- To characterize the course and outcome of rehabilitation following hip fracture (HF) in elderly diabetic patients. A non-interventional prospective cohort study of 224 elderly diabetic patients who were hospitalized for rehabilitation, compared to 738 non-diabetic patients. Stepwise multiple regression was used to assess the relative contribution of the variables to the variance of the rehabilitation outcome. The absolute efficacy of rehabilitation was lower in diabetics vs. non-diabetics (28.8 +/- 20.0 vs. 31.8 +/- 18.4, respectively, p = 0.04). This difference was caused by a significantly lower pre-event Functional Independence Measure (FIM) (112 +/- 17 vs. 115 +/- 14, p = 0.002) and a significantly higher rate of prior stroke with motor impairment (16% vs. 7%, p < 0.0001). No significant difference was found between the study groups for various rehabilitation-related complications, including mortality. None of the 12 specific variables associated with diabetes or its complications was significantly and independently associated with the outcome of rehabilitation among diabetics. The outcome of rehabilitation in elderly diabetic patients following HF is significantly worse than in non-diabetics. This result stems from the poor pre-event functional state of these patients. The results of this study can help to predict the outcome of rehabilitation in elderly diabetic patients who are candidates for rehabilitation following surgical repair of HF.