While the cost and burden of appropriate care for older Americans is steadily increasing, the number of professionals specializing in providing care for the geriatric population--physicians, nurses and nurse aides--is declining. Recent advances in sensor, communication and information technologies, however, have led to the creation of unique tools enabling remote diagnosis and management of chronic disease, emergency conditions and the delivery of health care.

In-home monitoring enables the evaluation and reporting of individualized health status to providers and caregivers, which allows for more timely and individually targeted preventive interventions. In fact, it may be one of the solutions to delivering care to the growing elder population, while providing hospitals and health systems with a unique opportunity to extend the delivery of care into the community.

Unobtrusive Support

Home environment health monitoring can be accomplished three ways: through ambulatory monitors that utilize wearable sensors and devices to record physiological signals; through sensors embedded in the home environment and furnishings to passively and unobtrusively collect behavioral and physiological data; and through a combination of those two methods. The passive approach is likely to attain higher levels of compliance, because the patient will not be asked to comply with any protocols or do anything that changes his or her daily routine.

The Medical Automation Research Center (MARC) at the University of Virginia Health System has developed an in-home monitoring system comprised of wireless motion sensors in every room and the shower area, a stove-top temperature sensor, a bed sensor system and a passive floor vibration-based gait monitor and fall detector. The bed sensor detects an individual’s presence through movement signals, monitors the pulse from the movement caused by heart contractions and quantifies breathing rate and depth. Pulse is computed from a pad signal while the monitored individual is resting in bed; movement artifacts, which can prevent pulse measurements, provide information on restlessness and sleep quality. All the sensing components wirelessly transmit information to a computer-based data manager in the home that requires no intervention from the patient.

The data manager collects information from separate sensor modules, processes the alert conditions locally and logs the collected data. If alert conditions are met, the data manager uses the phone line to immediately page a caregiver. Alerts are also recorded in the data log. The data manager monitors five alert conditions that include possible forgotten stove burner in the on position, possible fall, serious fall, and high or low pulse. The notification subsystem, unlike many emergency pendants, does not require user activation.

The data manger connects to the MARC’s data analysis server, where information is uploaded and automatically processed by activity inference software. The software detects key activities of daily living (ADLs) including meal preparation, showering and bathroom visits. This data is then integrated into an expanded personal health record to form a comprehensive health and wellness report.

Professional caregivers and providers can access summary assessment reports listing all the residents under their care with a priority score reflecting each monitored individual’s potential need for attention. The scoring algorithm quantifies changes in activity levels relative to the individuals norms, missed or increased specific ADLs (such as an increase in bathroom visits) and the occurrence of alert notifications. The prioritization method was developed in collaboration with care staff at a Volunteers of America National Services assisted-living facility in Minneapolis, Minn., using a user-centered, case-controlled design approach. The prioritization scheme has proven satisfactory to caregivers and facility management staff in a series of field pilots in different care settings.

Interventions, Costs Decline

Pilots of MARC’s passive in-home monitoring system reveal several benefits. In the Volunteers of America assisted-living facility mentioned above, the system assisted in rapid assessment and early detection of health conditions that might otherwise have been missed, and it proved to be an effective care coordination and management tool for professional caregivers. Moreover, monitored elders experienced a significant increase in their perceived quality of life following three months of monitoring, possibly due to an increased sense of security or improved care quality, despite the short period of the pilot and the small sample size of 15 older adults.

Professional caregivers used the information gathered to avert potential illnesses. For example, at the earliest signs of increased restlessness coupled with more frequent nightly bathroom visits, an indicator of a potential urinary tract infection, caregivers intervened by increasing fluid intake, which can clear up the infection without pharmaceutical or clinical intervention.

In a controlled three-month follow-up pilot, the technology resulted in a statistically significant reduction of billable interventions (aggregate physician visits, emergency department visits and hospital days), and therefore a statistically significant reduction in overall cost of care to payer. The study showed that the initial cost of the monitoring system, even as a research prototype, would produce a return on investment in care cost savings in six weeks of deployment.

Professional caregivers’ workloads at the pilot sites were significantly lower than the workloads of peers at the control sites who were not using the technology, despite the fact that the former were serving a significantly larger number of care recipients. The difference in the ratio of caregivers to patients indicates that caregivers at the monitoring sites achieved higher levels of efficiency than their control site counterparts, which could be due to the effective utilization of the monitoring technology. The difference in efficiency-normalized workload scale scores between the control and monitoring sites was statistically significant.

In a pilot with a home health agency managed by the Evangelical Lutheran Good Samaritan Society in Windom, Minn., the technology resulted in a statistically significant increase in perceived quality of life of monitored older adults after four months of monitoring. Providing wellness and activity reports to professional and informal caregivers may have contributed to improvement in care, as evidenced by the decrease in the number of combined physically and mentally unhealthy days for the adults after monitoring. Additionally, there was a statistically significant decrease in the measured indices of caregiver strain in informal caregivers.

Use of MARC’s monitoring system in independent senior housing apartments in Waconia, Minn., managed by the Good Samaritan Society, increased informal caregivers’ level of engagement in caring for their loved ones. Demonstrated by a statistically significant increase in informal care hours per week, the monitoring system and the availability of wellness and health status reports increased caregivers’ participation without any corresponding increase in strain or burden. The study also showed that there was no appreciable impact on patients’ quality of life, possibly because the reports were not provided to professional caregivers to provide any interventions.

When aided by in-home health status monitoring technology, professional caregivers, health care providers and elder care service providers can play a more meaningful and cost-effective role. The technology may present a new paradigm for geriatric care that would allow care providers to extend their services into the communities they serve. The approach has a significant value proposition to all the stakeholders in the process of caring for elders.