Evaluating the Impact of a Web-Based Risk Assessment System (CareSage) and Tailored Interventions on Health Care Utilization: Protocol for a Randomized Controlled Trial

Abstract

BackgroundSoaring health care costs and a rapidly aging population, with multiple comorbidities, necessitates the development of innovative strategies to deliver high-quality, value-based care.ObjectiveThe goal of this study is to evaluate the impact of a risk assessment system (CareSage) and targeted interventions on health care utilization.MethodsThis is a two-arm randomized controlled trial recruiting 370 participants from a pool of high-risk patients receiving care at a home health agency. CareSage is a risk assessment system that utilizes both real-time data collected via a Personal Emergency Response Service and historical patient data collected from the electronic medical records. All patients will first be observed for 3 months (observation period) to allow the CareSage algorithm to calibrate based on patient data. During the next 6 months (intervention period), CareSage will use a predictive algorithm to classify patients in the intervention group as “high” or “low” risk for emergency transport every 30 days. All patients flagged as “high risk” by CareSage will receive nurse triage calls to assess their needs and personalized interventions including patient education, home visits, and tele-monitoring. The primary outcome is the number of 180-day emergency department visits. Secondary outcomes include the number of 90-day emergency department visits, total medical expenses, 180-day mortality rates, time to first readmission, total number of readmissions and avoidable readmissions, 30-, 90-, and 180-day readmission rates, as well as cost of intervention per patient. The two study groups will be compared using the Student t test (two-tailed) for normally distributed and Mann Whitney U test for skewed continuous variables, respectively. The chi-square test will be used for categorical variables. Time to event (readmission) and 180-day mortality between the two study groups will be compared by using the Kaplan-Meier survival plots and the log-rank test. Cox proportional hazard regression will be used to compute hazard ratio and compare outcomes between the two groups.ResultsWe are actively enrolling participants and the study is expected to be completed by end of 2018; results are expected to be published in early 2019.ConclusionsInnovative solutions for identifying high-risk patients and personalizing interventions based on individual risk and needs may help facilitate the delivery of value-based care, improve long-term patient health outcomes and decrease health care costs.Trial RegistrationClinicalTrials.gov NCT03126565; https://clinicaltrials.gov/ct2/show/NCT03126565 (Archived by WebCite at http://www.webcitation.org/6ymDuAwQA).

Predictive Modeling of 30-Day Emergency Hospital Transport of Patients Using a Personal Emergency Response System: Prognostic Retrospective Study

“…An example of such an approach is a predictive model to detect impending deterioration of patients outside the intensive care unit [31], which was implemented in routine clinical care in 2 community hospitals [32]. In an ongoing randomized clinical trial, we are developing workflows that integrate daily PERS–based risk of 30-day emergency hospital transport with care pathways [33]. In this randomized clinical trial, we are using the predictive model described herein to predict patients’ risk for 30-day emergency transport, followed by a nurse assessment and tailored interventions for high-risk patients.…”

Predictive Modeling of 30-Day Emergency Hospital Transport of German Patients Using a Personal Emergency Response: Retrospective Study and Comparison with the United States

“…Case managers and health professionals should integrate risk prediction of patients into their clinical workflows to obtain the clinical and financial benefits from predictive models, which requires a detailed guideline that clarifies how the algorithm will inform care [27]. In a recently completed randomized clinical trial, we developed workflows that integrate daily PERS-based risk of 30-day emergency hospital transport with care pathways [17], resulting in 49% fewer EMS encounters in the intervention group [18]. In a currently running prospective study in Germany [19], the predictive model described herein is used to predict subscribers risk for 30-day emergency transport, followed by a case manager assessment, and tailored interventions for high-risk subscribers.…”

“…As a next step, we designed and executed a 2-arm randomized control trial, which demonstrated that PERS-based risk prediction with targeted interventions could reduce health care use and costs [17,18]. In this study, a study nurse contacted high-risk subscribers, conducted additional triaging and, if deemed necessary, provided them with interventions including educational support, nurse home visits, or primary care physician referral.…”

A randomized trial examining the effect of predictive analytics and tailored interventions on the cost of care

“…That is why such a predictive model was embedded in a Stepped-Care intervention as a first step, triggering nurse-driven interventions as a second step. To evaluate the impact of the Stepped-Care intervention on the healthcare utilization and costs of older patients using PERS we implemented a randomized controlled trial with 370 patients 6 , 7 .…”