Helmet protection from head injuries among recreational bicyclists

Wasserman RC, Buccini RV. American Journal of Sports Medicine, 1990;18(1):96-7

BRIEF NOTES
This is not a complete Commentary but a summary of observations and criticisms that have been made relating to this paper

Summary of paper (from authors' abstract)

Data was collected from questionnaries returned by mail from 191 recreational bicyclists who reported having fallen and struck their heads in a cycling mishap. Information was collected on the nature of the mishap, the extent of injury and whether the rider was wearing a helmet at the time of the fall. 57% of riders were wearing a helmet during the mishap. Helmet wearers were significantly older than those not wearing helmets. Helmet wearers experienced significantly fewer skull fractures (1% v 11%) and facial soft tissue injuries (5% v 18%) than non-wearers. No other variables accounted for differences in injuries.

Peer criticism

Based on Towner et al, 2002:

Injuries and helmet use were self-reported and therefore subject to reporting bias.
The cases (helmet wearers) were significantly older than non-wearers and suffered more arm and leg injuries. This suggests that the two groups of cyclists were not directly comparable.
Neck injuries were more frequent in helmeted riders.
Facial injuries were reduced in helmeted riders as well as head injuries, although it is not clear how helmets protect against lower face injuries.
Cyclists killed or seriously injured were not included.
Some controls were wearing non ANSI approved helmets, but it is not clear how many.
All the cyclists were cycling enthusiasts; there is no information about how these compared with the general population.
Helmet use was high and there is no informtion about how helmet wearing compared with the population norms.
Disparity between self-reported injury and admission to hospital. 1/3rd of helmeted and 40% of unhelmeted said they suffered concussion although overall only 50% had medically attended injuries and only 25% of these had been admitted to hospital.
Some cyclists are counted more than once in the injury tables.
Small sample study (109 cases and 82 controls).

Also:

Possible lack of detachment on the part of the researchers, whose paper ends: "The problem is how to get more bicyclists to wear bicycle helmets".

Conclusion

There are too many discrepancies and uncertainties for this to be a reliable study.

References

Towner et al, 2002

Towner E, Dowswell T, Burkes M, Dickinson H, Towner J, Hayes M, 2002. Bicycle helmets - a review of their effectiveness: a critical review of the literature. Department for Transport Road Safety Research Report 30.