1. Introduction
This is a review of the South Australian Office of Road Safety report 8/94 subtitled “Evaluation of the Compulsory Helmet Wearing Legislation for Bicyclists in South Australia”. This report (herein referred to as SAORS 8/94) shows that authorities in South Australia have been committed to the policy of helmet wearing for cyclists since 1984, if not earlier. At that time, indeed even now, the efficacy of helmet wearing as a means of reducing injury to cyclists was far from established by scientific studies. The authorities in South Australia relied on a study by Dorsch and others, but its findings were preliminary only. The commitment to the policy was therefore premature. Clearly, if a study were now to reveal that fact, and that the helmets legislation has not worked, or even been counter- productive, persons who were party to the decision to introduce it would be severely embarrassed at the least. The South Australian Office of Road Safety is such a person, being an arm of the Department of Transport. SAORS 8/94 is therefore not a disinterested assessment, as becomes evident from scrutiny of it.
That the authorities in South Australia, and elsewhere in Australia, had no serious doubts about the efficacy of helmet wearing is evident from their failure to establish systems to monitor the effects of helmets legislation, in particular the effect of discouraging cycling.
SAORS 8/94 is based on fairly limited data from which the authors concluded that the helmets legislation has been successful in reducing head injuries and recommended that it be continued. This recommendation appears to have been based as much on the author’s unquestioning faith in helmets as on any findings in their report. Legislators in other Australian States and Territories have since cited the SAORS 8/94 study in support of their own helmets legislation, but its findings need to be seriously questioned. Comment is first made on the study by Dorsch and others, then on the three measures by which SAORS 8/94 evaluates the effect of the helmets law, namely the number of cyclists (exposure), helmet wearing rates and the number of hospital admissions for cyclist casualties.
2. The Dorsch Study
Page 1 of SAORS 8/94 notes a study (Dorsch, 1984) which had shown “the chance of death in a cycling accident decreased by a factor of 19” when wearing a hard-shell helmet as opposed to no helmet. However, data from NSW from 1992-94 show that 80% of cyclists killed on the roads in NSW (RTA, 1992-94) were wearing helmets (excluding cases where helmet wearing status was unknown), which is similar to the proportion of cyclists observed wearing helmets (76% for children and 85% for adults). This is hardly even close to a factor of 19 – more like a factor of one.
The House of Representatives Standing Committee on Transport Safety noted in its 1985 report that the 1984 paper on research by Dorsch and others “has received almost universal acceptance by bicycle groups who have been working for many years to have bicycle helmets widely accepted on the basis of their effectiveness in reducing head injuries”. Like the Committee itself, bicycle groups assumed in advance of evidence the efficacy of helmets in protecting against injury and welcomed any findings, however tentative, that favoured their assumption. Certainly, the Dorsch paper has contributed to knowledge, but has been largely discounted by subsequent researchers.
Dr. Dorsch gave evidence to the House of Representatives Standing Committee on Transport Safety 1985. On page 901(a) of evidence she said: “One has to be very careful in making estimates of how effective universal bicycle helmet usage would be in reducing deaths and serious injuries … People who are wearing what we regarded as the good hard helmet … had 19 times less risk of suffering a fatal head injury. That was a hypothetical procedure. … It was based largely on an adult group of cyclists and because we went through a rather hypothetical statistical procedure to arrive at those numbers, I think one would have to be very careful in generalising those findings to very young bicyclists. … One has to be very careful in making estimates of how effective universal bicycle helmet usage would be in reducing deaths and serious injuries … In our paper we did, sure, put estimates on it but as a very hypothetical procedure. I was a bit distressed by some of the reports I had seen that suggested that 75% of deaths could be prevented by everyone wearing very good hard helmets. “. Here we point out that the Australian Standard for bicycle helmets was amended in 1990 to allow helmets with a soft shell. This amendment was made to make helmet wearing more acceptable, but was contrary to the recommendations of commissioned research.
The authors (Dorsch and others) introduced their published paper of 1987 with the words, “In the past, evaluation of helmet efficacy has been based on laboratory tests of limited relevance to real crashes. … Helmets for bicyclists could do much to reduce deaths and injuries among crash-involved riders. While few people would doubt this assertion, there are currently no quantitative data demonstrating the efficacy of bicycle helmets in real crashes.”
The authors sent 1321 questionnaires to members of bicycling clubs in South Australia, seeking information about their most recent crash. Out of 866 usable responses, “197 bicyclists reported a crash within the last five years in which they had struck their head or helmet. … it was estimated that the risk of death from head injury was considerably reduced for helmeted relative to unhelmeted bicyclists. … Further research is needed to confirm and refine our findings.”
The authors themselves acknowledged limitations of their study. Their sample was mainly drawn from members of bicycle racing clubs, and therefore would be far from representative of the generally younger and slower-travelling general population of cyclists. Thompson and others (1989), noted their findings and added, “Because of methodologic limitations, none of the available studies have produced compelling evidence of the effectiveness of bicycle helmets.” Hillman (1993) made comments to similar effect.
The Dorsch study also found 62% of cyclists who reported hitting their heads were wearing helmets at the time, but the percentage of helmet wearers would have been extremely low; by 1990, after 4 years of promotion, it had reached only 15% for cyclists over 15. This suggests that helmeted cyclists are more likely to have an accident and/or strike their heads. This could also be related to an observation that helmeted cyclists are often very keen to recount the incident of how their helmet “saved their life”, and may therefore have been over- represented in the questionnaire because they were more likely to respond. Any over-representation of helmeted cyclists in the data may have distorted the results.
3. Exposure (i.e. Reduction in number of cyclists)
South Australian authorities did no systematic monitoring of numbers of cyclists – as they say, “due to the disparate nature of the results from different sources, it is not possible to be conclusive about the effect of the requirement to wear bicycle helmets on the number of cyclists”. They guessed, on page 32, that the decrease in the number of cyclists was the same as the decrease in the “non preventable injuries”, but it was substantially greater in NSW where non-head injuries to children declined by 13 per cent (RTA), but cyclist numbers declined by 39 per cent (Walker). As their guess is also a factor in their calculations of “helmet effect”, those too are only guesses.
SAORS 8/94 cited cycling surveys and hospital data to determine the effects of reduced exposure to cycling on a change in the occurrence of head injuries. The authors admitted that due to the inadequacy of the various sources it is not possible to be conclusive about the effect of helmet legislation on number of cyclists. In other words, the South Australian Government was negligent. They should have ensured proper and comprehensive surveys were in place before and after the law was introduced. Indeed the second recommendation of the report was that “regular and more representative surveys be conducted”, but it’s a bit too late for that now!
In a letter dated 2 February 1996, SA ORS advised that they proposed to conduct future bicycle counts at suburban checkpoints, in addition to those already conducted at schools and around the city of Adelaide. It is no longer possible to determine accurately the extent of any decline in number of cyclists due to helmet legislation. The proposed surveys could be expected to show an increase in cycling from year to year, but this should not be used to claim that cycling has recovered from any loss due to compulsory helmets, as cycling around Australia was reported to have been increasing rapidly before any helmet laws were enacted. The Government should still be encouraged to conduct these surveys however, as they may prove highly useful for monitoring trends should the helmets law be repealed.
Comments on survey methodology: Self-reporting surveys are at best unreliable. The only reliable measure of numbers of cyclists is to count them as they pass given checkpoints. The exact same checkpoints should be used for both before and after surveys, and where possible even the same people employed at the same checkpoints. The checkpoints should be strategically placed so as to get a true representation of all cyclists (for example, some surveys in the past have counted only commuters, or only children entering schools). The surveys should be conducted at exactly the same time of the year since there have been shown to be great variations from month to month, and as much as possible in the same weather conditions. This was done in NSW for children (although, as is typical in these surveys, the real purpose was to record the helmet wearing rate) and, compared to the survey in April 1991, just pre-law, revealed a reduction in cyclists of 36 per cent by April 1992 and 43 per cent by April 1993.
3.1. Health Omnibus Surveys – Frequency of Cycling
These surveys showed no significant decline in cycling. The findings are questionable for a number of reasons which the authors of SAORS 8/94 failed to identify.
1) The before and after surveys were conducted in 1990 and 1993. 1991 was when the law came in, but what happened with 1992? Thus we have a three year gap between the before and after periods during which any number of factors other than a helmets law may have influenced cycling.
2) The data were all self-reported. Reliability is therefore extremely questionable.
3) The questions asked fail to quantify cycling sufficiently. Suppose that a person cycled every weekday before the law, but afterwards only once per week? They would still be in the “once per week” category and no decline would have been recorded for them, but their decline in cycling would actually have been 80 per cent. The questions were totally inadequate to record any decline in cycling due to compulsory helmets.
4) Tables 6a and 6b record that about half of bicycle trips and the greatest increases for bicycle trips for children under 15 years were on their own property, where they don’t have to wear a helmet!
In a letter dated 2 February 1996, SA ORS advised that they would redevelop relevant questions for inclusion in the annual Health Omnibus Survey.
3.2. City of Adelaide Bicycle Cordon Count
These surveys recorded very similar results for the years 1985 till 1992 (except for 1988 which was attributed to changed survey conditions for that particular year). The authors noted that 1991 (immediately after the law was introduced) actually recorded a slight increase. No mention was made of weather conditions though. The authors also failed to do a test for significance. The 2.9% increase in 1991 could well be within the bounds of normal variance as there appears to be some degree of variation in the data from year to year.
In Table 7, the survey for 1993 showed a 15 per cent drop in cycling, though the authors took great pains to try and explain this away to the weather and the impending Grand Prix. The surveys were delayed by the weather and the authors reasoned that fewer people may have cycled due to uncertainty about the weather. They failed to reason that (bearing in mind the surveys were eventually conducted in fine weather), cyclists may have been keen to get out and cycle having been kept indoors by the previously bad weather and further, because of the delays in the survey, it was getting closer to the summer months when the weather gets warmer and more people ride their bicycles.
Since the Grand Prix was close to the time of the surveys the authors reasoned that fewer people may have cycled due to uncertainty about road closures or detours. They failed to reason that people may have wanted to cycle near the Grand Prix circuit out of curiosity, and that bicycles are less susceptible to road closures and traffic jams than cars. Those people who do not wear helmets may also have cycled less because of the presence of police officers accompanying the road closures and detours.
Another observation which could be made is about the lack of a decline immediately after the law, followed by a 15 per cent decline two years after the law. This could perhaps be related to the characteristics of the commuter cyclist. The authors pointed out that this survey tended to record mainly commuter cyclists in city traffic. Commuter cyclists in city traffic are probably more likely than the average cyclist to wear a helmet in the first place and more likely to start wearing a helmet because of a law, and less likely than recreational cyclists to reduce or stop cycling. Immediately after the law the non-helmet wearing commuter cyclist may have continued to ride as before either with or without a helmet. Two years later, some of those wearing helmets against their will may have become discouraged with the reality of wearing a helmet which they find uncomfortable, and most of those not wearing helmets have by now felt the wrath of the law, or even the wrath of people who make it their business to yell at people for not wearing a helmet. The net effect could be a delayed decline in cycling. People who have given up cycling under these circumstances are unlikely to take it up again so long as there is still a requirement to wear a helmet.
It would seem that almost anything can be read into these surveys. The reasoning which the authors applied (in contrast to ours) to the results of the cordon count would seem to indicate that they had a bias which was to dismiss the possibility that the law has reduced cycling.
3.3. Observational Studies of Helmet Wearing Among South Australian School Children
The authors note that Harrison’s (1994) study of school children showed a 38 per cent decline in cycling from September 1988 to March 1994. Due to seasonal factors, however, cycling is more popular in March than September in southern Australia; at checkpoints in Western Australia in 1993 and the Australian Capital Territory in 1993/94, less than half the numbers of cyclists were counted in September as in March. Given also the general increasing trend of cycling shown in other states in the period before compulsory helmets were introduced and that there is a six year before/after period, the decline due to the helmets law could well have been much greater than 38 per cent.
These facts are in contrast to the arguments of the SAORS 8/94 authors, who on page 31 attempt to pass this decline off as a result of a decrease in popularity of bicycles, and increasing concern of parents about the safety of children travelling to school alone. One very likely reason for bicycles becoming less popular, would be the fear put into people as a result of Government helmet propaganda (for example, television commercials showing eggs being hit by hammers). As for the other point, surely children would be at much less risk from “the Bogey man” with the increased speed and mobility of a bicycle compared with walking to school or even walking to a bus stop, so are the authors saying that 38 per cent of parents have decided to drive their children to school rather than let them ride their bike? This would mean an increase in motor traffic, in particular around schools. It can be argued that cars stopping to drop children off create a dangerous safety environment for children. Further, a one person- bicycle trip would have been replaced by three person-motor vehicle trips since the parent must drop their child off and then return home. Also, some children who walk instead of ride will suffer injury.
3.4. Other Cycling Surveys
Given the data available, the authors have rightly pointed out that “it is not possible to be conclusive about the effect of the requirement to wear bicycle helmets on the number of cyclists”. However there are some pre and post law cyclist counts from other states which provide data with considerable reliability, and they all indicate rather large declines. It may be unreasonable to assume that South Australians are significantly different in their “acceptance” of the helmets law than other Australians.
Here is a brief summary of surveys that we are aware of:
1) Pre-law in Victoria, some students at schools which compelled them to wear helmets chose to give up cycling instead (evidence given to House of Representatives Standing Committee on Transport Safety, 1985, page 1078).
2) According to Vic Roads report IR 90-15 (1991), the number of cyclists described as adult commuters declined by nearly 60 per cent between March and July 1990. This figure is invalid because of the different times of the year involved (the decline is probably attributed to the usual reduction of cycling in winter months), although the July survey was taken in fine sunny weather.
3) Morgan and others (1991), in surveys in March 1990 and March 1991, showed declines of 30 per cent in recreational cycling and 32 per cent in commuter cycling in Victoria.
4) Surveys in Melbourne by Monash University Accident Research Centre (MUARC) showed total bicycle use by children had decreased by 36 per cent. Finch and others provide data showing the following decreases in numbers of bicyclists observed during the first year of the law: adults 29 per cent, teenagers 46 per cent and children 24 per cent.
5) A baseline survey by Walker for the RTA of NSW showed an average post-law decline for cyclists under 16 of 39 per cent (see above).
6) Surveys by Walker for the RTA of NSW were inconclusive for adult cyclists. A decline was shown since April 1991, but this was post-law.
7) In 1990 and 1991, branches of the Royal Automobile Club of Queensland conducted surveys, which showed a reduction of 22 per cent for schools in total, but because of changes in survey conditions the decline was probably more in the order of 30 to 50 per cent. Further, this decline occurred before the law was enforced.
8) In their survey, Healy and Maisey (1992) commented that the numbers of children cycling to primary schools and numbers of recreational cyclists (in WA) declined from 1991 to February 1992.
9) Heathcote (1993) presents limited data which show some decline in numbers of WA children cycling to school. Heathcote’s limited observations of “commuter” cyclists indicate an increase in numbers after the helmet law, but his observations of cyclists classed as recreational show a decline of over 50 per cent. Data from automatic counter surveys conducted by Main Roads showed a decline of 38 per cent from October-December 1991 to October-December 1992.
10) WA Main Roads also counted numbers of cyclists crossing the Narrows and Causeway bridges on weekdays, which show sharp declines in cycling.
11) Surveys by the Road Safety Council of the Northern Territory showed there was little change pre-law to post-law in the numbers of children cycling to primary school, but a 36 per cent decline in cycling to high school. The decline in numbers of “commuter” cyclists was of the order of 50 per cent.
12) Members of the Cyclists’ Rights Action Group carried out a survey of 325 cyclists in the ACT in May 1992, pre-law, and found that 28 per cent (or 90 respondents) would cycle less if helmets were compulsory.
13) Ratcliffe (1993) reported that mean weekday cycle path daily volumes in the ACT were recorded in 1992 to be about one third lower than the similar period in 1991, with mean weekend daily volumes declining by about half.
14) ACTUS followed up with another survey in 1993/94 which found a 34 per cent increase from 1992, almost back to pre-law figures. However, the 1993/94 surveys were conducted mainly in February, whereas the 1992 surveys were in December. Comparisons of those few sites with figures available for the same months, show a total count of 7763 for 1992/93 versus 7810 for 1993/94, an increase of 0.6 per cent.
4. Helmet Wearing Rates
Tables 8 through 13 of SAORS 8/94 are all self reported. If asked after the law if they wore a helmet, how many people would have lied, not wanting to admit to a lawbreaking activity, and said “yes” when the answer is in fact “no”, or even “yes (but only when I think I can’t get away with not wearing one)”. In any case it is only an indication of helmet wearing, and nothing to do with helmet law acceptance.
5. Hospital Admission Data
A number of points arise from the hospital admission data, and subsequent analysis of the effect of the helmets law.
5.1. Hospital Admissions for Under 15’s
The number of admissions for under 15’s was much the same in the first year of the law as for the first three pre-law years of the data. This is in spite of increased helmet wearing rates and reduction in number of cyclists. The number of admissions in the second year of the law actually increased. So with fewer cyclists wearing more helmets, hospital admissions were the same or more. Helmet laws were largely meant to protect the children, yet it appears that they have done children more harm than good.
5.2. Hospital Admissions for Over 15’s
The number of admissions for over 15’s showed a solid decrease after the law. This is inconclusive however in light of the absence of an accurate measure of the decline in cyclists. It is also established that older children and adults are more likely to have accidents involving motor vehicles than are young children (Hillman, 1993) so the fact that hospital admissions for over 15’s declined but under 15’s did not is likely to be related to improved traffic conditions and less motor vehicle involvement.
5.3. Improved Traffic Conditions
There is no account taken by the authors of the effect of improved traffic conditions, as indicated in Figure 19 of their report by the general decreasing trend of all road crash casualties. It is noted on page 31 that casualties remained high until around June 1991 at which point they showed a substantial decrease. This was at the same time that the helmets law was introduced. The authors rightly suggest that the decrease was possibly due to a combination of factors, including speed cameras and 0.05 BAC limit. This may be coincidental, but the requirement for compulsory bicycle helmets and the 0.05 BAC limit were both part of the Federal Government’s 10-point black spots road funding package in 1989.
Commenting on the decline as shown in Figure 19 of their report, the authors say on page 31 that “It would appear from this that the reduction in number of bicycle casualties admitted to hospital may not be attributed to any indirect effect of helmet legislation”. This statement is rather tentative and open to misinterpretation, but it seems they are claiming that the reduction in all bicycle casualties (not just head injuries) might have nothing to do with any reduction in cycling, i.e. that cycling did not in fact decline at all and the real reason there were fewer (non-head) injuries was because of the safer roads!
5.4. Involvement of Motor Vehicles
From Table 21 (Bicycle Accidents Resulting in Hospital Admission Involving A Motor Vehicle) the authors note that the majority (75-80%) of accidents resulting in hospital admissions do not involve a motor vehicle. This is correct, but, as noted by Hillman (1993): “The great majority of accidents are minor. They characteristically involve cyclists losing control of their cycles and falling off. Other vehicles are not involved. Injuries are rarely serious; admission to hospital usually reveals only short-term concussion. But when accidents are serious, they generally involve damage to the head following collision with a motor vehicle. This is true for the large majority of fatalities and about one half of the serious injuries. Adult cyclists are most frequently involved.” In Australia in 1988, a moving motor vehicle was involved in 92 per cent of fatalities to cyclists.
The SAORS 8/94 authors noted the decreasing trend of percentages for motor vehicle involvement in Table 21, but failed to realise that this is likely to be a direct result of the roads being made safer for cyclists. As a result of less motor vehicle involvement, it can be expected that the percentage of non-head injuries will increase in proportion to injuries involving the head. In South Australia, North et al. (1993) noted “We have recently observed an apparent fall in the number of patients suffering from head injury due to road trauma …. The largest drop in patient numbers was observed in motor cyclists, falling from an average of 24 per year previously, to only five in 1992.” This is unlikely to be related to helmet wearing since compulsory helmets for motor cyclists has been in effect for two decades. It appears to be well established that there was a decreasing trend in head injuries for all road users in South Australia, not just cyclists. To determine this effect on cyclists, a comparison between cyclists and pedestrians would be appropriate.
For Victoria, Robinson (1996) found that the percentage of head injuries in claims to the Motor Accidents Board for pedestrians and cyclists aged 0-16 from 1980-85 shows a correlation of 0.94, statistically significant with P < 0.02. Major initiatives directed at drink-driving and speeding were introduced in Victoria in December 1989 and March 1990 (Cameron et al. 1994), and pedestrian fatalities fell by 42 per cent from 159 in 1989 to 92 the following year. It is likely these initiatives also reduced impact speeds and hence head accelerations and the proportion of victims with head injuries. It would seem that any reduction in head injuries would have had more to do with RBT and anti-speeding campaigns than with the use of helmets.
As Robinson (1996) pointed out, “the cause of any common trends in percentage head injured for pedestrians and cyclists may relate to impact speed of motor vehicles”. Janssen and Wismans (1985) observed a 40 per cent reduction in maximum head acceleration of dummy pedestrians hit by vehicle fronts, compared with reductions of 30%, 22% and 10% for chest, pelvis and feet, when vehicle impact speed was reduced from 40 to 30 kph. The effect for dummies on bicycles was similar. Maximum head acceleration reduced by 50 per cent, compared with 30%, 16% and 36% for chest, pelvis and feet.
It is well established that a reduction in motor vehicle involvement and/or speed will reduce the incidence of head injuries significantly more than non-head injuries.
As mentioned earlier, Table 21 shows that about 80 per cent of cycling accidents resulting in hospitalisation do not involve a motor vehicle and further that motor vehicle involvement had declined. However, it is also well documented that over 90 per cent of serious head injuries to cyclists do involve a motor vehicle (Hillman, 1993). Therefore, even without helmets, we can reasonably expect the number of serious head injuries to decrease in proportion to other injuries for cyclists. In fact, from a mere 2 percentage-point change in motor vehicle involvement (say from 20% to 18%) we could expect about a 10 per cent reduction in serious head injuries compared to other injuries! It should also be pointed out that, in addition to a lower frequency of motor vehicle involvement, the average speed of the motor vehicle in those accidents might also have been reduced thanks to speed cameras and other measures which were in force, therefore further contributing to the reduction in head injuries.
From Table 21 of SAORS 8/94, motor vehicle involvement declined from 21.3 per cent in 1989/90 to 18.9 per cent in 1992/93, or by 11.3 per cent (2.4 percentage points) over the same period as the two year before/after period used to obtain the 24.7 per cent reduction in PPI (Potentially Preventable Injuries, ie injuries which a helmet might have prevented – see page 30 of SAORS 8/94). For the one year before/after period, with only a 12.1 per cent reduction in PPI (see page 29), the decline in motor vehicle involvement was from 19.7 per cent to 18.3 per cent, or by 7.1 per cent (1.4 percentage points). These figures of 24.7 versus 2.4, and 12.1 versus 1.4, are roughly proportional and provide an alternative explanation for the reduction in head injuries than simply the use of helmets.
The authors of SAORS 8/94 did not consider any of the above when they calculated their figures of a 12.1 per cent and 24.7 per cent reduction in potentially preventable head injuries, which they attributed entirely to helmets. They also failed to explain why there was such a great disparity between the one- and two-year figures, one of which was twice as large as the other! It is unlikely that the efficacy of helmets would have changed by so much in the space of just one or two years.
5.5. Cyclist Risk Taking
Another possible explanation for the data is that cyclists may be taking more risks now that they are wearing helmets. Risk compensation is fairly well documented. Hillman (1993) cites research on risk compensation and notes that “discussion of this subject of behavioural adaptation no longer centres on whether it occurs but on how complete it is”. By comparing head with non-head injuries and attributing the difference to the benefits of helmet wearing, the SAORS 8/94 authors have ignored the possibility that cyclists may be taking more risks and therefore having more accidents and/or more severe accidents. The net effect of this may be more non-head injuries with head injuries remaining the same due to the increased risk taking cancelling out the (limited) benefit of bicycle helmets. This is clearly not a beneficial or desirable outcome of the helmets legislation.
There is certainly no shortage of data to support the hypothesis that cyclists are taking more risks due to helmet wearing. Here are a few examples :
1) Rodgers (1988) studied over 8 million cases of injury and death to cyclists over 15 years in the USA. He concluded as follows: “There is no evidence that hard shell helmets have reduced the head injury and fatality rates. The most surprising finding is that the bicycle- related fatality rate is positively and significantly correlated with increased helmet use.”
2) The South Australian under 15 hospital admissions data showed no decline and even an increase in head injuries, in spite of any declines in cyclist numbers.
3) As previously stated, non-head injuries to children in NSW declined by substantially less than the decline in number of cyclists. This result happened despite declines in injuries to pedestrians and motorists: an indication of safer roads.
5.6. No Tests of Significance
There were no tests of significance done on the important data in Tables 24 and 25, from which conclusions of helmet effect were drawn. This would be unacceptable for any respectable scientific journal.
5.7. Classification of Injuries
The injuries listed in Appendix 1 as being potentially preventable by a helmet are biased. How could a helmet protect nasal bones, upper facial bones, eye, eyeball, nose and forehead? Research commissioned by the Federal Office of Road Safety recommended that the standard for bicycle helmets, AS2063, should be amended to extend protection to the temporal area, but it was not done in 1990, when other amendments to the standard were made. Another study (Acton, 1996) found that bicycle helmets were not preventing facial injuries in children.
It should also be pointed out that many of the injuries listed as being preventable by helmets are of a superficial nature (e.g. “open wound of scalp, nose”) and that helmet wearing is likely to result in an increase in the more serious diffuse brain injury (National Health and Medical Research Council, 1994).
6. Conclusions
1) There are no data from South Australia which adequately measures the true change in number of cyclists. Proper studies should have been set up but were not. It is no longer possible to determine with any certainty the effect legislation has had on the number of cyclists in South Australia.
2) High helmet wearing rates are seen as evidence of the effectiveness of the law, which seems to imply that the purpose of the law is simply to force cyclists to wear helmets at all costs, as opposed to saving lives and reducing injury.
3) When placed under scrutiny, the conclusions reached that the helmets law has reduced head injuries by 12.1 per cent to 24.7 per cent are without proper foundation. Further, these figures were derived without proper statistical tests of significance. It is not even established that rates of head injury to those still cycling after the helmets law declined; indeed, interstate experience indicates that the reverse is more likely. In any case, reduced head injury rates are more likely to be a result of improved road conditions due to factors including anti- speeding and RBT measures, than due to helmets.
4) Bicycle helmet legislation should be repealed in South Australia and elsewhere. There has been much harm done (loss of cycling benefits, infringement of civil liberties) for little or no gain. It is about time Governments were honest and admitted that the helmets legislation has been a failure. Recommending that legislation be continued whilst having (at best) flimsy evidence of its effectiveness is a case of reverse-onus – it is the Government that should have conclusive proof to support legislation.
5) The South Australian Office of Road Safety Report 8/94 is very limited in its value for the purposes of evaluating the effect of compulsory bicycle helmet legislation. Its conclusions were evidently motivated by the political desire to justify the current compulsory helmets policy of the Government.
6) The Dorsch study has been disregarded by most researchers.
7. Acknowledgements
Special thanks to:
D.L. Robinson, UNE
The South Australian Office of Road Safety
8. References
ACT Bicycle Volumes; Trends, Variations, Survey Results, Helmet Wearing And Cyclist’s Characteristics, 1991 – 1993/94; Report prepared by the Traffic Monitoring Unit, ACT City Services, Department of Urban Services, PO Box 158 Canberra ACT 2601, Australia.
Acton, Caroline H C, Nixon, James W, Clark, Ronald C ; Bicycle riding and oral/ maxillofacial trauma in young children, MJA 1996; 165: 249.
Cameron, M.;Vulcan, A.P.; Finch, C.F.; Newstead, S.V. Mandatory bicycle helmet use following a decade of helmet promotion in Victoria, Australia – an evaluation. Accident Analysis and Prevention, Vol 26, No 3, 1994a.
Dorsch, M.M., Wodward, A.J. and Somers, R.L., Do bicycle helmets reduce severity of head injury in real crashes?, Acc. Anal. & Prev. 19, 3, pp. 183-190, 1987
Finch, C.F., Heiman, L. and Neiger, D., Bicycle use and helmet wearing rates in Melbourne, 1987 to 1992: the influence of the helmet wearing law, Monash University Accident Research Centre report no. 45, February 1993
Harrison, R., Observational Study of Bicycle Helmet Wearing Amongst South Australian Schoolchildren, for the Office of Road Safety, Department of Transport, South Australia, April 1994.
Healy, M. and Maisey, G., The impact of helmet wearing legislation and promotion on bicyclists in Western Australia, Traffic Board of Western Australia, Perth, 1992
Heathcote, B., Bicyclist helmet wearing in Western Australia: a 1993 review, Traffic Board of Western Australia, Perth, 1993
Hillman, M., Cycle helmets, the case for and against, Blackmore Press, Longmead, Shaftesbury, Dorset, 1993
Janssen, E.G.; Wismans, J.S.H.M. Experimental and mathematical simulation of pedestrian-vehicle and cyclist-vehicle accidents. Proceedings of the 10th International Technical Conference on Experimental Safety Vehicles, Oxford, July 1985.
Marshall, J.; White, M.; Evaluation of the Compulsory Helmet Wearing Legislation for Bicyclists in South Australia, South Australian Department of Transport, Office of Road Safety Report Series 8/94, November 1994.
North, B.; Oatey, P.; Jones, N.; Simpson, D.; McLean, J.; Head injuries from road accidents – a diminishing problem? Medical Journal of Australia, 158:433; 1993
Ratcliffe, P., Bicycling in the ACT – a survey of bicycle riding and helmet wearing in 1992, ACT Department of Urban Services, Canberra, 1993.
Road Safety Council of the Northern Territory, Bicycle helmet wearing in the Northern Territory, GPO Box 1176, Darwin, 1992
Road Safety Council of the Northern Territory, 1993 bicycle helmet survey report, GPO Box 1176, Darwin, 1993
Robinson, D.L.; Head Injuries and Bicycle Helmet Laws, Accident Analysis and Prevention, volume 28, number 4, pages 463-75 (1996).
Rodgers, G.B., Reducing bicycle accidents: a reevaluation of the impacts of the CPSC bicycle standard and helmet use, Journal of Products Liability, 11, pp. 307-317, 1988
Smith, N.C. and Milthorpe, F.W., An observational survey of law compliance and helmet wearing by bicyclists in New South Wales – 1993, for the New South Wales Roads and Traffic Authority, Sydney, 1993
Thompson, R.S., Rivara, F.P. and Thompson, D.C., A case-control study of the effectiveness of bicycle safety helmets, The New England Journal of Medicine, 320: 21, 1989
Vic Roads, Initial Effects of Mandatory Bicycle Helmet Wearing Legislation”, July 1990, report IR 90-15, Melbourne, 1990
Walker, M.B., Law compliance and helmet use among cyclists in New South Wales, April 1991, for the New South Wales Roads and Traffic Authority, Road Safety Bureau consultant report CR 1/91 Sydney, 1991
Walker, M.B., Law compliance among cyclists in New South Wales, April 1992, a third survey, for the New South Wales Roads and Traffic Authority, Network Efficiency Branch, Sydney, 1991
Wikman, J. and Sims, C., Bicycle helmet wearing survey 1990, Royal Automobile Club of Queensland, Brisbane
Wikman, J. and Sims, C., Bicycle helmet wearing survey 1991, Royal Automobile Club of Queensland, Brisbane
