Each year in the United States, 280 children die from bicycle crashes and 144,000 are treated for head injuries from bicycling. Although bicycle helmets reduce the risk of head injury by 85 percent, few children wear them. To help guide the choice of strategy to promote helmet use among children ages 5 to 16 years, the cost effectiveness of legislative, communitywide, and school-based approaches was assessed. A societal perspective was used, only direct costs were included, and a 4-year period after program startup was examined. National age-specific injury rates and an attributable risk model were used to estimate the expected number of bicycle-related head injuries and deaths in localities with and without a program. The percentage of children who wore helmets increased from 4 to 47 in the legislative program, from 5 to 33 in the community program, and from 2 to 8 in the school program. Two programs had similar cost effectiveness ratios per head injury avoided. The legislative program had a $36,643 cost and the community-based one, $37,732, while the school-based program had a cost of $144,498 per head injury avoided. The community program obtained its 33 percent usage gradually over the 4 years, while the legislative program resulted in an immediate increase in usage, thus, considering program characteristics and overall results, the legislative program appears to be the most cost-effective. The cost of helmets was the most influential factor on the cost-effectiveness ratio. The year 2000 health objectives call for use of helmets by 50 percent of bicyclists. Since helmet use in all these programs is less than 50 percent, new or combinations of approaches may be required to achieve the objective.