Background/Purpose: Despite statements by the American Academy of Pediatrics (AAP) and the US Consumer Product Safety Commission (CPSC) against the use of all-terrain vehicles (ATVs) by children under the age of 16 years, nearly half of ATV-related injuries and over 35% of all ATV-related deaths continue to occur in this age group. Because ATV and bicycle crashes have been associated with serious injury in children, the authors compared the demographics, mechanism of injury, injury severity, and outcome of children with ATV- and bicycle-related injuries. Further, the authors sought to identify whether ATV-related injuries elicited changes in risk-taking behavior. Methods: A retrospective, comparative analysis of 109 children admitted for ATV-related injuries and 994 children admitted for bicycle-related injuries to a level 1 pediatric trauma center between January 1991 and June 2000 was performed. A phone survey was conducted to determine self-reported changes in safety behaviors or use patterns after ATV injury. Results: Mean age was 11.1 [plusmn] 3.5 years (range, 2 to 18 years) for ATV crashes versus 9.4 [plusmn] 3.3 years (range, 1 to 17 years) for bicycle crashes ( P [lt ] .05). Ninety-three percent of ATV crashes occurred in children less than 16 years of age; 31% in children [le ]10 years of age; and 7% in children [le ]5 years of age. Male-to-female ratio was about 3:1 for both groups. White race accounted for 97% of ATV injuries compared with 79% of bicycle injuries ( P [lt ] .05). Falls from ATVs or bicycles were the most common mechanism of injury (41% v 59%, respectively). Collisions with motor vehicles were more common for bicyclists (32% v 10%), whereas collisions with stationary objects were more common among ATV riders (27% v 9%). Sixteen percent of ATV crashes were caused by a roll-over mechanism. Mean injury severity score (ISS) were significantly higher for victims of ATV crashes (8.3 ATV v 6.7 bicycle; P [lt ] .05). ATV-related trauma was associated with multiple injuries, more operative interventions, and longer hospital stays. Location and distribution of injuries were similar for both groups. Helmet use was low in both groups but higher for ATV riders (23% v 8%; P [lt ] .5). Mortality rate was similar for both groups (0.9% for ATV riders v 0.7% for bicyclists). There was a 39% response for the phone survey post-ATV injury. Twenty-three of 43 (53%) respondents owned the ATV, and 70% of these received safety information at the time of purchase. However, only 14% of injured riders received any formal training before riding ATVs. Postinjury, 60% of children continued to ride, although 42% reported decreased riding time. Fifty-four percent of children reportedly wore helmets preinjury, and there were no changes in helmet usage postinjury. There were no differences in pre- and postinjury parental supervision (61% v 65%). Conclusions: Both ATV and bicycle-related injuries occur predominantly in boys, but ATV victims are older and almost all are white. Almost all ATV injuries occurred in children under the age of 16 years. Although both ATV and bicycle crashes cause severe injuries in children, injury severity is higher for ATV crashes in terms of multiple injuries, need for operative intervention, and longer length of stay. Despite severe injuries, the majority of children injured by ATVs continue to ride, albeit fewer hours per day, and safety behaviors are unaltered. These data reinforce the current AAP stance that legislation prohibiting the use of ATVs in children under the age of 16 years without a valid driver's license should be pursued and enforced aggressively.