Project abstract (1000 Characters): COMPASS-CP is an interoperable, patient-centered application that was built utilizing SMART on FHIR for better for interoperability with any capable electronic health record (EHR). COMPASS-CP captures patient-reported measures of functional and social determinates of health (PROMs) at the point of care. In real-time, COMPASS-CP uses algorithms to assess PROMs and immediately generate actionable, individualized care plans. COMPASS-CP houses a directory of community resources to support patients in making their care plans “actionable”. The care plan is centered around the patients’ needs.

With the help of an interdisciplinary team, COMPASS-CP is now integrated into Wake Forest Baptist Medical Center’s EHR, Epic, in an interoperable manner. COMPASS-CP is contributing to sustainable care planning for post-acute stroke patients by improving the care of patients and the efficiency of clinicians for enhanced patient engagement.

Project rationale, impact and innovation (3500 Characters): Over 800,000 individuals have a stroke each year. By 2030, total direct annual stroke-related medical costs are expected to more than double, from $71.55 billion to $184.13 billion. Post-acute care is the next frontier for improving stroke care and controlling Medicare spending. Improved care coordination, secondary prevention, and patient self-management could profoundly impact the social and economic burden of stroke. Stroke exemplifies a complex co-morbidity, 85% of Medicare beneficiaries with stroke have 4 or more chronic health conditions. Not surprisingly, 62% of stroke patients are either readmitted to the hospital or die within 1 year. Optimal stroke management requires new chronic care models, integrated across the continuum of care, guided by patient-reported assessments of the patient and caregiver’s capacity to manage health, individualized care plans to improve patient/caregiver engagement for self-management and shared decision-making. The care plans must be the roadmap for patient coaching and management across all providers.

In addition, CMS has set new directions for the management of patients with complex needs like stroke. New payment and care delivery models include transitional care management after hospital discharge, chronic care management, and Comprehensive Care Management Plus. In addition, CMS has identified the stroke episode bundle as one of the top utilizers of post-acute care targeted for spending reduction. The Medicare Access and CHIP reauthorization act of 2015 (MACRA) requires implementation of a merit-based incentive payment system (MIPS) to improve quality of care and value of healthcare delivery. MIPS requires care coordination, development of a care plan for complex patients and beneficiary engagement for self-management and shared decision making. Individualized care plans are a primary requirement across all of these initiatives. The care plan must be available to all patients and all health providers to manage care coordination and optimize the patient’s ability to manage their own health and independence. CMS’s proposed rule emphasizes the importance of collecting patient-reported data and the ability to impact care at the individual level. The care plan must include relevant health issues, engage patients to establish their goals of care, and identify functional and psychosocial needs, and resources to meet those needs. An individualized care plan is required to support 5 functions: (1) access and continuity, (2) risk-stratified care management, (3) planned care for chronic conditions and preventive care, (4) patient and caregiver engagement and (5) comprehensiveness and coordination of care. CMS payment incentives for chronic care management have not been fully utilized; CMS estimates 70% of beneficiaries - roughly 35 million patients - are eligible for chronic care services but CMS has only received reimbursement requests for a few hundred thousand beneficiaries. The barriers to efficient and effective implementation of the CMS requirements for care plans include (1) ineffective capture of patient-reported social and functional determinants of health, (2) information does not inform decision making, (3) excessive time to write a comprehensive care plan, (4) care plans are not actionable, and there is (5) a lack of interoperability of health informatics systems across the continuum of care.

Project design and implementation (7000 characters): An interdisciplinary team at Wake Forest Baptist Medical Center (WFBMC) has solved the barriers to chronic care management with an interoperable electronic application that utilizes HL7 Fast Healthcare Interoperability Resources (FHIR®) to access information in health information technology systems. Such a tool can be used across disease states, is easily scalable and will provide opportunities for health systems to integrate care and reimbursement across the delivery system. The solution is COMPASS-CP, a patient-centered electronic application that captures the social and functional determinants of an individual’s health and their goals of care at the point of care. Designed to be administered by a clinician, caregiver or provider in a clinical or home setting, the tool also assesses caregiver abilities and resources critical for patients during the post-stroke care period by identifying potential barriers to managing their health and recovery. The web-based application’s unique algorithm generates a personalized patient care plan, called an eCare Plan, in real time from patient and caregiver reported responses, immediately identifying and recommending specific areas where self-management and access to needed care interventions or support services are required. The purpose of the application is to facilitate a comprehensive, dynamic, holistic care plan as a vehicle for resolving communication issues and achieving highly patient-centered care. The ultimate goals are to align patient and caregivers’ goals for care with their care needs and recommend the resources to achieve those needs to optimize the patient capacity to manage their recovery, independence and health. The application also aims to reduce hospital readmissions (reducing readmission penalties), and optimize payments from the MIPS value-based payments and reward participation in advanced alternative payment models. The prototype for COMPASS-CP was developed as part of a Patient Centered Research Outcomes Institute (PCORI) pragmatic trial to implement and evaluate post-acute stroke comprehensive stroke care10-11. This prototype was implemented within 7 to 14 days post-stroke for approximately 8000 patients in a cluster-randomized trial of 41 hospitals in North Carolina.
Today, COMPASS-CP which is fully developed and interoperable through SMART on FHIR is implemented within Epic at Wake Forest Baptist Medical Center.

Project evaluation and sustainability (3500 characters): We evaluated clinician satisfaction with the prototype COMPASS-CP application and described the social and functional determinants of health interfering with recovery from stroke 14 days after hospital discharge.
We surveyed 56 clinicians (nurses, NPs or PAs) from 20 hospitals in North Carolina were surveyed to assess user satisfaction. We analyzed responses from 871 stroke and TIA patients PROMs.
We learned that clinicians 79% (N=44) completed the survey. Users agreed (66%) that they were satisfied with the application and 58% reported it made their job easier. Clinicians (75%) reported the application identified important factors impacting the patient’s recovery that they may have otherwise missed. Patients (N=871) completed COMPASS-CP assessments. The patient sample has a mean age of 67 (13.7 SD); 68% stroke, 32% TIA; 49% female; 18% African American; 62% Rankin <3; 20% rated their health as “poor” or “fair”.

The conclusion was that COMPASS-CP prototype application meets an immediate need to incorporate PROMs into the clinical workflow; to identify social and functional determinants of health, and develop patient-centered care plans for stroke patients, and has high user-satisfaction.

Twitter project summary (140 characters): COMPASS-CP captures the social/functional determinants of health, cardiovascular & stroke risk factors & goals of care at the point of care

How is FHIR used in the App being demonstrated (500 characters)? : SMART on FHIR, COMPASS-CP is being launched from Epic that is being authenticated through the Oauth2. The FHIR Patient Resource pushes the patient MRN and demographics, which creates the patient record. Patient zip code from FHIR drives the filtering and recommending the proper community resources that are geographically close to the patient. Additionally, the FHIR Condition resource is also being utilized to pull the patient diagnosis into the application that derives the care plan generation

1. What FHIR release does your application use? (500 characters)?: COMPASS CP has been validated with Cerner & Epic to support implementation and scalability. We have developed production & installation guides for on-site IT teams to install the app into local EHRs, & a clinician’s user guide, & clinical workflow outlines.
We currently use FHIR DSTU2 patient and condition resources along with the SMART framework for easy integration. With the new conditions are being added, additional resources are being incorporated such as medication, CarePlan & observation.

What is the data source for the FHIR resources and how are the FHIR resources accessed? (500 characters): FHIR patient resource that pulls from patient demographic. Additionally, FHIR condition resources pulls the patient diagnoses. We are using SMART on FHIR utilizing Epic interconnect.

Any other information about the project we should know about (1500 characters)?: COMPASS-CP was initially developed through the PCORI-funded pragmatic trial of COMprehensive Post-Acute Stroke Services (COMPASS Study). The COMPASS Study, currently ongoing, has the participation of 40 hospitals across the state of North Carolina. It has been designed with input from key stakeholders - hospital administrators, clinician-users and patients. For hospital administrators, COMPASS-CP meets MACRA requirements for care planning and care planning for complex patients. It also meets CMS billing requirements for transitional care. A small group of clinicians developed the questions, algorithms and piloted the application. The initial design was to support clinicians for stroke patients, recently, it has been modified to support other health conditions (e.g. congestive heart failure and cancer), which require chronic management and care planning.

Authors:

Umit Topaloglu (Presenter)
Wake Forest Baptist Health Systems