Dynamic models of human falls for prediction of hip fracture risk

Hip fractures are a health problem of enormous proportions: In the United States, approximately 280,000 hip fractures occur each year, resulting in estimated annual costs of over 7 billion dollars. About 90% of all hip fractures are caused by falls. This research is an investigation of the dynamics of human falls from standing height. The aim is to identify risk factors for hip fracture from falling. One of the practical limitations in studying falls is the potential danger of performing fall experiments with human subjects, in particular with elderly individuals. Therefore, a series of mathematical models of falling was developed, ranging from a simple point mass to a 15-segment rigid body model which included the action of lower extremity muscles. To validate these models, experimental data on fall kinematics were obtained from six young athletes falling on a thick foam mattress. To investigate the effects of age and muscular activity on fall dynamics, an apparatus was built to measure dynamic knee joint properties in vivo. Six young ($<$ 30 years) and six elderly ($>$ 65 years) women were tested. Knee-joint parameters, such as stiffness and damping, were obtained at two knee flexion angles and at different levels of muscular activity. A commercially available 15-link rigid body model was modified such that the measured knee joint properties could be used directly as input. Good agreement was obtained between predicted and experimentally obtained hip trajectories, hip impact velocities and body configurations at impact. Age-related differences in dynamic knee-joint properties were small and did not have any significant effects on fall dynamics. Also, according to our predictions, falling relaxed reduces the injury potential of a fall. Compared to falling in a muscle-active state, a muscle-relaxed fall resulted in impact that occurred earlier and closer to the feet. However, impact usually occurred first on the knee or side of leg, rather than the hip, thereby reducing the hip impact velocity and hip impact force. (Copies available exclusively from MIT Libraries, Rm. 14-0551, Cambridge, MA 02139-4307. Ph. 617-253-5668; Fax 617-253-1690.)