Video capture of the circumstances of falls in elderly people residing in long-term care: an observational study

Stephen N Robinovitch, Fabio Feldman, Yijian Yang, Rebecca Schonnop, Pet Ming Leung, Thiago Sarraf, Joanie Sims-Gould, Marie Loughin, Stephen N Robinovitch, Fabio Feldman, Yijian Yang, Rebecca Schonnop, Pet Ming Leung, Thiago Sarraf, Joanie Sims-Gould, Marie Loughin

Abstract

Background: Falls in elderly people are a major health burden, especially in the long-term care environment. Yet little objective evidence is available for how and why falls occur in this population. We aimed to provide such evidence by analysing real-life falls in long-term care captured on video.

Methods: We did this observational study between April 20, 2007, and June 23, 2010, in two long-term care facilities in British Columbia, Canada. Digital video cameras were installed in common areas (dining rooms, lounges, hallways). When a fall occurred, facility staff completed an incident report and contacted our teams so that we could collect video footage. A team reviewed each fall video with a validated questionnaire that probed the cause of imbalance and activity at the time of falling. We then tested whether differences existed in the proportion of participants falling due to the various causes, and while engaging in various activities, with generalised linear models, repeated measures logistic regression, and log-linear Poisson regression.

Findings: We captured 227 falls from 130 individuals (mean age 78 years, SD 10). The most frequent cause of falling was incorrect weight shifting, which accounted for 41% (93 of 227) of falls, followed by trip or stumble (48, 21%), hit or bump (25, 11%), loss of support (25, 11%), and collapse (24, 11%). Slipping accounted for only 3% (six) of falls. The three activities associated with the highest proportion of falls were forward walking (54 of 227 falls, 24%), standing quietly (29 falls, 13%), and sitting down (28 falls, 12%). Compared with previous reports from the long-term care setting, we identified a higher occurrence of falls during standing and transferring, a lower occurrence during walking, and a larger proportion due to centre-of-mass perturbations than base-of-support perturbations.

Interpretation: By providing insight into the sequences of events that most commonly lead to falls, our results should lead to more valid and effective approaches for balance assessment and fall prevention in long-term care.

Funding: Canadian Institutes for Health Research.

Conflict of interest statement

Conflicts of interest

We declare that we have no conflicts of interest.

Copyright © 2013 Elsevier Ltd. All rights reserved.

Figures

Figure. Example falls by elderly adults
Figure. Example falls by elderly adults
(A) Incorrect weight shifting while standing and turning. While initiating a turn, this woman rotates her walker and upper body 180 degrees, while her feet remain stationary (typical of Parkinson-like freezing). Despite eventual steps, a backward fall ensues. (B) Incorrect weight shifting while walking forward. While stepping around his dog, this man establishes too narrow a base of support, causing a sideways fall. (C) Trip while walking and turning. While playing ball, this man initiates a turn by crossing his left leg in front of his right. He loses balance during the next step, after the toe of his right foot collides with his left heel, resulting in a backward fall. (D) Trip while walking forward. This woman seems to attempt to steer around the foot of a lifting device, but trips on the obstacle. (E) Loss of support with external object while sitting down. Note that the wheelchair rolls backward on contact, and is unable to provide the support necessary to complete the transfer. Videos 1–5 show these falls in full.

Source: PubMed

3
Předplatit