Recently there has been a greater awareness of the increased risk of certain injuries associated with air bag deployment, especially the risks to small occupants, often women. These injuries include serious eye and upper extremity injuries and even fatalities. This study investigates the interaction of a deploying air bag with cadaveric upper extremities in a typical driving posture; testing concentrates on female occupants. The goals of this investigation are to determine the risk of upper extremity injury caused by primary contact with a deploying air bag and to elucidate the mechanisms of these upper extremity injuries. Five air bags were used that are representative of a wide range of air bag ‘aggressivities’ in the current automobile fleet. This air bag ‘aggressivity’ was quantified using the response of a dummy forearm under air bag deployment.For this study, sixteen excised cadaveric upper extremities were mounted to a four degree-of-freedom universal joint that functioned as a shoulder. The upper extremity position was a ‘natural’ driving posture in a one-hand turn crossover maneuver. In this posture, the forearm was pronated with the humerus normal to the plane of the steering wheel and the forearm positioned directly on the air bag module so that the distal third of the forearm was over the air bag tear seam. This position represents a ‘worst case’ for deployment of an air bag into a forearm.Injuries sustained during testing included ulna nightstick fractures and multiple fractures. A nightstick fracture is a diaphaseal ulna fracture unaccompanied by a radius fracture . The results indicate that the air bag/flap interaction may be significant in the production of some injuries, but is not necessary to produce peak moments in the forearm. These primary contact injuries occurred within 13 ms of the deployment near the site of interaction with the flap. In addition, these results show that the risk of injury increases with increasing air bag aggressivity. The testing also suggests that there is an aggressivity among those air bags installed in the current automobile fleet below which the risk of forearm injury for occupants is low. Further, the results imply that there is a forearm bone strength above which the risk of injury is low even for an aggressive air bag deployment. Results also suggest that humerus position, forearm pronation angle. and forearm position relative to the air bag module affect the risk of injury from air bag deployment.Comparison tests were performed with the SAE 5th Percentile Female Instrumented Arm using four of the air bags tested in the cadaveric series. These tests were performed using the SAE arm mounted on a Hybrid III dummy and the universal joint test fixture. Repeated tests revealed no significant difference between the forearm response on the fixture and on the dummy. Resulting forearm moments were used to develop logistic injury risk functions for ulna and ulndradius fractures of female forearms in the worst case position under air bag deployment. For 50% risk of ulna or ulndradius fractures, the dummy forearm moment is 61 N-m (+/- 13 N-m standard deviation). For 50% risk of fracture of both the ulna and the radius, the dummy forearm moment is 91 N-m (+/- 14 N-m standard deviation).