Abstract Background and Aims The hospitalization rate of patients undergoing hemodialysis (HD) is high, due to complications related to renal failure and hemodialysis treatment. Therefore, it is crucial to provide rehabilitation for hospitalized HD patients to improve their activities of daily living (ADL) and physical functions, so as to enable them to be discharged home, resume outpatient dialysis, and avoid re-hospitalization. However, the effectiveness of rehabilitation to improve the ADL and physical functions may differ between hospitalized patients undergoing HD (HD patients) and hospitalized patients not undergoing HD (non-HD patients), depending on their pathological conditions and influence of the HD treatment. The aim of the present study was to compare the outcomes of rehabilitation between hospitalized HD patients and non-HD patients. Method This was a retrospective cohort study. The inclusion criteria were hospitalized patients who received rehabilitation between April 2017 and February 2023. The exclusion criteria were patients who discontinued rehabilitation, were transferred to other hospitals, or died during the rehabilitation period. Rehabilitation consisted of physical therapy and/or occupational therapy provided six times a week, for both hospitalized HD and non-HD patients. The rehabilitation program comprised balance training, aerobic exercises, resistance training, and ADL training to prepare the patients for discharge to their homes. We measured several outcomes in the study, including the daily duration of rehabilitation (min/day), length of hospital stay (days), Barthel index (BI), as the ADL index, at the time of admission and discharge, grip strength and isometric knee extension strength (IKES) as muscle strength indices, and the 10-meter walking speed, timed up and go test (TUG), and short physical performance battery (SPPB) as mobility function indices. For the statistical analysis, missing values were completed by using multiple imputation methods. The inverse probability weighting (IPTW) method was used to match the HD group and non-HD group to adjust for confounding factors. The outcomes between the groups were then compared using a generalized linear model analysis. All procedures in the present study were conducted with the approval of the ethics committee of Seirei Christopher University. Results Of 1002 patients who were registered for the study, 100 patients who discontinued rehabilitation, were transferred to other hospitals, or died were excluded, and the data of the remaining 902 patients (765 of the non-HD group and 137 of the HD group) were included in the analysis in this study. The results of the generalized linear model analysis revealed a lower daily duration of rehabilitation [43.3(0.6) vs. 38.8(1.2) min/day, p=0.0003] and longer length of hospital stay [48.5(0.5) vs. 58.1(2.3) days, p=0.0108] in the HD group. In addition, the 10-meter walking speed [0.75(0.02) vs. 0.66(0.03) m/s, p=0.003] and scores on the TUG [20.8(0.7) vs. 24.3(1.0) sec, p=0.0021] and SPPB [6.3(4.57) vs. 4.7(0.6) points, p=0.0415] at discharge were also significantly lower in the HD group as compared with the non-HD group. There was no difference in the ability for ADL or muscle strength at discharge between the two groups. Conclusion This study was the first to demonstrate a lower daily duration of rehabilitation and longer hospital stay in hospitalized HD patients. Furthermore, rehabilitation had a lower impact on the mobility indices in HD patients as compared with non-HD patients. Although there was no difference in the ability for ADL or muscle strength at discharge between the two groups, HD patients exhibited a significantly lower ability for mobility functions, including a lower gait speed and lower scores on the TUG, and SPPB assessments. Therefore, rehabilitation efforts in HD patients need to be improved by securing more time for inpatient rehabilitation and promoting mobility function improvement.