BIKE HELMETS, CRASH SAFETY, AND PUBLIC HEALTH: From Anecdote to Evidence

I wear a bike helmet.  Always.  Every time I get on a bike.  I don’t think that the helmet will keep me from having an accident, just that it will reduce the odds of serious head injury in particular types of situations.  Small odds but a big benefit.

It’s likely that people who cycle like I do – regular commuters with enough experience and confidence to ride within busy traffic – suffer the most severe injuries.  I don’t want to be one of those statistics.  As my daughter (the doctor!) says about helmetless speedsters, “I hope they’re carrying an organ donor card.”

But avoiding injury– staying safe — is not my main motivation for cycling.  In addition to being cheaper and often faster than any other mode of urban commuting (as well as less polluting and more energy efficient), it helps me control my weight, stay fit, sleep better at night, have more energy the rest of the day, almost always puts me in a better mood – and is simply fun to do.  It keeps me healthy – body and soul.  I think it would be good for society if more of us biked instead of drove for at least the 25% of daily trips that are less than a mile long, if not for the 40% that are less than two miles and the 50% of daily commutes of less than five miles.

Safety and health:  two goals – the issue is how to pursue both at the same time.  Safety usually is given first place in policy discussions – although it seems as much from fear of getting sued as anything else. And given the profit-driven insanity of both our health insurance and liability systems, I don’t blame bike clubs or sponsors of cycling events for requiring that all participants wear helmets.

However, regardless of my personal proclivity, I don’t think it’s a good idea for governments to require that everyone wear a helmet.  While bicyclists have lots of “solo” accidents, serious injuries almost entire occur when they’re hit by a motorized vehicle, usually a truck, van, or bus.  Safety increases when drivers, including those in cars, become more aware of and accommodating to the presence of bicyclists.

Based on my own decades of bicycling around Boston, it seems that the biggest improvement in drivers’ acceptance of my presence – and therefore of my safety – happens when there are more cyclists on the road.  I’ve also seen the “numbers vs. accident rate” graphs from other cities, which reinforce the “safety comes from numbers” message.  And while it seems to be that a higher percentage of cyclists are wearing helmets these days than when I started, I’ve always assumed that requiring helmets would discourage some percentage of people from using a bike, which would both reduce safety and the public health benefits of physical activity.

“Assumed.”  “Thought.” “Felt.” “Experienced.”  – But I didn’t know.  Fortunately, it turns out that a lot of relevant research has been done on the topic.  With the help of Anne Lusk (Research Scientist, Harvard School of Public Health) and Price Armstrong (Program Manager, Massachusetts Bicycle Coalition/MassBike), we’ve pulled together 35 annotated citations, one of the most extensive lists I’ve found anywhere.   (The annotated links are visible by clicking on “Continue Reading…” below.)

The results are clear:  bike helmets reduce the severity of head injuries in the small percentage of bike accidents (better described as car crashes) that cause such an injury.  But helmets do not reduce the frequency of accidents or the percentage of head injuries caused by those accidents.  Worse, the passage of mandatory helmet use laws have actually been associated with increased accident rates because they led to significant decreases in the overall number of bicyclists, undermining the “safety comes from numbers” reality, with particularly disastrous impact on bike share programs.

I’m aware that simply opposing mandatory helmet laws is not enough.  Safety and health are both legitimate goals.  And we do not yet have enough evidence to fully evaluate strategies for pursuing them in mutually reinforcing ways.  However, it is time for the bicycling, public health, and public safety communities to move beyond traditional assumptions and find ways to further test and analyze the impact of the following:

1)      Stepping up efforts to create safer, low-traffic-stress bicycle facilities that have been shown to increase the number of “traffic intolerant” bicyclists and reduce crashes, such as traffic-separated cycle tracks and protected or buffered bike lanes, and low-traffic “neighborways.”

2)      Conducting a campaign educating drivers about safer methods of interacting with bicyclists, and educating cyclists about safer ways of interacting with cars, including the congestion-reducing impact of growing bicyclist numbers.  The cyclist component could be part of an effort to publicize the value, and increase the availability, of bicycling skill training workshops and classes – whose graduates are also more likely to wear helmets.  (San Francisco messages around the need for “co-existing” on the road rather than “share the road” on the grounds that it lets some car drivers think they have a choice to not share “their space” with intrusive cyclists.)

3)      Conduct a public relations campaign to encourage bicycling because of its personal and societal health, environmental, economic development, and travel-time reducing benefits — including carefully worded encouragement for voluntary helmet wearing which has been shown to both increase the percentage of people using helmets but also discourage cycling (although not as much as a mandatory helmet requirement) because it reinforces the public perception of bicycling as dangerous.

Hopefully, we can accumulate enough data to make intelligent choices that encourage more “ordinary” and traffic-intolerant people to regularly bicycle (including children and the elderly) while reducing the likelihood of injury-causing accidents, and lowering the severity of head and other injuries if an accident occurs.

(Note: I do support requiring and enforcing laws mandating front & back night lights since it reduces the risk of getting hit by a car by making it easier for drivers to see cyclists, and there is no evidence that their requirement will reduce the number of bicyclists.  Personally, I’m also fanatic about wearing shiny yellow jackets with as much reflective tape as will stick on.  It kills any hope of being a fashionista, but it makes me feel safer.)

—————————————

The following citations were gathered by Anne Lusk (Research Scientist, Harvard School of Public Health), Price Armstrong (Program Manager, Massachusetts Bicycle Coalition/ MassBike), and Steven E. Miller (Board of Directors, LivableStreets Alliance).  Since these were collected another list of citations has gone on the web at http://cyclehelmets.org/

1) There are statistically powerful correlations connecting increased number of cyclists on the roads and increased miles of appropriate bicycle facilities with reduced rates of cyclist-car accidents (and sometimes even reduced absolute numbers).  There is some evidence that increased numbers of cyclists leads to lower overall numbers of car accidents and injuries as well.

2) The presence, or the establishment, of legally enforceable helmet requirements has been shown to measurably reduce the number of people using bicycles, probably because of a combination of monetary, status, and convenience costs.  Furthermore, the resulting lower bike rates among young people persist into adulthood.  However, helmet laws do lead to a higher percentage of helmet use among those still riding, possibly because the cyclists most likely to continue riding with a mandatory helmet law are the ones who were already wearing helmets.

  • Carpenter and Stehr (2010) found that rates of bicycling among youth in states that enacted mandatory helmet laws declined by 4-5%. They also note that these lower rates of biking persist even after the youth are old enough to be exempted from the laws.

    *CARRS (2010) found in Melbourne, Australia that observed rates of bicycling fell 10% for children ages 5-11 and 44% for children ages 12-17; For every one teenager who began to wear a helmet, more than 10 others abandoned their bicycles.  Among adults in Queensland, data indicate that the number of trips by bike declined roughly 10% after the helmet laws came into effect;  although cycling rates had been increasing dramatically over the previous 15 years, the years after the new law saw a 36 percent decrease in ridership.

3) Most fatalities and serious injuries suffered by bicyclists occur when they are hit by a car.  There is no question that wearing an appropriately shaped, fitted, and well-constructed helmet while bicycling reduces the severity of head injury should a crash occur, although about two-thirds of serious injuries are to other parts of the body – meaning that the presence of a helmet wouldn’t influence the outcome.   In addition, although there is limited evidence about the impact of wearing a helmet on the likelihood of having an accident, some studies suggest that the passage of required helmet laws actually increases the likelihood of injury because of the reduced numbers of cyclists on the road.

  • Collisions with motor vehicles cause nearly all deaths and debilitating head injuries among cyclists.
    • Kraus JF, Fife D, Conroy C. “Incidence, severity, and outcomes of brain injuries involving bicycles. Am J Public Health 1987;77:76-8.

    *Dorothy Robinson (2005) found that after controlling for companion road safety campaigns, the bike injury rate actually increased after enactment of the 1990 helmet law. Her interpretation is that the “Safety in Numbers” principle was to blame – that is, because the helmet requirement reduced the number of bicyclists on the roads, the roads were relatively more dangerous.

  • Robinson (2006) found that in no country with mandatory helmet laws was there any solid evidence that they were effective in reducing the injury rate. She concluded that traffic law enforcement campaigns and increasing numbers of bicyclists were more effective strategies.
  • A Denver study, led by Dr. Jeffry Kashuk of the University of Colorado, presented at the American College of Surgeons 2009 annual meeting, found that two-thirds of bicycle accidents requiring hospitalization did not result in head injuries, and that the growing severity of bike accident injuries was mostly due to the increased percentage of on-road bike commuters over off-road recreational cyclists and the lack of appropriate “bikeways to support this increase.”
  • Carpenter and Stehr (2010) found that youth helmet use rates increased in US states enacting helmet laws by 20-34%, and fatalities decreased by roughly 19%.

    *However, New Zealand advocates felt that the reduction in the number of accidents after the passage of a helmet law was primarily due to other factors.

  • And at least one anecdotal study suggests that drivers pass closer to bicyclists wearing helmets.
  • Karkhaneh et al. (2006) found in an international review of 14 studies that bicycle helmet use changed from a baseline of 4-59% to a post-helmet law rate of 37-91%.
    • Karkhaneh, M, J‐C Kalenga, B E Hagel, and B H Rowe. Effectiveness of bicycle helmet legislation to increase helmet use: a systematic review. (2006) Injury Prevention, 12: 76-82. Retrieved from http://www.ncbi.nlm.nih.gov/pmc/articles/PMC2564454/ 4 February 2012.

    *The Centre for Accident Research and Road Safety (CARRS, 2010) found in Queensland, Australia that helmet use rates increased 20-30% among all age groups after a mandatory helmet use law was enacted in 1991.

4) Multiple studies show that increased cycling leads to a broad spectrum of individual and environmental health benefits – and that those benefits significantly outweigh any negative impacts from bicycle accidents. 

  • Jacobsen, PL says that increased numbers of bicyclists and pedestrians, independent of other variables, accounted for a significant decline in per capita injuries and fatalities.

    *CARRS (2010) notes, “In 1999, the British Medical Association’s (BMA) Board of Education and Science concluded that ‘cyclists are advised to wear helmets but legislation to make them compulsory is likely to reduce the number of people choosing to cycle and would not be in the in interests of health.’ (Carnall, 1999, p.1505).”

  • De Hartog et al. (2010) found a net benefit from biking of 3 – 14 months extended life despite increased exposure to air pollution (up to 40 days lost) and crash-related injury or fatality (5-9 days lost).
  • Some additional citations include:
    • Gotschi T. Costs and benefits of bicycling investments in Portland, Oregon. J Phys Act Health 2011;8 Suppl 1:S49-58.
    • Saelensminde K. Cost-benefit analyses of walking and cycling track networks taking into account insecurity, health effects and external costs of motorized traffic. Transportation Research Part A 2004;38:593-606.
    • Rojas-Rueda D, de Nazelle A, Tainio M, Nieuwenhuijsen MJ. The health risks and benefits of cycling in urban environments compared with car use: health impact assessment study. Bmj 2011;343:d4521.
  • According to a study published in 2006 by the British Medical Journal, cycling is not significantly more dangerous than either walking or driving. The study estimates that on average it takes 8,000 years of normal cycling to produce a serious head injury, and it takes 22,000 years to produce one death.

 5) Required helmet laws significantly impact bike share programs – cities having helmet laws have the least successful programs.  No other USA city with, or contemplating, a bike share program currently has, or intends to have, a helmet law.  (Hubway, as a multi-municipality system, has another layer of complexity – will a person renting a bike in Cambridge be subject to arrest if they enter Boston without a helmet?)

6) As a result of reduced numbers, helmet laws undermine the public health benefits of a city’s investment in bicycle facilities.  This is particularly true for low-income or minority populations and for infrequent riders who have been shown to be least likely to wear helmets, are least able to afford the infraction tickets, and most unwilling to risk future interaction with the police.  Low-income families will also be disproportionately hurt because they have the lowest rates of car ownership, transit costs are going up, and bicycling is the only low-cost way of getting around town for work or errands.

  • All published cost-benefit analyses of injury rates before and after helmet laws show the cost of helmets exceeded any estimated savings in healthcare costs.

7) Requiring bicyclist to wear a helmet is not equivalent to requiring helmets for motorcycling or seat belts for car driving.  There is no public health benefit from increased use of either of those motorized modes, so reducing the cost of injuries is the only issue.   Finally, a helmet ordinance in the City of Boston would be inconsistent with both state law and the local ordinances of surrounding communities, making enforcement extremely problematic, perhaps impossible. Even within Boston, from a law enforcement perspective, it will be expensive to enforce a helmet law and will divert limited resources towards more “stop and frisk” kinds of interactions as well as away from more important targets such as speeding, intersection violations, and other major causes of injury.  If the goal is increasing the health of the community, policies which support more biking should take precedence over those mandating helmet use.

* It may be useful to explore the “Health Economic Assessment Tool” (HEAT) for walking and cycling; Page 16 discusses the Relative Risk of bicycling and walking.

http://www.euro.who.int/__data/assets/pdf_file/0003/155631/E96097.pdf

* A 2006 article about helmets and the issues draws on the Australian experience.

http://www.cycle-helmets.com/robinson-bmj.pdf

* The UK-based Bicycle helmet Research Foundation came to a similar conclusion:

http://cyclehelmets.org/1139.html

To summarize:

There are statistically powerful correlations connecting increased number of cyclists on the roads and increased miles of appropriate bike facilities with reduced rates of cyclist-car accidents (and sometimes even reduced absolute numbers).  There is some evidence that increased numbers of bicyclists leads to lower overall numbers of car accidents and injuries as well, primarily because of reduced speeds.

The presence, or the establishment, of legally enforceable helmet requirements has been shown to measurably reduce the number of people using bicycles, probably because of a combination of monetary, status, and convenience costs. Furthermore, the resulting lower bike rates among young people persist into adulthood.  However, helmet laws do lead to a higher percentage of helmet use among those still riding – possibly because the cyclists most likely to continue riding with a mandatory helmet law are the ones who were already wearing helmets.

Most fatalities and serious injuries suffered by bicyclists occur when they are hit by a car.  There is no question that wearing an appropriately shaped, fitted, and well-constructed helmet while bicycling reduces the severity of injury should a crash occur, although about two-thirds of serious injuries are to other parts of the body – meaning that the presence of a helmet wouldn’t influence the outcome.   In addition, although there is limited evidence about the impact of wearing a helmet on the likelihood of having an accident, some studies suggest that the passage of required helmet laws actually increases the likelihood of injury because of the reduced numbers of cyclists on the road.

Multiple studies show that increased cycling leads to a broad spectrum of individual and environmental health benefits – and that those benefits significantly outweigh any negative impacts from bicycle accidents.

Required helmet laws significantly impact bike share programs – cities having helmet laws have the least successful programs.  No other USA city with, or contemplating, a bike share program currently has, or intends to have, a helmet law.  (Hubway, as a multi-city system has another layer of complexity: will a person renting a bike in Cambridge be subject to arrest if they enter Boston without a helmet?)

As a result of reduced numbers, helmet laws undermine the public health benefits of a city’s investment in bicycle facilities.  This is particularly true for low-income or minority populations who have been shown to be least likely to wear helmets, are least able to afford the tickets, and most unwilling to risk future interaction with the police.  Low-income families will also be disproportionately hurt because they have the lowest rates of car ownership, transit costs are going up, and bicycling is the only low-cost way of getting around town for work or errands.

Requiring bicyclist to wear a helmet is not equivalent to requiring helmets for motorcycling or seat belts for car driving.  There is no public health benefit from increased use of either of those motorized modes, so reducing the cost of injuries is the only issue.

Finally, a helmet ordinance in the City of Boston would be inconsistent with both state law and the local ordinances of surrounding communities, making enforcement extremely problematic, perhaps impossible. Even within Boston, from a law enforcement perspective, it will be expensive to enforce a helmet law and will divert limited resources towards more “stop and frisk” kinds of interactions as well as away from more important targets such as speeding, intersection violations, and other major causes of injury.

If the goal is increasing the health of the community, policies which support more biking should take precedence over those mandating helmet use.

Be the first to comment

Please check your e-mail for a link to activate your account.