Approximately 280,000 hip fractures occur in the United States annually, over 90 percent of which are caused by falls. This research investigates the mechanics surrounding impact to the hip during a fall, with the aim of identifying factors which distinguish injurious and non-injurious falls. Two basic questions have been addressed:​ 1) what force is applied to the femur during impact to the hip during a fall? and 2) can a hip padding device lower impact force in a fall to a value below the fracture range of the elderly cadaveric femur? To address question (1), "pelvis release experiments" were conducted to measure the dynamic response of the living human body to a step input in displacement applied to the hip (simulating a safe fall). A single-degree-of-freedom vibratory model was then fit to the measured response, and used to predict impact forces in falls from standing height. Validation tests of the pelvis release method with a nonlinear mechanical system showed that model predictions of peak impact force in high energy collisions were within 11 percent of actual values. Results with human subjects show that during impact to the hip, the effective mass of the body is approximately one-half body mass, and that the frequency and amplitude of the response is governed by compressive stiffness and damping elements directly in-line with the hip load vector, as opposed to beam-like flexural stiffness and damping elements peripheral to the hip. Peak estimated hip impact forces in falls were well within the range of force required to fracture the elderly femur in-vitro, and insensitive to the position of the trunk at impact. Impact tests on isolated trochanteric soft tissue samples revealed that in a fall involving 140 Joules of impact energy, up to 80 Joules of energy is absorbed in the soft tissues overlying the hip, and a three-fold increase in soft tissue thickness may decrease peak force by 25 percent. To address question (2), a novel hip pad was designed to shunt impact energy away from the femur and into the surrounding soft tissues. This pad, which covers the skin adjacent but not directly overlying the proximal femur, contains a shear-thickening or dilatant suspension of electrostatically stabilized microparticles in glycol. Appropriate selection of particle size and volume fraction resulted in a pad which remains in a liquid state during normal activities (and is thus comfortable to the user), but dramatically increases in viscosity under shear rates associated with impact, thus providing effective energy shunting. In simulated falls with an impact pendulum and surrogate pelvis, this pad provided twice the force attenuation of the next best among seven available padding systems, and was the only pad capable of lowering femoral impact force below the elderly fracture range.