Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
July 2002

Helmet UseA Survey of 4 Common Childhood Leisure Activities

Author Affiliations

From the Departments of Family and Community Medicine (Drs Forjuoh and Fiesinger and Ms Mason) and Physical Medicine and Rehabilitation (Dr Schuchmann), Scott and White Memorial Hospital, Scott, Sherwood and Brindley Foundation, Texas A & M University System Health Science Center, College of Medicine, Temple.


Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002

Arch Pediatr Adolesc Med. 2002;156(7):656-661. doi:10.1001/archpedi.156.7.656

Objective  To compare the distribution and observation of helmet use patterns among 4 common childhood leisure activities—bicycle riding, in-line skating, skateboarding, and scooter riding—in a setting with no helmet use legislation.

Design  A cross-sectional, unobtrusive, direct observation of children engaged in 1 of the 4 activities was conducted during an 8-week period.

Setting  Eight communities of at least 1000 persons each in a 2-county Central Texas area.

Main Outcome Measures  Prevalence of activity, helmet use, and correct helmet use.

Results  Of 841 children observed, most rode bicycles (74.8%); only 4.2% rode skateboards. They included 127 children estimated to be younger than 6 years (15.1%) and 495 estimated to be aged between 6 and 12 years (58.9%). Most were male (67.3%) and observed in urban communities (61.7%). The locations of highest concentration of the 4 activities were minor streets for bicycle riding, playgrounds for in-line skating, and sidewalks for skateboarding and scooter riding. The rate of overall helmet use was 13.6% and did not vary significantly by activity: in-line skating, 18.2%; skateboarding, 14.3%; bicycle riding, 13.5%; and scooter riding, 11.5%. However, the observed correct helmet use prevalence of 72.6% differed significantly by activity. Most children riding scooters wore their helmets incorrectly. Helmet use was positively associated with helmet use by accompanying children and/or adults.

Conclusions  While bicycle riding was the most popular activity observed, helmet use was most prevalent among in-line skaters and least prevalent among scooter riders, many of whom also wore them incorrectly. Observed helmet use was influenced mostly by adult and peer helmet use, indicating that public education and interventions should target children and their parents as well as scooter riders.

SAFETY PROMOTION to children has always been a matter of concern to physicians, public health practitioners, policy makers, and parents. Safety issues abound wherever children are, be it at play, in school, on the street, or at home. While playing represents one of the natural daily activities of children, such leisure activity is not always without risk of injury. Many of the consumer products used by children at play may be associated with some injury risk. Recently, scooters have been touted as a new cause of pediatric injuries in the United States,14 although bicycling probably remains one of the most common childhood activities worldwide. In the United States, in-line skating and skateboarding are other common childhood leisure activities.5,6

All 4 activities—bicycle riding, in-line skating, skateboarding, and scooter riding—are associated with some potential for injury resulting from a fall or collision with a moving or stationary object. Bicycle collisions result in more than 500 000 emergency department visits and more than 700 deaths in the United States annually.7 More than half of bicycle-related injuries occur in persons younger than 15 years.8 In 1997 alone, there were an estimated 153 000 in-line skating–related injuries in the United States among an estimated 37 million participants.7 Similar injury statistics have been reported among those who ride skateboards and scooters.35,7

Recently, some press releases have suggested that helmets may give a false sense of security. However, we believe that helmets offer the only proven form of physical protection from head and traumatic brain injuries for many sporting and leisure activities associated with a propensity for falls and collisions. Indeed, helmets represent a common intervention recommended for the prevention of head and traumatic brain injuries potentially associated with these 4 common childhood activities.6,913 For example, one of the objectives of the Healthy People 2010 national initiative is to increase helmet use among all bicyclists.13 In addition, many health professional bodies, such as the American Academy of Pediatrics, strongly recommend that children always use approved helmets for bicycling, skating, and scooter riding. Despite the general consensus on the recommendation of helmets for all 4 activities, tracking of helmet use has focused on bicyclists and in-line skaters.6,11,12,14 In particular, only a few studies have reported helmet use among scooter riders.3,4 Besides, to our knowledge, no study has compared distribution or helmet use among these 4 childhood leisure activities in the same setting. This information may be needed for physician anticipatory guidance and education of children and their parents.

In this study, we sought to compare the distribution and helmet use patterns among these 4 activities, for which helmets are recommended, in a 2-county Central Texas area. We designed our study to answer the following questions: (1) What is the distribution of these 4 leisure activities among children in the same setting? (2) Does helmet use differ among these 4 activities? (3) When helmets are used, are they used correctly? (4) Does helmet use for each activity depend on the location of activity?


We used a cross-sectional design to conduct an unobtrusive, direct observation of the 4 common childhood leisure activities in 8 communities in a 2-county area of Central Texas. Each of the 8 communities is made up of 1000 or more people. The survey was conducted during an 8-week period between May and July 2001. The study protocol was reviewed and approved to be conducted without informed consent by the institutional review board of the Scott and White Memorial Hospital (Temple, Tex).

The median family income for the Central Texas area is approximately $33 915. The area is predominantly rural. There is neither a statewide helmet law nor any municipal ordinance for helmet use. Nonetheless, sporadic bicycle helmet promotion programs led by area police and fire departments and other organizations have taken place in the area. Examples of these programs include bicycle rodeos, helmet giveaways, and safety presentations at schools and church organizations.


Trained observers were distributed throughout the 8 communities to observe school-aged children's use of helmets while riding a bicycle, in-line skating, skateboarding, or riding a scooter. These 8 communities were selected randomly from among the 25 communities in the area. Observers were instructed not to stop any child while observing so as not to intrude on personal privacy. The observers were mostly undergraduate students attending area colleges, who were provided a day's training on clues to proper identification of subjects and data recording, including how to identify correct use of helmets and to differentiate between a boy and a girl using hair style, bicycle colors, riding behaviors, and the presence or absence of crossbars. Observers (n = 19) were assigned to preselected observation sites (n = 70) throughout the 8 communities to maximize the coverage area. These sites represented areas reported by key community informants and law enforcement personnel to be those used most frequently by children for the activities being studied.

Observers recorded information on activity type (bicycle riding, in-line skating, skateboarding, or scooter riding); location of activity; age group; sex; whether a helmet was used and if yes, whether it was used correctly or incorrectly; whether a child was riding alone, with other children, or with both children and adults; and whether at least one of the companions, all of them, or none of them wore helmets. Location of activity included a minor street, which was defined as a dead-end or cul-de-sac street, an alley, or a 1-way street. The community of observation was noted as rural (<5000 people), suburban (5000–50 000 people), or urban (>50 000 people). If there were 2 or more children in a group, each child was observed independently of the others. Information was also collected about day of the week, illumination, and the day's temperature. We used an instrument developed by one of us (S.F.). This instrument had been used previously in Pittsburgh and had been adapted by the Pennsylvania Department of Health for statewide use. A copy of the survey instrument is available on request.

Observers were instructed to estimate the age of each child as approximately younger than 6 years; between 6 and 12 years; or older than 12 years. Observers were further instructed to mark "not sure" if they could not determine a child's age or sex. A helmet was considered to be used correctly if it was positioned on the head in a straight, horizontal line and strapped visibly below the chin. A helmet was considered to be used incorrectly if it was tilted forward or backward or not strapped under the chin. Observations were conducted only once at each site for approximately 30 to 60 minutes to reduce the likelihood of counting riders multiple times.


We computed activity prevalence, helmet use prevalence, and correct helmet use prevalence overall, as well as the children's demographic characteristics and other variables. We assessed group differences for statistical significance using the χ2 test or the Fisher exact test. Statistical significance was set at P<.05. We then estimated the magnitude of observed associations using relative risks (RRs) and 95% confidence intervals (CIs) while controlling for confounding variables, such as age, through multivariate logistic regression analysis. Because of small numbers and unstable estimates, multivariate analyses were not performed for helmet use for in-line skating and skateboarding or for correct helmet use.


During the 8-week study, 841 school-aged children were observed engaging in 1 of the 4 activities. They included 127 children (15.1%) estimated to be younger than 6 years and 495 (58.9%) estimated to be between 6 and 12 years old. Most subjects were male (67.3%) and observed in urban communities (61.7%). Twenty-five percent were observed in suburban communities and 15% in rural communities.


Most children (74.8%) were observed riding a bicycle, 14.5% were riding a scooter, and only 4.2% and 6.5% were skateboarding and in-line skating, respectively (Table 1). Engagement in individual activity did not vary by weather conditions or whether a child was riding alone or with company, but differed significantly by a child's age group, community, and location of activity. Most children who were riding a bicycle or riding a scooter were estimated to be between 6 and 12 years old. Skateboarding was least popular among children estimated to be younger than 6 years; 51% of children engaged in this activity were estimated to be older than 12 years. For all activities, boys were represented more than girls. All activities were most prevalent in urban communities. While skateboarding and scooter riding were mostly predominant on sidewalks, bicycle riding was concentrated on minor streets and in-line skating on playgrounds. Most children riding bicycles were in the company of other children, while most children skateboarding were riding alone (Table 1).

Table 1. 
Image not available
Characteristics of Children Engaged in 4 Leisure Activities for Which Helmets Are Recommended*

The overall helmet use among the children for all 4 activities was 13.6%. Additionally, a helmet was found hanging on the bicycle handlebars of 9 children or carried by them (1.1%). These children were considered to not be using helmets and subsequently analyzed as such. Helmet use did not vary significantly by activity (P = .69) but was most prevalent among in-line skaters (18.2%) and least prevalent among scooter riders (11.5%) (Figure 1).

Image not available

Helmet use prevalence and correct helmet-wearing prevalence by activity. The asterisk indicates that observations were based on those wearing helmets (n = 114). Differences were significant.

Significant differences were observed for overall helmet use by age group, sex, location of activity, and companions' helmet use. Helmet use was most prevalent among children approximately younger than 6 years (33.1%) and least prevalent among those approximately 6 to 12 years old (9.5%). Girls wore helmets at a higher rate than boys (17.0% vs 11.8%). Helmet use was most prevalent at bicycle paths (29.4%), playgrounds (19.4%), and parking lots (17.5%), and least prevalent on minor streets (9.1%). Children observed to be with adults were 6 times as likely to use helmets as those observed to be alone.

Table 2 presents data on children's helmet use by demographic and selected characteristics and activity. Helmet use for bicycle riding was most prevalent among girls and children estimated to be younger than 6 years. Children riding a bicycle on bicycle paths, playgrounds, or parking lots had the highest helmet-use rates. Children riding a bicycle with adults were 12 times as likely to be wearing helmets compared with children riding a bicycle alone (RR, 12.44; 95% CI, 5.80-26.70). Helmet use for in-line skating, on the other hand, was most prevalent among children estimated to be older than 12 years and those observed on major streets. Additionally, all of the children in-line skating with adults wore helmets. While no child was observed to be skateboarding with an adult, there was a 100% helmet use prevalence among children younger than 6 years riding skateboards. Helmet use prevalence by demographic characteristics for scooter riding mirrored those of bicycle riding except that no child was observed riding a scooter on bicycle paths.

Table 2. 
Image not available
Relative Risk (95% Confidence Intervals) of Helmet Use by Children for Selected Characteristics and Activity*

After multivariate adjustment, riding with adults (adjusted RR, 10.01; 95% CI, 4.28-23.38) was the variable with the strongest association with helmet use among bicycle riders, while riding on a minor street (adjusted RR, 0.37; 95% CI, 0.17-0.82) and older age were variables significantly associated with not using helmets among bicycle riders. Riding with adults (adjusted RR, 11.95; 95% CI, 1.37-104.45) was the only variable that had a significant association with helmet use among scooter riders (Table 3).

Table 3. 
Image not available
Adjusted Relative Risks (95% Confidence Intervals) of Helmet Use by Children for Selected Characteristics and Activity*

Of those wearing helmets, 72.6% were observed to wear them correctly, 26.5% incorrectly, and in one child, the observer was unsure. While observed helmet use did not vary significantly by activity, the observed manner of wearing helmets differed significantly by activity (P = .04). Most children riding scooters (53.3%) wore their helmets incorrectly. The manner of helmet wearing among the other 3 activities was similar, with approximately one fifth of the children wearing them incorrectly (Figure 1). For each activity, however, the manner of helmet wearing did not vary significantly by demographic characteristic.


Helmet use was generally affected by whether a child's companions were also using one. All children riding a bicycle or a scooter wore helmets when all other accompanying persons wore them. For children observed in-line skating, helmet use was the same whether one companion wore a helmet or all of them wore helmets. For children skateboarding, helmet use did not seem to be influenced by a companion's helmet use.

Adult helmet use was also significantly associated with children's helmet use. All children in-line skating with adults who were using helmets, assumed to be mostly the children's parents, used helmets compared with 16% of children skating alone. Also, 53% of children riding a bicycle with adults who were using helmets used helmets compared with 7.8% riding alone.


Despite the documented rising popularity of the unpowered scooter among children, bicycle riding remains the most popular of the 4 childhood leisure activities studied in our 2-county area in Central Texas. Nearly 3 in 4 children observed rode a bicycle compared with a scooter, in-line skates, or a skateboard. A recent US Consumer Product Safety Commission publication reported a 700% rise in popularity of the unpowered scooter in the last few years, with its associated injuries.2 Similar sentiments have been confirmed by the Centers for Disease Control and Prevention (Atlanta, Ga).1,15 Helmet use rates were generally low overall in our area and did not vary significantly by the 4 activities studied. However, helmet use was most prevalent among in-line skaters and least prevalent among scooter riders, many of whom also wore their helmets incorrectly. Perhaps the most interesting finding was the huge association of adult and peer helmet use with children's helmet use.

To our knowledge, this is the first study to compare child engagement and helmet use patterns among these 4 activities, for which helmets are recommended, in the same setting. This is also the first study, to our knowledge, to accurately determine the observed helmet use prevalence in Central Texas using a large sample of subjects. We were not surprised to find that bicycle riding was the most prevalent activity of the 4 activities studied. There are obviously more bicycles in circulation, which may be passed on to younger siblings, than scooters. The sex distribution for these activities was, however, similar to that reported for other settings.4,6

The low helmet use prevalence of 14% observed in child bicycle riders in our area is similar to that reported in settings without any bicycle helmet legislation or any interventions designed to increase helmet use.16,17 For in-line skaters, the helmet use prevalence of 18.2% found in our study is higher than that found in other settings.12,18 In a similar unobtrusive, observational survey of 1548 in-line skaters, only 2.6% of the skaters were observed to be wearing a helmet.18 In another study, Jacques and Grzesiak12 found that 9% of 89 in-line skaters observed in Oakland County, Michigan, wore helmets. Schuster and Israeli9 reported that many of the skaters surveyed in New York City (NY) did not wear helmets or knee pads even though they apparently owned this equipment. The higher rate of helmet use among the skaters in our area may be related to increasing awareness. While we could not find any observational survey of helmet use among scooter riders, only 2 (13.3%) of the 15 injured scooter riders in the pediatric emergency department in a municipal hospital were documented to have a helmet.3

The finding of varying helmet use by location for the same activity merits some discussion. On one hand, the higher rate of helmet use prevalence in our setting among bicycle riders on bicycle paths, playgrounds, and parking lots may be due to the presence of companions, including adults who may have been parents of these children. Children riding a bicycle at these locations were more likely to be in the company of others than were those observed at other locations. On the other hand, the lower helmet use prevalence observed among bicycle riders on minor streets and sidewalks may reflect a false sense of security associated with riding in these locations in less populated areas with less traffic. That skaters were more likely to wear helmets on major streets than in other locations was no surprise.

The significant influence on child helmet use by friends and adults found in our study has been demonstrated by prior studies.1921 Our study has only quantified this observed association. Although it is acknowledged that not all parents may ride or engage in these activities with their children, it is clear that their practices and attitudes may influence their children's actions.22

Findings of this study should be interpreted in the light of a few limitations. First, the study was limited in the accuracy of subject age owing to its unobtrusive design. Therefore, findings based on age may not be very accurate and not generalizable to the source population. Moreover, bicycle riding was often done in groups of 4 or more persons. This would make it difficult for observers to accurately record all of the required information in a timely manner. Nonetheless, the percentage of missing data for each variable was very small. For example, observers recorded "not sure" for sex in only 4 (0.5%) of the cases observed in a group. Second, the observation of helmet wearing could be skewed since it was based on observer opinion. Third, the study used spot locations in the 2 counties. Finally, we did not collect any socioeconomic data. Despite these limitations, our findings provide some accurate information on helmet use that to our knowledge was nonexistent in this area. This study also raises questions that merit further investigation, such as why helmet use varies by age group and why different activities show different age group relationships with helmet use.

In conclusion, bicycle riding remains the most prevalent among the 4 common childhood leisure activities studied in the 2-county area of Central Texas despite the reported resurgence in popularity of the scooter. The proportion of children wearing helmets is still low, especially for scooter riders. Therefore, tracking of helmet use should not only focus on bicycle riding and in-line skating, but also on other activities for which helmets are recommended, such as scooter riding and skateboarding. This study also shows that when parents or adults take the safety lead by wearing helmets themselves, child compliance may improve. It is not enough to target children alone in the promotion of helmet use. Adults need to be encouraged and educated about the importance of helmet use. Even if parents do not ride with their children, merely insisting that their children wear helmets may lead to child compliance.22

Back to top
Article Information

Accepted for publication March 29, 2002.

This work was funded by a grant from the Texas Department of Transportation (Austin), which disclaims responsibility for any interpretations or conclusions.

What This Study Adds

The distribution and helmet use patterns of the 4 common childhood leisure activities—bicycle riding, in-line skating, skateboarding, and scooter riding—in the same setting has not been determined. In addition, helmet use tracking has focused on bicyclists and in-line skaters. This information may be needed for physician anticipatory guidance and education of children and their parents. Using an unobtrusive, direct observational survey of school-aged children, this study determined that bicycle riding remains the most popular among the 4 activities despite the documented rising popularity of the unpowered scooter. Although rate of helmet use did not differ significantly by activity, it was lowest among scooter riders, many of whom also wore helmets incorrectly. Helmet use tracking and promotional efforts should include scooter riders.

Corresponding author and reprints: Samuel N. Forjuoh, MB, ChB, DrPH, Department of Family Medicine, Scott and White Santa Fe Center, 1402 W Ave H, Temple, TX 76504 (e-mail:

Centers for Disease Control and Prevention, Unpowered scooter-related injuries--United States, 1998-2000. MMWR Morb Mortal Wkly Rep. 2000;491108- 1110
US Consumer Product Safety Commission, Scooter data. Available at March 14, 2002
Levine  DAPlatt  SLFoltin  GL Scooter injuries in children. Pediatrics. 2001;107e64Available athttp://www.pediatrics.orgAccess verified April 11, 2002Article
Abbott  MBHoffinger  SANguyen  DMWeintraub  DL Scooter injuries: a new pediatric morbidity. Pediatrics. 2001;108e2Available athttp://www.pediatrics.orgAccess verified April 11, 2002Article
Osberg  JSSchneps  SEDi Scala  CLi  G Skateboarding: more dangerous than roller skating or in-line skating. Arch Pediatr Adolesc Med. 1998;152985- 991Article
Frankovich  RJPetrella  RJLattanzio  CN In-line skating injuries: patterns and protective equipment use. Physician Sports Med. 2001;2957- 62
National Safety Council, Injury Facts.  Itasca, Ill National Safety Council1999;
Frank  EFrankel  PMullins  RJTaylor  N Injuries resulting from bicycle collisions. Acad Emerg Med. 1995;2200- 203Article
Schuster  MIsraeli  A Survey of injuries and protective gear worn by in-line skaters in public parks. Am J Phys Med Rehabil. 1999;787- 10Article
Attewell  RGGlase  KMcFadden  M Bicycle helmet efficacy: a meta-analysis. Accid Anal Prev. 2001;33345- 352Article
Cook  ASheikh  A Cycle injury trends: helmets are most likely explanation [letter]. BMJ. 2001;3221427- 1428Article
Jacques  LBGrzesiak  E Personal protective equipment use by in-line roller skaters. J Fam Pract. 1994;38486- 488
US Department of health and Human Services, Healthy People 2010. Conference edVols 1 and 2 Washington, DC US Dept of Health and Human Services2000;
Beirness  DJFoss  RDDesmond  KJ Use of protective equipment by in-line skaters: an observational study. Inj Prev. 2001;751- 55Article
Centers for Disease Control and Prevention, Unpowered scooter-related injuries in the United States, 1998-2000. JAMA. 2001;28536- 37
Bergman  ABRivara  FPRichards  DDRogers  LW The Seattle children's bicycle helmet campaign. AJDC. 1990;144727- 731
Farley  CHaddad  SBrown  B The effects of a 4-year program promoting bicycle helmet use among children in Quebec. Am J Public Health. 1996;8646- 51Article
Young  CCMark  DH In-line skating: an observational study of protective equipment used by skaters. Arch Fam Med. 1995;419- 23Article
Finnoff  JTLaskowski  ERAltman  KLDiehl  NN Barriers to bicycle helmet use. Pediatrics. 2001;108e4Available at verified April 11, 2002.Article
DiGuiseppi  CGRivara  FPKoepsell  TDPolissar  L Bicycle helmet use by children: evaluation of a community-wide helmet campaign. JAMA. 1989;2622256- 2261Article
DiGuiseppi  CGRivara  FPKoepsell  TD Attitudes toward bicycle helmet ownership and use by school-age children. AJDC. 1990;14483- 86
Miller  PABinns  HJChristoffel  KKPediatric Practice Research Group, Children's bicycle helmet attitudes and use: association with parental rules. Arch Pediatr Adolesc Med. 1996;1501259- 1264Article