Thin-shell finite element formulations are known to suffer from pronounced membrane locking. Despite numerous strategies proposed in the literature to address this issue, there remains a clear need for approaches that are both numerically robust and straightforward to implement. A particularly appealing recent development is the hybrid discretization technique introduced by Sauer et al. [1] for alleviating membrane locking in isogeometric Kirchhoff-Love shells. Their method couples standard quadratic NURBS-based shell elements with linear Lagrange membrane elements. Importantly, the approach introduces no additional degrees of freedom, ensuring that the global system size is unchanged. In this work, the hybrid membrane-bending interpolation method is further refined to improve computational efficiency while preserving solution accuracy. Several adaptations are introduced to achieve this goal. First, reduced integration is employed, replacing the standard 3×3 Gauss quadrature with a 2×2 scheme for the bending contributions, reducing evaluation cost. Second, the formulation is reorganized such that membrane and bending terms are handled consistently within a unified element routine. Finally, the approach is extended to accommodate multi-patch configurations, enabling the analysis of arbitrarily shaped geometries. To ensure proper coupling across patch interfaces, continuity is enforced using a penalty-based constraint inspired by the work of Duong et al. [2]. With these enhancements, the hybrid method remains theoretically equivalent to its original version while becoming more practical and broadly applicable. To assess the performance of the modified hybrid formulation, a series of established benchmark problems is investigated and compared against standard isogeometric shell formulations. In particular, extensive L2-norm convergence studies show that the modified hybrid formulation retains the intended locking-free behavior and clear accuracy advantage over classical IGA approaches that were already observed for the original version, especially for membrane stress measures in highly slender shells. Furthermore, the use of reduced integration significantly increases computational efficiency without compromising accuracy.