Addressing Disparities in the Propensity Score Distributions for Treatment Comparisons from Observational Studies