Does the evaluation of gait quality during daily life provide insight into fall risk? A novel approach using 3-day accelerometer recordings

Neurorehabil Neural Repair. 2013 Oct;27(8):742-52. doi: 10.1177/1545968313491004. Epub 2013 Jun 17.


Background: Many approaches are used to evaluate fall risk. While their properties and performance vary, most reflect performance at a specific moment or are based on subjective self-report.

Objective: To quantify fall risk in the home setting using an accelerometer.

Methods: Seventy-one community-living older adults were studied. In the laboratory, fall risk was assessed using performance-based tests of mobility (eg, Timed Up and Go) and usual walking abilities were quantified. Subsequently, subjects wore a triaxial accelerometer on their lower back for 3 consecutive days. Acceleration-derived measures were extracted from segments that reflected ambulation. These included total activity duration, number of steps taken, and the amplitude and width at the dominant frequency in the power spectral density, that is, parameters reflecting step-to-step variability. Afterwards, self-report of falls was collected for 6 months to explore the predictive value.

Results: Based on a history of 2 or more falls, subjects were classified as fallers or nonfallers. The number of steps during the 3 days was similar (P = .42) in the fallers (7842.1 ± 6135.6) and nonfallers (9055.3 ± 6444.7). Compared with the nonfallers, step-to-step consistency was lower in the fallers in the vertical axis (amplitude fallers, 0.58 ± 0.22 psd; nonfallers, 0.71 ± 0.18 psd; P = .008); in the mediolateral axis, step-to-step consistency was higher in the fallers (P = .014). The 3-day measures improved the identification of past and future falls status (P < .005), compared to performance-based tests.

Conclusions: Accelerometer-derived measures based on 3-day recordings are useful for evaluating fall risk as older adults perform daily living activities in their everyday home environment.

Keywords: accelerometer; activities of daily living; activity monitoring; aging; fall risk; gait; mHealth.

Publication types

  • Research Support, N.I.H., Extramural
  • Research Support, Non-U.S. Gov't

MeSH terms

  • Accelerometry*
  • Accidental Falls / statistics & numerical data*
  • Aged
  • Aged, 80 and over
  • Female
  • Gait*
  • Humans
  • Male
  • Monitoring, Physiologic
  • Risk Assessment