon runners dangerous drivers?

Graeme D Ruxton   (22 December 2007)

Comparing marathon deaths with Road Traffic Accident Deaths

Dan S TUNSTALL PEDOE   (1 January 2008)

Are marathon runners dangerous drivers? 22 December 2007
Graeme D Ruxton, Professor University of glasgow G12 8QQ Send response to journal: Re: Are marathon runners dangerous drivers?	Redelmeier and Greenwald (R&G) make a convincing case that the mortality risk associated with marathon running is low. However, their conclusions about associated road traffic accidents (that road closures associated with marathons leads to a reduction in motor vehicle fatalities) are less convincing. Of course, closing roads to vehicles will stop accidents on those roads at those times, the key issue is whether the road closures merely shift the accidents to other places and times. R&G observe that “analysis of counties outside the course that compared observed and expected crash deaths during hours of road closure should no countervailing increase in fatal crashes”. However, the rate of fatalities in road traffic accidents per county per day is so low that picking up even a moderate change would be very difficult, as R&G’s own analysis reveals: despite his large sample the confidence internal for their analysis above spans changes from a decrease of 6% to an increase of 7%. A similar problem arises with their failure to find a shift in time. As well as this methodological problem, there is a problem of identifying the underling mechanism. If road closures related to marathons actually cause a reduction rather than just a distribution of road fatalities, then we must ask how does this reduction occur? Is it because people cancel their journeys to stay home and watch coverage of the marathon on TV? It seems more likely that the marathon would encourage more journeys (among participants and spectators) than it stops. It seems unlikely that the marathon would lead to safer driving, indeed the re-routing of drivers to unfamiliar roads is more liable to increase risk of accident. Thus, if R&G are correct that marathons lead to a reduction rather than a redistribution of road traffic accidents, perhaps the reason is because marathon runners are intrinsically dangerous drivers and the key function to society of marathons is to keep these people off our roads for a few hours! Competing interests: None declared
Comparing marathon deaths with Road Traffic Accident Deaths 1 January 2008
Dan S TUNSTALL PEDOE, Emeritus Medical Director London Marathon (Medicak Director 1981-2007) retired from Homerton and St Bartholomews Hospitals home 29 Meynell Cresc LOndon E9 7AS Send response to journal: Re: Comparing marathon deaths with Road Traffic Accident Deaths	Redelmeier and Greenwald (R and G) have some interesting retrospectively collected data on large US marathons but have neglected mention of the London marathon death risk data collected prospectively over the last 27 years which has been compared with different forms of vehicular road usage risk in Europe and published in BJSM and elsewhere.(refs 1,2,3,4) We have found that on a "time of exposure basis" running the London marathon has a comparable death risk to many daily activities. It has a smaller death risk than riding a motorbicycle but has about three time greater risk than cycling for the same length of time amd about 10-12 times greater risk than exposure while being a motorist. These calculations are based on an average assumed exposure of 6 hours to the marathon and its environs and the comparison is made with European Transport statistics . Our current data gives a slightly higher death rate from cardiac causes (8 in 676000 ) compared to the US figures of 0.8 per 100,000 and also includes data on successful cardiac resuscitation (6) and 3 non cardiac deaths. Cardiac arrest occured at the finish in less than one third of cases and was distributed from 6 miles onwards. The mortality on US roads is given by R and G as 12 deaths per 42 kms per 1000 hours, an extraordinary figure even if the 1000 was supposed to be 100,000. Their derived figures seem as mythical as their comment that runnng a marathon can lead to suddden death "as documented 2500 years ago". There is no contemporaneous account of Pheidippides death after the battle of Marathon The first allusion to it is 600 years later in Roman histories. Pheidippides is said by Herodotus the Greek historian in his history of the battle written within a life span of the event to have run from Athens to Sparta and back to try and get help for the Athenians and Herodotus says Pheidippides met the god Pan on the way. Herodotus makes no mention of his death after running from Marathon to Athens less than a fifth of the distance. A myth should not be the basis of a scientific dscussion. Whatever the true US road death figures, marathon running.has a comparatively low but not negligible risk but encourages a healthy life style and if we aim to keep death off the roads perhaps we should encourage more people to run to work but if possible along car free routes eg canal tow paths. . 1. Tunstall Pedoe, D. S. . "Sudden cardiac death in sport:spectre or preventable risk?" Br J Sports Med 2000 34(2): 137-40. 2.Tunstall Pedoe, D. S. . "Sudden death risk in older athletes: increasing the denominator." Br J Sports Med 200438(6): 671-2. 3. Tunstall Pedoe,DS (2000) “Morbidity and mortality in the London Marathon “p 197-207in “Marathon Medicine” editor Dan Tunstall Pedoe RSM press London 2000 4Tunstall Pedoe DS. Marathon cardiac deaths : the London experience. Sports Med.2007 2;37(4-5):448-50. Competing interests: None declared