BACKGROUND : Vitamin D depletion is associated with unfavourable outcomes after hip fracture. However, the classes of vitamin D status currently in use, which are defined according to serum calcifediol levels, have not been validated for their predictive capability of the functional recovery. OBJECTIVE : To investigate the association between serum calcifediol categorized into 4 classes and the functional recovery after hip fracture. METHODS : Prospective, short-term observational study. METHODS : Rehabilitation hospital in Italy. METHODS : We evaluated 1350 of 1412 inpatients with hip fracture. METHODS : Serum calcifediol was measured by an immunoenzymatic assay 14.7±4.4 (mean±SD) days after surgery and categorized into 4 classes: I class <12 ng/mL; II class 12-20 ng/mL; III class 21-29 ng/mL; IV class ≥30ng/mL. The functional outcome was assessed by using the Barthel Index. RESULTS : We found a significant difference in Barthel index scores at the end of inpatient rehabilitation across the 4 classes of vitamin D status: χ2 (3, N.=1350) 27.2; P<0.001. The difference persisted after adjustment for 8 covariates (P=0.004). By comparing pairs of classes, we found that Barthel index scores were lower in the 829 patients of the I class than in the 275 of the II (P=0.005) who had in turn Barthel index scores lower than the 132 patients of the III class (P=0.038). Conversely, no significant differences emerged between the patients of the III class and the 114 patients of the IV class (P=0.421). The results did not materially change when Barthel Index effectiveness was substituted for Barthel Index scores as the outcome measure. CONCLUSIONS : Calcifediol levels below 12ng/mL were associated with a worse recovery than those between 12 and 20ng/mL that were in turn associated with a worse recovery than those between 21 and 29 ng/mL. Conversely, no significant differences were found between the patients with calcifediol between 21 and 29ng/mL and those with calcifediol ≥30 ng/mL. CONCLUSIONS : Despite caution due to the observational design, our study suggests that vitamin D depletion should be treated after hip fracture to optimize the functional outcome, with a target level for serum calcifediol of 21-29ng/mL and no further advantages associated with calcifediol levels of 30ng/mL or higher.