The Promise of Smartphone Fall Detection Solutions for Falls in Older Adults
Michael Kwame Poku · Journal of the American Geriatrics Society · 2015
To the Editor: Falls are a leading cause of morbidity and mortality in older adults. One in three older adults falls at least once annually.1 Older adult who have fallen may face negative sequela including ongoing medical problems, limited mobility, persistent fear of falling, and lack of independence.2, 3 Emerging technologies may quickly and accurately detect falls to allow for prompt intervention. Such technology could curb physical and psychological damages that result from falls in addition to the associated costs, which can reach $40,000 per fall in severe cases.3 These solutions could bolster older adults' confidence to live independently and reassure families and caregivers, decreasing the need for more-expensive institutional care. Smartphone-embedded solutions are promising given their predicted growth in adoption and that many smartphones are already equipped with the necessary technology (e.g., accelerometers, gyroscopes) to enable accurate fall detection. At least three dozen small-scale studies have been conducted in simulated environments evaluating wearable technology for fall detection, and the results are promising; median values for studies that recorded accuracy, sensitivity, and specificity are 96%, 97.5%, and 96.7%, respectively.3 Nevertheless, a substantial challenge to incorporating wearable technologies is getting older adults to adhere to wearing them. Opportunely, smartphones are positioned to address this. In 2014, 20% to 25% of U.S. older adults owned a smartphone, and some projections suggest that one in two U.S. elderly adults will have a smartphone by 2017.4-6 Integrating fall detection technology into older adults' lives through their smartphones could boost adherence rates, but because of the developing nature of the technology, there are gaps that must be addressed before these solutions are adopted widely. Policymakers and experts need to develop standards regarding privacy, security, and performance. Furthermore, although many smartphone models incorporate the accelerometer and gyroscope technology necessary for fall detection, these components may need upgrading to achieve reliable results. Likewise, additional studies should be conducted to validate the efficacy and value of these fall detection sensors. These studies would ideally be large scale, longitudinal in nature, and conducted in real-world environments and would use older adults as subjects.7 Another problem is lack of consensus regarding various elements pertaining to falls. A consistent framework curated by relevant stakeholders is needed to address pertinent questions: What constitutes a fall, what level of accuracy is adequate for fall detection devices, what data (if any) is permissible to sell to thirdparties, and what are the necessary privacy and security standards? An additional barrier is power consumption. It is likely that continuous fall detection monitoring would quickly deplete smartphone battery life. Successful solutions will use methods to activate fall sensing only when necessary and will take advantage of innovators attempting to optimize and improve smartphone power management.8 Even if these issues are resolved, there will be gaps of coverage when older adults are not holding or wearing their smartphones, but perfect should not be the enemy of good. If cost-effective fall detection solutions are embedded into smartphones, many falls could be intervened on earlier to the benefit of older adults and families. In addition, combining multiple types of detection technology platforms could address gaps in coverage. Challenges aside, there are positive signs that smartphone fall detection solutions will continue to evolve. Researchers are increasingly investigating smartphone fall detection algorithms and techniques. The efforts of other stakeholders will be necessary to bolster this technology, and there are already many positive signs. Conflicts of Interest: None to report. Author Contributions: MP: all aspects of concept, design, analysis, and preparation. Sponsor's Role: There is no sponsor.