• Center on Health Equity and Access
  • Clinical
  • Health Care Cost
  • Health Care Delivery
  • Insurance
  • Policy
  • Technology
  • Value-Based Care

Using Wearable Sensors at Home Can Detect Falls Among Patients With Parkinson Disease


Using a body-worn sensor to detect falls at home is feasible for elderly individuals, including those with Parkinson disease, who have a higher risk of falling, according to new study findings.

Using a body-worn sensor to detect falls at home is feasible for elderly individuals, including those with Parkinson disease (PD), who have a higher risk of falling, according to new study findings.

Falls are common and dangerous in elderly adults, with an occurrence rate of 19% to 49%, and those with PD are at higher risk due to both motor (eg, impaired balance, gait deficits, freezing episodes) and nonmotor (eg, cognitive impairment) symptoms. According to authors of the research in Movement Disorders, falls at home have historically been tracked with self-reported diaries, but these are often inaccurate.

The study examined the use of a more advanced fall-detection technology—wearable sensors worn around the patient’s neck that can automatically detect falls and summon help if needed. A sample of 2063 patients with PD who subscribed to a commercial service offering one such device was matched with a control group of 2063 subscribers without PD based on age, gender, number of self-reported medical conditions, and whether they lived alone or not.

Outcome measures were the differences between the PD and control groups in incidence rate of any fall, incidence rate of a new fall after enrollment in the sensor service, and 1-year cumulative incidence of falling. Falls were identified by the push of a button or detected by the sensor device, then confirmed with a follow-up call immediately after. False alarms, accidental button pushes, and near falls were excluded.

Over a 2.5-year window, 6436 fall events occurred in the total data set; 70% were detected automatically and 30% were reported via button push. The proportion of confirmed falls reported automatically was higher in the PD group than in the control group (73.5% vs 60.1%).

The participants with PD also had a higher incidence rate of falls (2.1 vs 0.7 falls per person-year; P <.0001) and were more likely to be classified as recurrent fallers based on their having more than 2 falls in their follow-up year (29.6% vs 14.5%; P <.0001) compared with the control group. The incidence of a new fall after enrollment was higher among those with PD (hazard ratio [HR], 1.8; 95% CI, 1.6-2.0). The HRs for falling after enrollment were higher for those with PD across strata of age and gender (women: HR, 1.69; men: HR, 1.89). Fall rates were highest in participants with PD older than 78.6 years, who fell an average of 2.7 times per person-year compared with 0.8 among controls in that age group.

These findings confirm that “PD is associated with a high incidence rate of falls in daily life, emphasizing the need for fall-prevention programs tailored to this specific population,” wrote the authors. They also noted the advantages of the sensor-based fall detection system, which produced fewer false-positives than detectors using only an accelerometer and provided more accuracy than self-reported fall diaries.

The researchers noted that the analysis was performed among only individuals who had the means and motivation to sign up for a fall-detection service, so the results are mainly generalizable to elderly individuals who recognize their risk of falling and subscribe to such services. Nonetheless, the volume of fall events collected in this study “highlights the potential of using body-worn sensors for long-term home monitoring,” they concluded.


Silva de Lima AL, Smits T, Darweesh SKL, et al. Home-based monitoring of falls using wearable sensors in Parkinson’s disease [published online August 26, 2019]. Mov Disord. doi: 10.1002/mds.27830.

Related Videos
Nicki Niemann, MD, neurologist at the Muhammad Ali Parkinson Center.
Michael S. Fitts, The University of Alabama at Birmingham,
Michael Fitts, Michael J. Fox Foundation
Related Content
© 2024 MJH Life Sciences
All rights reserved.