5042.0: Wednesday, November 15, 2000 - 8:54 AM

Abstract #2298

Predictors of self reported child bicycle helmet use: A derivation/validation random school survey

Gregory L. Larkin, MD, MSPH1, Deborah I Friedman, MPH2, Mary P. McKay, MD2, Samuel N Forjuoh, MD, DrPH3, and Jeffrey H Coben, MD2. (1) Emergency Medicine Association of Pittsburgh/Mercy Hospital of Pittsburgh, 13 Pride Street 2nd Floor, Pittsburgh, PA 15219, 412-232-5772, glarkin+@pitt.edu, (2) Department of Emergency Medicine, Center for Violence and Injury Control, Allegheny General Hospital, 320 East North Avenue, Pittsburgh, PA 15212, (3) College of Medicine, Scott & White, Texas A&M University, Santa Fe Center, 1402 West Avenue H, Temple, TX 76504

Objective: To define predictors of reporting bike helmet use in school age children.

Methods: Design:Prospective, two-staged, derivation/validation random school survey. Setting:Urban community with a state mandate for bike helmets in children under 12. Subjects:Two cross-sectional cohorts of students in grades 3-8 from a census-based, socioeconomically stratified, random sample of public schools. Derivation cohort:819 students whose parents independently filled out a similar survey. Validation cohort:1437 students sampled from the same schools. Uni- and multivariate logistic regression was used to determine predictors of self-reported bicycle helmet use.

Results: Children who denied riding a bike were excluded (Derivation Cohort, n=70; Validation Cohort, n=186).

Candidate Predictor

Predictive Value Odds Ratio(95% CI)




Race (white vs. nonwhite)

3.59 (2.57,5.03)

2.99 (2.35,3.81)

Age (elementary vs. middle)

2.88 (2.03,4.08)

2.51 (1.94,3.26)

Agree:Bike helmet laws are good

5.23 (3.67,7.51)

4.67 (2.85,4.66)

Knowledge of bike helmet law

4.22 (3.02,5.91)

2.37 (1.88,2.99)

Child usually wears seat belt

4.73 (3.31,6,76)

4.16 (3.20,5.41)

Friends usually wear helmets


11.15 (8.19,15.17)

Stepwise multivariate logistic regression model derived from first cohort revealed six predictors of reporting helmet use: white race, younger age, belief that helmet laws are good, knowledge of the local helmet law, regular seat belt use, and having friends who usually wear helmets. Other candidate predictors including gender and childís or friendís prior injury fell out of the derivation.

Conclusions: These results suggest childrenís report of bike helmet use is strongly influenced by their peersí actions and by their knowledge and beliefs about laws and related safety issues.

Learning Objectives: Recognize the predictors or self reported bicycle helmet use. Identify potentially effective methods of increasing bike helmet use Describe strengths of relationship between helmet use and age,race,knowledge and peer behavior

Keywords: Bike Helmets, Children

Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.

The 128th Annual Meeting of APHA