Traditionally, in the design of isolation bearings for bridge applications, bearing rotations due to seismic conditions are neglected. This comes from the assumption that boundary conditions are rigid and, as such, the bearing end-plates remain parallel as the bearings shear. However, for bridges with tall piers, the pier flexibility may lead to rotations at the base of the bearings, and flexibility of the bridge deck will lead to rotation at the top of the bearing. Thus, typical assumptions about bearing end conditions may be invalid. Furthermore, there is no guidance on the necessary stiffness of the framing elements to assume typical end conditions. This paper presents experimental studies on the behavior of natural and lead rubber bearings placed at a column top. The subassembly is quasistatically tested with increasingly smaller columns, resulting in rotation at the bottom of the bearings. In combination, increasing rotational demands are applied at the top of the bearing, so that both top and bottom end conditions assumptions are challenged. The experiments find significant decreases in bearing stiffness at the fixed end conditions are relaxed. Additional stability tests were conducted with the natural rubber bearing mounted on top of the columns to compare against numerical bearing stability models that assume parallel end plate conditions or end plates with permanent (static) rotation. The findings from the experimental research are used to evaluate potential changes in performance of a bridge with tall piers. The bridge model is developed in OpenSees and analyzed under design and maximum considered earthquake levels to determine the seismic demands on the isolation bearings particularly focusing on the rotational demands. The bearing models are then updated to account for the rotational flexibility and the changes in bridge displacement demands under seismic motion as well as force demands into the piers and deck are evaluated.