Knee Injury Air Bag Risk Assessment for Children

Principal Investigator: John H. Bolte IV, PhD, The Ohio State University

Below is an executive summary of this project. Please note that this summary describes results and interpretation that may not be final. Final interpretation of results will be in the peer-reviewed literature.


Occasionally, devices designed to protect adults in MVCs can be physically hazardous to children. One example is frontal air bags. Although proven to save adult lives, they also have contributed to the deaths of 159 children between 1993 and 2003. This study was designed to determine what risks, if any, knee bolster air bags pose to child occupants restrained in the front seat. The study was conducted in three steps: 1) to determine the real world population most likely at risk by deployment of this type of air bag and the possible injury mechanisms this group would encounter; 2) to determine how the injury mechanisms can be properly evaluated in automobile testing; 3) to determine the level of risk for pediatric lower extremity injuries due to knee bolster air bags. 

In order to ensure that the air bag testing would simulate realistic situations, data was collected from several real world crash sources and included typical children’s ages, seating positions, types of restraint, common injuries, and injury mechanisms in front seat crashes. Several pieces of instrumentation were added to the Hybrid III 6-year-old anthropomorphic test device’s lower extremities based on the expected injury mechanisms, including accelerometers, angular rate sensors, and force sensors. For each trial, 67 channels of data and three high-speed camera views were recorded. 

Static tests were conducted using a stationary knee air bag module positioned in front of a vehicle seat. Five air bag deployment trials were completed using the Hybrid III 6-year-old ATD in a different position for each: properly restrained in a booster seat; sitting on the edge of the seat with feet on the floor; sitting on the edge of the seat with knees at the air bag module; leaning back in the seat with toes on the air bag module; and leaning back with heels on the air bag module.

Although the majority of the parameters studied did not reflect a high chance of injury, the test with subjects on the edge of the seat indicated the possibility of tibia fracture. 

Better data collection methods, such as tibia load cells, should be implemented so that more reliable conclusions can be drawn. Also, the two tests involving subjects leaning back in the seat resulted in extremely high ankle rotation rates. In order to collect meaningful ankle data in the future, the biofidelity of the ankle joint on ATDs must be improved.

IAB Mentor

Doug Longhitano, Honda R&D Americas Inc.