The Barthel Index (BI) is widely used to determine eligibility criteria for inpatient rehabilitation and to monitor patients' recovery, irrespective of the illnesses that affect them. The culturally adapted Italian version of the Barthel Index (IcaBI) was recently validated. This paper reports the structural validity and inter-rater reliability of the IcaBI and its responsiveness to the results of inpatient rehabilitation. The IcaBI was administered to a cohort of 264 patients hospitalized in two rehabilitation centers in Rome, Italy. Factor analysis using principal component analysis revealed a monofactorial structure for neurological patients and, after removal of item 1 "feeding", also for orthopedic patients. Substantial to optimal inter-rater reliability was found (0.74 > intraclass correlation coefficient < 0.96). The IcaBI was found to be accurate (area under the curve= 0.72) with a minimal clinically important change score of 35 points. This work confirms that IcaBI is a useful tool for measuring disability in health and social care settings along the continuum of care. Further studies are needed to assess its criterion validity, interpretability and responsiveness in other specific disease conditions.