Why I don't support mandatory helment laws.

Please skip to 4.0 Conclusion for the abridged version of this long post. 1.0 Introduction I wear a helmet when I bike, and cannot imagine cycling without one when either road riding, mountain biking or on a family ride. I am uncomfortable without a helmet and feel vulnerable to injury by cars and my own stupidity - I have fallen off my bike often enough to know that for me a helmet is probably a good idea. Furthermore, I am trying to instill this habit with my children; they always have a helmet on when they get onto any self propelled wheeled vehicle. To my mind, there is no downside to wearing a helmet as long as you cycle within your ability.

But, having said the above, I am not certain that mandatory helmet laws are effective in preventing head injuries. In fact, I wonder if the bicycle helmet debate distracts us from a harder discussion on road safety and risk. I think that it is easy to have a public debate about a fringe group of cyclists and how they should be safer, rather than focus on road safety which would include slower residential speed limits, impact road and urban design, and be more difficult to reach consensus. After all bike helmets save lives, right? I am not sure, and so I am looking to answer the following questions:

A. Do mandatory helmet laws prevent head injuries?

B. What are the consequences, intended or otherwise, of mandatory helmet laws?

C. Are there better ways to protect cyclists?

I am writing within context of Alberta, which has had a helmet law since May 2002 that applies to children under the age of 18. Furthermore, this is a first attempt at understanding a complex issue. Please help me understand you position, comment below!


2.0 Evidence Based Policy I don’t think that it is correct to assume that helmets save lives, just as I don’t believe that climate change is a hoax because it is −20C in Edmonton while I write this. Where there is a hypotheses related to policy, there should be a way of collecting data to test that hypotheses. In other words, good public policy is not made from individual cases, anecdote, or from a gut feeling, but rather should be made from a position of informed choice where all viable options and outcomes are examined through the lens of data analysis. The final policy choices may be made for political reasons, but at least these are informed decisions. Medicine is an example where evidence plays a big role in determining best practices. As discussed here, evidence based medicine (EBM) is the “…conscientious, explicit and judicious use of current best evidence in making decisions…”.

Every time (I hope, but perhaps this is a pipe dream) you go to an emergency department, see a medical specialist, or have a procedure done, you rely on this process to get the best results and safest treatment (otherwise you see a naturopath). Evidence based decision processes do not have to be limited to medicine, we can expect the same rigor for public policy, in general, and within the helmet debate specifically. To that end, I plan on examining the evidence that addresses my questions.

3.0 What does the data say? Three types of data should be considered when assessing whether cyclists, by law, should be required to wear helmets - (1) bicycle / vehicle collisions and (1a) bicycle incidents that are comprised of falls or collisions with stationary objects or other bikes, and (3) emergency room visits related to bicycling accidents. These data should be looked at within an Albertan and international perspective. To give a relative risk, pedestrian data will also be examined. Caveat: I don’t have total numbers of cyclists and pedestrians - so their is some interpretation going on here based on totals numbers.

3.1 Bike, pedestrian and vehicle interaction in Alberta The following data are taken from Alberta Traffic Collision Statistics 2010 (found here [PDF]) , and represent the number of incidents for 2010.

Road User Killed (N) Killed (%) Injured (N) Injured (%) Total (N) Total (%)
Drivers 185 53.8 11012 60.3 11197 60.2
Passangers 65 18.9 4528 24.8 4593 24.7
Pedestrians 35 10.2 1129 6.2 1164 6.3
Motorcyclists 31 9.0 683 3.7 714 3.8
Bicyclists 6 1.7 467 2.5 467 2.5
Other 22 6.4 450 2.4 462 2.5
Total 344 100 18253 100 18597 100

These data were copied from Table 3.1 of the Alberta Traffic Collision Statistics, 2010

3.1.1 Observations I am surprised that more pedestrians were killed and injured than bicyclists in 2010. 1164 pedestrians (6.3% of the total) as compared to 467 bicyclists (or 2.5% of the total). These are interesting numbers, but not really relevant unless we know the total number of pedestrians in Alberta vs. bicyclists. Or even better, the total number of kilometers walked by pedestrians vs. the total number of kilometers cycled by bicyclist. In this latter case, we could then calculate injuries and deaths per kilometer travelled for each case. It would also be interesting to factor motorcycles into this mix. A side note. The raw numbers clearly indicate that about 2.5 times more pedestrians were killed than cyclists in 2010 and yet no one is suggesting that pedestrians wear helmets.

3.1.2 Other observations from the Traffic Collision 2010 report From the same report, Table 9.4 indicates that “Casualty rates per 10 000 population were highest for persons between the ages of 15 and 19”. The lowest rates were for those under 5 and over 55. 27.1% of bike accidents due to the cyclist disobeying a traffic signal, and 13.5% due to the cyclist failing to yield the right of way at an uncontrolled intersection.

From Table 9.3, 9.2 and 9.1 - most accidents occurred on weekdays between 3 − 7PM (rush hour), in June. From Section 8 of the report; Pedestrians faced the similar hazards, with most vehicle, pedestrian interactions occurring during rush hour, in October. It is interesting to note the language of the report points to the actions of the drivers being responsible for pedestrian injury (rather than pedestrians being responsible for their own injury) or death, versus the action of the bicyclist being responsible for their own injury or death.

3.2 Cycling rates and ER admittance The section will draw upon a PhD thesis, from the University of Alberta, written by Mohammad Karhaneh in 2011. Commentary and insight on this thesis was found here.  . Karhaneh notes the following based on pre- and post-law observational studies completed in 2006:

  • There was a large decrease, 56%, in children’s cycling from 2000 to 2006.
  • There was also a decrease in teenage (13-17YO) cycling at 27%
  • A 21% increase in cycling for adults.

In his dissertation, Karhaneh simply notes a decease in the number of people seen in the ER as a result of cycling injury. He does not create a ratio of cycling injury to the number or rate of cycling in the population (from cyclehelmets.org). In contrast to the decline in cycling, the rate of injury seems to have gone up after a mandatory helmet law was legislated. As cyclinghelmets.org notes:

“The number of children treated in emergency rooms for non-head injuries was an average of 1,762 per year in 2003-6 compared with 1,676 in 1999-2002, despite there having been a fall of 56% in children cycling over the period. For teenagers the average number of injuries rose from 870 to 1,101 per year while the amount of cycling went down by 27%.”

4.0 Conclusion

This is a complex issue that involves many factors, and I have only offered a slice of the available data, with some commentary that makes sense to me provided by cyclehelmets.org.  I have not touched on most of the available data, nor have I explored the physical limits of helmets as protective equipment (see here [PDF] for an overview). I have also neglected to look at injury rates in other jurisdictions that have slower residential speed limits, dedicated cycling lanes and a stronger bike culture.  I could complete a PhD on this question, and in fact some have - perhaps this complexity and depth of information is a barrier.

To summarize, it seems that helmets are of use for a limited population in certain circumstances, as in children cycling at slower speeds who fall a shorter distance onto smooth ground.  As soon as the speeds increase, and the falls become more complex (uneven ground, impacts with other moving objects such as cars, etc.), the protective benefits  of a helmet are negated.  Male children 15 - 19 seem to be the most at risk of injury - the data indicates that they are, as a population, more reckless, and take greater risks beyond the capacity of a helmet to manage.  In fact, it is speculated, that if they were not wearing a helmet, that 15-19 YO males might not take so many risks. In other words, the helmet was perceived to offer more protection than it does. This finding can be extrapolated more widely to account for the general increase in head injury found in Alberta AFTER the introduction of bike helmet laws.  People take more risks when they think they are protected + the limited protection afforded by a helment = more cycling injuries.

I also got the sense from the data that a culture of cycling was protective against injury for those who commute. Dedicated bike lanes, more bikes on the road and drivers who were more aware seems to make roadways safer. I also wonder about residential speed limits for overall road safety. What if we were limited to 30KM/HR rather than 50?

Finally, the data also seems to indicate that the perception of cycling as a dangerous activity dissuades people from riding, leading to two unfortunate outcomes: (1) those who do not cycle, don't gain the health benefits,  and this could be significant given increases in obesity, and; (2) spontaneous Bixi Bike  type trips are less likely to happen in those jurisdictions that have helmet laws.

In general the rate of head injuries is declining, but this is not consistent across the country, nor is it attributable to legislation as some provinces with legislation experienced a decline while others did not.”  Middaugh-Bonney T, Pike I, Brussoni M, Piedt S, Macpherson A, 2010. Bicycle-related head injury rate in Canada over the past 10 years. Injury Prevention 2010;16:A228.