Real-time Symptom Monitoring Using ePROs to Prevent Adverse Events During Care Transitions

Adverse Event, Multiple Chronic Conditions

Digital Health, Patient Reported Outcomes, Symptom Monitoring, Predictive Model, Care Transitions, Post Acute Care

This study aims to predict and minimize post-discharge adverse events (AEs) during care transitions through early identification and escalation of patient-reported symptoms to inpatient and ambulatory clinicians by way of predictive algorithms and clinically integrated digital health apps. We will (1) develop and prospectively validate a predictive model of post-discharge AEs for patients with multiple chronic conditions (MCC); (2) combine, adapt, extend, and iteratively refine our EHR-integrated digital health infrastructure in a series of design sessions with patient and clinician participants; (3) conduct a RCT to evaluate the impact of ePRO monitoring on post-discharge AEs for MCC patients discharged from the general medicine service across Brigham Health; and (4) use mixed methods to evaluate barriers and facilitators of implementation and use as we develop a plan for sustainability, scale, and dissemination.

Adverse events (AE) during care transitions range from 19-28% and may lead to readmissions, representing an ongoing threat to patient safety. Early identification and escalation of patient-reported symptoms to inpatient and ambulatory clinicians is critical, especially for patients with multiple chronic conditions (MCC). Clinically integrated digital health apps have the potential to more accurately predict post-discharge AEs and improve communication for patients, their caregivers, and the care team. Such tools can provide individualized risk assessments of AEs by systematically collecting relevant patient-reported outcomes (PROs) and leveraging standardized application programming interfaces (API) to combine them with electronic health record (EHR) data. While patient-reported outcomes (PROs) are increasingly used in ambulatory settings, their use for real-time symptom monitoring and escalation during transitions from the hospital is novel and potentially transformative-by both empowering patients to better understand their individualized risks of post-discharge AEs, and improving monitoring while transitioning out of the hospital. Our proposed intervention is grounded in evidence-based frameworks for care transitions, and scaling and spread of digital health tools. To inform our intervention, we propose developing and validating a predictive model of post-discharge AEs for 450 MCC patients using relevant PRO questionnaires and electronic health record (EHR) derived variables during our baseline pre-implementation period. Simultaneously, we will combine, adapt, extend, and refine our previously developed EHR-integrated hospital and ambulatory-focused digital health infrastructure to support MCC patients in real-time symptom monitoring using PROs when transitioning out of the hospital. Our intervention uses interoperable, data exchange standards and APIs to seamlessly integrate with existing vendor patient portal offerings, thereby addressing critical gaps and supporting the complete continuum of care. Our multidisciplinary team uses principles of user-centered design and agile software development to rapidly identify, design, develop, refine, and implement requirements from patients and clinicians. Our team will rigorously evaluate this intervention in a large-scale randomized controlled trial of 850 in which we compare our real-time symptom monitoring intervention (425) to usual care (425) for patients with MCCs transitioning out of the hospital. Finally, we will conduct a robust mixed methods evaluation to generate new knowledge and best practices for disseminating, implementing, and using this interoperable intervention at similar institutions with different EHR vendors

No related papers found yet.

United States