Developing practical recommendations for the use of propensity scores: discussion of 'A critical appraisal of propensity score matching in the medical literature between 1996 and 2003' by Peter Austin, Statistics in Medicine.

I first wish to thank Dr Austin for an important paper on the use of propensity score methods in practice. It follows a series of papers by Dr Austin highlighting issues that applied researchers face in using propensity score methods—an area of research that deserves more attention. As Dr Austin shows, there has been considerable misuse and misunderstanding of propensity score methods. Papers such as his are important tools to ensure researchers use propensity scores in the most appropriate ways. In this discussion I focus on two primary topics. First, I fully endorse and elaborate on Dr Austin’s discussion of balance diagnostics of propensity score methods. Second, I discuss his statement regarding the requirement of analyses that account for the paired nature of the data after propensity score matching.