A cycling enthusiast asked me about helmets. It seems compellingly obvious to me that a head impacting the pavement without a helmet is likely to sustain more damage than a head protected by a helmet. He challenged that, citing a BMJ article by Ben Goldacre that questioned whether the evidence showed that helmets do any good. He said I was making a non-evidence-based assumption and challenged me to actually look at the evidence, so I did.
Goldacre says there is a:
complex contradictory mess of evidence on the impact of bicycle helmets. Like most places where there’s controversy and disagreement, this is a great opportunity to walk through the benefits and shortcomings of different epidemiological techniques, from case control studies to modeling.
He proceeds to give a lesson in epidemiology. He points out that there are a lot of emotion involved, and that epidemiologic studies, because of their inherent imperfections, are probably not capable of resolving the debate.
There are basically two questions:
- What is the effect of wearing a helmet for the individual?
- What is the effect of a public policy that promotes or requires helmet use?
He shows why research has difficulty answering both of those questions, and covers some of the possible confounding factors:
- If the controls are cyclists with other injuries, the analysis assumes that wearing a helmet doesn’t affect overall accident risk.
- People who choose to wear helmets may be different from those who don’t (i.e., more cautious).
- People who are forced by legislation to wear helmets may be different:
- They may not wear helmets correctly
- Their behavior may change through “risk compensation” (“I’m protected, so I can take more chances.”)
- Drivers may give larger clearance to cyclists without a helmet
- Even if helmets have reduced injury rates, legislation might not have public health benefits overall, because of changes in cycling rates.
- Regular cyclists tend to live longer because of the health benefits of exercise.
- Reduction in risk is greater where crashes are more common, such as for children.
- Legislation that compels compliance may selectively reduce cycling in the subgroup of cyclists who tend to cycle more slowly and have less equipment like helmets.
- In countries like Denmark and the Netherlands, low injury rates without helmets are achieved through interventions like good infrastructure, legislation to protect cyclists, and a culture of routine non-sporty, non-risky cycling.
- Risks and benefits are exaggerated or discounted due to political, cultural, and psychological factors.
Some other problems: studies focus on lives saved rather than injuries mitigated, and the documented effect of legislation is admittedly small. There is probably under-reporting of cycling accidents. Helmets may not prevent concussions; the scalp is protected from a direct impact but the brain is still vigorously jostled. Other measures including training may do more to mitigate injuries than helmets.
I reviewed relevant studies in PubMed and systematic reviews.
Here is an impressive list of cycling helmet safety studies.
Systematic reviews have shown efficacy.
Impact simulations have verified that helmets reduce impact to the head.
Individual studies reach varying conclusions; but overall, I interpret the existing body of evidence as showing that the science of protection is clear: helmets offer a significant benefit. The advisability of helmet laws is an entirely different question. The impact of legislation has not been as impressive as hoped, and decisions about laws are not made simply on the basis of science. Those decisions can (and should) be informed by science, but ultimately they depend on societal attitudes and other factors.
Physics and common sense
We don’t need to do randomized controlled studies to know that parachutes save lives or that it’s a good idea to suture wounds and immobilize broken bones, and we shouldn’t need randomized controlled studies to know that helmets are protective. The principles of physics and a little common sense tell us that helmets must provide some protection, even if the degree of that protection can be disputed. If you personally had the choice of hitting your head on the pavement with or without the protection of a helmet, which would you choose? It seems to me to be a no-brainer. (If nothing else, a helmet would prevent skin abrasions and contamination of wounds with dirt and gravel.) We have long accepted the wisdom of using helmets in a variety of sports, from show jumping to football. Construction workers use hard hats. Soldiers wear helmets in combat. Eggs are sold in protective containers rather than dropped individually into shopping bags; the principle is the same. It seems to me that the reaction against helmets for cycling is more emotional than evidence-based. Helmets clearly offer at least some degree of insurance for the individual. Whether individuals should be compelled by legislation to protect themselves is another matter entirely, a decision that I am willing to leave to others.