Assessing residual bias within a real-world analysis a priori: A case study of cardiovascular (CV) risk profile in individuals treated for overactive bladder (OAB)