Article révisé par les pairs
Résumé : Usual walking speed (WS) is a relatively easy and reproducible tool for detecting mobility impairment. For some reasons, however, geriatric patients might not be able to perform walking tests. Therefore, a subjective assessment could be an alternative method to screen for mobility impairment. In the present paper, we explore the use of the mobility item from the Mini Nutritional Assessment-short form (MNA-sf) to assess mobility and its congruence with walking speed in hospitalized and ambulatory patients. We analyzed retrospective data from 357 patients and found a highly significant correlation between WS and the MNA-sf mobility item. After dichotomization of the MNA-sf mobility score (mobility impairment ≤1 and no impairment >1), AUC for ROC curves showed that the mobility item derived from the MNA-sf reflects fairly well the mobility of geriatric hospitalized patients (AUC = 0.773), while it performs better in ambulatory patients (AUC = 0.838).