GERIATRIC PATIENTS' MOBILITY STATUS AS REFLECTED BY THE RELEVANT ITEMS OF THE BARTHEL INDEX AND IN‐HOSPITAL FALLS

Abstract

To the Editor: Although falls and fall-related injuries substantially contribute to adverse events in older hospitalized patients,1, 2 available data of in-hospital fall prevention are scarce.3, 4 The identification of patients most eligible for preventive interventions is crucial.5 Combinations of risk factors for falls are more the rule than the exception in hospitalized patients, of which mobility status itself is often extremely variable interindividually and, furthermore, can change intraindividually during the hospital stay. This may imply methodological problems of risk assessment. One thousand five hundred ninety-six fall events were recently analyzed in a large geriatric in-hospital cohort.6 The frequency of falls varied substantially over time and between patients of different diagnostic groups, with Parkinson patients showing the highest fall rate, although there was no clear relationship with activity of daily living status as measured using the Barthel Index (BI)7 except that patients who fell had significantly lower total BI scores than patients without falls. Observations on the relationship between patients' mobility status as reflected by the relevant BI items (transfer, walking, stair climbing) and falls from another in-hospital cohort of the years 2003 to 2005 (6,040 patients; mean age±standard deviation, 80.7±8.5; 68.9% women) are reported here. In general, the mean value of the total BI scores was lower in patients with than without falls (39.3±20.8 vs 48.3±25.6; P<.05). The percentage of patients with falls was lowest in those with the highest BI mobility-item score levels. This held true for admission and discharge (Table 1). Furthermore, from admission to discharge, the number of patients with falls predominantly increased by variable degrees, from 1.7 to 5.1%, corresponding to improving score levels of the items “transfer” and “walking.” There were only modest changes for “stair climbing.” As could be expected, actual mobility status was related to falls in hospitalized geriatric patients. Patients' functional competence in moving safely changed over time. Even as patients reached higher levels of mobility, their fall risk transiently increased. Therefore, hospital in-patients' mobility status as assessed on admission should not be regarded as a “static or constant variable.” This may have implications for adequate fall-risk assessment at any time during a hospital stay, which is a prerequisite for the targeting and adaptation of preventive interventions.