An open letter to Dr. Patrick J. Crocker, Chief of Emergency Medicine at Brackenridge & Children's Hospital of Austin, in response to the Oct. 11, 2006 press release announcing Crocker's plan of conducting a bicycle helmet study to gain ammunition for his support of a mandatory bike helmet law.

Dear Dr. Crocker,

The press release announcing your study says, "The initiative was created to test the hypothesis that an all-ages, citywide bicycle helmet ordinance will substantially mitigate preventable head injuries due to bicycle accidents...."

But it is impossible for you to come to any such conclusion because there are important factors that your study cannot measure. What's more, cyclists such as Patrick Goetz, myself, and others have explained these factors quite clearly, but you are apparently ignoring them. This is extremely disappointing. For its intended conclusion, your study is worthless.

I will detail the various factors your study fails to consider below, but let me lead with the most significant one:

Bicycle helmet laws indisputably discourage bicycling. This ultimately makes cycling more dangerous for those cyclists who remain, since cycling safety is extremely highly correlated with the number of cyclists on the road. Your study does not (and can not) consider this problem, and thus it is impossible for your study to credibly support the conclusion that mandated helmet usage means a reduction in injuries.

Support for a helmet law assumes that once the law is passed, the only thing that will change is that previously unhelmeted cyclists will start wearing them. This view is dreadfully simplistic and naïve. The unintended consequences of such legislation not only negate any public health benefit, they likely increase the danger to people who continue to cycle. As you said in your August 10 letter to the City Council, "read that last line again".

These dangers include, but are not limited to:

• a more dangerous environment for cyclists when there are fewer of them on the road.
• a more dangerous environment for everybody when former cyclists start driving instead.
• the possibility that helmet-wearers may be more likely to be struck by a motor vehicle (study described below).
• the fact that the health benefit from cycling exceeds the risk from a mortality standpoint.

Cycling advocates have explained most of these issues in detail at the very same council meeting where your own testimonial was read into the public record in your absence. This hearing aired on the City's public information television channel and a transcript is available on the city's website. I also know that Rob D'Amico wrote to you directly about his concerns.

If you did not pay attention to these points raised by myself and others, that is quite disturbing. Since you are not only attempting to influence public policy but are also conducting original research to support that effort, it behooves you to notice when problems with both that policy and the research behind it are pointed out very plainly. Perhaps you disagree with the concerns raised, but if so you have not bothered to even acknowledge the existence of such dissent, much less answer it thoughtfully. I'm afraid these things conspire to make your research non-credible even before it begins.

Again, helmet efficacy, even if proven, is completely different from expecting a decline in injuries by enacting a law. You are confusing a good thing with a law requiring a good thing. A law has many unintended consequences which can turn a good thing into a bad thing, even from a public health perspective.

Here's an example of how unintended consequences work in legislation. In the 16th century the crown got the idea to levy a tax based on how many of those new-fangled glass windows people had in their homes. They figured it would be an easy way to raise money, and their revenue analysis consisted of this: estimating the number of windows that were out there, and multiplying that by the amount of tax. (Much like estimating the number of injuries prevented by passing a helmet law.) But here's what actually happened when the tax was levied, as explained by a member of Canadian Parliament:

Lo and behold, if we go to small English towns today and look at some of the Tudor homes, we will see over and over again these very clear window shaped spaces, that have been plastered over, and filled in or bricked in. Why? It was because 500 years ago people understood that they could [avoid] taxes legally and so they bricked in those beautiful glass windows all across England.¹

Likewise, a bicycle helmet law will result in numerous unintended consequences, yet your limited research relies on the mistaken assumption that once a law is passed, nothing will change except that previously unhelmeted cyclists will start wearing them. But, as with the window tax, people will take other, unexpected actions in response to the law. Let us go over those unintended consequences.

1. Helmet laws discourage bicycling. (indirect effects)

Every community which has passed a helmet law that was enforced has seen the number of people who bike for transportation plummet.

• In the Australian Northern Territory, bike commuting plummeted 50%.
• In the capital territory, the number of cyclists dropped between 33 and 50%.
• In New Zealand, it fell by 22%.
• In British Columbia it dropped 28%, and dropped 40 to 60% in Nova Scotia.
• In Denmark, cycling by children fell 30%.

See CycleHelmets.org for the source studies² and even more examples of cycling reduction.

The reduction in cycling is probably not just because some cyclists prefer to not wear helmets, it is likely also because helmet laws paint a picture of bicycling as a dangerous, inherently unsafe activity, which make people less inclined to take it up. But the reasons that cycling declines is beside the point; for whatever reason(s), the fact is that helmet laws discourage cycling.

And when people don't bicycle, they drive cars instead. This has a number of public health implications, such as increased pollution (auto emissions kill 30,000 Americans every year³), more global warming and climate change (some would say that a major threat to the existence of every living thing on the planet could be considered a public heath issue), and more dependence on foreign oil (over 225 Texas soldiers have been killed in Iraq, far greater than the number of bicyclists killed here).

Less abstractly, from a mortality perspective the health benefits from cycling outweigh the risks. Research from New Zealand suggests the benefit vs. risk is as much as 20:1.⁴ This is yet another way that helmet laws, by discouraging cycling, have an overall negative public health benefit.

Dr. Crocker, your simple research does not (and can not) take into account the reduction in cycling caused by helmet laws, nor the public health problems that result.

2. A reduction in cycling increases the risk for the cyclists who remain.

The evidence is clear, convincing, and overwhelming that communities with the highest rate of bicycling also have the lowest rate of cycling injuries.⁵ The assumed reason is the safety in numbers principle: The more cyclists on the road, the more drivers are used to seeing and interacting with them. A reduction in cycling participation increases the risk for the cyclists who remain on the roads.

In fact, such a reduction increases the risk to pedestrians and even other motorists, because when people don't bike, they drive instead, where they're more likely to kill other road users. A helmet law has the unintended consequence of making the roads a little more dangerous for everyone.

Dr. Crocker, your simple research cannot measure the increased risk to remaining cyclists resulting from the decimation of the cycling population.

3. Helmet-wearing may promote collisions.

Recent research from England shows that motorists give less room to when passing helmeted cyclists vs. unhelmeted ones.⁶ The researcher conducting this study was actually hit twice by motor vehicles while biking, each time while he was wearing a helmet. While we've always known that wearing a helmet won't prevent collisions, we now know that helmet-wearing may actually promote injury by promoting collisions.

Dr. Crocker, your simple research apparently ignores any increased risk of collision faced by helmeted cyclists, if it attempts to measure it at all.

 

While failure to account for unintended consequences is the major flaw in your research, it is not the only one. Here are others.

4. Case-control studies suffer from self-selection bias.

Cyclists who voluntarily wear helmets are apparently more cautious as a group than those who don't wear helmets, or wear them only when forced to do by law. The same studies that show that helmets are 85% effective in preventing head injuries also show that helmets are 72% effective in preventing leg injuries.⁷ Obviously bicycle helmets can't prevent broken arms and legs. A reasonable conclusion is therefore that helmeted cyclists suffered less injury because cyclists who choose to wear helmets are more careful in general, and that we can't necessarily expect the same kind of benefits by requiring people to wear helmets who are not already inclined to do so.

Dr. Crocker, your over-simplistic case-control study cannot control for such confounding variables. Any data you collect is therefore ultimately useless.

5. Population studies have already disproved your hypothesis.

In the 1990's, helmet use among cyclists in the U.S. skyrocketed. And so head injuries among cyclists went down, right? No, in fact, as helmet use increased, the rate of head injuries among cyclists went up. As you like to say, read that last line again.

As the New York Times reported, head injuries among bicyclists went up 51% even as the use of helmets became widespread.⁸ What can account for this? There are a number of possible explanations, the first two as mentioned previously:

• The number of cyclists in the U.S. has been declining. Fewer cyclists on the road means more danger for those cyclists who remain.
• Motorists give less room when passing helmeted cyclists vs. unhelmeted ones, likely leading to more collisions with helmeted vs. unhelmeted cyclists.
• Helmets might give riders a false sense of security, causing them to take more risks and suffer more injury. (This view is at odds with the idea above that voluntarily-helmeted cyclists are safer, but I'm just throwing out a possible explanation for the increase in head injuries.)

Whatever the reason(s), they may be largely beside the point. The point is that increased helmet use in the U.S. did not reduce head injuries among cyclists; for whatever reason(s), it had the opposite effect. If you arrive at your desired conclusion that helmets save lives, how will you explain that nationwide statistics covering thousands of cyclists suggest a very different result than that provided by your tiny, geographically limited study?

Dr. Crocker, even if your simple research finds helmets to be effective in preventing injury when examined from a limited case-control perspective, it cannot measure the true outcome in the real world, which is the only outcome that matters.

6. Does not consider non-compliance and non-enforcement.

In your August 10, 2006 letter to the Austin city council, you say that one reason we need a helmet law is that only 5% of child bicyclists treated at your trauma center were helmeted. You say, "This is one area where a helmet law could provide a significant safety factor for our children."

Dr. Crocker, how can you not be aware that Austin has had a helmet law for children on the books since 1996?! I might also wonder why you feel qualified to weigh in on a public policy matter you obviously know precious little about. But more to the point, if we already have a helmet law for kids, and only 5% of kids are wearing helmets, then clearly legislation has not achieved its intended goal.

Dr. Crocker, your simple view that head injuries will be prevented by a helmet law fails to consider that requiring helmets by law doesn't necessarily get them onto people's heads.

7. Fails to put its findings into context.

Cyclists account for only 2% of road injuries and deaths.⁹ Two percent. This suggests that a much greater public health benefit exists by requiring motorists to wear crash helmets. If helmets are not uncomfortable or inconvenient for bicyclists, then they're even less uncomfortable and inconvenient for motorists, who have the benefit of air conditioning and lockable storage for the helmet. Any argument you make for cyclists wearing helmets is an even stronger argument for motorists to wear helmets, too. Yet you're not making that argument. And you haven't explained why you're singling out cyclists among all transportation users for mandated helmet wearing.

Dr. Crocker, your simple research fails to consider the group that most suffers from injury and death on our roadways: motorists. Why are you ignoring this greater potential for reducing transportation-related injuries?

I have focused only on the public health problems caused by helmet laws, but there are other downsides to be sure, which I will mention only briefly:

• Overzealous enforcement. When Austin had an adult helmet law in 1996, many cyclists were arrested and went to jail for being unhelmeted; not just tickets, but arrest and jail.
• Racial profiling. From 1997-99, over 90% of the no-helmet tickets given to kids went to black and Hispanic kids.
• Lack of justice for cycling victims Motorists frequently face little to no penalties for killing or injuring cyclists, even when the motorists are clearly at fault.¹⁰ This is evidently the result of a societal bias against cyclists, and a presumption of guilt on the part of the cyclist. Many people feel that cyclists are "asking for it" just by being on the roadway. If it is already difficult to hold at-fault motorists responsible for killing or injuring cyclists, imagine how much harder it would be when an unhelmeted cyclist gets hit and there's a helmet law on the books. Insurance companies will have a cakewalk in denying payment to surviving cyclists or their families. Even if the cyclist is helmeted, the existence of the law alone still furthers the idea that it's the cyclist who is truly culpable.

If you really want to promote bicycle safety, may I suggest your support of the following measures:

• A safe-passing law. Require motorists to give cyclists at least three feet of clearance when passing, as is done in other locales.
• Getting cars out of bike lanes. Austin, Texas has the dubious distinction of allowing cars to park in many of its bicycle lanes -- legally. Bike lanes are useless if cyclists can't actually use them.
• Striping more bike lanes. More bike lanes encourages more biking, and gives cyclists dedicated space on the roadway.
• Creating bike-only thoroughfares. The risk to cyclists is from motorists. Providing a car-free space enhances safety -- and encourages more bicycling (which in turn enhances safety).
• Requiring bike safety to be taught in public schools.
• Providing free bike safety classes to the general public.
• Banning cell phone use while driving.
• Holding at-fault motorists responsible when they injure and kill cyclists.
• Promoting bicycling as a healthy and community friendly activity. The more people who ride bikes, the safer bicycling becomes for everyone, since motorists quickly learn to look for and expect to find bicyclists on the roadways. The large percentage of transportation bicyclists is likely the single most important reason why cities like Amsterdam and Copenhagen enjoy bicyclist injury/fatality rates that are 1/10 or less than that of the United States -- and this with virtually no bicyclists in these communities wearing helmets. When one adds to this the public health benefits of bicycling, actively promoting bicycling is likely one of the most effective preventative health care measures that could be adopted. Many public health officials have already declared obesity to be the single most critical public health problem in America today, and bicycling is an excellent way to control obesity and prevent heart disease, another top killer.

But returning to the issue of the problems with your study: Dr. Crocker, as a respected medical professional who is attempting to influence public policy through research, it behooves you to address the concerns raised about that research. I believe you owe it to the community to end your silence on these issues.

Sincerely,

Michael Bluejay
Internationally-published bicycle safety author
BicycleSafe.com

1. Mr. Jason Kenney, Canada's 37th Parliament, 1st Session, March 2, 2001

2. How helmet promotion and laws affect cycle use, CycleHelmets.org

3. From the Eugene/Springfield (OR) Bicycle Map (1998?), which further credits the American Lung Association, Oregon Traffic Commission, Association of Commuter Transportation, American Automobile Association, and City of Eugene. Reported at BicycleUniverse.info

4. New Zealand bicycle helmet law-do the costs outweigh the benefits? Taylor M, Scuffham P. 2002. Injury Prevention: 2002;8:317-320

5. CycleHelmets.org, graph on home page

6. Drivers give unhelmeted cyclists less room, Reuters, Sept. 12, 2006

7. Head injuries rising despite bike helmets, Julian Barnes, New York Times, July 29, 2001

8. The Dr. Patrick J. Crocker letter, Patrick Goetz, League of Bicycling Voters, October 2006

9. Traffic safety facts 1996: Pedalcyclists, U.S. Dept. of Transportation

10. No justice for cyclists, BicycleAustin.info