Bone resorption around hip stems is a disturbing phenomenon, although its clinical significance and its eventual effects on replacement longevity are as yet uncertain. The relationship between implant flexibility and the extent of bone loss, frequently established in clinical patient series and animal experiments, does suggest that the changes in bone morphology are an effect of stress shielding and a subsequent adaptive remodeling process. This relationship was investigated using strain-adaptive bone-remodeling theory in combination with finite element models to simulate the bone remodeling process. The effects of stem material flexibility, bone flexibility, and bone reactivity on the process and its eventual outcome were studied. Stem flexibility was also related to proximal implant/bone interface stresses. The results sustain the hypothesis that the resorptive processes are an effect of bone adaptation to stress shielding. The effects of stem flexibility are confirmed by the simulation analysis. It was also established that individual differences in bone reactivity and mechanical bone quality (density and stiffness) may account for the individual variations found in patients and animal experiments. Flexible stems reduce stress shielding and bone resorption. However, they increase proximal interface stresses. Hence, the cure against bone resorption they represent may develop into increased loosening rates because of interface debonding and micromotion. The methods presented in this paper can be used to establish optimal stem-design characteristics or check the adequacy of designs in preclinical testing procedures.