Sorry this a bit long but I thought it would be better than a link:
What evidence is there that cycle helmets reduce serious injury?
There have been many predictions that cycle helmets are effective in reducing serious injuries. Most of these predictions come from case-control studies, which are based on small research populations and have been criticised for methodological limitations.
On the other hand, large population data, from sources such as traffic casualty statistics and hospital treatment records, do not support these predictions. These sources show no improvement in serious injury trends as helmet use has become more common. Indeed, sometimes they suggest that the number or severity of injuries has increased.
In Great Britain, there was no detectable improvement in fatalities, serious injuries or the average severity of injuries to cyclists over the period 1985 to 2001, during which helmet use rose from close to zero to approx 22%. Injury severity increased as helmet use became more common (BHRF, 1071). A study of road traffic casualties has found no association between differing patterns of helmet wearing rates and casualty rates for adults and children. Similarly, boys and girls have identical percent head injury rates but markedly different levels of helmet use (Hewson, 2005; Hewson, 2005b).
In Greater London, cyclist injuries became more serious as helmet use increased in the mid 1990s (BHRF, 1072). In 2001, although about half of cyclists wore helmets, the severity of injuries was significantly higher than in 1981 and fatalities were highest since 1989. In Edinburgh, also with approx 50% helmet wearing, casualties have become more serious as helmet use has increased (BHRF, 1247). In the Lothian region (close to Edinburgh), wearing a helmet has made no significant difference to outcome in the case of the more serious head injuries measured by need of follow-up or hospital admission (Scottish Exec, 2005).
In the USA, cyclists suffered more head injuries in 2001 than in 1991 although helmet use had increased from 18% to 50%. There is no clear information that cycle use increased during this period and some evidence that it may have fallen. (BHRF, 1041)
In Australia, helmet laws caused head injuries to fall by 11% to 21%. But cycle use fell by 30% to 60%, suggesting that those who continued to cycle were more at risk (BHRF, 1096). In New Zealand, large increases in helmet use have not brought any reduction in the proportion of serious head injuries. Some reduction in mild concussions and lacerations has been balanced by an increase in neck injuries (##10017). An analysis of enforced helmet laws in Australia, New Zealand and Canada has found no clear evidence of benefit and increased risks for cyclists post-legislation (Robinson, 2006).
Analysis by Erke and Elvik, 2007 showed an increased accident risk per cycling-km for cyclists wearing a helmet. In Australia and New Zealand the increase was estimated to be around 14%.
In Germany, research found no significant difference in the level of head-trauma in cycling crashes between cyclists who wore a helmet and those who did not (Möllman, Rieger and Wassmann, 2004).
More generally, concerns have been expressed that helmets may increase the risk of the most serious types of head injury typical of road crashes and which involve rotational forces (BHRF, 1039).