In Reply The aim of any prediction model for readmission is to identify the patients who are most likely to benefit from intensive transitional care interventions. In our study,1 we use risk for potentially avoidable readmissions as a proxy for that outcome. While this is different from the Centers for Medicare and Medicaid Services (CMS) definition, the outcome is closer to the desired goal than all-cause readmissions. Of note, we have tested the HOSPITAL score for the particular diagnoses that are part of the CMS Readmissions Reduction Program, and the HOSPITAL score showed good performance (manuscript in preparation).
Donzé JD, Schnipper JL. Further Limitations of the HOSPITAL Score in US Hospitals—Reply. JAMA Intern Med. 2016;176(8):1233. doi:10.1001/jamainternmed.2016.3822