Potential of Kinect-assessed stepping test for assessing fall risk in community-dwelling older women

Abstract

Background

Previously, we have shown that movement path distance and displacement during a 20-sec stepping test (ST) discriminates between independent and dependent-living older adults. The present study aims to determine whether movement and displacement characteristics during ST are an indicator of risk of falling in independent-living, community-dwelling older women

Methods

Independent-living older women recruited for this cross-sectional study were divided into a fall (F: n = 154) or no fall (NF: n = 847) history group. Each participant completed one trial of an eyes open, ST. ST outcome measures included head total movement distance (TMD), maximum movement displacement of the head (MMD), total knee movement distance (KMD), maximum height of the left and right knees (MKH), and step number (STEP) as determined with a KINECT^TM infrared depth sensor. Ratio of KMD/TMD was calculated to index upper- body sway relative to the lower-body.

Results

Age, height, body mass, and BMI were not different between groups. KMD (F: median: 4.812 m, interquartile range (IQR): 3.720–7.718 m; NF: 5.309 m, 4.266–6.600 m), TMD / KMD (F: 0.145 m, 0.107–0.196 m; NF: 0.127 m, 0.100–0.170 m) and MKH (F: 0.073 m, 0.052–0.106 m, NF: 0.091 m, 0.060–0.135 m) were different between F and NF (Mann-Whitney, P < 0.05). Logistic regression revealed an association between falls and KMD (OR 1.232, 95 % CI 1.108–1.370).

Conclusions

Characteristics of ST movement and displacement assessed with Kinect^TM provide useful indicators for fall risk assessment among independent-living, community-dwelling older women.