Dear Editor

Franklin and Robinson are correct to question the complexity of the evidence on helmet wearing among children.[1] As a brief report our paper was unable to explore this in detail but we are grateful for the opportunity to do so here. The helmet wearing surveys suggested that helmet wearing fell among children between 1994 and 1996.[2] Analysing accident data for the years 1995/96 alone shows a corresponding increase in % head injuries (%HI). This increase is not seen in adult data. From 1996 onwards helmet wearing increased slightly and non-significantly among children and %HI remained almost constant (figure 1).

Fig. 1 Head injury in child cyclists, 1995-2001

These trends need to be interpreted very cautiously, particularly the sudden change in %HI in 1996. However, the suggestion is clearly that head injury rates track helmet wearing closely and this was the conclusion that we tried to summarise in our paper.[3] Franklin and Robinson examine helmet-wearing rates in detail, but the sensational headline is premature without the other half of the analysis - injury rates.

Raven also jumps to a premature conclusion.[4] His first mistake is to state that we admit a fundamental error in our paper. Rather, we admitted an arithmetic error in the discussion section of our paper that has no effect on the main results and so can hardly be described as fundamental - a word used only by Annan.[5] His reliance on Annans' letter and failure to pay sufficient attention to our original paper also leads to a second mistake. We estimated the reduction in head injuries among pedestrians and cyclists from regression models, and gave 95% confidence intervals for each. A high protective effect of helmets therefore remains consistent with these data.

References

1. Franklin J, Robinson D L. Another serious error in Cook & Sheikh's analysis [electronic response to Cook A and Sheikh. Trends in serious head injuries among English cyclists and pedestrians] injuryprevention.com 2003 http://ip.bmjjournals.com/cgi/eletters/9/3/266#101

2. Bryan-Brown K, Christie N. Cycle helmet wearing in 1999. Transport Research Laboratory Report 487, 2001.

3. Cook A, Sheikh A. Trends in serious head injuries among English cyclists and pedestrians. Inj Prev 2003; 9: 266-267.

4. Raven T. Re: Authors' reply [electronic response to Cook A and Sheikh A. Trends in serious head injuries among English cyclists and pedestrians] injuryprevention.com 2003 http://ip.bmjjournals.com/cgi/eletters/9/3/266#103

5. Annan J D. Fundamental error in "Trends in serious head injuries..." Cook and Sheikh 2003 [electronic response to Cook A and Sheikh A. Trends in serious head injuries among English cyclists and pedestrians] injuryprevention.com 2003 http://ip.bmjjournals.com/cgi/eletters/9/3/266#59

Dear Editor

There is another serious problem with Cook and Sheikh's paper.[1] The authors cite a TRL report [2] stating that, on major roads, helmet wearing (%HW) increased from 16.0%, in 1994, to 17.6% in 1996 then 21.8% in 1999. The TRL report continues: "this was due to an increase among adult cyclists wearing helmets: there was no change amongst child cyclists."[2]

In fact, Table 3 shows %HW of girls was unchanged, but boys' %HW declined steadily (16.0%, 13.3% and 12.7% in 1994, 1996 and 1999; c²=3.51, p=0.06, for 1994 vs 1999).[2] Another TRL survey in 1999 found that child %HW was even lower on quieter roads, 9.7%.

Yet Cook and Sheikh reported almost identical declining trends in percentages of UK hospital admissions with head injury (%HI) for adult (8.09 percentage points) and child (8.32) cyclists, with more children (1625) injured than adults (1129).[1] Most injured child riders are male; boys (75% of child cyclists in the TRL surveys) ride more and also tend to take more risks. The TRL surveys show boys' %HW was low and decreasing, and that, overall, children's %HW fell from 17.6% in 1994 to 15.0% in 1999.

If Cook and Sheikh believe the fall in adult %HI was due to increased helmet wearing, they must, logically, also believe the fall in children's %HI was due to reduced %HW. Instead of campaigning for mandatory helmet laws, they'd call for child helmets to be banned!

It's much more plausible that %HI of child and adult cyclists in the UK followed similar trends (fairly similar to %HI trends in pedestrians), unrelated to helmet wearing.[3] The most informative data on the efficacy of helmet laws comes from Australia and New Zealand; large increases in %HW did not produce any obvious response in %HI.[4]

Readers interested in reducing injuries to cyclists should therefore consider proven measures such as traffic calming, cyclist and motorist education, reducing the incidence of speeding and drink-driving,[4] and encouraging people to cycle [5] as a popular, healthy, normal, everyday activity that doesn’t need special equipment such as helmets.

References

(1) Cook A, Sheikh A. Trends in serious head injuries among English cyclists and pedestrians. Inj Prev 2003; 9: 266-267.

(2) Bryan-Brown KCN. Cycle helmet wearing in 1999. Transport Research Laboratory.

(3) Robinson D. Reasons for trends in cyclist injury data. Injury Prevention 2004; 10: 126-127.

(4) Robinson DL. Head injuries and bicycle helmet laws. Accid Anal Prev 1996; 28: 463-475.

(5) Jacobsen PL. Safety in numbers: more walkers and bicyclists, safer walking and bicycling. Inj Prev 2003; 9: 205-9.

Dear Editor

To focus on injury mitigation in cyclists to the exclusion of addressing the causes of crashes, as is the trend in public debate at present,[1] risks fundamental errors - not least the post hoc fallacy of assuming that cycling head injuries are the result of failure to wear helmets, rather than of the types of crashes cyclists experience.

As a result of this obsession we have arrived at the absurd position where the death of a child cyclist is publicly attributed not to the fact that he rode off the pavement into traffic on a bike with defective brakes, but to the fact that when he did so he was not wearing a helmet.

Recent analysis of Department of Health data on child hospital admissions for England for the period 1995/96 to 2002/03[2] showed that:

• the proportion of head injuries in child cyclists on the road is hardly different from that of child pedestrians (49% against 46%, with helmet wearing rates of 15%[3])
• the risk of head injury in off-road cycling is an order of magnitude lower than in road cycling;
• cycling is far from being the leading cause of head injury admission, being behind trips and falls and even assault.

Why is it, then, that cycling is seen as a uniquely dangerous activity, when a dispassionate look at these and many other statistics indicates very clearly that it is not? There are probably a number of factors at work:

• head injuries raise the spectre of intellectual disablement, which of course cannot be "fixed" by modern medicine, even though this is very rare - the fact that such injuries are now thought to be mainly the result of rotational forces which helmets do not mitigate (and may even aggravate) adds a touch of irony
• even trivial head injuries can bleed spectacularly, which combined with the fear factor, and justifiable anxiety over the cosmetic outcome, increases the likelihood of attendance at A&E “just in case”, even though in most cases treatment is limited to basic first aid - so that nurses, for example, "see a lot of cycling head injuries";
• there exists a substantial industry whose expensive product will not sell unless a culture of fear is maintained, and the protective effect of its product "sexed up" - few people would spend the price of a modern helmet if they were told bluntly that they are tested only for the equivalent of a fall from a stationary riding position, yet this is the literal truth (note the wide disparity between claims made by manufacturers and by helmet advocacy groups, who still quote as gospel the flawed 1989 Thompson, Rivara and Thompson study,[4] despite well- documented and acknowledged criticisms[5])
• the culture of fear extends in particular to the danger of motor traffic, with some justification as the estimated 10% of child cycling which is on-road results in half of all child cyclist admissions and almost all the deaths
• there is a false belief that nothing can be done about the source of this danger (i.e. drivers cannot be made to drive more carefully), and no amount of riding skill can reduce the danger; but cycle trainers report widespread ignorance of riding techniques which can avert many of the more common sorts of crash[6], and surely only political will is lacking in challenging driving behaviour
• fundamentally, most of those campaigning for helmets are not cyclists and have little understanding of the vast range of different activities and scales of risk which that term encompasses - it is as if all outdoor activity from afternoon walks in the park to free-climbing were considered under a single umbrella.

There is a pressing need to return debate to the sources of danger, and means of its reduction. Motor vehicles account for around a tenth of child injury admissions but half of all injury fatalities[7]. This increased risk is shared by cyclists, (far more numerous) pedestrians, and motor vehicle occupants. The danger is inherent in the source, not the victims’ activities, and no proper study of head injury in cyclists should ignore this fact.

It is also time to recognise that opposition to helmet compulsion, and to a lesser extent promotion, is not purely libertarian, but based on robust evidence. As road safety minister David Jamieson recently acknowledged, the Government knows of no case where cyclist safety has improved with increasing helmet use. There can be no justification for the continuing dominance of the cycle safety agenda by this single issue.

Most importantly, the principal indicator for cyclist safety is numbers cycling.[8] By deterring participation,[9] helmet promotion may actually reduce safety, in the name of mitigating a minority of those injuries cyclists may experience.

Perhaps if we all slowed down and drove more carefully the perceived need for helmets would evaporate - along with the terrible toll of road traffic fatalities.

References

For example: 1. Sheikh A, Cook A, Ashcroft R. Journal of the Royal Society of Medicine 2004;97:262-265

2. Data provided to CTC, the national cyclists’ organisation, by the Department of Health

3. Cycle helmet wearing in 2002 , Gregory K, Inwood C, Sexton B, 2003 TRL Report 578, p247

4. Thompson, Rivara & Thompson. New England Journal of Medicine 1989, Vol 320 No 21 p1361-7

5. For example: Robinson, http://www.cyclehelmets.org/papers/c2010.pdf

6. cf. Cyclecraft, John Franklin, The Stationery Office, 1997

7. Office of National Statistics, Deaths by age, sex and underlying cause, 2003 registrations: Health Statistics Quarterly 22,

8. Assessing the actual risks faced by cyclists Wardlaw M Dec 2002 Traffic Engineering and Control.

9. Cycle Helmet Wearing in 1996. Bryan-Brown K & Taylor SB, TRL Report 286.

Dear Editor

The main conclusion of Cook and Sheikh,[1] that a bicycle helmet prevents 60% of head injuries, is incorrect due to a fundamental error in the way they have treated their percentages. A correct analysis demonstrates unequivocally that there must be major confounding factors in their data set that they have failed to take into account, and therefore any estimate of helmet effectiveness is purely speculative.

Assuming that their basic analysis of the data is correct (although the numbers they quote in the text do not actually appear to match the figure plotted), they arrive at a figure of a 3.6% for the reduction in the head injury (HI) rate for cyclists, over and above the "background" reduction that pedestrians have also seen. They assume that this drop in HI is due to increased helmet-wearing. However, this reduction is presented in terms of the number of percentage points, and relative to the baseline value of 27.9% HI for cyclists in 1995-6 it actually represents a 3.6/27.9 = 13% drop in the HI rate.

The decrease in the number of helmetless cyclists over the same interval is 5.8 percentage points from a baseline of 84% unhelmeted, giving the percentage drop as 5.8/84 = 7%. Cook and Sheikh calculate helmet effectiveness to be given by the ratio 3.6/5.8 = 60%. However the correct expression to use is 13/7 = 186%. In other words, "helmet effectiveness" is so high that each helmet does not just save its wearer, but a non-wearer too. At this rate, head injuries would be eliminated completely if just a little over half of all cyclists wore them! This is clearly ludicrous.

A more reasonable conclusion to draw from this would be that there are some other factors that are responsible for the large drop in HI rate, and therefore any attempt to attribute some part of the total 30% (8.49/27.9) change to the provably marginal impact of a very small number of extra helmet wearers is at best highly speculative and fraught with inaccuracy.

What makes this all the more poignant is the fact that the authors have recently produced a book entitled "Basic skills in statistics"!

Reference

1. A Cook and A Sheikh Trends in serious head injuries among English cyclists and pedestrians Inj Prev 2003; 9: 266-267.