Peer and Adult Companion Helmet Use Is Associated With Bicycle Helmet Use by Children
http://www.100md.com
《小儿科》
Division of Pediatric Medicine (Pediatric Outcomes Research Team), University of Toronto Faculty of Medicine and Hospital for Sick Children
Bloorview Research Institute and the University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Objective. To examine whether the helmet status of riding companions (child or adult) influences helmet use by children (5–14 years).
Methods. This study is part of a larger prospective observational survey that was conducted in a defined urban region of Toronto, Canada (1990–1997, 1999, and 2001). Trained observers collected data on children who were riding bicycles in April through October each year. Observation sites included school yards, parks, major intersections, and residential streets.
Results. Of the 2094 children who were observed riding a bicycle, 50% were riding alone, 36% were riding with at least 1 child companion, and 14% were riding with at least 1 adult companion. Compared with riding alone, children were less likely to wear a helmet when riding with nonhelmeted child companions (relative risk [RR]: 0.29; 95% confidence interval [CI]: 0.22–0.38) and were more than twice as likely to wear a helmet when riding with either helmeted child (RR: 2.18; 95% CI: 1.97–2.42) or adult companions (RR: 2.67; 95% CI: 2.44–2.93). Compared with children who were riding with nonhelmeted child companions, children were more likely to wear a helmet when riding with nonhelmeted adults (RR: 3.93; 95% CI: 2.86–5.40), helmeted adults (RR: 9.18; 95% CI: 7.04–11.98), or helmeted children (RR: 7.49; 95% CI: 5.72–9.82). Adjusted analyses showed no difference between companion helmet status and helmet use by children before and after legislation.
Conclusions. Helmeted and nonhelmeted adult riding companions were positively associated with child helmet use. The association between child riding companions and child helmet use was positive for helmeted child riding companions and negative for nonhelmeted child riding companions. Efforts need to be made toward improving adult helmet use and children's perceptions and attitudes toward helmet use.
Key Words: bicycle helmet children legislation
Abbreviations: RR, relative risk CI, confidence interval OR, odds ratio
An estimated 7 per 1000 children who are 1 to 14 years of age have a bicycle-related injury each year in the United States.1 Although fatalities are uncommon (5/100000/y), there are an estimated 51 hospital admissions and 1437 emergency department visits for every bicycle-related death.1
The most common cause of bicycle-related mortality is from head injury.2, 3 Furthermore, head and traumatic brain injury occur in one third of all nonfatal bicycle-related crashes.4 A systematic review has shown that bicycle helmets reduce the risk for head and brain injury by 65% to 88%.5 Both educational and legislative initiatives have been effective in increasing short-term bicycle helmet use by children.6–11
Barriers to bicycle helmet use have been examined in several cross-sectional surveys. Results from these studies suggest that parents overestimate helmet use by their children and are also poor role models for helmet use.12–14 Factors that are associated with bicycle helmet use by children include parental rules, parental involvement, parental helmet use, parental helmet ownership, peer helmet use, and children's attitudes toward helmets.15–19 A significant limitation in all of these studies, however, is the use of self-reported measures of helmet use. Given that self-report is prone to social desirability bias, observational surveys have been considered the most valid method for measuring helmet use. The purpose of this observational study was to examine the influence of riding companions (child and adult) on helmet use by children.
METHODS
Data were selected from a larger study that was conducted in East York, a health district within metropolitan Toronto. The larger observational study used population-based survey methods and random sampling of multiple observation sites from low, middle, and high socioeconomic areas. Income areas were identified using census tract data. Observations were made at preselected sites (school yards, parks, residential streets, and major intersections), visited in random order and observed for a 1-hour period (morning, afternoon, or evening; weekend or weekday) during the months April through October for the years 1990–1997, 1999, and 2001. Observers were trained to use a standardized data collection form and to identify children within the 5- to 14-year age range. For eliminating the possibility of including children who were younger than 5 years, only children who were riding 2-wheeled bicycles were included. A pilot study showed that the data collected by observers were reliable and valid.20 Data that were collected on every child included gender, helmet use, site, area-level socioeconomic status, riding status, and the helmet status of riding companions. Detailed methods of the bicycle helmet study have been described elsewhere.9, 10 Ethical approval for the study was obtained from the Hospital for Sick Children Research Ethics Board, the East York Board of Education, and the Metropolitan Separate School Board.
For the purposes of this analysis, children's riding status was grouped into 3 mutually exclusive categories: riding alone, riding with at least 1 child companion, or riding with at least 1 adult companion. Simple random sampling (using a computer number generator) was used to select 1 child per category per site for each year of the study. Of note, some sites did not have children riding in all 3 categories. Therefore, unequal numbers of children across categories occurred when random sampling of a category was not possible.
For each category (alone, riding with 1 child, and riding with 1 adult), the probability that the randomly selected index child was wearing a bicycle helmet was calculated. In addition, for the categories that included riding with companions (child or adult), the probability that the index child was wearing a helmet was stratified by the helmet status of companions (no riding companion wearing a helmet and at least 1 companion wearing a helmet). Relative risks (RRs) and 95% confidence intervals (CIs) for categories were calculated using 2 reference groups: children riding alone and children riding with nonhelmeted child companion(s). RRs and 95% CIs were also adjusted for the potential confounding effect of provincial bicycle helmet legislation for children (prelegislation 1990–1995 vs postlegislation 1996–1997, 1999, and 2001). Adjusted RRs were calculated using the Mantel-Haenszel approach.21
RESULTS
Over the 9 year period, there were a total of 10935 observations. For the study period, the median number of sites was 111 (range: 88–128), and the median number of observations was 1070 (range: 550–1795). Random selection of 1 child per category per site for each year of the study resulted in 2094 independent observations. Of these, 50% (1048 of 2094) of the children were riding alone, 36% (754 of 2094) were riding with at least 1 child companion, and 14% (292 of 2094) were riding with at least 1 adult companion.
As shown in Table 1, the probability that the index child was wearing a helmet was 35% when riding alone, 45% when riding with child companions, and 67% when riding with adult companions. Furthermore, in rank order 95% (111 of 117) of children wore a helmet when riding with helmeted adults, 77% (209 of 270) wore a helmet when riding with helmeted children, 41% wore a helmet when riding with nonhelmeted adults, 35% wore a helmet when riding alone, and 10% wore a helmet when riding with nonhelmeted children.
Children who were riding with nonhelmeted child companions were 71% less likely to wear a helmet compared with children who were riding alone (RR: 0.29; 95% CI: 0.22–0.38). Children who were riding with helmeted child companions (RR: 2.18; 95% CI: 1.97–2.42) or helmeted adult companions (RR: 2.67; 95% CI: 2.44–2.93), however, were more than twice as likely to wear a helmet compared with children riding alone. There was no significant difference in helmet use between children who were riding with nonhelmeted adult companions and children who were riding alone (RR: 1.14; 95% CI: 0.94–1.39).
Children who were riding in groups with at least 1 child companion who was wearing a helmet were 7.49 times more likely to be wearing a helmet, compared with children who were riding with nonhelmeted child companions (95% CI: 5.72–9.82). Children who were riding in groups with at least 1 adult who was wearing a helmet were 9.18 times more likely to be wearing a helmet compared with children who were riding with nonhelmeted child companions (95% CI: 7.04–11.98). Even children who were riding with nonhelmeted adults were almost 4 times more likely to wear a helmet compared with children who were riding with nonhelmeted children (RR: 3.93; 95% CI: 2.86–5.40). RRs adjusted for the effect of helmet legislation showed similar associations between companion helmet status and helmet use by children (see Table 1).
DISCUSSION
This study showed that helmet use by riding companions is positively associated with helmet use by children. Helmeted adult companions showed a greater positive effect than helmeted child companions. Compared with riding alone, riding with nonhelmeted child companions had a negative effect on helmet use by children. Compared with children who were riding with nonhelmeted child companions, children who were riding alone, with helmeted companions (child or adult), or with nonhelmeted adults were more likely to wear a helmet.
Study strengths include direct observation of helmet use, prospective data collection, 9 years of data during pre- and postlegislation periods, and the use of a large and random sample. Cluster effects were minimized by randomly selecting 1 child per site per year for children who were riding alone, riding with 1 or more child companions, or riding with 1 or more adult companions. A limitation of the study was the inability to examine relationships between observed helmet use and other factors, such as child attitudes and family characteristics. For example, children who wear helmets might ride with friends who also wear helmets.
The findings from this prospective, observational study support results from cross-sectional surveys that found children's self-reported helmet use to be strongly associated with parental rules, parental encouragement, and helmet use by peers. In a survey conducted in 14 schools among children who were 8 to 16 years of age and living in Stockholm, 84% of children who reported using a helmet also reported that their parents told them to use a helmet.17 In a study that was conducted in Chicago-area pediatric practices using parental surveys and child interviews, children with a strict rule regarding helmet use were more likely to report helmet use most or all of the time compared with children with a partial rule or no rule (88% vs 19%; P < .001).15 When adjusting for age of the child and presence of siblings, the existence of a strict rule significantly increased the likelihood of helmet use (odds ratio [OR]: 46.5; 95% CI: 37.6–383.7). Moreover, in a school survey among 3082 children who were aged 10 to 16 years and living in the United Kingdom, self-reported helmet use was associated with parental encouragement (OR: 6.77; 95% CI: 4.09–11.14) and having their closest friend wear a helmet (OR: 2.80; 95% CI: 2.00–3.92).18
One study has made direct observations of bicycle helmet use by children and adults who were riding in a group.22 The helmet status of groups of riders (including at least 1 adult and 1 child per group) were recorded over 3 separate weekends using a convenience sample of bicyclists on a bicycle trail in southeast New England. The results for 134 groups showed a significant and positive correlation between no helmet use by adults and no helmet use by children (r = 0.397, P = .001). No association between adult and child helmet use or between proper helmet use by adults and children was noted. Although the study measured observed helmet use, the study collected data over a short time frame, used a nonrepresentative sample, and did not examine helmet use by children when riding alone or with other children. In addition, the authors did not take into account clustering in the analysis.
Results from this study and previous cross-sectional surveys suggest that there is a strong association between adult helmet use and child helmet use. In this study, 95% of children who were riding with a helmeted adult wore a helmet, whereas only 41% of children who were riding with nonhelmeted adults wore a helmet. Adults influence children's behavior not only through setting rules but also through role modeling. Public educational efforts need to encourage adults to wear a helmet when riding with children. Bicycle helmet legislation that excludes adults may diminish the opportunity for adults to role-model positively health behaviors for children. Future research is needed to formulate and evaluate effective intervention strategies that are aimed at increasing children's helmet use indirectly by influencing the behavior and the perception of helmet use in adults.
Child companions exert a positive influence on helmet use by other children only when they themselves wear a helmet, whereas nonhelmeted children deter other children from wearing a helmet. Children were more likely to wear a helmet when riding alone compared with children who were riding with nonhelmeted child companion(s). Because children commonly ride with and are greatly influenced by other children, continued efforts need to be made toward improving children's attitudes and perceptions about helmets. In addition, future research to examine adult and peer influence on children's helmet use by age group may provide important information for targeting intervention strategies. This study included children from a broad age range (5–14 years). Given the observational nature of the study, it was not possible to ascertain the exact age of children or conduct stratified analyses.
ACKNOWLEDGMENTS
This study was supported by grants from the Ontario Ministry of Transportation (grant 9210) and the Ontario Ministry of Culture, Tourism and Recreation (grant 92-70-0408). The Pediatric Outcomes Research Team is supported by a grant from the Hospital for Sick Children Foundation.
FOOTNOTES
Accepted May 2, 2005.
No conflict of interest declared.
REFERENCES
Powell EC, Tanz RR. Cycling injuries treated in emergency departments: need for bicycle helmets among preschoolers. Arch Pediatr Adolesc Med. 2000;154 :1096 –1100
Spence LJ, Dykes EH, Bohn DJ, Wesson DE. Fatal bicycle accidents in children: a plea for prevention. J Pediatr Surg. 1993;28 :214 –216
Sosin D, Sacks J, Webb K. Pediatric head injuries and deaths from bicycling in the United States. Pediatrics. 1996;98 :868 –870
Sacks JJ, Holmgreen P, Smith SM, Sosin DM. Bicycle-associated head injuries and deaths in the United States from 1984 through 1988. How many are preventable JAMA. 1991;266 :3016 –3018
Thompson DC, Rivara FP, Thompson R. Helmets for preventing head and facial injuries in bicyclists. Cochrane Database Syst Rev. 2000;(2):CD001855
Cote TR, Sacks JJ, Lambert-Huber DA, et al. Bicycle helmet use among Maryland children: effect of legislation and education. Pediatrics. 1992;89 :1216 –1220
Cameron MH, Vulcan AP, Finch CF, Newstead SV. Mandatory bicycle helmet use following a decade of helmet promotion in Victoria, Australia—an evaluation. Accid Anal Prev. 1994;26 :325 –337
Leblanc JC, Beattie TL, Culligan C. Effect of legislation on the use of bicycle helmets. CMAJ. 2002;166 :592 –595
Liller KD, Nearns J, Cabrera M, Joly B, Noland V, McDermott R. Children's bicycle helmet use and injuries in Hillsborough County, Florida before and after helmet legislation. Inj Prev. 2003;9 :177 –179
Ehrlich PF, Helmkamp JC, Williams JM, Haque A, Furbee PM. Matched analysis of parent's and children's attitudes and practices towards motor vehicle and bicycle safety: an important information gap. Inj Control Saf Promot. 2004;11 :23 –28
Ortega HW, Shields BJ, Smith GA. Bicycle-related injuries to children and parental attitudes regarding bicycle safety. Clin Pediatr. 2004;43 :251 –259
Forjuoh SN, Schuchmann JA, Fiesinger T, Mason S. Parent–child concordance on reported barriers to helmet use by children. Med Sci Monit. 2003;9 :436 –441
Miller PA, Binns HJ, Christoffel KK. Children's bicycle helmet attitudes and use: association with parental rules. Arch Adolesc Pediatr Med. 1996;150 :1259 –1264
Berg P, Westerling R. Bicycle helmet use among schoolchildren—the influence of parental involvement and children's attitudes. Inj Prev. 2001;7 :218 –222
Cryer PC, Davidson LL, Rahman M, Ching V, Goodall JB. Rates of, and the factors affecting, cycle helmet use among secondary schoolchildren in East Sussex and Kent. Inj Prev. 1998;4 :106 –110
Hu X, Wesson DE, Parkin PC, Chipman ML, Spence LJ. Current bicycle helmet ownership, use and related factors among school-aged children in metropolitan Toronto. Can J Public Health. 1994;85 :121 –124
Parkin P, Morris B, Chipman M, Miller C, Hu X, Wesson D. Measurement of bicycle helmet use by direct observation: accuracy of a measurement methodology . Paediatr Perinat Epidemiol. 1991;5 :A22 –A23
Dawson, B, Trapp RG. Basic and Clinical Biostatistics. 4th ed. McGraw-Hill: New York, NY; 2000
Twomey JG, Bevis MC, McGibbon CA. Associations between adult and child bicycle helmet use. MCN Am J Matern Child Nurs. 2001;26 :272 –277(Amina Khambalia, MSc, Col)
Bloorview Research Institute and the University of Toronto, Toronto, Ontario, Canada
ABSTRACT
Objective. To examine whether the helmet status of riding companions (child or adult) influences helmet use by children (5–14 years).
Methods. This study is part of a larger prospective observational survey that was conducted in a defined urban region of Toronto, Canada (1990–1997, 1999, and 2001). Trained observers collected data on children who were riding bicycles in April through October each year. Observation sites included school yards, parks, major intersections, and residential streets.
Results. Of the 2094 children who were observed riding a bicycle, 50% were riding alone, 36% were riding with at least 1 child companion, and 14% were riding with at least 1 adult companion. Compared with riding alone, children were less likely to wear a helmet when riding with nonhelmeted child companions (relative risk [RR]: 0.29; 95% confidence interval [CI]: 0.22–0.38) and were more than twice as likely to wear a helmet when riding with either helmeted child (RR: 2.18; 95% CI: 1.97–2.42) or adult companions (RR: 2.67; 95% CI: 2.44–2.93). Compared with children who were riding with nonhelmeted child companions, children were more likely to wear a helmet when riding with nonhelmeted adults (RR: 3.93; 95% CI: 2.86–5.40), helmeted adults (RR: 9.18; 95% CI: 7.04–11.98), or helmeted children (RR: 7.49; 95% CI: 5.72–9.82). Adjusted analyses showed no difference between companion helmet status and helmet use by children before and after legislation.
Conclusions. Helmeted and nonhelmeted adult riding companions were positively associated with child helmet use. The association between child riding companions and child helmet use was positive for helmeted child riding companions and negative for nonhelmeted child riding companions. Efforts need to be made toward improving adult helmet use and children's perceptions and attitudes toward helmet use.
Key Words: bicycle helmet children legislation
Abbreviations: RR, relative risk CI, confidence interval OR, odds ratio
An estimated 7 per 1000 children who are 1 to 14 years of age have a bicycle-related injury each year in the United States.1 Although fatalities are uncommon (5/100000/y), there are an estimated 51 hospital admissions and 1437 emergency department visits for every bicycle-related death.1
The most common cause of bicycle-related mortality is from head injury.2, 3 Furthermore, head and traumatic brain injury occur in one third of all nonfatal bicycle-related crashes.4 A systematic review has shown that bicycle helmets reduce the risk for head and brain injury by 65% to 88%.5 Both educational and legislative initiatives have been effective in increasing short-term bicycle helmet use by children.6–11
Barriers to bicycle helmet use have been examined in several cross-sectional surveys. Results from these studies suggest that parents overestimate helmet use by their children and are also poor role models for helmet use.12–14 Factors that are associated with bicycle helmet use by children include parental rules, parental involvement, parental helmet use, parental helmet ownership, peer helmet use, and children's attitudes toward helmets.15–19 A significant limitation in all of these studies, however, is the use of self-reported measures of helmet use. Given that self-report is prone to social desirability bias, observational surveys have been considered the most valid method for measuring helmet use. The purpose of this observational study was to examine the influence of riding companions (child and adult) on helmet use by children.
METHODS
Data were selected from a larger study that was conducted in East York, a health district within metropolitan Toronto. The larger observational study used population-based survey methods and random sampling of multiple observation sites from low, middle, and high socioeconomic areas. Income areas were identified using census tract data. Observations were made at preselected sites (school yards, parks, residential streets, and major intersections), visited in random order and observed for a 1-hour period (morning, afternoon, or evening; weekend or weekday) during the months April through October for the years 1990–1997, 1999, and 2001. Observers were trained to use a standardized data collection form and to identify children within the 5- to 14-year age range. For eliminating the possibility of including children who were younger than 5 years, only children who were riding 2-wheeled bicycles were included. A pilot study showed that the data collected by observers were reliable and valid.20 Data that were collected on every child included gender, helmet use, site, area-level socioeconomic status, riding status, and the helmet status of riding companions. Detailed methods of the bicycle helmet study have been described elsewhere.9, 10 Ethical approval for the study was obtained from the Hospital for Sick Children Research Ethics Board, the East York Board of Education, and the Metropolitan Separate School Board.
For the purposes of this analysis, children's riding status was grouped into 3 mutually exclusive categories: riding alone, riding with at least 1 child companion, or riding with at least 1 adult companion. Simple random sampling (using a computer number generator) was used to select 1 child per category per site for each year of the study. Of note, some sites did not have children riding in all 3 categories. Therefore, unequal numbers of children across categories occurred when random sampling of a category was not possible.
For each category (alone, riding with 1 child, and riding with 1 adult), the probability that the randomly selected index child was wearing a bicycle helmet was calculated. In addition, for the categories that included riding with companions (child or adult), the probability that the index child was wearing a helmet was stratified by the helmet status of companions (no riding companion wearing a helmet and at least 1 companion wearing a helmet). Relative risks (RRs) and 95% confidence intervals (CIs) for categories were calculated using 2 reference groups: children riding alone and children riding with nonhelmeted child companion(s). RRs and 95% CIs were also adjusted for the potential confounding effect of provincial bicycle helmet legislation for children (prelegislation 1990–1995 vs postlegislation 1996–1997, 1999, and 2001). Adjusted RRs were calculated using the Mantel-Haenszel approach.21
RESULTS
Over the 9 year period, there were a total of 10935 observations. For the study period, the median number of sites was 111 (range: 88–128), and the median number of observations was 1070 (range: 550–1795). Random selection of 1 child per category per site for each year of the study resulted in 2094 independent observations. Of these, 50% (1048 of 2094) of the children were riding alone, 36% (754 of 2094) were riding with at least 1 child companion, and 14% (292 of 2094) were riding with at least 1 adult companion.
As shown in Table 1, the probability that the index child was wearing a helmet was 35% when riding alone, 45% when riding with child companions, and 67% when riding with adult companions. Furthermore, in rank order 95% (111 of 117) of children wore a helmet when riding with helmeted adults, 77% (209 of 270) wore a helmet when riding with helmeted children, 41% wore a helmet when riding with nonhelmeted adults, 35% wore a helmet when riding alone, and 10% wore a helmet when riding with nonhelmeted children.
Children who were riding with nonhelmeted child companions were 71% less likely to wear a helmet compared with children who were riding alone (RR: 0.29; 95% CI: 0.22–0.38). Children who were riding with helmeted child companions (RR: 2.18; 95% CI: 1.97–2.42) or helmeted adult companions (RR: 2.67; 95% CI: 2.44–2.93), however, were more than twice as likely to wear a helmet compared with children riding alone. There was no significant difference in helmet use between children who were riding with nonhelmeted adult companions and children who were riding alone (RR: 1.14; 95% CI: 0.94–1.39).
Children who were riding in groups with at least 1 child companion who was wearing a helmet were 7.49 times more likely to be wearing a helmet, compared with children who were riding with nonhelmeted child companions (95% CI: 5.72–9.82). Children who were riding in groups with at least 1 adult who was wearing a helmet were 9.18 times more likely to be wearing a helmet compared with children who were riding with nonhelmeted child companions (95% CI: 7.04–11.98). Even children who were riding with nonhelmeted adults were almost 4 times more likely to wear a helmet compared with children who were riding with nonhelmeted children (RR: 3.93; 95% CI: 2.86–5.40). RRs adjusted for the effect of helmet legislation showed similar associations between companion helmet status and helmet use by children (see Table 1).
DISCUSSION
This study showed that helmet use by riding companions is positively associated with helmet use by children. Helmeted adult companions showed a greater positive effect than helmeted child companions. Compared with riding alone, riding with nonhelmeted child companions had a negative effect on helmet use by children. Compared with children who were riding with nonhelmeted child companions, children who were riding alone, with helmeted companions (child or adult), or with nonhelmeted adults were more likely to wear a helmet.
Study strengths include direct observation of helmet use, prospective data collection, 9 years of data during pre- and postlegislation periods, and the use of a large and random sample. Cluster effects were minimized by randomly selecting 1 child per site per year for children who were riding alone, riding with 1 or more child companions, or riding with 1 or more adult companions. A limitation of the study was the inability to examine relationships between observed helmet use and other factors, such as child attitudes and family characteristics. For example, children who wear helmets might ride with friends who also wear helmets.
The findings from this prospective, observational study support results from cross-sectional surveys that found children's self-reported helmet use to be strongly associated with parental rules, parental encouragement, and helmet use by peers. In a survey conducted in 14 schools among children who were 8 to 16 years of age and living in Stockholm, 84% of children who reported using a helmet also reported that their parents told them to use a helmet.17 In a study that was conducted in Chicago-area pediatric practices using parental surveys and child interviews, children with a strict rule regarding helmet use were more likely to report helmet use most or all of the time compared with children with a partial rule or no rule (88% vs 19%; P < .001).15 When adjusting for age of the child and presence of siblings, the existence of a strict rule significantly increased the likelihood of helmet use (odds ratio [OR]: 46.5; 95% CI: 37.6–383.7). Moreover, in a school survey among 3082 children who were aged 10 to 16 years and living in the United Kingdom, self-reported helmet use was associated with parental encouragement (OR: 6.77; 95% CI: 4.09–11.14) and having their closest friend wear a helmet (OR: 2.80; 95% CI: 2.00–3.92).18
One study has made direct observations of bicycle helmet use by children and adults who were riding in a group.22 The helmet status of groups of riders (including at least 1 adult and 1 child per group) were recorded over 3 separate weekends using a convenience sample of bicyclists on a bicycle trail in southeast New England. The results for 134 groups showed a significant and positive correlation between no helmet use by adults and no helmet use by children (r = 0.397, P = .001). No association between adult and child helmet use or between proper helmet use by adults and children was noted. Although the study measured observed helmet use, the study collected data over a short time frame, used a nonrepresentative sample, and did not examine helmet use by children when riding alone or with other children. In addition, the authors did not take into account clustering in the analysis.
Results from this study and previous cross-sectional surveys suggest that there is a strong association between adult helmet use and child helmet use. In this study, 95% of children who were riding with a helmeted adult wore a helmet, whereas only 41% of children who were riding with nonhelmeted adults wore a helmet. Adults influence children's behavior not only through setting rules but also through role modeling. Public educational efforts need to encourage adults to wear a helmet when riding with children. Bicycle helmet legislation that excludes adults may diminish the opportunity for adults to role-model positively health behaviors for children. Future research is needed to formulate and evaluate effective intervention strategies that are aimed at increasing children's helmet use indirectly by influencing the behavior and the perception of helmet use in adults.
Child companions exert a positive influence on helmet use by other children only when they themselves wear a helmet, whereas nonhelmeted children deter other children from wearing a helmet. Children were more likely to wear a helmet when riding alone compared with children who were riding with nonhelmeted child companion(s). Because children commonly ride with and are greatly influenced by other children, continued efforts need to be made toward improving children's attitudes and perceptions about helmets. In addition, future research to examine adult and peer influence on children's helmet use by age group may provide important information for targeting intervention strategies. This study included children from a broad age range (5–14 years). Given the observational nature of the study, it was not possible to ascertain the exact age of children or conduct stratified analyses.
ACKNOWLEDGMENTS
This study was supported by grants from the Ontario Ministry of Transportation (grant 9210) and the Ontario Ministry of Culture, Tourism and Recreation (grant 92-70-0408). The Pediatric Outcomes Research Team is supported by a grant from the Hospital for Sick Children Foundation.
FOOTNOTES
Accepted May 2, 2005.
No conflict of interest declared.
REFERENCES
Powell EC, Tanz RR. Cycling injuries treated in emergency departments: need for bicycle helmets among preschoolers. Arch Pediatr Adolesc Med. 2000;154 :1096 –1100
Spence LJ, Dykes EH, Bohn DJ, Wesson DE. Fatal bicycle accidents in children: a plea for prevention. J Pediatr Surg. 1993;28 :214 –216
Sosin D, Sacks J, Webb K. Pediatric head injuries and deaths from bicycling in the United States. Pediatrics. 1996;98 :868 –870
Sacks JJ, Holmgreen P, Smith SM, Sosin DM. Bicycle-associated head injuries and deaths in the United States from 1984 through 1988. How many are preventable JAMA. 1991;266 :3016 –3018
Thompson DC, Rivara FP, Thompson R. Helmets for preventing head and facial injuries in bicyclists. Cochrane Database Syst Rev. 2000;(2):CD001855
Cote TR, Sacks JJ, Lambert-Huber DA, et al. Bicycle helmet use among Maryland children: effect of legislation and education. Pediatrics. 1992;89 :1216 –1220
Cameron MH, Vulcan AP, Finch CF, Newstead SV. Mandatory bicycle helmet use following a decade of helmet promotion in Victoria, Australia—an evaluation. Accid Anal Prev. 1994;26 :325 –337
Leblanc JC, Beattie TL, Culligan C. Effect of legislation on the use of bicycle helmets. CMAJ. 2002;166 :592 –595
Liller KD, Nearns J, Cabrera M, Joly B, Noland V, McDermott R. Children's bicycle helmet use and injuries in Hillsborough County, Florida before and after helmet legislation. Inj Prev. 2003;9 :177 –179
Ehrlich PF, Helmkamp JC, Williams JM, Haque A, Furbee PM. Matched analysis of parent's and children's attitudes and practices towards motor vehicle and bicycle safety: an important information gap. Inj Control Saf Promot. 2004;11 :23 –28
Ortega HW, Shields BJ, Smith GA. Bicycle-related injuries to children and parental attitudes regarding bicycle safety. Clin Pediatr. 2004;43 :251 –259
Forjuoh SN, Schuchmann JA, Fiesinger T, Mason S. Parent–child concordance on reported barriers to helmet use by children. Med Sci Monit. 2003;9 :436 –441
Miller PA, Binns HJ, Christoffel KK. Children's bicycle helmet attitudes and use: association with parental rules. Arch Adolesc Pediatr Med. 1996;150 :1259 –1264
Berg P, Westerling R. Bicycle helmet use among schoolchildren—the influence of parental involvement and children's attitudes. Inj Prev. 2001;7 :218 –222
Cryer PC, Davidson LL, Rahman M, Ching V, Goodall JB. Rates of, and the factors affecting, cycle helmet use among secondary schoolchildren in East Sussex and Kent. Inj Prev. 1998;4 :106 –110
Hu X, Wesson DE, Parkin PC, Chipman ML, Spence LJ. Current bicycle helmet ownership, use and related factors among school-aged children in metropolitan Toronto. Can J Public Health. 1994;85 :121 –124
Parkin P, Morris B, Chipman M, Miller C, Hu X, Wesson D. Measurement of bicycle helmet use by direct observation: accuracy of a measurement methodology . Paediatr Perinat Epidemiol. 1991;5 :A22 –A23
Dawson, B, Trapp RG. Basic and Clinical Biostatistics. 4th ed. McGraw-Hill: New York, NY; 2000
Twomey JG, Bevis MC, McGibbon CA. Associations between adult and child bicycle helmet use. MCN Am J Matern Child Nurs. 2001;26 :272 –277(Amina Khambalia, MSc, Col)