Data for Healthy Decisions: Computation for Passive Monitoring of Medium-Risk Individuals at Home

Many senior adults living independently are at some risk of severe health problems, and have adult caregivers who do not live with them. There is a need for positive indicators of well being to be available to those caregivers. Technology that requires the senior adult to actively perform tasks, such as filling out logs or hooking up sensors, may not be effective, due to the burden on the individual leading to noncompliance. What is needed is technology that is largely passive, installed unobtrusively in the home, that can generate the data needed for calculating indicators of well being. Such data is useful to remote caregivers as well as to medical professionals. One example of such a system, a prototype used to passively monitor and compute nocturnal trips to the bathroom, is presented. Data were collected from 7 senior adults living alone over a four-week period. Computational challenges were significant. Effectiveness of such technology requires social acceptance, ease of use, data security, and calculation of reliable, actionable metrics. KeywordsComputational social science; social analytics,