Introduction: Multi-trauma rehabilitation is delivered in a variety of hospital settings. However, it is unclear whether the proximity of rehabilitation to acute services has an effect on rehabilitation outcomes. Objective: To evaluate whether the primary outcomes of an inpatient multi-trauma rehabilitation program (functional outcome and length of rehabilitation stay) are impacted when rehabilitation is delivered in a unit co-located in an acute hospital compared with a unit located in a freestanding hospital. To also compare these outcomes at a national level using data provided by the Australasian Rehabilitation Outcomes Centre (AROC). Design: Observational, retrospective audit study. Setting: An inpatient, orthopedic, multi-trauma rehabilitation unit that re-located from an acute co-located facility to a freestanding facility. Patients: Patients following multi-trauma injury admitted to the co-located rehabilitation unit (n = 216) or after its relocation to the freestanding rehabilitation unit (n = 186). Methods: Data were audited from the patients’ hospital medical records including demographics, injury characteristics, and rehabilitation outcome measures (Functional Independence Measure [FIM] and length of rehabilitation stay). Outcome Measures: The primary outcome variables were motor FIM change (change in function between admission and discharge), FIM efficiency (functional gain per inpatient day), and length of rehabilitation stay. Results: There were no statistically significant differences between the two settings in terms of motor FIM change (adjusted for admission motor FIM score) and motor FIM efficiency. In general, there was no statistically significant difference in length of rehabilitation stay between settings (median: 26 vs 27 days). At a national level, the majority of facilities offering inpatient multi-trauma rehabilitation are co-located. Nationally, freestanding units resulted in a slightly greater motor FIM change (difference between median changes adjusted for baseline = −.5, 95% confidence interval [CI] = −2.5, −0.6, P =.0012). Conclusions: There were no differences observed in outcomes between multidisciplinary rehabilitation programs in either setting. Optimizing the individual components of a rehabilitation program and improving staff skill sets should be a focus going forward.