Kinect-based choice reaching and stepping reaction time tests for clinical and in-home assessment of fall risk in older people: a prospective study

Ejupi, Andreas; Gschwind, Yves J.; Brodie, Matthew A.; Zagler, Wolfgang L.; Lord, Stephen R.; Delbaere, Kim · 2016 · OpenAlex-citations

DOI: 10.1186/s11556-016-0162-2

archive: archived pipeline: cataloged verified

Get this paper ↗ (DOI — opens at the source; we link to it, we don't host it)

Summary

This prospective study addresses the need for accessible, low-cost tools to assess fall risk in older adults, specifically focusing on the speed of protective reactions. Falls are a major public health issue, and slow voluntary reaction times are a known risk factor. While laboratory-based tests exist, they are not feasible for regular monitoring due to resource constraints. The authors developed and evaluated Kinect-based choice reaching and stepping reaction time tests (Kinect-based CRTs) to determine if they could differentiate between older fallers and non-fallers and if they were feasible for in-home administration. The study involved 94 community-dwelling older adults (aged 65+) from retirement villages in Sydney, Australia. Participants underwent laboratory assessments using Microsoft Kinect sensors to measure choice reaching and stepping reaction times. The Kinect tracked skeletal landmarks to calculate reaction time (cue to movement initiation), movement time (initiation to target hit), and total time. Participants were followed for six months to record fall occurrences. Additionally, a subgroup of 20 participants performed the tests in their homes to assess feasibility and compare results with laboratory data. Signal processing algorithms extracted features from the 30 Hz skeleton data, and statistical analyses included ANOVA for group differences and Pearson’s correlations for validity and reliability. The results showed that 19 participants (20.2%) experienced at least one fall during the follow-up period. The Kinect-based reaching reaction time test significantly differentiated fallers from non-fallers; fallers exhibited slower reaction times (797 ± 136 ms vs. 714 ± 89 ms), movement times (392 ± 50 ms vs. 358 ± 51 ms), and total times (1189 ± 170 ms vs. 1072 ± 109 ms). In contrast, the stepping reaction time test did not significantly discriminate between the two groups. The tests demonstrated convergent validity, correlating significantly with traditional clinical reaction time measures. Furthermore, the in-home assessments showed strong correlations with laboratory results (r = 0.689 for reaching; r = 0.860 for stepping), indicating high feasibility and consistency across settings. The study concludes that Kinect-based CRTs are a feasible and effective method for assessing fall risk in both clinical and home environments. The reaching reaction time test, in particular, shows promise as a screening tool for identifying older adults at risk of falling. The ability to administer these tests at home supports the development of sensor-based self-assessments, enabling regular monitoring of fall risk and evaluation of prevention interventions without the logistical burdens of traditional laboratory testing. Further validation in larger cohorts is recommended to confirm these findings.

Provenance

The full processing record for this entry. Every stage of this paper's journey through the pipeline is logged — what ran, with which tool and model, how many attempts it took, and when it last completed.

Summary generated by qwen3.6-27b-prismaquant on 2026-06-26; verification: verified.

Topics

Ranked by relevance to this paper. Hover a topic for its definition.

• perception reaction time