Predictors of inpatient rehabilitation after total knee replacement: an analysis of private hospital claims data

In reply: We welcome the debate about the choice of rehabilitation at home as an effective treatment option. Our study1 supports the position statement of the Rehabilitation Medicine Society of Australia and New Zealand (RMSANZ)2 and the comments of the Australasian Faculty of Rehabilitation Medicine of the Royal Australasian College of Physicians which recommend inpatient rehabilitation be based on patient need. We considered many clinically relevant patient factors identified in the position statement, including age, socio‐economic status, comorbidities, living alone and surgical complications.1,2 Large interhospital variation in inpatient rehabilitation persisted after adjustment for these factors. It seems implausible, given the large sample, that unobserved patient factors could explain the wide variation (76% in one large volume private hospital, 10% in another1), but we acknowledge our study could be improved with function and obesity measures in particular.