• Non-helmeted riders are 14 times more likely to be involved in a fatal crash than helmeted riders.
  
• Head injuries account for more than 60 percent of bicycle-related deaths.
  
• A very high percentage of cyclists' brain injuries can be prevented by a helmet, estimated by different studies at anywhere from 45 to 88 per cent.

Bicycle Deaths by Helmet Use
1994-2006
Year No Helmet Helmet Total*
Num
1994 776 (97%) 19 (2%) 796
1995 783 (95%) 34 (4%) 828
1996 731 (96%) 27 (4%) 761
1997 785 (97%) 23 (3%) 811
1998 741 (98%) 16 (2%) 757
1999 698 (93%) 42 (6%) 750
2000 622 (90%) 50 (7%) 689
2001 616 (84%) 60 (8%) 729
2002 589 (89%) 54 (8%) 663
2003 535 (85%) 58 (9%) 626
2004 602 (83%) 87 (12%) 722
2005 676 (86%) 77 (10%) 784
2006 730 (95%) 37 (5%) 770

...

• Nearly all bicyclists who died (97%) were not wearing a helmet.
• Riding without a bicycle helmet significantly increases the risk of sustaining a head injury in the event of a crash. Nonhelmeted riders are 14 times more likely to be involved in a fatal crash than helmeted riders.

...

Here is a study from Western Australia that shows that helmet use has reduced the incidence and severity of head injuries there. It is based on hospital data, and shows that the number of closed head injuries was cut in half with increased helmet use over time, though other injuries did not change significantly in number. The head injuries were less serious, and hospital stays were shorter.