Proportions of hospitalised cyclists (n = 1859) that sustained at least one injury to particular body regions following a collision with a motor vehicle, NSW 2001–2009; a) all injuries and b) serious injuries (SRR ≤ 0.965). From Bambach et al.

A new paper attracted a lot of attention earlier this month because of a finding that cyclists who ride without a helmet are more likely to take risks.

The Conversation did its own investigation and ran the story under the heading Crash data shows cyclists with no helmets more likely to ride drunk. The Sydney Morning Herald took a more sober approach – it reported Cyclists without helmets ‘likely to be risk-takers’.

Risky behaviour is one of the issues addressed in the paper but these reports distracted attention from a couple of other important results.

The key purpose of the research reported in the paper was to examine the effectiveness of helmets in minimising head injuries in the event of a collision between a bicycle and a motor vehicle.

The paper documents an investigation undertaken by four UNSW researchers, M Bambach, R Mitchell, R Grzebieta, and J Olivier. It’s published in the April 2013 edition of Accident prevention and analysis, a leading journal in this area. There’s an ungated version available here (the file has to be saved first).

Bambach et al examined 6,475 collisions between cyclists and motor vehicles in NSW where the accident was reported to police, hospital data was available if applicable, and the helmet-wearing status of the cyclist was recorded.

The authors were able to combine rich information about the type and severity of injuries with data about the circumstances of the accident.

Their headline finding is that helmets confer a large protective effect in the event of a collision with a motor vehicle.

The odds of sustaining a ‘moderate’ head injury in a collision are 1.9 times greater if the cyclist doesn’t wear a helmet.

However the odds of suffering a head injury classified as ‘serious’ are 2.6 times greater if the rider’s unhelmeted. In the case of a ‘severe’ head injury they’re 3.9 times greater.

When the researchers broke head injuries down by type, the estimated odds of suffering a ‘serious’ or ‘severe’ skull fracture if no helmet is worn is 4.6 times greater.

These findings go against the meme that helmets only offer protection in the event of a minor accident, like simply falling off a stationary bike or while pedalling at low speed.

They contradict the myth that a helmet is useless in an accident with a car. The authors say the research shows the benefit from wearing a helmet increases with the severity of the injury.

It’s also very likely the better odds offered by helmets are under-stated. That’s because the number of riders who were involved in a collision with a motor vehicle but avoided a head injury because they wore a helmet can’t be known.

There’s also some comfort in the finding that more than 90% of cyclists who were in a collision with a car didn’t sustain a head injury.

Those that did, however, were more likely to have collided with a larger vehicle; or to be cycling on a road with higher maximum speeds; or to have disobeyed a traffic signal; or to not be wearing a helmet.

Worryingly, the researchers also found children who were involved in a collision were much more likely to be unhelmeted. Children aged 12 years or less comprised 19% of all those who weren’t wearing a helmet at the time of their accident but just 7% of those who were.

This pattern was similar for those aged 13-19 years. Teenagers made up 35% of all those who weren’t wearing a helmet when they collided with a motor vehicle but 11% of those who were.

Overall around half of all children and teenagers were not wearing a helmet at the time of the accident. This compares with 23% of those aged 20-29, 15% aged 30-39, and 12% aged 40-49 years.

Since it’s about accidents involving motor vehicles, this research can’t tell us if children and teenagers are more inclined to cycle without a helmet. However it shows non-helmet wearers in this age group are significantly over-represented in accidents.

Both children and teenagers who weren’t wearing a helmet were also more likely to have sustained a head injury than their peers who were. They make up 35% of those with a head injury compared to 24-27% of the controls.

As noted at the start, riders who weren’t wearing a helmet were more likely to have been engaged in risky behaviours at the time of the accident.

The net result is unhelmeted riders were involved in more severe crashes, but the authors say the difference in severity is small.

This research bears on the argument about whether or not Australia’s mandatory helmet laws make sense. I interpret the findings as reinforcing the good sense of wearing a helmet when cycling, especially on roads.

I would of course be free to make that choice even if the helmet law were repealed. The question of whether or not helmets should continue to be mandatory must be considered in the context of other arguments, especially the claim that the law deters cycling.

A more pressing issue though is the over-representation of under 20s in accidents with cars; their higher rates of head injury; and the remarkably low helmet-wearing rates of those in this age groups who’re involved in collisions with motor vehicles.