Interaction between cognitive and motor disorders for risk screening of incident falls: results of an elderly population-based observational cohort study.

Abstract

This study aims to examine (1) the association of "Emergency Room Evaluation and Recommendations" (ER2) cognitive and motor items with incident falls (i.e., ≥ 1), their recurrence (i.e., ≥ 2) and post-fall fractures and (2) the performance criteria (i.e., sensitivity, specificity) of the greater identified association for each incident fall outcome in older community dwellers. 7147 participants (80.5 ± 3.8; 100% female) of the EPIDémiologie de l'OStéoporose (EPIDOS) observational population-based cohort study were recruited in France. Inability to name the day's date and the use of a walking aid and/or anhistory of falls were recorded at baseline. Incident outcomes, which were ≥ 1 fall, ≥ 2 falls and post-fall fractures, were collected every 4months over a period of 4years. The overall incidence of ≥ 1 fall was 26.4%, 6.4% for ≥ 2 falls, and 19.1% for post-fall fractures. Cox regressions revealed that the use of a walking aid and/or anhistory of falls [Hazard ratio (HR) ≥ 1.03 with P ≤ 0.011], inability to name the day's date (HR ≥ 1.05 with P ≤ 0.003), and their combination (HR ≥ 1.37 with P ≤ 0.002) were significantly associated with both incident falls, regardless of their recurrence, and post-fall fractures. A significant positive association between ER2 cognitive and motor items, both, respectively, and in combination, with an overall incidence of falls, regardless of their recurrence, as well as with post-fall fractures was demonstrated. However, the low sensitivity and high specificity of the combination of ER2 items suggest that these items cannot be used for risk screening of fall outcomes in the older population.