Computer simulations, dummy experiments with a new enhanced upper extremity and small female cadaver experiments were used to analyze the small female upper extremity response under side airbag loading. After establishing a worst case initial position, three tests were performed with the fifth percentile female hybrid III anthropometric test dummy and six experiments with small female cadaver subjects. A new fifth percentile female enhanced upper extremity was developed for the dummy experiments that included a two-axis wrist load cell in addition to the existing six-axis load cells in both the forearm and humerus. Forearm pronation was also included in the new dummy upper extremity to increase the biofidelity of the interaction with the handgrip. Instrumentation for both the cadaver and dummy tests included accelerometers and MHD angular rate sensors on the forearm, humerus, upper and lower spine. In order to quantify the applied loads to the cadaver hand and wrist from the door mounted handgrip, the handgrip was mounted to the door through a five-axis load cell and instrumented with accelerometers for inertial compensation. All six of the cadaver tests resulted in upper extremity injuries including comminuted mid-shaft humerus fractures, osteochondral fractures of the elbow joint surfaces, a transverse fracture of the distal radius and an osteochondral fracture of the lunate carpal bone. The results from the 6 cadaver tests presented in this study were combined with the results from 12 previous cadaver tests. A multivariate logistic regression analysis was performed to investigate the correlation between observed injuries and measured occupant response. Using inertially compensated force measurements from the dummy mid-shaft forearm load cell, the linear combination of elbow axial force and shear force was significantly (P=0.05) correlated to the observed elbow injuries.