The long-term goal of our research is to understand the mechanism of osteoarthritis (OA) initiation and progress through experimental and theoretical approaches. In previous theoretical models, joint contact mechanics was implemented without consideration of the fluid boundary conditions and with constant permeability. The primary purpose of this study was to investigate the effect of fluid boundary conditions at the articular surfaces on the contact mechanics, in terms of load sharing and fluid flow properties using variable permeability. The tested conditions included totally sealed surfaces, open surfaces, and open surfaces with variable permeability. While the sealed surface model failed to predict relaxation times and load sharing properly, the class of open surface models (open surfaces with constant permeability, and surfaces with variable permeability) gave good agreement with experiments, in terms of relaxation time and load sharing between the solid and the fluid phase. In particular, the variable permeability model was judged to be the most realistic of the three models, from a biological and physical point of view. This model was then used to simulate joint contact in the early and late stages of OA. In the early stages of OA, the model predicted a decrease in peak contact pressure and an increase in contact area, while in the late stages of OA, peak pressures were increased and contact areas were decreased compared to normal. These findings agree well with experimental observations.