Entrenchment of Helmet Laws in Australian Road Rules
by Bill Curnow
18 August 1997
Under an agreement by all governments in Australia, the National Road Transport Commission is developing Australian Road Rules that all road users in Australia will need to follow. Draft rules were released for public comment in 1994. In the light of comments received, a proposed rule to compel horse-riders to wear helmets was deleted, but, despite 62 per cent of replies opposing compulsory helmets for adult cyclists, it was retained, in rule 14.14. If approved by ministers, this rule will have the effect of entrenching compulsory helmet laws throughout Australia.
The reason given for retaining rule 14.14 was “safety considerations and the experience of States and Territories where helmet wearing is presently required.” CRAG requested the supporting evidence. In response, the NRTC cited two studies:
- Marshall, J. and White, M. (1994) Evaluation of the compulsory helmet wearing legislation for bicycles in South Australia, Office of Road Safety, SA Department of Transport; and
- King, M. and Fraine, G., Bicycle helmet legislation and enforcement in Queensland 1991-1993: Effects on helmet wearing and crashes, Road User Behaviour Section, Road Transport and Safety Division, Queensland Transport.
The NRTC cited Marshall and White as claiming: “a 12.1 per cent decrease in hospital admissions for cycling injuries that are potentially preventable by the use of a bicycle helmet” and King and Fraine as claiming that helmet laws have had some effect in reducing injuries beyond that associated with any decline in cycling.
CRAG then made the following submission to the NRTC, including analysis showing that these claims are not supported by the evidence. The submission read as follows:
CRAG’s Submission to the NRTC
In response to your letter of 27 September 1996 about bicycle helmet laws, I am writing to provide you with some more information on studies which show that injuries to cyclists have not declined in proportion to the decline in cycling, and to comment on the two studies which you cited. First, however, I want to say that there has never been a sound basis for the helmet laws.
No sound basis
There are, I suggest, three basic requirements for any legislation in a democratic society: first, the existence of a social ill requiring a legislative remedy; second, proven efficacy of the proposed remedy; and third, compliance with international standards for laws. Governments seem to have implicitly assumed head injury to cyclists to be a social ill, but adduced no evidence of any increase in it. In fact, deaths of cyclists in Australia had been declining by about 1 per cent a year in the 20 years before the first helmet laws; hospitalisations according to police data changed little from 1988 to 1990 before helmet laws; and, in the ACT, admissions of cyclists to public hospitals for head injury in the two years before the helmets law were much lower than a decade before.
The efficacy of helmet wearing in preventing head injury had not been established by proper scientific studies, and no consideration was given to the likely adverse effect of discouragement of cycling. The complex matter of efficacy is discussed in the paper attached as A.
International standards for laws in democracies proscribe laws to compel self-protection, there being no question of harm to another person, and laws which discriminate arbitrarily – see the Universal Declaration of Human Rights 1949 and the International Covenant on Civil and Political Rights, to which Australia has acceded. The helmet laws contravene these standards by compelling people to adopt particular means to protect their own persons. Also, the stated rationale for the helmet laws is to save on the social cost of caring for injured cyclists. Apart from the point of principle – where do you draw the line on restricting individual behaviour to save on costs of health care? – 17 times as many motorists as cyclists died from head injury in 1988, and FORS report CR59 recommends that motorists should wear protective headgear. To compel only cyclists to do so is arbitrary discrimination.
States and territories did not give due weight to these requirements. Enactment of helmet laws was of course to their financial advantage following Prime Minister Hawke’s 1989 offer of Federal funds for roads as a quid pro quo, but belief in the efficacy of helmet wearing as a safety measure would seem to have been decisive. Such efficacy may seem obvious, and government propaganda from the early 1980s reinforced that impression, but, as shown in the attachments hereto, it has no sound basis.
Helmet laws counter-productive
Dorothy Robinson, a biometrician at the University of New England, has adduced evidence in “Head injuries and bicycle helmet laws”, Accident Analysis and Prevention vol. 28, No. 4 pp. 463-475, July 1996, which shows that the greatest effect of the helmet laws has been to discourage cycling. For NSW, she found that neither the number of injuries to the head, nor to other parts of the body, declined as much as the estimated amount of cycling. For Victoria, her Table 5 suggests that, “for the same child cycle use as before the law, there would now be no fewer head injuries and more total injuries”.
Papers which Dorothy Robinson and Bruce Robinson, no relation, will present at the Velo Australis international bicycle conference in Fremantle on 30 October [1996] will contain further evidence on these lines.
I attach as B my own analysis of data for NSW which supports the following appraisal: “Contrary to the general trend to improved road safety, for those still cycling after the law the apparent risk of head injury increased by up to 25 per cent and other injury by up to 69 per cent.”.
This is consistent with a statistical study of 8 million casualties over 15 years in the USA by Rodgers, Journal of Product Liability, Vol. 11, pp. 307-317, 1988. He concluded that “there is no evidence that hard shell helmets have reduced the head injury and fatality rates” and “the bicycle-related fatality rate is positively and significantly correlated with increased helmet use.” My appraisal is also consistent with current theory of head injury as referred to in, for example, the National Health and Medical Research Council report cited in my letter of 3 September. Quoting from it:
“The mechanism of the production of an injury to the head is complex. It involves not only the effects of a direct impact to the skull and its coverings but, more importantly, the effects of the relative motion of the brain within the skull in response to that impact. This relative motion of the brain creates the shearing and rotational forces on the individual neurones which results in axonal stretching and diffuse brain injury. Concussion differs from the more severe diffuse axonal injury in that it represents the mild end of the spectrum of head injury, where the symptoms are transient and the injury does not necessarily result in structural brain damage.
“The addition of a helmet to the head will increase both the size and mass of the head. This means that blows that would have been glancing become more solid and thus transmit increased rotational forces to the brain. Because helmets distribute the force from focal impacts across a larger area, this may result in reduced fracture/laceration injuries but may increase diffuse brain injury.”
Not only has the risk of head injury to cyclists increased, but it is likely that the incidence of brain injury has increased disproportionately.
Studies to evaluate compulsory helmet laws- general
To ascertain the efficacy of compulsory helmet wearing as a public health measure would require a rigorous scientific study based on the testing of a coherent theory including mechanisms of brain injury, vide the NHMRC’s report, this being the injury that protagonists of compulsory helmet wearing profess to be mainly concerned about. Detailed and accurate data are also required. Such data were not available when the helmet laws were decided upon. As the Victorian Government’s submission to the House of Representatives Standing Committee on Transport Safety in 1985 stated: “the incidence of bicycle helmet use has not yet reached a sufficiently high level anywhere in the world for a scientific examination of helmet effectiveness in injury reduction to be undertaken”. Nevertheless, that inquiry recommended compulsory helmet wearing.
The decision-making process that led to the helmet laws did not include any evaluations based on theory of brain injury. The testing of helmets in accordance with the Australian Standard is of direct impacts, not impacts that generate rotational forces. Also, as our FOI requests have shown, the Federal Department of Transport made no evaluation of the efficacy of helmets and gave no advice on it to the ministers of the then Australian Transport Advisory Council who decided on compulsory helmets. Nor did the Department seek advice from the NHMRC or any other health authority. The helmet laws are a response to undue public fears and unfounded belief in the value of helmets.
Despite their unsound basis, the compulsory helmet laws in Australia, the world’s first, provided the best opportunity to that time to evaluate the efficacy of helmets in practice. Changes in cyclists’ behaviour in response to the laws, in particular by cycling less could have been measured. Monitoring systems could have been established to measure this, helmet wearing and casualties, including fine detail of head injuries. All data could have been matched and comparable throughout Australia. Then an accurate scientific evaluation might have been possible. It did not happen. Instead, state transport authorities, parties to introducing the laws, have made evaluations which are simply belated attempts to justify them. Their data, often fragmentary, were collected for other purposes. Such an evaluation is contained in a leaflet issued by the Roads and Traffic Authority of NSW in 1994, copy attached as C. Such also are the empirical studies by Marshall and White in South Australia and King and Fraine in Queensland.
NSW leaflet
The leaflet is full of inaccuracies and misleading statements, for example:
- The 1988 report of the Staysafe Inquiry referred to in the first sentence of paragraph 2 noted “Head injuries sustained by 85% of bicyclist fatalities”. In Australia in that year the figure was similar, 80%. What reason can there be then for attributing the statement in the second sentence to 1989 research in America, other than to make it appear that that research, indicating risk reductions of 85% and 88%, supported the Staysafe Inquiry? Also, the American research was empirical only and has been criticised – see for example Robinson’s paper cited above.
- In the section headed “Effectiveness of wearing bicycle helmets”, the numbers in brackets in the second sentence are not reductions in the percentages of head injuries as stated; they are percentage reductions in numbers of head injuries.
- All statements in the section are distortions because no account is taken of the post-law decline in cycling. It is not even mentioned. When account is taken of reduced cycling, the apparent risk of head injury per cyclist increased, contrary to the conclusion of the last sentence – unless “positive effect” is interpreted as an increase in the head injury rate!
Studies in South Australia and Queensland
Detailed comments on the study in South Australia by Marshall and White are attached as D. A few points are:
- The decision by Cabinet in 1984 was premature. Note the reference to Dorsch’s estimate of the chance of death decreasing by a factor of 19. In evidence to a House of Representatives committee in 1985, Mr Hallion, an officer of the S.A. Department of Transport, said studies using Dorsch’s results “show that universal helmet usage would reduce bicycle fatalities by up to 75 per cent and serious injuries up to 40 per cent”, but Dorsch herself refuted this in later evidence – see Attachment A hereto.
- 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”, though it was substantially greater in NSW – see Attachment B As their guess is a factor in their calculations of “helmet effect”, those too are only guesses, and the finding cited in your letter is unsupported.
- There is insufficient detail about head injuries to be of use for an evaluation in the terms of the NHMRC report. No tests of statistical significance were done.
The study in Queensland by King and Fraine began by assuming that the road safety benefits of helmet wearing are well established. Not so; the studies they cited all have serious deficiencies – see Attachments A and B.
King and Fraine also lacked accurate data on the post-law decline in cycling. They too assumed, implicitly, that it was equal to the decline in injuries other than to the head. Their data on injuries are fragmentary, many of the numbers are small and no tests were done for statistical significance. Also, they chose to use restricted data from the QISPP, not comprehensive data from the Queensland Road Crash Database from which the following table has been compiled.
Table: Cyclists Killed or Hospitalised with Head Injury (QLD)
YEAR 1990 1991 1992 1993 1994 1995p
1) Concussion 19 21 23 24 29 30
2) other intracranial 39 28 42 20 20 17
Subtotal - 1+2 58 49 65 44 49 47
3) Skull fracture 16 20 23 15 14 15
4) Open wound to head 47 27 33 20 14 18
Other injury 202 205 209 162 163 163
Simply because the legislation discouraged cycling, a commensurate decrease in casualties could be expected. Queensland authorities did not measure the decline in cycling, but it is likely to have been about 40 per cent, similar to NSW and Victoria. It would explain the decrease in skull fractures and open wounds to the head without assuming any benefit from helmet wearing.
The numbers for brain injury, concussion plus other intracranial, were higher in 1992 than in 1990, and from 1993-95 little different than in 1991. With fewer people cycling, the apparent risk of brain injury to the remaining cyclists therefore increased substantially. This is consistent with scientific understanding of the mechanism of brain injury as cited by the NHMRC, and with experience in NSW – Attachment B.
Lack of evidence of reductions in head injury has been noted elsewhere. In New Zealand, researchers found “increased helmet wearing has had little association with serious head injuries to cyclists as a percentage of all serious injuries to cyclists.” In Victoria, the proportion of head injured cases in the third year of the helmet law was found to be no different from the trend predicted by the model using pre-law wearing rate trends.
Consequently, the conclusion of King and Fraine that the helmet legislation has been effective in reducing the incidence of head injuries is not supported. I can provide further comment on their study if you wish.
Requests for inquiries
We have made several requests for the compulsory helmets policy/laws to be reviewed, but governments have adopted a defensive attitude.
In the ACT, we were able to present to the Chief Minister personally evidence that the helmets law is counter-productive. In response, she told us she “would be very happy for a legislative assembly committee to be established to examine this matter” and that it would be necessary to gain the support of MLAs for this. Some have stated their support, but Mr De Domenico, who is the Minister for Urban Services and responsible for the helmets law, responded as follows:
“The proposed Australian Road Rules … are being developed … and will automatically become ACT law under an existing Heads of Government agreement. Any amendment which may be made to the bicycle helmet laws in the Traffic Act 1937 will be superseded at the time of introduction of the ARRs. In light of the above, I am unable to support your request for the establishment of a review committee to examine this law.”
Mr De Domenico’s support for the helmets law goes back to its introduction in April 1992, when he was in opposition. The then minister in his second reading speech stated “There is overwhelming evidence that shows the lifesaving benefits of bicycle helmets.” Our FOI inquiries have since revealed that his department had done no evaluation of the efficacy of helmet wearing. Its only evidence was a short paper it received from the Federal Office of Road Safety a month before which claimed such “overwhelming evidence”. In the debate on the bill no MLA questioned this. We have pointed out the lack of evidence to Mr De Domenico, but to no effect.
Mr Whitecross, now leader of the Labor Party that introduced the helmet law when in government in 1992, said “Labor will not support an enquiry until strong evidence is produced that cycle helmets increase the risk of injury”. This is a reverse onus; governments have never produced evidence that helmets reduce the risk.
Conclusion
There has never been a sound basis for compulsory helmet wearing for cyclists. Governments acted prematurely without proper advice, in particular on the critical matter of the efficacy of helmet wearing and its adverse effects.
The helmet laws do not meet international standards for laws, as they infringe civil liberties and discriminate unfairly compared to other road users.
Compulsory helmet wearing has proved to be a counter-productive measure. Its main effect has been to discourage cycling, but, in states where sufficient data on the post-law decline of cycling are available, injuries have declined less than commensurately. This means that the apparent risk of injury to the remaining cyclists has increased. The apparent risk of head injury has not declined, and probably increased, and it is likely that the risk of serious brain injury has increased disproportionately.
There is evidence that transport authorities, for example the Roads and Traffic Authority of NSW, are suppressing and misrepresenting the truth about the effects of compulsory helmet wearing. This is a serious matter as the body of cyclists is largely composed of children and the disadvantaged. Such persons are not capable of making a critical evaluation of the merits of helmet wearing and necessarily have to trust in the advice of government authorities. This is as it should be, but authorities have betrayed this trust. It is clear then that radical reforms of their mode of operation are needed. A particular need is for a strengthening of capacity in science.
Another matter of concern, resulting perhaps from the lack of knowledge of scientific method, is the refusal of transport authorities to reconsider the helmet laws in the light of new evidence. Consequently, the original errors are being compounded, and the nation is the loser.
Finally, transport authorities seem to have the attitude that the onus is on opponents of the helmet laws to show that they are harmful. This attitude is the opposite of democratic practice. The onus should be on governments to show beyond all reasonable doubt that compulsory helmet wearing has substantial net benefits. Even if the studies in South Australia and Queensland were sound, they would not be enough when other studies show there is net harm.
Attachments
Attachment A – Efficacy of Helmet Wearing For Cyclists.
Attachment B – Cycling and Injuries in New South Wales.
Attachment C – NSW Leaflet.
Attachment D – Review of South Australian ORS Report 8/94.
The NRTC’s Response
The NRTC’s response noted that the Australian Road Rules are based on existing law. Only where there were differences in laws between jurisdictions, or a strong desire by the parties involved have the ARRs diverged from existing practice. Representatives of state and territory road authorities have indicated a desire to maintain compulsory helmet wearing laws.
CRAG then pointed out that our submission discredited states’ claims that their experience is that helmet laws increase the safety of cyclists. It seemed, however, that the NRTC had a co-ordinating role only and could not act directly pursuant to our submission. In the interests of facilitating evaluation of available knowledge on the efficacy of helmets, CRAG therefore requested the NRTC to circulate our submission to transport authorities and invite their comments. The NRTC did this. CRAG suggested that if the serious doubts raised about the efficacy of the helmet laws are not refuted, then no responsible authority concerned about the safety and welfare of cyclists could support their being entrenched as Australian Road Rules.
Subsequently, CRAG made requests direct to state and territory road authorities that their comments should include the evaluation, upon which their compulsory helmet wearing law is based, of the evidence of efficacy of helmets in protecting cyclists from injury. CRAG reported the responses received to the NTC on 14 August 1997, as follows.
State and Territory Responses
The responses received are as follows, in summary:
Victoria, by letter of 11 July 1997: “… your letter of 20 June requesting the information on which the decision to introduce compulsory helmet wearing was made. Please find enclosed a copy of the Regulatory Impact Statement that was prepared when the compulsory helmet law was introduced in Victoria in 1990.” I replied pointing out that that statement did not include any evaluation of the evidence of efficacy; it simply assumed, on page 24, that helmet wearing is efficacious. Victoria’s response, by letter of 31 July, was simply to say “The evaluations of efficacy used in justifying the introduction in Victoria of compulsory helmet wearing law were those published papers and reports which were available at the time.”
Comment: This, of course, is a non-answer. I might add that published papers at the time included many which cast substantial doubt on the efficacy of helmets, or on published evaluations of their efficacy – see papers cited in my submission to you of 29 October 1996, by Rodgers, Dorsch, Corner et al and the Victorian Government itself. In other words, the literature shows there was much disagreement at the time about the efficacy of helmets in protecting cyclists from injury. There still is.
Queensland, by letters of 20 June and 11 July 1997: “There appeared to have been a reduction in cycling following the introduction of helmet laws, but it was noted that there was a much greater reduction in injuries to the head than injuries to other parts of the body. We consider this evidence to be sufficiently compelling to justify retention of bicycle helmet laws.”
Comment: Another non-answer; their (unsound) evaluation is post-law.
Western Australia, by letter of 8 August 1997, also could provide only a post-law evaluation: the Report on compulsory helmet wearing for bicyclists, and other bicycling issues by the Select Committee on Road Safety of the Parliament of Western Australia, dated 12 May 1994. The report’s conclusion that helmet wearing reduces the severity of head injuries and saves lives is not supported by the data in it.
Northern Territory responded by email on 23 July, simply to say it had introduced compulsory helmet wearing as part of the Federal Government’s Blackspots Program, and went on to describe the legislation. There was no evaluation.
New South Wales, South Australia, Tasmania, and the ACT have provided no response, despite reminders. Also, in NSW, official claims have been promulgated to the effect that helmets protect from brain damage, but I have received no substantive reply, despite reminders, to my letter of 26 May 1997, copy attached, requesting the supporting knowledge.
In sum, no government in Australia has provided the evaluation, upon which the compulsory helmet wearing law is based, of the evidence of efficacy of helmets in protecting cyclists from injury. Against this, theory of brain injury and experimental evidence cited by Corner et al and the National Health and Medical Research Council, see references 31 and 24 in Attachment A to our submission of 29.10.96, suggest there is no good reason to expect helmets to protect from serious injury, and that they may aggravate brain injury. The obvious deduction is that governments do not have sound evidence of the efficacy of helmets. How, then, can helmet wearing laws be justified?
Conclusions – as at 18 August 1997
Governments in Australia have not shown that the wearing of helmets is beneficial in reducing injury to cyclists.
Indeed, theory and experimental evidence suggest that helmet wearing is likely to increase brain injury.
Surveys, where they have been carried out, have shown that numbers of cyclists declined by around 40 per cent following the introduction of compulsory helmet wearing laws, but injury and head injury to cyclists declined less than commensurately, indicating increased risk of injury.
The helmet wearing laws are therefore inimical to health and safety. The proposed entrenchment of them would be counter-productive.