Falls are a prevalent and significant health problem among older adults, and there is a particular need for ways to reduce the number of falls among long-term care residents. Many technological innovations have been proposed to address the problem of falls, and one of the more promising areas appears to be sensor-based technology.
A study in the International Journal of Medical Informatics reviews the existing research literature on sensor technology to prevent falls. The review found that such systems have had limited usefulness due to constraints in how widely sensors can be practically used, and that current systems have too high a rate of false alarms to be widely useful for long-term care and nursing staff.
The review examined 12 studies based in residential care settings for older adults. To be included in the review, studies were required to be designed to examine how a sensor-based system was used to prevent or reduce falls, to focus on older adult patients in either a nursing home or hospital setting, and to provide quantitative falls-related outcome data. Three of the studies were randomized control trials, eight were pre- and post- studies that compared falls rates before and after the installation of a sensor system, and one was a non-random control study.
The systems in the studies involved either wearable sensors (alarm devices worn by a resident that are intended to alert staff when the resident may have fallen) or non-wearable sensors; the latter involved weight-detecting sensors placed on or near beds or chairs, or infrared systems that detected resident movement. The authors note that most of these systems only allow for monitoring in a limited number of settings, while many falls occur away from beds, chairs, or other sensor-monitored location. Further, wearable sensors can be perceived as uncomfortable, cumbersome, or otherwise unpleasant by wearers.
Most of the studies did not see significant reductions in falls rate or injury rate resulting from the sensor systems. Three studies did see significant reductions in falls and in injuries. In one study, infrared sensors were associated with a decreased rate of nighttime falls, though this was not seen in other studies involving infrared sensors. In another, wristband sensors were associated with overall reductions in falls, though other wearable sensor interventions were unsuccessful. One multifactorial intervention saw significant decreases in both falls and falls-related injuries, but as the intervention included other non-sensor components, such as individual risk assessment and environmental modification, it is unclear to what extent the sensors contributed to the outcome.
A limitation in the sensor system research is that few of the studies included information about false alarms. One study reported that 16 percent of the system alerts were false alarms, while four other studies reported significant problems involving false alarms (although no quantitative information was included). In many studies, care staff did not find the sensor systems to be practically useful, or reported that they did not have the time to learn to use the systems. Installation, programming, and incorporating the sensor systems in care protocols appeared to be significant barriers to the use of a system, and one site removed the sensors because staff felt they were associated with lower quality of care.
The authors note that technological advances may improve the quality of future sensor systems, but argue that future systems will need to be implemented better, and that the design of such systems will need to incorporate the input of users (residents/patients and staff), and be both non-intrusive and useful across a wider range of settings than current systems.
Kosse NM, Brands K, Bauer JM, et al. Sensor technologies aiming at fall prevention in institutionalized old adults: A synthesis of current knowledge. International Journal of Medical Informatics (2013); 82: 743–752.