Shafi S, Gilbert JC, Loghmanee F, Allen JE, Caty MG, Glick PL, Carden S, Azizhan RG.
J Paediatric Surgery, 1998 Feb;33(2):317-21
208 child bicycle crash victims admitted to the regional pediatric trauma center in Buffalo, NY from 1993 to 1995 were studied retrospectively. Head injuries were classified as concussion alone, skull fractures, intracranial hemorrhages, cerebral contusions or diffuse cerebral edema alone. Only 31 children (15%) wore helmets at the time of the crash. Helmet use rose from 2% to 26% during the period following the enactment of a helmet law on June 1 1994. The proportion of children suffering head injury was similar in both groups. However the type of head injury was different. Helmet wearers were more likely to sustain concussion alone, were less likely to have skull fractures and exhibited a trend toward less intracranial hemorrhages, cerebral contusions and cerebral edema. Excluding the isolated concussions, head injuries were noted in only one helmet wearer and 30 non-wearers. None of the 3 children who died had worn helmets and all died of mulitple head injuries.
It seems improbable that helmets could protect from serious head injuries but not less serious ones; this also conflicts with reseach that suggests the opposite. The sample size was too small for reliable comparisons.
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.