AT EASE: Automated Technology for Elder Assessment, Safety, and Environmental monitoring

Background : Independent Living Residences (ILRs) for elders are becoming a popular and less expensive option than assisted living facilities. ILRs, however, operate with minimal professional staffing and new ways are needed to provide oversight of increasingly frail and/or confused elders who live alone. Objective : To gain an understanding of the elders, families and staff concerns in ILRs and to investigate whether remote residential monitoring, using off-the-shelf wireless sensors, might address these concerns. Methods : Mixed methods approach. Phase 1 qualitative research involved eight sets of focus groups comprised of either elderly residents from ILRs, relatives of residents, building managers and superintendents, or affiliated nurse practitioners (NPs) providing residential services. Phase 2 implementation built the system in response to end users input. Detailed testing of signal reliability and validity occurred. Phase 3 involved an 18 month implementation period with 10 sets of end users using the system for at least 4 months each. End user sets –residents, family members, building staff and nurse practitioners were assessed pre and post implementations. Results : Across the end user subgroups concerns varied but they all shared in common worries about the safety and well-being of the residents. Specific memory related issues included medications, meals, and shutting off the toilet/bath water. The Automated Technology for Elder Assessment, Safety and Environment (AT EASE) remote home monitoring system was developed to uniquely tailor the type of sensor and activities monitored to the individual’s particular concern(s). Initial reliability trials uncovered severe signal interference for the X10 based wireless sensor system. This resulted in a redesign and successful deployment of a Zigbee based system. Moreover for the first time alert information was triaged to multiple parties, authorized as recipients, and occurred without any data security breaches. For example, only the building superintendent would receive the toilet overflow alert, the family member the missed meal or medication alert and the nurse practitioner when several medications were missed. Conclusions : Multiple safety and well being concerns arose that could be addressed through sensor based residential monitoring but most residents underestimated their personal vulnerability. Key to acceptance was the residents' perceived need and usefulness of the system to maintain independence and prevent being relocated to a more restrictive environment. Families found the system easy to use and were very satisfied. Building staff highly valued the water overflow alerts and having an additional means to oversee residents’ safety that did not increase their workload. NPs were non-adopters (feared information overload) favoring personal interaction with residents to medication compliance monitoring. End users favored passive alert notices, (over proactively monitoring the website site for residents’ status) that were few in number, and valid. They did not want to monitor the monitoring data and believed that accurate alerting was critical and achieved.