Improving bicycle safety: The role of paediatricians and family physicians

Abstract

Cycling is a complex activity requiring motor, sensory and cognitive skills that develop at different rates from childhood to adolescence. While children can successfully ride a two-wheeled bicycle at age five or six, judgment of road hazards are poor at that age and matures slowly until adult-like judgment is reached in early adolescence. Safe cycling depends on the care, skills and judgment of cyclists and motorists; roadway design that promotes safe coexistence of bicycles and motor vehicles; and the use of safety devices, including bicycle helmets, lights and reflective tape. Whereas, research into optimal roadway design and educational programs for drivers to improve road safety has yielded contradictory results, the benefits of bicycle helmet use and programs to enhance their use have been clearly shown. This paper has the following objectives for paediatricians and family physicians:

1. To understand the relationship between bicycle safety and children’s motor and cognitive skills.

2. To understand the effectiveness and limitations of strategies to improve bicycle safety.

3. To describe activities to promote bicycle safety that physicians can undertake in clinical settings and in the community.

CYCLING SAFETY

Children and parents are thrilled when they can throw away the training wheels on a two-wheeled bike. However, cycling is not without its risks. Among Canadian children age zero to 19 years, there were 21 bicycle-related deaths in 2001 (2.6 per one million children, age zero to 19 years per year). Fortunately, this is a marked decline from 88 bicycle-related deaths in 1984 (11.6/million/year) and 46 bicycle-related deaths per year between 1990 and 1992 (6.0/million/year) (1,2).

Although children have sufficient motor control to ride a bicycle by age five or six, they do not yet have the skills to assess the risks of riding on roadways. A child’s judgment about traffic safety is poor between five- to seven-years-old and shows a gradient of increasingly adult-like judgment from five through 14 years of age (3–5). For example, nine-year-olds showed a higher level of ability than seven-year-olds to identify dangerous road crossing sites, but were not as good as 11-year-olds (6). In another study where five- to 12-year-old children were asked to identify the distance between them and a motor vehicle that allowed for safe crossing of the street, eight- and nine-year-olds scored significantly better than five-year-olds but they could not make safe decisions when vehicles exceeded 60 km/h. Most children relied primarily on the distance between themselves and the vehicle to assess safety and could not factor in the speed of vehicles adequately (5).

In general, parents have limited understanding about the preventability of cycling injuries and their role in prevention. In one national survey, parents were well informed about the risk of injuries to motor vehicle occupants; however, they knew little about the risk of injuries to pedestrians and cyclists (7). In another survey, parents believed that injuries were a natural consequence of childhood and made choices that placed their children at risk because of convenience, and because of their own goals or the belief that they could keep their children safe in hazardous situations (8). With respect to injury prevention and normal child development, a 1982 survey (9) of 118 parents found that parents had poor knowledge of both normal childhood development and injury prevention.

In addition to attitudes and behaviours of children and parents, risk of injury while cycling depends on responsible motorist behaviour, roadway design that promotes safe coexistence of bicycles and motor vehicles and the effectiveness of personal safety devices, including bicycle helmets, safety lights and reflective tape. Unfortunately, evidence supporting education and roadway design is contradictory. For example, one meta-analysis failed to find convincing evidence that driver education for adolescents reduced collisions and identified some programs where the risk appeared to increase (10). Some urban roadway designs may increase the likelihood of a cyclist/motor vehicle collision. For example, separate bicycle paths may encourage wrong-way riding and may lead to collisions at intersections (11). As a result, many experts believe that cyclists and motorists should share roads designed for both so that each is continually aware of the other.

The use of bicycle helmets, on the other hand, has proved to be one of the most effective and least costly means for reducing injury and death for which there is strong evidence from case-control, ecological and cross-sectional studies. Although by no means the sole solution to improving bicycle safety, the demonstrated effectiveness and ease of implementation of bicycle helmets should give it a high ranking in counselling directed towards injury prevention.

UNDERSTANDING THE EFFECTIVENESS AND LIMITATIONS OF BICYCLE HELMETS

The relationship between bicycle helmet use and prevention of head injuries has been studied for approximately 20 years. In 1989, the New England Journal of Medicine published a carefully conducted and powerful case-control study (12) that showed an 85% reduction in head injuries when helmets were worn while cycling. Recently, two meta-analyses have summarized this and many other studies. The first was a Cochrane review that restricted analysis to five high quality case-control studies. These studies demonstrated a “large protective effect of helmets” that supported a causal mechanism and could not be explained by selection bias, observation bias, confounding or chance. Helmets conferred a 63% to 88% reduction in the risk of head and brain injury for all ages, a 65% reduction of injuries to upper and mid-facial areas, a 69% protection from crashes involving motor vehicles and a 68% protection from all other causes (13). The second meta-analysis included cross-sectional surveys of injured and noninjured cyclists in addition to case-control studies (14). It presented overall odds ratios (ORs) for head and brain injuries similar to those of the Cochrane review (13) and also had sufficient data to calculate an OR for fatalities. Based on four deaths among 2461 helmeted cyclists and 43 deaths among 4794 non-helmeted cyclists, the OR was 0.27 (eg, a 73% reduction) with a 95% CI of 0.10 to 0.71. The OR for neck injuries was 1.36 and approached statistical significance (95% CI 1.00 to 1.86) – an approximate 36% increase in neck injuries if the result was not due to chance. However, these studies evaluated helmets designed in the 1980s and results may not apply to more recent thinner helmets that are now widespread. Even if one accepts the estimated 36% increase in neck injuries, this is more than offset by the 60% to 88% decrease in head and brain injuries and the rarity of neck injuries (about 1% of all injuries occurring to cyclists).

PROMOTION OF HELMET WEARING

Helmets will have no effect unless they are worn and worn correctly. Strategies used to increase helmet wearing include educational campaigns, promotion by professional organizations and legislation. These strategies are more effective when combined than when used alone. For example, during a multi-year education program, voluntary helmet use among Seattle school-aged children increased from 5.5% in 1987 to 40.2% in 1992 (15). From 1983 to 1990 in Melbourne, Australia, voluntary helmet use rose from 5% to 70% in primary school children, 2% to 20% in secondary students, and 27% to 40% in adults (16). Legislation alone has been followed by large gains in helmet use. After legislation was proclaimed in Nova Scotia in 1997, helmet use rates rose dramatically, from below 40% in 1995 and 1996, to 75% in 1997, to above 83% in 1998 and 1999 (17). Two studies assessed the synergy between education campaigns and helmet legislation. In Maryland, self-reported helmet use rose from 11% in 1990 before legislation to 37% in 1991 after legislation. This contrasted with a neighbouring county that used educational strategies but no legislation where rates increased from 8% to 13% over the same time period (18). In Cleveland, Ohio, helmet-wearing rates were compared in two suburbs with neither helmet-use bylaws nor educational campaigns. One suburb used education alone and another suburb had both bylaws and an educational campaign. In the community with both strategies, 67.6% of children reported wearing helmets compared with 37% in the community with an educational campaign alone and 18% and 22% in the 2 communities with neither strategy (19).

This increase in helmet use has been followed by a decrease in serious head injuries. In Australia head injuries decreased by 48% in the first and 70% in the second year after the introduction of legislation (16) and in Seattle, they decreased by two-thirds in five- to 14-year-olds from the beginning to the end of the education program mentioned above (15). This decrease in head injuries, as rates of helmet wearing increased, provides corroborating evidence that the strong association between helmet use and head injuries found in case-control studies is causal.

As of 2003, legislation requiring helmet use while cycling is in place in six Canadian provinces (Table 1). Currently, such legislation exists in 20 American states, and in all the states of Australia and New Zealand, but it is rare in Europe in spite of the popularity of cycling (20).

TABLE 1.

Bicycle helmet legislation in Canada

Province	Year proclaimed	Comments
British Columbia	1996	All ages
Alberta	2002	For those under 18 only
Saskatchewan	No	Legislation permits municipalities to pass bylaws
Manitoba	No
Ontario	1995	For those under 18 only
Quebec	No	Legislation permits municipalities to pass bylaws
New Brunswick	1995	All ages
Nova Scotia	1997	All ages
Prince Edward Island	2003	All ages
Newfoundland	No	Legislation permits municipalities to pass bylaws
Yukon	No
Northwest Territories	No
Nunavut	No	Being considered

COUNTERING OPPOSITION TO HELMET WEARING

Paediatricians must be prepared to counter two arguments made by opponents to bicycle helmet legislation: first, that wearing helmets gives the rider a false sense of security and may actually increase risk taking activity (the ‘risk homeostasis theory’); and second, that forcing riders to wear helmets through legislation has actually led to a marked decrease in riding rates. It is claimed that decreased rider-ship has led to the decrease in head injury rates following the introduction of helmet legislation (21). It has also been claimed that “cycle helmets inevitably damage public health” (22) and that “helmet laws save a few brains but destroy many hearts” due to presumed decreased ridership and the assumption that people who stop cycling will substitute this with less healthful sedentary activities (23). Although the evidence for both of these explanations is weak and contradictory, it is nevertheless cited by those opposed to helmet legislation.

Risk homeostasis implies that the adjustment of our risk behaviour depends on our perception of risk. For example, a person would adapt their behaviour toward greater risk when wearing a helmet if they perceived themselves to be safer as a result of wearing the helmet. Wilde (24), the originator of this theory, hypothesized that all human beings have a drive to live at a certain level of risk and will adjust their behaviour to stay close to that level. The theory suggests that all injury prevention gains conferred by improving technology would be perfectly offset by increased risk activity when it is possible to modify risk. Evidence from many domains demonstrates that risk homeostasis does not exist, although there are several studies that illustrate changes in behaviour depending on perception of risk, such as the evidence that some risk compensation occurs. In particular, one example showed that taxi drivers used shorter stopping distances when antilock brakes were installed (25). Parents tested in simulated play situations stated that they would allow children to increase bicycle speed if they wore protective gear (26). Counter-examples also exist. For example, the introduction of helmets in snowboarding has resulted in reduced head injuries but no concomitant increase in neck injuries as would result if helmeted snowboarders took greater degrees of risk (27). A recent study examined 23,000 cyclists involved in traffic crashes and found that committing a speed-related infraction was not associated with the frequency of helmet use (28). A Cochrane review states that cyclists who wear helmets would have to increase their risk-taking behaviour fourfold to nullify the reduction in head injuries conferred by wearing bicycle helmets (13). If this change in risk-taking behaviour existed, all case-control studies of the effectiveness of helmet use would have missed this outcome in spite of adjusting for rider experience, crash severity and cycle speed (29). In summary, although risk compensation has been demonstrated in specific settings, there is substantial evidence that risk homeostasis does not exist.

With respect to ridership effects, a single study (16) reported in 1993 showed a decrease in ridership rates in children and adolescents, but not in adults, two years following the introduction of legislation in Australia. On the other hand, ridership actually increased in children and adolescents one-year following legislation in East York, Ontario (30).

WHAT THE PHYSICIAN CAN DO TO PROMOTE BICYCLE HELMET SAFETY

Physicians involved in primary care can improve children’s use of bicycle helmets through anticipatory counselling. A randomized-controlled trial of counselling following presentation to the emergency room with a bicycle injury did not find that counselling increased subsequent bicycle helmet purchase (31). However, in a recent randomized-controlled trial in a clinical setting, 3525 children and their families were randomized to an intervention focusing either on tobacco and alcohol use, gun safety, or bicycle-helmet and seatbelt use. An office-systems approach was used where other office staff reinforced the primary message of the paediatrician with pamphlets and posters. Compared with the group receiving counselling about alcohol and tobacco, the number of children who did not use bicycle helmets during the previous 12 months decreased by 25% in the group receiving counselling (32).

The American Academy of Pediatrics recommends that paediatricians promote helmet use for both children and their parents, emphasize proper use of helmets and promote all aspects of bicycle safety. They also encourage paediatricians to act as advocates to promote mandatory helmet wearing (33).

In addition, it is recommended that physicians review in some detail the role of parents in injury prevention, the variety of skills required to cycle safely, and the gradual increase in children’s judgment and skill assessing risky cycling situations as they age. Specifically, paediatricians and family physicians can do the following:

• Review the development of children’s motor and cognitive skills with respect to bicycle safety. Rather than specifying a ‘safe’ age for cycling, counselling should aim to convey the child’s gradual acquisition of judgment and skills required to reduce risk and the pivotal role of parents in assessing risk based on their children’s development and the safety of roadways where they might cycle;

• Promote the use of safety measures such as lights, reflective tape and helmets for both children and their role models (parents and caregivers); and

• Illustrate correct positioning of helmets and their supporting straps.

In addition to individual counselling, physicians can be influential public health advocates. Statements recommending helmet use by professional bodies such as the Canadian Paediatric Society are important but require dissemination and promotion. For example, helmet legislation exists due to advocacy by professional bodies, collaboration amongst professionals interested in safety, public demand, and the lobbying of local government officials and politicians. As respected child health advocates, paediatricians and family physicians can and should play an important role in promoting measures that increase correct helmet use. From a public health perspective, physicians can do the following:

• Write and meet members of legislative assemblies and provincial cabinet ministers;

• Encourage parents to write and meet these politicians;

• Collaborate with organizations that promote child injury prevention such as local public health groups and organizations concerned with head injury; and

• Periodically contact media, for example, following a bicycle-related injury in the community.