Lee AJ, Mann MP. Archives of Disease in Childhood, 2003;88:465-466
The authors – members of the Bicycle Helmet Initiative Trust – argue that head injury when cycling is a major problem for children in the UK, that cycle helmets are effective in preventing most of these injuries and that legislation is needed to maximise cycle helmet wearing by children. They cite the work of one team of researchers and a meta-analysis in support of helmet effectiveness, and their own work which claimed a reduction in head injuries seen by one A&E department when helmet use in the locality was increased through an education programme. The laws introduced in other countries are seen as precedents that should be followed in the UK.
The paper states "Each year more than 50 young people under 15 years are killed in cycling related accidents, with 70-80% of these deaths caused by traumatic brain injury". The reference given for this statement is a Westminster briefing in a previous issue of the same journal (ADC, 1996). Only one paragraph of the briefing refers to child cycling as follows:
"Each year in the 1990s the number of children killed in road cycling accidents in England has been between 37 and 50 and the number of non- fatal cycling injuries between 7349 and 8132. In Scotland there are 3- 7 deaths and 462-625 non-fatal injuries per year." (Hansard, 1996)
An average of much less than 50 fatalities a year has been reported by Lee and Mann as "each year more than 50 young people … are killed". Their reference says nothing about deaths involving brain injury. In the last year (2002) for which information was available at the time of publication of their paper, there were just 19 child cyclist fatalities, of which only 10 (53%) involved head injury (Ladyman, 2003), but the authors make no reference to these more recent statistics which are the continuation of a long-term downward trend not associated with helmet wearing..
Lee and Mann say that "Hospital based figures show that more than 100 000 British children each year are injured in cycling accidents". However, the reference cited (Sacks, Holmgreen, Smith and Sosin, 1991) is a paper concerned only with injuries in the United States and which does not refer in any way to Great Britain.. Only about 3% of child cycling casualties result in hospital admission and the great majority of injuries suffered by child cyclists do not involve significant injuries to the head and are therefore irrelevant to the consideration of a helmet law.
The authors then state that "approximately 1000 children a year receive a moderate or severe brain injury from cycle accidents that leads to permanent neurological disability". This is contradicted by Department of Health data (BHRF, 1100), which shows that at most around 500 children suffer serious head injuries each year when cycling and most of these have no lasting consequence. The authors fail to put the number of serious head injuries into context, as the number is very small relative to the 6 million children in England alone who cycle regularly.
The authors state that 85% of child cyclist crashes occur 'off-road' and only 2% of hospital-treated casualties have been involved in a collision with another vehicle. However, the great majority of serious head injuries - and almost all of those with long-term consequences - do involve another vehicle (Kraus, Fife and Conroy, 1987) and they typically involve forces much greater than those capable of being mitigated by a cycle helmet (Walker, 2005). Few people would support a helmet law if they knew its outcome would be primarily to prevent minor injuries.
The reference cited in support of cycle helmet effectiveness (Thompson, Rivara and Thompson, 1996c) was financed by the Snell Foundation, an organisation whose income is in part derived from cycle helmet sales. It is one of a series of papers by a team of authors whose work has been criticised widely for methodological shortcomings (e.g. BHRF, 1068). The papers are based on non-randomised case-control studies that are readily subject to bias (BHRF, 1052). Predictions such as those cited by Lee and Mann have not been approached anywhere in the world. There is no justification for the premise that levels of protection would have been much higher (at precisely 85% for head injuries and 88% for brain injuries, which exactly coincide with earlier, disputed research by the same team, BHRF, 1068) if different controls had been used. Moreover, the meta analysis cited (Thompson, Rivara and Thompson, 2002-9) as having shown substantial protection to the face is the work of the same team of authors whose own papers dominate the analysis.
Lee and Mann seem unaware that case control studies need to be randomised to be a reliable method for analysing interventions and outcomes, and that all cycle helmet case control studies are non-randomised and vulnerable to bias (BHRF, 1052). The meta analysis by Attewell et al (Attewell, Glase and McFadden, 2001) considered only case control studies and did not look at all at the wider evidence on cycle helmet effectiveness. Most of the papers analysed have themselves been the subject of peer criticism. Lee and Mann state that "those who died would have survived if a helmet had been worn", but none of the papers they cite provides evidence that this would have been the case. There is no population-level evidence from any country where helmet use has become significant that cyclist fatalities have declined as a consequence (BHRF, 1012).
This relates to work carried out by Lee and Mann themselves. Although helmet use may have increased amongst the groups targeted by the BHIT initiatives, no reliable evidence has been presented that this resulted in a reduction in head injuries relative to cycle use. There are many deficiencies in this research, as detailed in commentary (Lee, Mann and Takriti, 2000).
The paper cited by Scuffham (Scuffham, Alsop, Cryer and Langley, 2000) acknowledged there were significant time trends in the New Zealand data, and if time trends are fitted, they swamp the helmet effect, leaving no significant estimated benefit of helmets (BHRF, 1237; BHRF, 1241). Lee and Mann do not mention other published research (Robinson, 2001) that reported that, if time trends were not responsible for the effect, the models fitted implied that voluntary helmet wearing was 15 times more effective at preventing head injuries than helmet wearing because of legislation. This strongly suggests that most of the effects previously attributed to helmet wearing were in fact time trends or other artefacts of the data.
Scuffham himself subsequently reported (Taylor and Scuffham, 2002) that the New Zealand law had not been cost-effective for most age groups (including those children most at risk). Lee and Mann do not mention this, nor a similar analysis for Western Australia (Hendrie, Legge, Rosman and Kirov, 1999) showing that, even under the most optimistic estimate of the benefit of helmets, the laws was not cost effective.
The authors do not provide a reference to support their assertion that cycle related deaths and head injuries fell in Canada as a result of laws there, but other evidence suggests that this was not the case (BHRF, 1096; Burdett, Can).
The authors acknowledge initial concerns about a reduction in cycle use in Australia following its laws but do not make it clear that after more than a decade in most parts of that country cycle use has still not recovered, particularly among children (BHRF, 1096). Their statement that helmet legislation did not reduce the number of children cycling in Ontario fails to note that the law there was not enforced and that, most probably for the same reason, there was also no net increase in helmet wearing (BHRF, 1102).
Criticisms of the Lee and Mann paper were published in a subsequent issue of Archives of Disease in Childhood and also on the journal's website. A paediatrician (Waterston, 2003) argued against any measure that would discourage children from cycling as the health benefits of more people cycling without helmets are much greater than those of fewer people cycling with helmets. Another critic (Wardlaw, 2004b) expressed concern at the lack of balance in the paper and the adequacy of the peer review process in allowing so many errors to pass.
The paper contains large errors of fact, exaggeration and misrepresentation of the evidence presented. There is no reference at all to the considerable body of evidence that does not support cycle helmet legislation. The paper is not of sufficient rigour to inform debate on its subject.
Westminster briefing. Arch Dis Child 1996;75:463.
Attewell RG, Glase K, McFadden M, 2001. Bicycle helmet efficacy: a meta-analysis. Accident Analysis & Prevention 2001-05 v33 n3 p345-52.
Robinson DL, . Costs and benefits of the NZ helmet law. .
Burdett A, . Cyclist fatalities in Canada. OCBC .
Hansard: 18 July 1996, Col 660 (England); 23 July 1996, Col 230 (Scotland).. .
Hendrie D, Legge M, Rosman D, Kirov C, 1999. An Economic Evaluation of the Mandatory Bicycle Helmet Legislation in Western Australia. Road Accident Prevention Research Unit .
Kraus JF, Fife D, Conroy C, 1987. Incidence, severity and outcomes of brain injuries involving bicycles. American Journal of Public Health 1987;77:76-78..
Answer to Parliamentary Question from Brian Jenkins MP by Dr Ladyman, Dept of Health minister. . Hansard 17-8W 10.11.03 .
Lee AJ, Mann NP, Takriti R, 2000. A hospital led promotion campaign aimed to increase bicycle helmet wearing among children aged 11-15 living in West Berkshire 1992-98. Injury Prevention 2000;6:151-153.
Robinson DL, 2001. Changes in head injury with the New Zealand bicycle helmet law. Accident Analysis & Prevention 2001 Sep;33(5):687-91.
Sacks JJ, Holmgreen P, Smith SM, Sosin DM, 1991. Bicycle associated head injuries and deaths in the United States from 1984 through 1988: how many are preventable? . JAMA 1991;266:3016-3018.
Scuffham P, Alsop J, Cryer C, Langley JD, 2000. Head injuries to bicyclists and the New Zealand bicycle helmet law. Accident Analysis and Prevention 2000 Jul;32(4):565-73.
Taylor M, Scuffham P, 2002. New Zealand bicycle helmet law - do the costs outweigh the benefits?. Injury Prevention 2002;8:317-320.
Thompson DC, Rivara FP, Thompson RS, 1996. Circumstances and severity of bicycle injuries. Snell Memorial Foundation .
Thompson DC, Rivara FP, Thompson RS., 2002. Helmets for preventing head and facial injuries in bicyclists (Cochrane Review). Cochrane Database Syst Rev issue 4, 2002.
Walker B, 2005. Heads up - the science of helmets. Cycle Jun/Jul 2005.
Wardlaw MJ, 2004. Only wholeness leads to clarity. Archives of Disease in Childhood 2004;89:692-3.
Waterston T, 2003. Cycle helmets and legislation. Archives of Disease in Childhood Elec letters 3rd Sep 2003.