Looking Again: Do Bicycle Helmets Work?

California became the sixth state in the US to pass bicycle helmet legislation for children under 18 years of age. The law became effective on 1 January 1994. In San Diego County, from July 1991 through June 1994, there were 725 bicycle-related injuries, including 41 deaths, treated at trauma centers in the county. A majority of the injuries (51.6%) were among those aged 19 years or less.. After an excellent introduction reviewing work to date on the efficacy of bicycle helmets, the authors present their study that uses an ecological trend design to evaluate the effect of the helmet law in San Diego County by measuring: (a) the reduction of bicycle-related head injuries, the ultimate goal of such legislation, and (b) the increase in helmet use.

Data and Analysis
The study population was all bicycle-related injury records in the county’s Trauma Registry from 1992 to 1996. The helmet law went into effect in 1994. The study goal was to determine the significance of the intervention by comparing outcomes both before and after the start of the law. According to the year 2000 US Census, there were 2,813,833 residents in San Diego County of which 723,661 were under the age of 18 ­– 26% of the population.

Variables that were analyzed included anatomic site of injury, injury severity (AIS score), calendar time of injury, use or non-use of a helmet and outcome. Age was recorded as either less than 18 years or 18 years and above. A total of 1116 bicycle trauma patients were in the registry between 1992 and 1996. There were 510 (45.6%) children less than 18 years; the adult category was 606 (54.4%). The majority were male (82.1%) and white (62.8%). Relatively few, 52 (4.7%), died due to their injuries. Two hundred three of the 1064 trauma patients (18.2%) were wearing a helmet at the time of injury. However, another 295 patients had data missing on helmet wearing. Among those patients where helmet use was known, 24.7% were wearing helmets. Of the 52 fatalities, 50 (96.2%) had serious head injuries; helmets had been used by 12.2% of these, not used by 63.9%, and in 23.9% helmet use was unknown.  Chi-square test evaluated the strength of bivariate associations between helmet use and selected demographic characteristics and injury patterns. The 295 patients for whom helmet use was unknown were excluded. The analysis showed that adults used helmets 31.2% of the time; children used them 18.2% of the time. Whites were most likely to have used a helmet (33.1%). Only 16.1% of patients with serious head injury used helmets, compared to 28.2% in those who did not have serious head injury. In the years before the helmet legislation, 13.2% of patients used helmets; 31.7% after the law went into effect.

The number of serious head injuries was plotted against the percentage of helmet use for each year of the study and for adults and children respectively. The time trends showed a significantly increasing trend in helmet use over time, with an average magnitude of increase of 43% per year. However, the trend for adults was only marginally significant; children showed a significant trend averaging 84% per year. There was no statistically significant trend for serious head injury. Patients injured in 1994 were 2.61 times as likely to wear a helmet compared to the pre-law period. Those injured in 1995 were 4.42 times as likely; those injured in 1996 were 2.86 times as likely to wear a helmet. Patients with serious head injury were 0.43 times as likely to have worn a helmet as those without serious head injury. There was no significant association between serious head injury and time over the study period despite increasing helmet use. However, there was an unexplained decrease in helmet use among children between 1995 and 1996, which might have contributed to this finding of no significant decrease of severe head injuries post-law.

The reader is urged to look at the full paper for a more detailed report of the study and for a discussion of its numerous limitations. Because of these limitations, the authors end by cautioning against concluding that helmet use did not have an effect on reducing serious head injuries.

Reference
Ji M, Gilchick RA, Bender SJ. Trends in helmet use and head injuries in San Diego County: The effect of bicycle helmet legislation. Accid Anal Prev 2006; 38:129-134.

back to Headlines