RESUMEN
BACKGROUND AND PURPOSE: The aim of this study is to observe the differences between fallers, common fallers, and non-fallers in stroke patients compared with the global ability in a rehabilitation setting. MATERIALS AND METHODS: An observational and prospective study has been carried out. A total of 476 subacute stroke patients have been observed. The main outcome measures were assessed using the Canadian Neurological Scale (CNS), Barthel Index (BI), Functional Ambulatory Category (FAC), and Trunk Control Test (TCT) at admission to the rehabilitation unit and after 90 days of the rehabilitation treatment (nearly 3 h for day for 5 days for week) at the discharge with intermediate evaluations after the first and second months. RESULTS: Out of 397 patients, 109 reported 1 or more falls (27.5%), of whom 67 fell 1 time (fallers) in the hospital (16.9%) and 42 fell 2 or more times (common fallers) (10.6%). For fallers, BI and FAC scores had a significant effect (p = 0.003 for both). Common fallers had statistically significant differences in BI (p = 0.002), FAC (p = 0.012), and TCT scores (0.023) compared with non-fallers. CONCLUSIONS: The severity of stroke may directly increase the risk of fall, and also indirectly, lengthening the hospitalization. Our study seems to suggest that patients with BI scores of between 21 and 30 on admission are more prone to fall in the first period of hospitalization, whereas in the second month, those with scores of between 11 and 20 on admission have a higher risk of falls. In the third month, patients with BI scores below 10 on admission are more susceptible to falls.