“Both read the Bible day and night,. But thou read’st black where I read white.”
– William Blake, “The Everlasting Gospel”
A group of academics from Sydney, the capital city with the lowest cycling rates in the country, have recently been engaging in an extensive debate about the effectiveness of mandatory helmet laws (MHL) in Australia. Their articles may be found in academic journals and their op-eds appear on such sites as the Sydney Morning Herald, The Age, ABC News, and The Conversation. Another commentator, Alan Davies, examines the research and history concerning the laws on his blog (now part of Crikey.com), The Urbanist. From the University of NSW, the main pro-MHL academics are Jake Olivier, Scott Walter, and Raphael Grzebieta – statisticians and road safety academics.
On the helmet choice side of the debate, from the University of Sydney, is Chris Rissel, a public health academic. Elsewhere in Sydney, at Macquarie University, Piet de Jong, a professor of actuarial studies and statistics, has written on the costs and benefits of a mandatory helmet law. Dorothy Robinson from the University of Newcastle and Simon Batterbury from the University of Melbourne are also actively involved in the online debate at The Conversation.
The problem with the UNSW studies is not in the statistics; a huge amount of work has been done in the paper (Bambach et al, 2013) to cross-reference injury and police data. Bambach et al note that previous case control studies have had limitations such as small sample sizes. The studies are purely observational and conducted in a mandatory helmet law environment where fines are enforced. Sydney is 64% of the population of the state of New South Wales and 2011 Census data shows that 0.9% of Sydney workers commuted to work by bicycle – 15,000 out of 1,646,000 on Census Day 2011, the lowest of all capital cities.
New South Wales as a state had the same 0.9% bicycle modal share of commuters in 2011, the lowest of any state except Tasmania. The study was conducted using data between 2001 and 2009 when cyclists were even more marginalised than today. American cycling expert John Pucher in 2010 “found the level of the hostility of enough Sydney motorists worse than I had seen anywhere in the world”. The vice president of Cycling NSW, Richard Birdsey, said in 2010 concerning Sydney that “[upright bikes are designed … for women] … If you are moving in the serious traffic here you need something a bit quick, something you can throw around, something more performance oriented.”
The paper refers to a “debate regarding the effectiveness of cycling helmets in preventing head injuries”. To some extent, the sides are arguing at cross purposes; the paper refers to the debate being about helmets but the commentary on the paper in the media is about helmet laws. Evidence that the focus of the paper is on helmets rather than helmet laws per se is that it does not cite any papers by Rissel, de Jong or Robinson.
The conclusion of the paper was that
“This case–control study of 6745 cyclist casualties resulting from collisions with motor vehicles has indicated that helmet use is significantly associated with reduced risk of head injury by up to 74%.”
“Given the large protective effect of helmets demonstrated in the present study, this issue should be addressed with preventative action.”
Presumably the preventative action would involve the stricter enforcement of existing mandatory helmet laws in NSW, as well as segregated cycling facilities. However, a well-known result in the Netherlands (Rijkswaterstaat, 2008), a country without MHLs (and helmet use among all cyclists <1%) found that “13.3% of cyclists admitted to hospital were wearing helmets when they were injured.”. Instead of drawing the conclusion that helmet use leads to injury, the most reasonable conclusion would be that the legal environment determines the behaviour of cyclists. Thus, in the Netherlands, the helmeted cyclists are the “risk takers”, the sports cyclists who are more likely to be injured, not the everyday cyclists.
The problem with the studies is more in the interpretation of the paper’s results; the “post-processing”. The media remarks made by the authors involve huge leaps of logic.
“…we had the information as to how it occurred, so in other words from the crash data we were able to determine whether a person was at fault or not.”
This fails to account for the possibility of the assignation of blame by the police being biased due to a lack of helmet wearing by the injured cyclist, as well as bias due to cycling no longer being a “normal” form of transport in Australia.
“The evidence says helmets work: they minimise the risk of injury.”
Again, as explained repeatedly by helmet freedom advocates, enforced mandatory helmet laws change the cycling population. Some potential cyclists will give up cycling.
Professor Rissel commented
“You’ve got helmets creating a barrier to cycling, particularly spontaneous, short-trip cycling”.
Thus, in an enforced MHL environment, cycling becomes associated almost exclusively with commuting and sports, or recreational cycling – the opposite of the Netherlands. In turn this has led to lower numbers cycling and may have to led poorer cycling facilities than would have existed without the law. This supposition is backed up by the statements of Pucher and Birdsey above. In particular, the spontaneous trips associated with bicycle sharing schemes are drastically curtailed, and this can be seen in the trips per bike per day figures for the Brisbane and Melbourne schemes. A 2012 study by Forrest (Table 9) showed that helmet laws were the biggest barrier to the use of the schemes among non-cyclists, regular cyclists and bike sharing scheme users. Brisbane’s CityCycle scheme has 150 stations and at least 1800 bikes and has the potential to revolutionise inner-city travel, but is being held back by a misguided law and lack of dedicated facilities.
Remarks on The Conversation web site, where both Olivier and Grzebieta responded to their critics, smacked of “argument from authority”. Grzebieta challenged one of his critics asking about his academic papers in the area and cites the “head of the Dutch road authority” as saying at a Canberra conference that “if they could get MHL through they would do it in an instant”. **
This conflicts with the 2012 recommendation cited in the same thread that it is more important to build safe cycling infrastructure than impose a helmet law. Wegman, the head of SWOV, the authority, does not mention helmets in a 2011 column. Indeed another SWOV paper from 1994 draws a careful distinction between “everyday cycling” and “recreational cycling” which is a key omission from the UNSW academics’ entire output.
** We have done some investigating and no such person visited Australia. Grzebieta is referring to a presentation given by Ms Nel Aland entitled, “What a small country can be good at (besides football)”, and she is on Linkedin. Her position is the ‘unit head of a department for analysis and development, of the air transport inspector at the governmental inspection for the environment and transport’. She is not the ‘head’ of a ‘Dutch Road Authority’ at all and if you search for ‘bicycle’ or ‘bicycle helmet’ (in Dutch, of course) on their site you will find no mention of them. The Dutch Minister for Transport recently confirmed in the news that she does not support a mandatory bicycle helmet law and the only organisation that is pro-helmet (but not pro-helmet-LAW!) is the SWOV… but they were also against priority for cyclists on roundabouts! In The Netherlands, this is an outlier viewpoint. No Dutch organisation supports mandatory bicycle helmet laws. None.
In a mandatory helmet law environment, there are all sorts of confounding factors when examining the accident rates and severity of accidents, comparing helmeted and non-helmeted cyclists. The study attempted to account for alcohol use and disobeying traffic signals. This issue has been commented on extensively at The Conversation. In an MHL environment, it may well be that a lack of helmet use is associated with other factors which were not measured in the police data – for example cyclist speed. And how exactly does a “footpath cyclist” have an accident with a motor vehicle as in the study’s controls? In megacities such as Tokyo footpath cycling is considered the norm and does not require helmet use.
In summary, both sides rely on the same data but reach different conclusions. The pro-helmet-law academics, who seem to be most prevalent in countries with existing helmet laws, rely on observational studies and make large leaps of logic to try to show the law is justified.
Helmet freedom supporters, in contrast, take a more international approach and concentrate on the public health benefits of cycling outweighing the costs and civil liberties issues.
The critical questions never answered satisfactorily by the coterie of UNSW academics are as follows:
- How is it that after more than twenty years of MHL in Australia and NZ, the other 191 countries in the world have not rushed to copy the laws? Aren’t good ideas supposed to travel? In fact, apart from these two examples, adult bicycle helmet laws are almost unique to regions of English speaking countries such as the United States and Canada, with the exception of Dubai. Naturally, these countries also have high car ownership rates.
- Why do foreign cycling experts such as John Pucher, Jan Gehl, and Mikael Colville-Andersen urge against bicycle helmet laws? Sure, this is an argument by appeal to foreign authority, but listening to foreigners can help overcome our blind spots.
- Why do Australian bicycle sharing schemes have such low usage rates compared to schemes worldwide? In fairness, one of the UNSW academics, Tim Churches, suggested the “tiny size” of the schemes at The Conversation but this is incorrect as Brisbane’s scheme is one of the larger schemes, with 150 stations.
- Would the academics recommend helmet use by pedestrians or car drivers? Do they wear helmets as car drivers themselves? If not, why not?