This study is being done to evaluate the impact of a system called CareSage. The CareSage system is a technology that uses clinical data to monitor, identify and target care to patients at risk of being transported to the hospital. The investigators want to see whether the CareSage system is effective in helping to identify hospital admissions which can be prevented.

The growing elderly population and rising rates of costly chronic disease have led to the development of preventative homecare management opportunities to improve health outcomes and reduce the number of patients who enter costly inpatient care. Partners HealthCare at Home (PHH) is a preventative homecare management system which offers general care as well as specialized services to help patients and their loved ones manage chronic conditions at home. Such continuity of care is achieved through a multidisciplinary clinical team and the integration of telemonitoring into the patient's care plan.

The Philips Lifeline Personal Emergency Response Service (PERS), a wearable button device worn on the wrist or as a pendant, is a telemonitoring device available to patients receiving care through PHH. When the patient presses the button, the patient is immediately connected with a Lifeline response agent. Philips Lifeline (PLL) has developed CareSage, a predictive analytics engine that combines continuous monitoring with predictive analytics. The clinical interface (CareSage platform) helps care teams monitor the patient's calculated risk scores for emergency transport. The algorithm used by this integrated risk assessment system was originally developed after studying a large cohort of the PERS subscribers (N = ~600,000). In Phase 1 of this study the algorithm was validated among a cohort (N = 3,335) of PHH patients to predict emergency transports in this population (AUC = .76).

In Phase 2 of this study, the investigators will conduct a prospective, randomized trial of 370 patients to assess the effect of the risk assessment CareSage platform and PHH tailored interventions on the rate of readmissions, quality of life, and the overall cost of medical care. The investigators hypothesize that a multidisciplinary intervention approach could significantly reduce the healthcare resource utilization in patient at high risk for hospitalization. The intervention will flag those patients at the highest risk for hospital transport, alert the patient's care team that an intervention may be needed, thereby targeting care at high risk patients who are most likely to get readmitted. By predicting which patients are at high risk of hospital transport which may lead to hospitalization, targeting interventions at those patients and engaging patients with their care team, the investigators hope to reduce readmissions, hospital days, and rates of mortality in high risk patients.

The goal of this 2-arm randomized controlled study is to assess the impact of the CareSage risk assessment platform on 90- and 180-day Emergency Department (ED) visits in a cohort of PHH patients. All patients will receive the PERS device and be followed for a total of 9 months, including an initial 3-month observation period and followed by a 6-month intervention period. At the beginning of the observation period (baseline), enrolled patients will be randomized into 2 groups: the Intervention Group or the Control Group. During the observation period the CareSage algorithm will calibrate using patient data (both groups) collected during this interval from the PERS device. Then, during the intervention period, patients in the intervention group will be actively monitored by the CareSage algorithm and will receive tailored PHH interventions if flagged as being at high risk for emergency transport. Patients in the control group will receive care as usual during the intervention period. All data collected during the study will be used to further strengthen the CareSage algorithm developed in the Phase I of this study.

Inclusion Criteria:

• Aged 18 years or older
• Receiving care from Partners Healthcare at Home (PHH)
• Total healthcare costs fall within the middle 51st-95th percentile of patients seen at Partners Healthcare from fiscal year 2016

Exclusion Criteria:

• Currently on admission in a hospital facility with planned discharge to a long-term-care facility or Skilled Nursing Facilities (SNFs).
• Severe dementia, Alzheimer's or other serious psychiatric illness (severe anxiety disorder or psychosis).
• Enrolled in the Integrated Care Management Program (iCMP) program.
• Continued telemonitoring (TM) support extending into the 6 months of study period post the observation period of 3 months.
• PHH patients with pacemakers and other implanted devices