Effect of helmet legislation

Non-enforced legislation
From October 1995, helmets became compulsory for child cyclists in the province of Ontario, Canada. The law was not enforced. No Toronto child has been ticketed for cycling without a helmet. After an initial increase, helmet use fell back to pre-law levels. Non-enforced legislation is of doubtful efficacy, except perhaps to teach children that laws need not be obeyed.

Enforced legislation
Cycle helmets became compulsory in New Zealand from 1st January 1994. The law was enforced with vigour, driving adult helmet use up from 40% to 90%+ where it has remained. However, there was no reduction in the severity of serious head injuries. Cycle use fell by 22% between 1993 and 1997.

In the state of Victoria, Australia, where cycle helmets also became compulsory, the number of cyclists hospitalised with head injuries, after road accidents, fell by 48% and 70% in the first and second years of the helmet law, respectively. However, the helmet law was introduced alongside campaigns against drink driving and speeding, and this also contributed to fewer accidents. For instance, pedestrian deaths also fell by 42% in the first year of the cycle helmet law. Road casualty data provide no evidence of death or serious head injury prevented by the helmet law. Cycle use fell by about one third as the law was enforced.

For Australia as a whole, head injury deaths in road accidents fell by less for cyclists than for other road users in the years when cycle helmet laws came into force. This is revealed by data from the Fatality File of the Australian Transport Safety Bureau. Between 1988 and 1994, the decline for cyclists was 30%, but it was 40% for pedestrians and car occupants. Given that walking and driving were not deterred by helmet laws, it would appear that cyclists faced higher than expected risk of death by head injury, not lower. The possible reasons are considered below.

These population-level outcomes are not consistent with documented hospital-based case-control studies, which have predicted that helmet use reduces the risk of serious head injury in a crash by 50-80%. This literature has been widely cited to establish a case for the compulsory wearing of cycle helmets. One would expect that mass helmet use should give an obvious reduction in the proportion of cycling injuries that are to the head, yet such a reduction is not observed in reality. In epidemiology, it is now recognised that case-control studies are prone to erroneous results when applied to self-selected behaviour. This is due to confounding by social factors - "selective recruitment". It is hard to evaluate what protection a cycle helmet may provide at the individual level, but the effect of mass helmet use cannot be discerned at the population level. This counter-intuitive result is not widely appreciated.

Enforced helmet legislation drives cycle use down. In Australia, cycle use was generally growing before the helmet laws of the early 1990s. Since then it has declined steadily in most states. In New South Wales, child cycle use had fallen by 44% by the second year of the helmet law. In Sydney, cycle use was still 48% down on pre-law levels five years after legislation. The state of Western Australia has made considerable efforts to promote cycling, yet per-capita cycle use has barely recovered to pre-law levels ten years after legislation, in contrast to big increases in the decade preceding the law. In Nova Scotia, Canada, cycle use dropped by 40%+ after legislation. In British Columbia, Canada, cycle use fell by an estimated 28% following legislation.