How poor methods and analyses result in misrepresentation of the Canadian experience

      Avery Burdett, November 3, 2005


      Canada's experience in implementing helmet laws is frequently cited by helmet law proponents as evidence that laws reduce head injuries and don't discourage cycling. A closer examination of these claims is required since helmet laws in Canada are a provincial jurisdiction and the conditions from province to province having helmet legislation are not all the same. The claims derive principally from three published studies - one covering the experience in a district of Toronto, Ontario(1), one on the effects in Halifax, Nova Scotia(2), and a third comparing data from legislating provinces with non-legislating provinces(3). An examination of the studies demonstrate that conclusions are not supported by the data they are based upon.

      East York, Toronto

      A small study of child cyclists aged 5 to 14 years of age in the Toronto Borough of East York, Mandatory helmet legislation and children's exposure to cycling, Macpherson et al, challenges the claim that helmet laws have the effect of reducing cycling even though such reductions in Australia have been well documented(4). The study concluded that Ontario's helmet law, applicable to cyclists under the age of 18 and introduced in October 1995, had not affected the level of cycling in East York.

      A later analysis of the East York study(4) identified a number of shortcomings including the observation methods applied and the exclusion of children from 15 to 17 years of age, the group most likely to quit cycling rather than wear helmets. Notwithstanding those shortcomings, the most serious criticism is that, unlike in Australia, the law was never enforced by police(5). An unenforced law is likely to have little or no effect(6). This explains why the number of children riding bicycles did not decrease. Young people often recognize intuitively when a doubtful adult discipline is foisted on them. It did not take them long in Ontario to figure out that if the cops do not care about helmet laws then why should they.

      Another study showed that although helmet use increased in 1996, the year after the law was introduced, within two years it had started to decline(7). By 1999 helmet use had dropped to pre-law levels(8). So the two trends diverged. While Ontario's head injury rate fell, East York's helmet use rate reversed itself and started to drop back to its pre-law rates.

      It is concluded that the East York experience is not relevant in arguing that helmet laws don't discourage cycling.

      Halifax, Nova Scotia

      A study of helmet use in the City of Halifax before, during and after local enforcement of Nova Scotia's universal helmet legislation (on September 1, 1977), Effect of legislation on the use of bicycle helmets, Leblanc et al, found that rates of helmet use increased. As well, the authors implied in their discussion that helmet legislation had reduced cyclist head injuries. The publication of the study immediately drew strong criticism(9).

      The very small number of head injuries, 25 out of a total of 1085 injuries over the three years, cited included those, such as dental injuries, that have no relevance in a discussion about protection afforded by helmets. Furthermore, lacerations, and mild concussions comprise the vast majority of head injuries; serious ones are very rare - 1% or 2%, so it's not clear that Halifax had much of a problem to deal with in the first place. A close examination of the data suggests that head injuries indeed declined but had done so as a result of reduced cycling (10,11) rather than increased helmet use. Although not explicitly stated, the authors imply in their discussion that the post-legislation reduction in head injury rate could be attributed to the increased helmet use rate.

      Another criticism is Leblanc's unsubstantiated claim that deaths due to head injuries are preventable when helmets are worn. No reference was offered to support this claim, in fact no credible evidence to this effect can be found. Transport Canada fatality data over a twenty seven year period show cyclists following an almost identical trend to (unhelmeted) pedestrians despite significant increases in helmet use among cyclists(12). Similar trends were derived by Kunich from US Department of Transport data(13). These data suggest that it isn't helmets but other factors which account for declining fatality trends.

      Comparison of results from legislating and non-legislating provinces

      The objective of this study, Impact of mandatory helmet legislation on bicycle-related head injuries in children: A population-based study, Macpherson et al,  was to measure the impact of helmet legislation on bicycle-related head injuries in Canadian children. It compared rates of head injury in regions with and without mandatory helmet legislation. The authors concluded that there was a strong association between helmet legislation and lower rates of head injury and that this finding supports the adoption of helmet legislation as an effective tool in the prevention of childhood bicycle-related head injuries. It should be pointed out that this report relies solely on the assumption that helmet legislation results in increased helmet use following its implementation. If that assumption is false then the report's findings are invalid regardless of other issues identified. The study shows in Canada, that assumption is based on studies of the unrepresentative Borough of East York(14), and the City of Halifax, Nova Scotia(2) the two locations mentioned in this article above.

      While the assumption may not be false, it has not been shown to be true either for the following reason. Toronto's and Halifax's implementations of their respective provincial helmet laws differed. The law was strictly enforced in Halifax but not in Toronto or the rest of Ontario. While the rate of helmet use jumped immediately following legislation in both locations(7) and was maintained in Halifax(2), by the second year a decline had started in East York(7) and by 1999 the rate had declined to pre-law levels(8).

      A separate analysis by Robinson(15) criticizes the study for failing to account for trends unrelated to helmet legislation. Data at the provincial level of detail were examined and they showed that together two provinces, British Columbia (BC) and Ontario, accounted for 9 out of 10 of the injuries in legislation provinces. The most significant changes in injury rates however did not coincide with helmet legislation. For Ontario, the largest fall in the head injury rate occurred in the second year following legislation (when helmet use had started to decline in at least one district of Toronto), and for BC, the largest single-year fall was in the year before the law's implementation. In that pre-law year, BC's head injury rate was 2.7% less than no-law provinces; a year after legislation, it was 2.5% less. Far from reducing the rate of head injuries, Robinson's analysis shows the effect relative to no-law provinces was a very slight increase.

      The failure to control for trends unrelated to helmet laws render Macpherson et al's findings unreliable.


      It is clear that the conclusions in all three of these oft-quoted Canadian studies are not supported by the data underpinning them, and therefore the reports are not reliable sources of evidence in favour of legislating of helmet use.

      1. A K Macpherson, P C Parkin and T M To, Mandatory helmet legislation and children's exposure to cycling, Injury Prevention 2001;7:228-230
      2. John C. LeBlanc, Tricia L. Beattie and Christopher Culligan, Effect of legislation on the use of bicycle helmets, CMAJ, March 5, 2002;166(5)
      3. A K Macpherson, T M To, C Macarthur, M L Chipman, J G Wright, and P C Parkin, Impact of Mandatory Helmet Legislation on Bicycle-Related Head Injuries in Children: A Population-Based Study, Pediatrics Vol. 110(5) November 2002, pp. e60
      4. D L Robinson, Helmet laws and cycle use, Injury Prevention 2003;9:380-381
      5. A. Burdett, Effectiveness of cycle helmets and the ethics of legislation, J. R. Soc. Med., October 1, 2004; 97(10):503
      6. Andrew Irvine; Brian H Rowe; Vic Sahai, Bicycle helmet-wearing variation and association factors in Ontario teenagers and adults, Canadian Journal of Public Health; Sep/Oct 2002; 93, 5;
      7. P. C. Parkin, A. Khambalia, L. Kmet, and C. Macarthur, Influence of Socioeconomic Status on the Effectiveness of Bicycle Helmet Legislation for Children: A Prospective Observational Study Pediatrics, September 1, 2003; 112(3): e192 - 196
      8. Macpherson AK, Macarthur C, To TM, Chipman ML, Wright JG, Parkin PC. Economic disparity in bicycle helmet use by children six years following the introduction of legislation. Injury Prevention. 2006;12:231235 Abstract
      9. Canadian Medical Association Journal, August 1, 2002; 167(4): 337 - 339
      10. M. L. Chipman, Butting heads over bicycle helmets, Can Medical Association Journal, August 1, 2002; 167(4): 337 - 339
      11. M. L. Chipman, Hats Off (Or Not?) to Helmet Legislation, Canadian Medical Association Journal, March 5, 2002; 166(5)
      12. Canadian Fatality Trends The Vehicular Cyclist website.
      13. T. Kunich, US Fatality Trends The Vehicular Cyclist website.
      14. Parkin PC, Macarthur C. The effect of legislation on bicycle helmet use by children [abstract]. Ambulatory Child Health.1997;3:203
      15. D L Robinson, Confusing trends with the effect of helmet laws, Pediatrics, eletters, 7 July 2003

December 2005
Return to: Home