Written in 1994
1. Introduction
This paper examines the rationale for compulsory wearing of helmets as stated by the then Federal Minister for Land Transport: that head injury to cyclists will be reduced, resulting in savings to the community in health care and other costs – in effect, an increase in public welfare.
Laboratory tests have shown that a helmet has the potential to mitigate injury to the head, but this is limited to impacts at low speed. It is a long way from the controlled conditions of the laboratory to the variable behaviour of people in the circumstances of accidents. Consequently, evidence of the efficacy of helmets in reducing injury to cyclists in practice is conflicting. Further, compulsion has other effects with potential to add to the costs of public health care.
It is true that rates of death and serious injury to cyclists in road accidents in Australia have declined. From 98 in 1989, the last full year before helmets became compulsory for cyclists in Victoria, fatalities declined by 58 per cent to 41 in 1992. There were 43 fatalities in the year ended October 1993. Data showing declines in serious injuries in New South Wales, Victoria and Western Australia are discussed below. The declines can be largely explained by factors other than the compulsory helmet laws. Improvements in roads and motorists’ behaviour have reduced rates of death and injury of all road users. More importantly, many people have responded to the compulsory helmet laws by giving up cycling, or, its statistical equivalent, cycling less. Substantial declines in numbers of cyclists have been measured by surveys in New South Wales, Victoria, Queensland, Western Australia, the Australian Capital Territory and the Northern Territory. The findings of these surveys are used here to estimate the total effect of the helmet laws in discouraging cycling.
Taking account of these other factors, the contribution of compulsory helmet laws to reducing rates of death and serious injury of cyclists is discussed below, together with the adverse consequences for public welfare.
2. Effect of improvements in roads and motorists’ behaviour
As a motor vehicle was involved in 92 per cent of fatal accidents to cyclists in Australia in 1988 (Attewell and Dowse, 1992), the behaviour of motorists is a critical factor in death and serious injury of cyclists. Following intensified random checks of speed and blood alcohol, and improvements in roads, total road fatalities declined by 30 per cent from 1989 to 1992, from 2804 to 1974. The decline has continued, to 1903 in the twelve months ended October 1993 (FORS, 1993).
The general decline of 30 per cent is reflected in deaths of pedestrians. It would therefore explain half or more of the 58 per cent decline in deaths of cyclists.
3. Decline in number of cyclists
3.1 Findings of surveys
3.11 Victoria
Before compulsory helmets were introduced in Victoria on 1 July 1990, some students at schools which compelled them to wear helmets chose to give up cycling instead (Ref?).
According to Vic Roads report IR 90-15 (1991), the first survey after helmets became compulsory in Victoria showed that the number of cyclists described as adult commuters declined from 5162 to 2098, or nearly 60 per cent, between March and July 1990. The author attributed this to the usual reduction of cycling in winter months, but the July survey was taken in fine weather. It would appear that the helmet law contributed to the decline.
Morgan and others (1991), reporting on surveys directed primarily to measure the use of helmets, made comparisons of matched samples of recreational and commuter cyclists in March 1990 and March 1991, at sites in Melbourne, Ballarat and Shepparton. These showed, in their Tables 17 and 18, declines of 30 per cent in recreational cycling and 32 per cent in commuter cycling from 1990 to 1991.
Monash University Accident Research Centre (MUARC) has conducted four surveys of bicycle use and helmet wearing rates in Melbourne, in November to January 1987/88 and successively in essentially the same 2-week periods in May/June of 1990 (pre-law), 1991 and 1992 (post-law) (Finch and others, 1993a). The 2-week periods were chosen to minimise the effect of different weather conditions. The 1991 survey found that total bicycle use by children had decreased by 36 per cent compared to 1990, with the largest decrease (44 per cent) being by the 12-17 year- old group (Cameron and others, 1992). This age group is included in the higher risk categories for fatalities to cyclists (FORS, 1991).
Finch and others provide, on page 36, 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. They go on to say that the numbers increased from 1991 to 1992 as follows: teenagers 6 per cent, children 20 per cent and adults 34 per cent. “As a result of these increases, the number of adult and child bicyclists was not much smaller than the pre-law numbers (a 9 per cent drop in children, 5 per cent in adults).” Later, on pages 43 and 44, they note that different weather may have been a factor in the change in numbers from 1991 to 1992, half of all weekend observation sessions coinciding with rain in 1991 but only 14 per cent in 1992. Another explanation they offer for the increase in numbers in 1992 is a bicycle rally passing through observation site 80. Information obtained from the authors is that the numbers of cyclists observed there were as follows: 1987/88, 129; 1990, 347; 1991, 72; 1992, 451. The 451 amounts to 18.2 per cent of the total number of 2477 cyclists observed in 1992 and site 80 alone accounts for 19 points of the 23 per cent increase in total numbers of cyclists 1991 to 1992.
More comment on the findings of Finch and others is contained in an associated submission from Dorothy Robinson.
3.12 New South Wales
In New South Wales, helmets have been compulsory since 1 January 1991 for adult cyclists, defined as 16 years and older, and since 1 July 1991 for those under 16. A survey of helmet wearing by cyclists was conducted for the Roads and Traffic Authority (RTA) in September 1990 and a second survey in March and April 1991 (Walker, 1991) to allow evaluation of the impact of the helmet law on adult cyclists and to provide a baseline measurement of helmet wearing by cyclists under 16 before helmets became compulsory for them.
The surveys showed a 6 per cent increase in the number of adult cyclists between September 1990 and April 1991. Walker concluded that “it appears clear that the new regulation has not deterred cyclists”, but this is far from clear. As well as the surveys being taken at different times of the year, the first “was conducted in overcast conditions in Sydney and, in some areas, was interrupted by rain whereas the second survey was conducted in sunny conditions.” A greater increase could have been expected.
Walker (1992) conducted a third survey in April 1992 for the main purpose of providing an assessment of the use of helmets by children under 16 before and after the law. It showed that the number of adult cyclists had decreased by 14 per cent and child cyclists by 38 per cent since April 1991. (The RTA’s summary of 14 August 1992 stated the latter figure as 36 per cent, by reason of including school children of 16 and 17.)
As the helmet law was in force for adult cyclists in both April 1991 and April 1992, the 14 per cent decrease in their numbers is not a measure of its discouraging effect on them. It may measure a delayed effect however, perhaps as people became more averse to wearing a helmet or the law was enforced against them.
A fourth survey in April 1993, essentially a replication of the previous surveys, showed a further decline of 11 per cent in numbers of child cyclists, to 45 per cent below 1991 (Smith and Millthorpe, 1993). The greatest decline was in the number of female students cycling to secondary schools – to less than a quarter of 1991.
3.13 Queensland
In 1990 and 1991, branches of the Royal Automobile Club of Queensland conducted surveys to measure the proportions of cyclists wearing helmets (Wikman and Sims, 1990, 1991). The surveys were mainly based on students riding to school and included Brisbane and centres ranging from Gold Coast to Cairns. Data were gathered for 1990 from 29 October to 24 November and for 1991 from 29 September to 24 November, the earlier start being chosen to improve the coverage of all age groups of students.
The wearing of helmets became compulsory on 1 July 1991, between the surveys. It could therefore be expected that the numbers of cyclists recorded, incidental to the purpose of measuring helmet wearing, would indicate the discouraging effect of the law. In fact, the 1991 report says, “Most surveyors commented that ridership was lower than previous years”, but adds, “To compensate for this it was requested that where possible more schools be surveyed”. Consequently, the decline in numbers of riders would have been greater than the 22 per cent shown for schools in total. Indeed, for most centres, where apparently it was not possible to survey more schools, declines of around 30 to 50 per cent were typical.
3.14 Western Australia
Compulsory helmet wearing was introduced on 1 January 1992, but not enforced until 1 July.
As in other states, in Western Australia the primary concern of surveys has been to measure rates of wearing of helmets. Such was the survey of Healy and Maisey (1992). They also commented that the numbers of children cycling to primary schools and numbers of recreational cyclists declined from 1991 to February 1992.
Heathcote (1993) did a wider-ranging survey, including data on numbers of cyclists in 1991 and 1992. The limited data he presents show some decline in numbers of children cycling to school. He suggested that the data do not show any change in the rate of decline, implying it is merely a continuation of a trend of previous years, but this is mere assumption, as he offered no explanation for the previous decline. His conclusion that the helmet legislation was not a significant factor in the decline is similarly no more than an assumption.
Heathcote’s limited observations of “commuter” cyclists indicate an increase in numbers after the helmet law, but his observations, also based on limited data, of cyclists classed as recreational show a decline of over 50 per cent. Therefore, data from automatic counter surveys conducted by Main Roads at the Narrows and Causeway bridges were used. These showed that the number of cyclists crossing on Sundays declined by 38 per cent from October-December 1991 to October- December 1992. Loath to attribute the cause of the decline to the law, Heathcote attempted to explain it by changes in leisure patterns, economic factors, population movements and the weather. These explanations are spurious; the first three could not change significantly in one year and, by his own statement on page 4, observations for all groups were taken only when the weather was fine and suitable for cycling.
Main Roads also counted numbers of cyclists crossing the Narrows and Causeway bridges on weekdays. These data for the period October 1991 to October 1993 are shown in graphical form in Figure 1 and 2, attached. The mean number of cyclists declined between October- December 1991 and October-December 1992 by 9 per cent at the Narrows bridge and 26 per cent at the Causeway bridge. Information from Western Australia Police shows that enforcement of the helmets law intensified in 1993, the rate of issue of infringement notices being 50 per cent higher than in the second half of 1992. Figures 1 and 2 show that this was matched by a further decline in cycling; for example, 10 per cent at the Narrows and 4 per cent at Causeway October-November 1993 compared to 1992. Sharper declines are evident in other months – 43 per cent from April 1992 to April 1993 and 52 per cent May 1992 to May 1993 for both bridges combined. Other factors may have contributed, but this evidence does support the hypothesis that the helmet law has resulted in a substantial reduction in cycling by “commuters” and schoolchildren.
3.15 Northern Territory
Incidental to surveys of bicycle helmet wearing by the Road Safety Council of the Northern Territory (1992, 1993), the following numbers of cyclists were recorded. The horizontal lines indicate the start of the helmet law, 1 January 1992 for cyclists 17 and over and 1 July 1992 for others.
Numbers of cyclists
Date of survey prim. school sec. school "commuter"
1990, August 252
1991, April 222
1991, August 350
---------------
1992, April 987 931 | 142
--------------------------------
1992, August 995 595 122
1993, May 823 570 131
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.
3.16 Australian Capital Territory
Members of the Cyclists’ Rights Action Group carried out a survey of 325 cyclists in May 1992, shortly before helmets became compulsory in July 1992. When asked, “Would you cycle less if helmets were compulsory?”, 90 or 28 per cent of respondents answered “yes”. Ratcliffe (1993) reported on surveys of bicycle traffic in the ACT before and after that, as follows:
For data over one week in November 1991 and December 1992 respectively: “Based on automatic counts of cycles for one week at 23 cycle path locations, mean weekday cycle path daily volumes 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.”
“Based on a survey conducted over a single morning … the number of morning peak period cyclists commuting to city and the other three town centres was measured to reduce by an average 21 per cent between November 1991 and November 1992.”
Such declines occurred despite very little enforcement of the law, only 50 traffic infringement notices having been issued from July 1992 to April 1993.
Claiming that it is “not possible with the existing data to draw firm conclusions about the effect of helmet legislation on cycle activity in the ACT”, Ratcliffe attempted to explain away the declines from 1991 to 1992 as not significant or due to the weather. Some preliminary analyses of variance were conducted on data from three of the 23 count stations. The results show that “only in a few cases was a significant difference estimated … . These results derive from the very high daily flow variability”. But Ratcliffe did not explain why tests for significance were not carried out on the much larger numbers for all stations in aggregate. The aggregate declines were from 7290 to 5176 on weekdays and from 5496 to 2876 at weekends. It is inconceivable that these large numbers would not test as significant.
As for the weather as a factor, all bicycle data were collected on rainless days, and detailed records show little variation of weather from 1991 to 1992 on collection days.
Finally, Ratcliffe’s summary of findings of surveys elsewhere is misleading. He said “Reports of bicycle activity in Victoria and New South Wales show that, for the periods before and after introduction of helmet legislation, there was no overall change to cycling in Victoria, and an average reduction of 26 per cent in NSW.” For Victoria, the “period before” is November 1987 to January 1988, 2« years before helmets became compulsory. The “period after” is May/June 1991. The authors of the survey that Ratcliffe cited, Cameron and others (1992), cautioned that the “comparison was made at different times of the year and nearly 3« years apart, and hence should be considered tenuous”, but Ratcliffe did not mention that. The only finding reported by Cameron and others which qualifies as a valid comparison of pre-law and post-law was a decline in cycling by children of 36 per cent between May/June 1990, just before the law came into effect, and May/June 1991. Ratcliffe did not mention it. (The survey of May/June 1990 did not include adults.)
In respect of NSW, Ratcliffe’s summary misrepresented certain of the survey findings. The data he cited (Walker, 1992, Table 9) are from surveys carried out in April 1991 and March/April 1992. In April 1991, the helmet law applied only to adults, and data from the survey then are pre-law only for children. Ratcliffe wrongly designated all 1991 data cited as belonging to the period before introduction of helmet legislation. Then he claimed there was an average reduction in cycling in NSW of 26 per cent. The average is of 14 per cent reduced cycling by adults and 37 per cent by children, but only the latter is a measurement of the decline in cycling from pre-law to post-law. On these figures, Walker had suggested “the new regulation has had the impact on numbers of bicycle riders under sixteen rather than any incidental changes … “, but Ratcliffe disregarded this, saying, “these results hide significant differences in cycling groups, with both reports being cautious in attributing all the changes to cycle activity to the helmet legislation”.
3.2 Analysis of findings of surveys
The surveys which show best the comparisons of numbers of cyclists pre-law and post-law are those of New South Wales and Victoria, and the best data are for child cyclists. But the measured decreases in numbers of child cyclists, 38 and 45 per cent below the base period in NSW in 1992 and 1993 respectively, and 36 per cent in Victoria in 1991, still under-estimate the discouraging effect of helmet laws.
First, some such effect would have occurred in anticipation of the law coming into force. Before and during the base periods for the relevant surveys, May/June 1990 in Victoria and April 1991 in NSW, publicity campaigns encouraged the wearing of helmets, and in April 1991 in NSW it was compulsory for adult cyclists. Many parents would in effect have anticipated the law by applying it to their own children. Tables 19 and 20 of Walker (1992) show helmet wearing by children increased from 12.7 per cent in September 1990 to 25 per cent in April 1991, but some children would have responded by giving up cycling or cycling less. The slight decrease in the number of child cyclists from September 1990 to April 1991 (Tables 19 and 20) where an increase could be expected provides evidence of this. As a result, the number of child cyclists recorded in the April 1991 base period of the third survey in NSW would have been lower because of the law. The declines in cycling measured subsequently would therefore under-estimate its full effect. Instead of the 36 and 38 per cent measured, the actual decline in the number of child cyclists pre-law to post-law is estimated at 40 per cent.
Second, the discouraging effect has been shown to continue post-law. For cyclists under 16 in NSW, Smith and Milthorpe (1993) measured a decline of 10 per cent in the second year – see their Table 3.2. For adult cyclists, Walker (1992) measured post-law declines of 14 per cent in the first year and a further 10 per cent in the second year – see his Table 3.10. Data from cycle counters operated by Main Roads in Western Australia also show post-law declines.
Third, any true measure of the effect of the law must be on the basis that it is enforced. The 45 per cent decrease in the number of child cyclists in NSW from 1991 to 1993 occurred without full enforcement – as 26 per cent of them were still not wearing helmets in April 1993. Table 3.2 of Smith and Milthorpe (1993) shows the numbers 6270 in April 1991, 3887 in 1992 and 3478 in April 1993 (after adjusting for cyclists at recreation areas in Albury not included before). Calculating from the table, those not wearing a helmet decreased from 4295 to 933 to 889. If the law were fully enforced, to ensure a wearing rate of virtually 100 per cent, some of the 889 persistent non-wearers would choose to cease cycling, or would cycle less. The number can be estimated from Table 3.2. Of the 4264 cycling without a helmet in 1991, 563 chose to don a helmet and keep on cycling until 1993, but 2792 – five times as many – had given up cycling by 1993. On the basis that five out of six, i.e. 740, of the 889 persistent non- wearers would choose similarly, only 2738 would still be cycling, a decline of 56 per cent from 1991.
The 56 per cent includes the decline of one year post-law. Some further decline seems likely. Adding the effect of anticipation, it is estimated that the long term effect of the law will be to discourage 60 per cent of children from cycling. As there are 3 million children aged 5-16 in Australia, perhaps a million aged say 10 or more would be discouraged, with consequent detriment to national welfare – see below. For Western Australia, the corresponding number would be over 100000.
4. Effect of compulsory helmet laws on death and injury
Deaths of cyclists in road accidents in Australia decreased from 98 in 1989 to 41 in 1992. Of these, deaths of children up to 16 were 44 and 17 respectively. How much of these decreases can be attributed to compulsory helmet laws?
With small numbers, random variations are relatively large; pre-law, the total of 98 fatalities in 1989 compares with 87 in 1988, and for cyclists aged up to 16 the difference is proportionately greater: 44 in 1989 and 33 in 1988 (FORS, 1989, 1990). Fatalities might well have been high by chance in 1989. Setting that point aside, the 30 per cent general decline in road fatalities from 1989 to 1992 would explain a decrease from 98 to 68 in deaths of cyclists in total, and from 44 to 31 in deaths of child cyclists. As the helmet law was not in force everywhere in Australia throughout 1992, its discouraging effect on cycling by children would be somewhat less than the 56 per cent estimated above. Putting it at 50 per cent would explain a further decrease in deaths to 16. The actual number was 17. It is therefore not evident that the helmet laws made any positive contribution to reducing deaths of child cyclists.
Have the helmet laws contributed to reducing serious injury to cyclists? Data are available for Victoria, New South Wales and Western Australia.
Victoria
In a study entitled “Head injury reductions in Victoria two years after introduction of mandatory helmet use”, Finch and others (1993b) of MUARC used data based on claims to the Transport Accident Commission for bicyclist casualties who were involved in collisions with a motor vehicle. The decrease in the number of bicyclists killed or admitted to hospital after sustaining head injuries was 48 per cent in the first post-law year (year ended June 1991) and it reached 70 per cent in the second year. The corresponding decreases for those killed or admitted to hospital who did not sustain head injuries were 23 per cent and 28 per cent.
Finch and others concluded that the introduction of the law had been accompanied by an immediate large reduction in the number of bicyclists with head injuries. “Apparently this has been achieved through a reduction in the number of bicyclists involved in crashes (at least partly through a decrease in bicycle use by children and teenagers) and a reduction in the risk of head injury of bicyclists involved in crashes”. They were understandably reluctant to conclude that the reduction in head injuries was an effect of the helmet law; the numbers are fairly small; considerable fluctuations have occurred in the past; and these reductions occurred at a time of improved behaviour of motorists due to initiatives directed at drink/driving and speeding, and of publicity about cyclists.
Vulcan and others of MUARC reported these and similar findings in Morbidity and Mortality Weekly Report 42, 18, 14 May 1993 of the US Department of Health and Human Services. The editors of the report commented: “The reduction in number of bicyclists with head injuries following implementation of the law in Victoria may reflect a combination of several factors, including the decline in bicycle use by children … the effect of educational efforts and publicity … initiatives to reduce motorists’ speeding and drinking and driving. Further assessment is needed to identify the most important components … and to measure the effectiveness of the program in reducing head injuries.”
Factors other than those suggested by Finch and others apparently were operating to reduce the proportion of head injuries in the total of injuries. Data for claims to the Transport Accident Commission by both cyclists and pedestrians in the years 1987-92 are attached as A. For cyclists, from the pre-law years 1988 and 1989 to the post-law years 1991 and 1992 claims for head injuries declined by 56 per cent and other injuries by 33 per cent – similar to the findings of Finch and others. But the data for pedestrians show this occurred for pedestrians too, the corresponding declines being 31 per cent and 14 per cent. The numbers appear to be large enough for this difference to be significant.
Reduced speed of motor vehicles and other changes in how they are driven may be an explanation for the difference. Discussion of the effect of changing speed on the seriousness of accidents in Corner and others (1987) shows that such effects are not simple. Research reported by Janssen and Wismans (1985) showed similarities in the impact of pedestrians’ and cyclists’ heads with vehicles, but that impact of a cyclist’s head with the windshield or even with the roof is much more likely than for a pedestrian.
The Victorian Injury Surveillance System (VISS), which monitors injuries to children under 15 years old, has published statistics for admissions of cyclists aged 5 to 14 years to five hospitals in Melbourne in the years 1989-92. Admissions for head injury declined by 32 per cent between 1989 and 1992, but this is unremarkable as greater rates of decline were shown by total road fatalities in Victoria, 49 per cent (FORS, 1989, 1990), and admissions of cyclists for other injuries, 35 per cent. From these data, it is not evident that the helmet laws made any contribution.
McDermott and others (1993) undertook an evaluation of the efficacy of helmet use by comparison of crashes and injuries in 366 helmeted and 1344 unhelmeted casualties treated from 1987-89 at Melbourne and Geelong hospitals. They noted that the relative risk of head injury in helmeted cyclists compared with unhelmeted involved in crashes cannot be estimated directly from this study. They concluded that approved helmets reduced the risk of head injury by at least 39 per cent, and the severity of injury to the head. There were several anomalies, however. First, they concluded that helmets reduced injuries to the face – not covered by a helmet. When questioned, McDermott was unable to offer a satisfactory explanation.
Second, they found that the frequency of neck injury was increased in helmeted casualties, and offered no explanation. Third, helmeted casualties had a greater proportion and more severe non-head injuries than unhelmeted. They offered a partial explanation for the greater proportion, but admitted to being unable to explain why the severity was greater.
All in all, the data adduced here underline how little is known about head injuries to cyclists and the effects of helmets. As data are accumulating, more unexplained effects are becoming evident and the value of helmets in reducing injury coming more into question. Further, indications are emerging (Robinson, 1993) consistent with the finding of Rodgers – see our Appendix 1 – of a positive correlation between use of helmets and casualties to cyclists.
A matter for concern is the reliance that can be placed on studies made in the various states and territories. Shortcomings of Ratcliffe’s study in the ACT have been pointed out above. The TAC data for pedestrians puts the findings Finch and others in a different light, but they made no mention of pedestrians in their report.
New South Wales
Governments have cited sharply reduced deaths of cyclists in New South Wales in 1991 and 1992 following the introduction of the helmets law as evidence of its efficacy. Robinson (1993) has analysed statistics from the Road Traffic Authority’s annual publication Road Traffic Accidents in NSW. The average number of deaths in 1991 and 1992, 8, was 59 per cent lower than the (pre-law) average of 1989 and 1990, 19.5, but the decline can be explained by other factors. First, with these small numbers, relatively large variations can occur by chance. Pre-law, there were 34 deaths in 1988 followed by 19 in 1989 and 20 in 1990. Second, at least 25 points of the decline can be explained by fewer cyclists being on the road. Third, the hazard to cyclists due to motor vehicles was lower, owing to breath testing and other factors. The decline of 32 per cent in deaths of pedestrians is a measure of the reduction in this hazard. It is noted that 12 of the 16 cyclists killed in 1991 and 1992 were wearing helmets.
The statistics for serious injuries tell a similar story. Total deaths and serious injuries declined at much the same rate from 1989/90 to 1991/92 for all road users; deaths and injuries to cyclists as a percentage of these have therefore remained fairly constant, as follows:
Year 87 88 89 90 91 92 Killed or seriously injured 3.9 3.8 3.7 4.4 4.0 3.8
This result has occurred despite the decline in numbers of cyclists on the road and increased wearing of helmets.
Robinson also used the NSW statistics for death and serious injury and for helmet wearing in 1992 to assess the protective effect of helmets. In that year, when 171 adult and 97 child cyclists were killed or seriously injured, Walker (1992) recorded 85 per cent of adult cyclists and 76 per cent of children in April as wearing helmets. Robinson reasoned that, if wearing a helmet had no protective effect, the proportions of dead and injured wearing a helmet would be the same, which she calculated as 81.7 per cent for adult and child cyclists combined. The actual proportion of 82.4 per cent is in close agreement and even slightly higher. There are minor uncertainties in the data, but it is clear that they do not demonstrate any reduction in the probability of death or serious injury from wearing a helmet.
Western Australia
For Western Australia, Healy and Maisey (1992) claimed that the growing rate of helmet wearing was reflected in reduced admissions to hospitals for intracranial injuries. On the basis of that, they evaluated costs and benefits of mandatory helmet wearing.
Healy and Maisey said on page 21 that the 5-12 age group indicates best the possible association between bicycle helmet wearing and admissions to hospital for intracranial injuries. In their figure 13, the plot of intracranial injuries to that group shows a steady decline from 113 in 1981 to 62 in 1990 against a rising trend in helmet wearing. But the data for helmet wearing are only for 1986 onward. Also, the plotting of injuries as 3-year averages masks substantial year-to-year fluctuations; as shown in their Table 8, injuries in 1988 were not appreciably lower than 1986. More important, they take no account of the declining rate of cycling; Table 7 of Heathcote (1993) shows that for the 5-12 age group cycling declined rather faster from 1988 to 1990 than admissions for intracranial injuries. Finally, the attempt to correlate declining intracranial injuries with increasing wearing of helmets is discredited by the data for the 13-17 age group. Despite little decline in their rate of cycling and a low wearing rate, intracranial injuries to them declined at least as much as the 5-12 year-olds. The basis of Healy and Maisey’s evaluation of costs and benefits is therefore unsound.
Heathcote (1993) presented, in Table 13, numbers of fatalities to cyclists in the years 1986-92, and in Table 14 and Appendix 10, numbers of cyclists suffering head injury in 1981-92. These show that fatalities declined from 8 in 1991 to 1 in 1992, and head injuries from 196 to 110. He concluded that the helmet law “reduced the number of head injuries and fatalities suffered by cyclists”. This conclusion is not sound. The decline in fatalities could well be due to chance: 1986 to 1987 showed a greater decline, from 14 to 4, and 5 fatalities occurred in the first 10 months of 1993 (advice from Western Australia Police).
The decline in head injuries from 1991 to 1992 would be statistically significant, but is not proved to be a “close approximation of head injuries that may be reduced due to helmet wearing”, as Heathcote suggested. In any case, the reduction occurred partly through the mechanism of discouraging cycling. Fewer cyclists were on the road, with reported crashes declining by 21 per cent and total injuries by 16 per cent. Head injuries declined even more, it is true, from 23 per cent to 15.4 per cent of total injuries, but the proportion of head injuries in the total has fluctuated greatly before, 1991 being the highest for 5 years and 1989 at 16.7 per cent being little higher than 1992.
The conclusion is that it is not possible to measure from these data what contribution the wearing of helmets made to reducing head injuries.
5. Negative effects of compulsory helmets on national welfare
No significant positive effects of helmet laws in reducing death or serious injury being evident, their negative effects on national welfare become relatively more important.
5.1 Distraction from other measures
First, the emphasis that governments place on the wearing of helmets distracts attention from measures for preventing accidents. Helmets of course can merely mitigate the effects of an impact and do not prevent accidents at all. Hillman (1993) discusses the suggestion that wearing a helmet may give cyclists, especially children, the false confidence to take more risks.
More attention should be given to education and training of child cyclists, and action should be taken to counter commercial promotion of cycling by children too young for it, and to dispel the impression created by advertising that bicycles are toys for doing tricks. Mathieson (1986) cited studies in America that showed that trained and experienced cyclists are 4 times less likely to be involved in an injury accident than untrained adults, and 7 times less likely than children.
As 92 per cent of fatal accidents to cyclists in 1988 involved a moving motor vehicle, it is imperative that action should be taken to protect cyclists from motorists, including more education to induce young motorists in particular to have greater care for cyclists, but governments have done little.
In Britain, the British Medical Association (1992) said that “resources are almost certainly better directed towards the prevention of accidents rather than the limitation of the damage cause by them. The introduction of mandatory wearing of cyclist helmets shifts responsibility for the safety of cyclists to the cyclists themselves … It seems more appropriate that responsibility should be borne by those causing the injuries, namely drivers of motor vehicles”. The BMA added: “The wearing of helmets must be placed in the context of other means of protecting cyclists, such as motor-vehicle speed reduction, improving the conduct of drivers, and developing cycle networks and enhancing road design for the benefit of cyclists.”
5.2 Change to mode of travel
Many children who reduce their rate of cycling will travel by car instead, or walk – and have accidents. As noted above, 42 child passengers in cars and 39 child pedestrians died of head injury in 1988 compared to 11 cyclists. The 1987 report of the Parliamentary Inquiry in Victoria which recommended compulsory helmets showed, for the years 1980-85, that the rate of death and injury per 100 000 children of 0-16 years was lower for cyclists than for pedestrians, in total and in all age groups except 12-16 years. Moreover, there was a consistently lower rate of head injury to cyclists than pedestrians. Deaths and injuries to children – and to adults – who change to another mode of travel rather than wear a helmet therefore should be debited to the helmets laws. Measuring them of course would be difficult.
5.3 Loss of skills in cycling
Children who opt out of cycling rather than wear a helmet will fail to learn the skills of cycling. Instead of beginning to develop road sense at an early age with a slow vehicle unlikely to cause injury to others, they will begin with a motor vehicle. Also, not being familiar with cycling themselves, they will be less aware of cyclists and less likely to be considerate of them. Increased danger for the remaining cyclists will result.
People who opt out of cycling in their youth will be unlikely to take it up in their adult life, and, not having developed skills in cycling at an early age, are likely to be prone to accidents. The outcome will be a society more dependent on motor transport.
5.4 Effects on health
Less cycling will mean loss of beneficial exercise. In its book the British Medical Association (1993) described health benefits gained through the exercise of cycling. It cited several studies to measure benefits in terms of life years. One study concluded that those who cycled 60 miles a week from the age of 35 could add 2« years to their life expectancy. The BMA concluded (p. 121) that “the benefits gained from regular cycling are likely to outweigh the loss of life through cycling accidents for the current population of regular cyclists”. The BMA went on “The net benefits of cycling as an effective and accessible form of exercise should therefore be promoted vigorously by government agencies, policy-makers, health authorities … cycling can make a very real contribution to improving the health of the nation.”
As estimated above, the helmet laws could result in a million children aged 10 or more in Australia and more than 100000 in Western Australia opting out of cycling and losing its benefits for their health. Cycling combines transport and exercise. The benefits of the exercise are therefore gained at little or no cost in time where people cycle to work, school, shops or other places they need to get to. Extra costs in terms of time will be incurred where travel, by motor vehicle, and exercise are separate activities. Is it not recognised that one of the most effective forms of preventive medicine is for teenage children, especially girls, to take more exercise? Cycling to school may be the best way, but, ironically, the discouraging effect of compulsory helmets is greatest for this group.
Some cyclists say that wearing a helmet aggravates migraine and other ailments. Skin cancer is of increasing concern in Australia, but helmets provide little protection against the sun. Transport authorities argue that peaks can be fitted to helmets or supplementary caps worn, but these do not protect the sides of the face. According to The Canberra Times of 28 August 1992, researchers at Dryburn Hospital in Durham, England, found that hats with brims at least 7.5 cm wide are needed “to provide reasonable protection around the nose and cheeks, where non-melanoma skin cancers commonly occur”. The Federal Minister for the Environment, Ros Kelly, in November 1993 advised Australians to adopt a sun-smart lifestyle to reduce the incidence of skin cancers. She pointed out that the thinning of the ozone layer increases the risk of skin cancer and cataracts. Compulsory helmet laws counteract her advice and their long-term effect is likely to be more skin cancer. In time, some of those affected may litigate to obtain compensation from governments.
6. Findings
A critical factor in death and serious injury to cyclists is the behaviour of motorists.
The general decline in road deaths and injuries following introduction of random checks of speed of motor vehicles and blood alcohol of drivers is a factor in a decline in deaths and injuries of cyclists in recent years.
Surveys carried out before and after introduction of compulsory helmet laws in New South Wales, Victoria, Queensland, Western Australia, the Northern Territory and the Australian Capital Territory have all shown increases in rates of helmet wearing, but also sharp declines in numbers of cyclists.
The earliest and among the best data of decline in numbers of cyclists are of child cyclists in Victoria, where total bicycle use by children decreased by 36 per cent in a year following introduction of the helmet law.
In New South Wales, surveys have shown a decline of 38 per cent in child cyclists in the first year and a further decline of 11 per cent in the second year of the helmet law – a total decline of 45 per cent.
For every child cyclist induced by the law in New South Wales to wear a helmet, five chose to give up cycling instead.
Many of those gave up cycling would perforce have adopted another mode of travel and have had accidents while travelling by that mode; any deaths or injuries thus incurred – number unknown – should be debited against the helmet law.
Based on data from New South Wales, it is estimated that the full effect of the helmet law with rigorous enforcement is a 60 per cent reduction in cycling by children, and the effect is likely to increase.
Some reductions in head injury may have occurred as a result of increased wearing of helmets, although the data are not conclusive and the number who may have benefited is very small compared to those giving up cycling in response to the law.
Indications are emerging of a positive correlation between use of helmets and casualties to cyclists.
A matter for concern is the reliance that can be placed on studies made in the various states and territories.
The helmets law has distracted from other measures to protect cyclists and will result in loss of skills and health benefits of those who give up cycling rather than wear a helmet, the loss of health benefits alone being likely to outweigh any benefits in mitigation of head injury.
References
Attewell, R.G. and Dowse, M.J., Fatal crash types. Analysis of 1988 fatality file, Federal Office of Road Safety report no. CR 105, Canberra 1992
British Medical Association, Cycling towards health and safety, Oxford University Press, Oxford, New York, 1992
Cameron, M., Heiman, L. and Neiger, D., Evaluation of the bicycle helmet wearing law in Victoria during its first 12 months, Report No. 32, Monash University Accident Research Centre, Melbourne, July 1992
Corner, J.P., Whitney, C.W., O’Rourke, N. and Morgan, D.E., Motorcycle and bicycle protective helmets: requirements resulting from a post crash study and experimental research, Federal Office of Road Safety report no. CR 55, Canberra, 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
Finch, C.F., Newstead, S.V., Cameron, M.H. and Vulcan, A.P., Head injury reductions in Victoria two years after introduction of mandatory bicycle helmet use, Monash University Accident Research Centre report no. 51, 1993
FORS, Road fatality statistics Australia, , Canberra, 1991.
FORS, Road Crash Statistics Australia, November 1989 and June 1990, Federal Office of Road Safety, Canberra
FORS, Day-to-Day Travel in Australia 1985-86, Federal Office of Road Safety, Canberra, 1989
Federal Office of Road Safety, Road Fatality Statistics Australia, monthly report, April 1993, Canberra
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, Policy Studies Institute, 100 Park Village East, London, 1993
Janssen, E.G. and Wismans, J.S.H.M., Experimental and mathematical simulation of pedestrian-vehicle and cyclist-vehicle accidents
Lambert, J., Number of cyclists, bicyclist trips and bicyclist accident reports in Victoria, 1986-1989, VICROADS internal report, May 1990, reference 18 cited in Victorian Bicycle Strategy, VICROADS, November 1991.
Mathieson, J.G., Gaps in current knowledge and effects on counter measures, Bikesafe 86 conference, Newcastle
McDermott, F.T., Lane, J.C., Brazenor, G.A. and Debney, E.A., The effectiveness of bicyclist helmets: a study of 1710 casualties, The Journal of Trauma, 34, 6, pp. 834-845, 1993
O’ Rourke, N.A., Costello, F., Yelland, J.D.N. and Stuart, G.G., Head injuries to children riding bicycles, Medical Journal of Australia, Vol. 146, 15th. June 1987
Parliament of Victoria, Social Development Committee, Safe Roads for Children, inquiry into child pedestrian and bicycle safety, second report, 1987, Tables 5, 6, 11 and 12
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., submission to Select Committee on Road Safety, Parliament of Western Australia, 1993
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
Victorian Injury Surveillance System, Hazard No. 6, Page 6., 1990
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