Any discussions of the law and the use of bicycle helmets will be helped by focusing on three questions. Firstly, how much do cycle helmets affect the risk of injury in a crash? Secondly, how much do laws requiring cyclists to wear helmets affect casualties? And, thirdly, should wearing of bicycle helmets be required by law?

Answers to the first two questions have objective answers that can be sought from specific empirical studies and what is already known about traffic safety.1 Three unrelated sources of evidence consistently show that cycle helmets reduce risk substantially in a crash. The science of biomechanics shows that a helmet reduces the peak acceleration forces that are associated with injury. Many published epidemiological studies, including that by Maimaris and colleagues in this issue of the BMJ and the references it cites,2 find that helmets reduce injury and the risk of death. Such studies fall short of the methodological ideal of comparing harm in matched treatment and control populations - so less direct methods must necessarily be used.

For example, examining the ratio of head injuries to nonhead injuries in cyclists wearing and not wearing helmets indicates that helmets reduce the risk of head injury …