Education Shared Resource

1. Bicycle Helmet Laws and Educational Campaigns: An Evaluation of Strategies to Increase Children's Helmet Use --------------------------- Andrew L. Dannenbeig MD,…
of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  • 1. Bicycle Helmet Laws and Educational Campaigns: An Evaluation of Strategies to Increase Children's Helmet Use --------------------------- Andrew L. Dannenbeig MD, MPH, Andrea C. Gielen, ScD, Peter L. Beilenson, MD, MPH, Modena H. Wdson, MD, MPH, and Alain Joffe, MD, MPH Inlodudion der way to promote bicycle helmet use; in a second adjacent county there were no Bicycling for both recreation and special initiatives related to helmets. transportation is a popular activity in the These three counties provide a unique op- United States. Bicycle-related injuries portunity to compare the effects of legis- are receiving increasing attention as a lation and education on the wearing of bi- public health issue; each year approxi- cycle helmets. mately 960 persons (55% of whom are The Johns Hopkins University Injury under age 20) die and 558 000 persons Prevention Center and the Maryland De- (79% of whom are under age 20) are partment of Health and Mental Hygiene treated in emergency departments as a collaborated on a multifaceted evaluation result of such injuries.' Head injury is the of the impact of the Howard County law. primary or contributing cause of death in Because of the low number of severe bi- 70% to 80% of all bicycle-related fatali- cycle-related head injuries expected each ties.2 Bicycle helmets have been shown year in the county, we focused on changes to be highly effective in reducing the risk in helmet use rates rather than injury rates. of bicycle-related head injuries,23 yet The first component of the evaluation, a only a small proportion of children wear mailed survey of school children in the helmets while riding.4-8 three counties, is described in this report. Education and legislation are among The second component, quot;beforequot; and the strategies available to persuade indi- quot;afterquot; roadside observations of helmet viduals to change their behavior to reduce use, is described elsewhere.'1 the risk of injury. Educational efforts to The mailed survey was designed to increase children's bicycle helmet use assess differences in helmet use rates over have been sporadic, have generally been time and across three conditions: legisla- conducted at school, and have seldom tion and education, education alone, and been evaluated. An intensive community- wide educational campaign in Seattle did Andrew L. Dannenberg, Andrea C. Gielen, not convince the majority of child bicy- and Modena H. Wilson are with the Injury clists to wear helmets.9'10 When education Prevention Center, and Peter L. Beilenson was with the General Preventive Medicine alone is insufficient, legislation mandating Residency Program, all at Johns Hopkins Uni- helmet use may be a more effective ap- versity School of Hygiene and Public Health, proach. Baltimore, Md. Modena H. Wilson and Alain In 1990, a law was passed in Howard Joffe are with the Department of Pediatrics, Johns Hopkins University School of Medi- County, Maryland, that requires persons cine. younger than 16 years of age riding bicy- Requests for reprints should be sent to cles on county roads and paths to wear an Andrew L. Dannenberg, MD, MPH, Johns approved safety helmet. As the first of its Hopkins Injury Prevention Center, 624 N kind in the United States, the law evoked Broadway, Room 545, Baltimore, MD 21205. This paper was accepted January 15, widespread interest from local and na- 1993. tional news media concerning the costs, Editor's Note. See related editorial by Ru- benefits, and social acceptability of such nyan (p 637), commentary by Dannenberg and legislation. In one adjacent county, a Vemick (p 644), and article by Ruch-Ross and O'Connor (p 728) in this issue. large-scale educational campaign was un- American Journal of Public Health 667
  • 2. Dannenbei et aL Statistical Methods sues at bicycle rodeos held at some but not no special initiatives. We examined a all middle schools; and (5) promotion of number of characteristics associated with All analyses, except those related to bicycle safety as part of local news media helmet use, so that children at high risk for bicycle ownership and use, excluded re- coverage of the new law. noncompliance with the law and for non- spondents who reported they did not ride a use of helmets in general can be identified bicycle. In addition, responses for specific Swvey Instrument and future efforts to promote helmet use questions that were missing, uncodeable, can be targeted appropriately. Study re- or answered quot;Don't know,quot; quot;No opin- After reviewing the literature, the in- sults have implications for promoting be- ion,quot; quot;Don't remember,quot; or quot;Doesn't ap- vestigators developed and pretested the havioral change in other areas of injury plyquot; were excluded from analyses except survey instrument, which focused on prevention. where indicated otherwise. knowledge, attitudes, and practices re- Univariate and bivariate data analy- lated to bicycles and helmets. Specific ses were done using the EPI INFO soft- questions included in this report focused Medhods ware package developed by the Centers on bicycle ownership and use, helmet for Disease Control. For most bivariate ownership and use, awareness of the hel- Setting results, chi-square analysis was used for met law, sources ofinformation about hel- The survey was conducted in three testing statistical significance. Consider- mets, the effect ofpeer pressure on helmet Maryland counties that are predominantly ing the large number of tests of signifi- use, and history of bicycle-related inju- cance performed in these analyses, results suburban and rural. The first adjacent ries. Questions on frequency of behavior county selected for comparison with for which P < .01 are considered statisti- were rated on a 4-point scale from quot;Al- Howard County was Montgomery cally significant and those for which waysquot; to quot;Never.quot; Questions soliciting County, which had extensive educational .05 < P < .01 are considered marginally attitudes were rated on a 5-point scale helmet promotion efforts in progress. In significant. Logistic regression analyses from quot;Strongly disagreequot; to quot;No opin- the second adjacent county, Baltimore were done using Statistical Analysis Sys- ionquot; to quot;Strongly agree.quot; The final instru- County (which excludes Baltimore City), tem software (SAS Institute, Cary, NC). ment included 43 questions and took 10 to there had been no substantial legislative or The study protocol was reviewed and ex- 15 minutes for children to complete. Cop- educational helmet promotion activities. empted by the Johns Hopkins School of ies of the survey instrument are available Based on US census data, the 1990 pop- Public Health Committee on Human Re- from the senior author. ulations ofBaltimore, Howard, and Mont- search because the survey contained no gomery counties were 692 000, 187 000, sensitive questions and was anonymous Sample and 757 OOOpersons, and the 1988 per cap- and voluntary. The survey was administered to chil- ita incomes were $20 418, $19 089, and dren in the fourth, seventh, and ninth $27 831, respectively. Results grades to span the age range from the low- The bicycle safety educational cam- est grade considered able to respond to a paign in Montgomery County in 1990 in- Response Rates mailed survey to the highest grade in cluded (1) a bicycle safety school poster Response rates were calculated using which almost all students would be cov- contest, to which more than 1000 children denominators that excluded surveys re- ered by the law in Howard County. After submitted posters; (2) distribution to ele- turned as undeliverable (58, 16, and 31 in stratification by geographic location mentary teachers and school nurses of Baltimore, Howard, and Montgomery within each of the three counties, a total of more than 2000 workbooks emphasizing counties, respectively). The overall re- 47 schools were randomly selected. In methods for teaching bicycle safety atvar- sponse rate was 48.4% (3494f7217). The each county, surveys were sent to the par- ious grade levels; (3) distribution through county-specific rates were 47.7% in Bal- ents of (1) all fourth-grade students in 9 schools, fairs, and other sites of approxi- timore County, 51.2% in Howard County, elementary schools (10 in Baltimore mately 10 000 coupons for discounts on and 46.5% in Montgomery County. For County), (2) all seventh-grade students in helmets; (4) promotion of bicycle safety, fourth-, seventh-, and ninth-grade stu- 3 middle schools (4 in Howard County), by means of videos, literature, and dis- dents, the response rates were, respec- and (3) all ninth-grade students in 3 high count coupons, at approximately 20 local tively, 48%, 53%, and 41% in Baltimore schools. The cover letter asked parents to fairs; and (5) promotion of bicycle safety County; 59%, 52%, and 43% in Howard have their child complete the survey with- on cable television and radio talk shows County; and 47%, 46%, and 47% in Mont- out help and to return it anonymously in and in local newspaper articles, as well as gomery County. Assuming student enroll- an enclosed preaddressed stamped enve- a proclamation on bicycle safety by the ment is evenly balanced by gender in each lope. Reminder postcards were sent 10 to county executive that received media at- county, response rates for boys and girls 20 days after the initial mailing. tention. were, respectively, 48% and 47% in Bal- Surveys were sent to 2712, 2278, and The bicycle safety educational cam- timore County, 50% and 52% in Howard 2332 students in Baltimore, Howard, and paign in Howard County in 1990 included County, and 44% and 49% in Montgomery Montgomery counties, respectively. Sur- (1) distnbution of a handbook on bicycle County. veys were mailed in May 1991 in Mont- safety to all elementary and middle school gomery County, in June 1991 in Howard Bicycle and Helmet Ownership students; (2) increased emphasis on bicy- County (8 months after the law went into cle safety as part of the health education and Use effect), and in October 1991 in Baltimore classes in all elementary schools; (3) in- Bicycle ownership was high among County. Mailing dates were determined corporation of a bicycle safety curriculum respondents in all counties, exceeding by dates when approval to conduct the into physical education courses at some 85% in each of the three grades in each survey was obtained from the public but not all middle schools; (4) promotion proportion of re- county (Table 1). The school administrators in each county. of bicycle helmets and related safety is- May 1993, Vol. 83, No. 5 668 American Journal of Public Health
  • 3. Bkycle Hele L ion spondents who rode bicycles was slightly higher than the proportion who owned them, presumably reflecting use of bicy- cles owned by siblings or friends. In the three counties combined, the proportion of respondents who rode their bicycles at least a few times per month was signifi- cantly higher among fourth-graders (91%) than among ninth-graders (66%) (P < .0001). Among respondents who ride a bicy- cle, helmet ownership was significantly higher in Howard County than in the other two counties (P < .0001). Helmet owner- ship was significantly higher among fourth- graders than among ninth-graders in each of the three counties (P < .0001). The pro- portion of bicyclists who reported owning a helmet but not wearing it the last time they rode a bicycle ranged from 4% among Montgomery County ninth-gade students (8% ownership and 4% use) to 25% among Howard County seventh-grade students (51% ownership and 26% use) (Table 1). We defined wearing a helmet on one's most recent bicycle ride as consis- creased from 12% to 38% for boys and into effect) and how often they wore it in tent with quot;alwaysquot; or quot;usuallyquot; wearing from 11% to 36% for girls. Although the the past month. The proportion of respon- a helmet while riding in the past month; changes in helmet use were smaller in Bal- dents who reported that they quot;alwaysquot; or similarly, not wearing a helmet on one's timore and Montgomery counties, the pat- quot;usuallyquot; wore a helmet increased from most recent ride was defined as consistent terns of differences by grade and gender 11.4% for last year to 37.5% for the past with quot;sometimesquot; or quot;neverquot; wearing a were similar to those found in Howard month (P < .0001) in Howard County helmet in the past month. When these def- County. (Table 2). The corresponding increases initions were used, the responses on hel- were from 8.4% to 12.6% in Montgomery met use on most recent ride and in past Awareness of the Law County (P < .01) and from 6.7% to 11.1% month were consistent in 96% of the sur- Respondents were asked whether in Baltimore County (P < .001). In How- veys (P < .0001). there was a law requiring helmet use by ard County, the absolute increase in hel- Changes in Heblet Use children where they lived. In Howard met use from last year to past month was County, 87% of the respondents who greater for fourth-graders (from 24% to Respondents were asked how often rode bicycles thought there was such a 61%) than for ninth-graders (from 4% to they wore a helmet when bicycling 1 year law, 4% thought there was not, and 10% 15%) (Figure 1); the overall use rate in- ago (before the Howard County law went American Journal of Public Health 669 May 1993, Vol. 83, No. 5
  • 4. _annenbe-g et aL Sources ofInformation about Helnets 1990 Respondents were asked to indicate 61% (from a list) all sources from which they E1991 had received information about bicycle helmets. School, parents, and television were the sources most commonly men- tioned by children who rode bicycles in all Baltimore Montgomery Howard three counties (Table 4). The proportions County County County of respondents in each grade mentioning school, parents, and friends were similar in Baltimore and Montgomery counties; these proportions were higher in Howard 25% County than in the other two counties. In each county, the proportion of children 20% who reported receiving information about helmets from parents and school was higher for fourth-graders than for ninth- 5% % 9% 5% graders (P < .0001). Other major sources id|Ss% 10 I I of such information included newspapers, magazines, and radio. The proportion of 0 children who reported receiving informa- 9 7 4 9 7 4 9 4 7 tion about helmets from physicians was Grade less than 20% in each grade in each Note. Percentages based on self-report in 1991 survey of 31 11 respondents who reported county. riding a bicycle in the last year (1990) and 2789 respondents who reported riding in the month To estimate the impact of various in- before completing the survey. Helmet use was defined as quot;alwaysquot; or quot;usually' wearing formation sources in promoting bicycle helmet while riding bicyde. helmets, we compared the sources re- ported by helmet users and nonusers (Ta- FIGURE 1-Change in percenage of schoolchildren wearing bicycle helmets, by grede, In three Maryland countIes. ble 5). Helmet users were more likely than nonusers to have heard about helmets from parents in all counties (P < .0001) or from school in Baltimore and Montgomery statement (P < .0001). In each county, re- did not know. The corresponding propor- counties (P < .01). The likelihood of hav- spondents whose friends wore helmets tions were 7%, 51%, and 41% in Mont- ing heard about helmets from television, gomery County, where a mandatory hel- were significantly more likely to wear hel- radio, magazines, or newspapers did not met law was under active consideration mets than were those whose friends did differ between helmet users and nonusers. during data collection in May 1991; the not wear helmets (P < .0001) (Table 3). In Baltimore County, 9% of the respon- law finally passed (but did not go into For respondents in grades 4, 7, and 9, hel- dents who indicated three or fewer infor- effect) in June 1991, after more than 90% met use rates were, respectively, 78%, mation sources were helmet users, com- of the survey forms had been returned. 71%, and 55% among those who reported pared with 18% of those who indicated The proportions in Baltimore County that all or most of their friends usually more than three sources (P < .0001). The (where no such law was under consider- wore helmets, compared with 20%o, 11%, corresponding helmet use rates were 31% ation) were 6%, 67%, and 27%, respec- and 6% among those who reported that and 400o in Howard County (P < .01) and tively. In Howard County, of respon- some or none oftheir friends usuallywore 10% and 19% in Montgomery County dents who were aware that there was a helmets (P < .0001). (P < .001). mandatory helmet law, 38% wore hel- Respondents who reported always mets on their last ride, compared with or usually using a seat belt were more Multivariate Analysis 14% of those who thought there was no likely to be helmet users than were those law (P < .01) and 17% of those who did The variables with the strongest bi- who reported sometimes or never using a not know if there was a law (P < .0001) variate associations with helmet use were seat belt in Baltimore County (P < .001) (Table 3). entered sequentially into a logistic regres- and Howard County (P < .0001), but not sionmodel. In this multivariate model (Ta- in Montgomery County (P = .20). Anal- Correlates of Helmet Use ble 6), the strongest predictors of helmet yses of both county-specific and grade- use were having all or most friends usually In the following analyses, helmet us- specific data showed that respondents wear helmets (odds ratio [OR] = 8.4), ers are defined as those bicyclists who re- who had been personally injured in a bike agreement that helmet laws are good ported wearing their helmets the last time accident or who had a friend who had (OR = 3.1), being in the fourth grade they rode. In both county-specific and been injured in a bike accident were not (OR = 2.4), and living in Howard County grade-specific analyses, respondents who significantly more likely to be helmet us- (OR = 2.3). Other significant predictors agreed with the statement quot;Laws that ers than were those who had no such ex- of helmet use included always or usually make children wear bike helmets are wearing a seat belt (OR = 2.1); hearing posure to personal or peer injury (Table goodquot; were more likely to be helmet users about helmets from parents, teachers, 3). than were those who disagreed with the May 1993, Vol. 83, No. 5 670 American Journal of Public Health
  • 5. egn Bicycle Helmet doctors, or friends (OR = 1.9); and hear- ing about helmets from four or more sources (OR = 1.4). Riding a bicycle at least a few times a week was a marginally significant predictor of helmet use (OR = 1.3). Diwussion Although a very high proportion of the school children who responded to this survey rode bicycles, most respondents reported that they did not routinely wear bicycle helmets. The large increase in the helmet use rate reported after the manda- tory helmet law went into effect in How- ard County suggests that such laws can have a major impact. By contrast, the in- crease in self-reported helmet use in Montgomery County during the time of educational efforts was substantially smaller and was not significantly different from that seen in Baltimore County, where no special county-wide initiatives to promote helmet use were undertaken. Considering that educational efforts pre- ceded and were subsequently expanded as part of the implementation of the How- ard County law, these results indicate that legislation combined with education in- creased helmet use substantially more than did education alone. It is important to note that a large proportion of Howard County children, especially
  • We Need Your Support
    Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

    Thanks to everyone for your continued support.

    No, Thanks