OBJECTIVES: The ACA mandated development of a national quality strategy (NQS). NQS aligns the nation toward Better Care, Healthy People/Healthy Communities, and Affordable Care.  To monitor progress, CMS tracks a number of hospital level quality metrics that provide an understanding of hospital level performance. These increasingly serve as an input to value based payments to hospitals by payers.  As hospitals may serve distinctly different patient populations, cross hospital comparisons on quality metrics and informed value based payments by hospital pose challenges. The objective of this analysis is to examine approaches that can provide for fair cross-hospital comparisons that factor in differences in the underlying hospitals and which in turn can underpin informed value based payments. METHODS: We leverage demographic information, baseline population health metrics, and information on hospital characteristics to segment hospitals into groups of “like” hospitals that can be compared directly on quality metrics.  This segmentation leverages a latent class approach incorporating covariates and indicator variables. We also explore approaches that further adjusts hospital quality metrics based on hospital characteristics prior to comparisons (similar to how price levels over time are compared by using market baskets with constant weights). We examine these adjustments both overall as well as by segments for “like” hospitals. RESULTS: We find that a latent class approach provides an intuitive framework that can be leveraged successfully to compare “like” hospitals on quality metrics without having to scale underlying quality metrics. Further enhancements can be made when information is available to successfully adjust quality metrics within segment. CONCLUSIONS: A latent class approach to hospital segmentation allows for more meaningful cross hospital comparisons on quality metrics and can be leveraged to inform the structure and appropriate level of value based payments for individual hospitals.