The claim

“The evidence is mixed. I have to say that in countries where they have made (cycle helmets) compulsory, it hasn’t always necessarily been good for cycling.”
Boris Johnson, 2 August 2012

The background

Britain’s Olympic hero Bradley Wiggins sparked passionate debate this week when he appeared to come out in favour of making it compulsory for cyclists to wear helmets on our roads.

The gold medallist was reacting to the news that a cyclist, named today as Dan Harris, died following a collision with a bus carrying journalists between Olympic venues.

At time of writing, we don’t know the precise circumstances surrounding Mr Harris’s death – and Wiggins has since gone into reverse gear over his remarks.

Whatever the details of the tragic case, it has inspired a discussion on cycling safety, and the merits of a change in the law on helmets.

The Mayor of London, Boris Johnson, said he was against making them compulsory – a view echoed by the Department for Transport. Are the politicians being complacent, or are they right to be cautious?

The analysis

The Mayor of London is right to say that the evidence on the effectiveness of wearing bike helmets is mixed.

It’s not just mixed – it’s a vast, confusing, misleading, often biased and occasionally flat-out contradictory body of work.

Traffic psychologist Dr Ian Walker from the University of Bath was only half joking when he told us: “It’s a ridiculously complex issue – seriously, that Higg’s Boson stuff and the Middle East conflicts look simple by comparison.”

Some studies look at hospital admission data and try to establish whether there is a difference in outcomes for helmet-wearers.

These kinds of studies have tended to be pro-helmet, with one major meta-review claiming that helmets reduced head injuries by 85 per cent and brain injuries by 88 per cent.

The main problem with research of this type is the difficulty of allowing for “confounding factors”.

This is the old “ice cream causes drowning” problem. If there is a positive correlation between sales of ice cream and drowning deaths at the beach, ice cream must increase the risk of drowning, right? Wrong – it’s just that more people swim in the summer months, and they buy more ice cream too.

Similarly, one study that initially claimed to find evidence that helmets led to fewer serious injuries realised that alcohol was the more likely causative factor.

A drunk cyclist was more likely to get hurt – and less likely to wear a helmet. But it was the booze, not the helmet, that had the biggest impact.

Read more: Bradley Wiggins denies he called for legally enforced helmets

There are any number of possible confounding factors in the helmet studies. It could be that helmet-wearers tend to be more cautious, and that’s the reason why they have fewer accidents. Or it could be that riders are more likely to wear headgear while cycling off-road, where the dangers may be greater.

Other studies focus instead on the whole population of cyclists in a country or region where helmets have been made compulsory. Injury rates are measured before and after.

These have tended to be more negative. Population studies from England, Australia, New Zealand and Canada suggest there is no reliable evidence that helmets bring injury and death rates down.

Again, there are numerous confounding factors that could skew these kind of studies. It could be that other road safety initiatives have coincided with the new helmet law, and it is these that are really behind the change in accident statistics. And so on…

Perhaps the most intriguing confounding variable is the fact that wearing a helmet can change the behaviour of both cyclist and driver.

Some studies have suggested that drivers tend to give bare-headed riders a wider berth when overtaking, making a collision less likely. It has also been claimed that riders who wear helmets tend to go faster and take more risks.

Predictably, the opposite has also been suggested.

The Transport Research Laboratory trawled through almost every major piece of academic research on the subject in 2009 and concluded that all of them had potentially fatal methodological weaknesses.

Even more depressingly, the team doubted that it will ever be possible to account for every confounding factor and make a “definitive claim” for the value of helmets based on hard science.

They did however have a stab at a different kind of analysis that looked at a much smaller group of accident victims and examined their head injuries in detail to see whether headgear would have helped.

Their tentative conclusion was the 10 to 16 per cent of deaths in that small sample (113 cyclists) could have been prevented if they had worn a helmet.

The verdict

The science behind helmet safety is ultimately inconclusive, but we do know that in countries or regions where their use has been made compulsory, there has generally been a big fall in the numbers of people cycling, at least initially.

This has led campaigners to suggest that the cost to society of fewer people exercising (premature death, increased rates of disease) far outweighs the dubious safety benefits.

The British government would do well to bear that in mind before considering a change in the law here.

But the Transport Research Laboratory’s findings suggest we shouldn’t be too quick to dismiss helmets as ineffective. The team did conclude that a helmeted head can fall at least four times as far as an unprotected one for the same risk of injury.

The bottom line is that helmets do provide some protection against impact and, while they won’t save the life of every cyclist involved in a collision, they will save some.

The study also rejects an assertion sometimes made by the anti-helmet lobby – that helmets can actually increase the risk of a rotational head injury.

Critics of the helmet also point out that fewer cyclists wear them in countries like Holland and Denmark, but there are lower rates of injury.

That’s a difficult comparison to make, because the difference could be all about infrastructure: more cycle lanes to keep bikes away from heavy traffic.

If that’s true, it could be that the whole debate about helmets could be a distraction from the real solution to cycle safety. As Dr Walker puts it: “If the answer is a bicycle helmet, we’ve not understood the problem.”

By Patrick Worrall