Improving Care Transitions Through Risk Reduction with Machine Learning Support

Simple item page
dc.contributor.authorCarver, M. Coletteen
dc.contributor.authorJones, Nateen
dc.contributor.authorDjuric, Danen
dc.contributor.authorButt, Carolineen
dc.contributor.authorMarkham, Carlaen
dc.contributor.authorBrookman, Jeremyen
dc.contributor.authorReece, Chandaen
dc.contributor.authorSmith, Jamieen
dc.date.accessioned2024-01-30T13:14:00Zen
dc.date.available2024-01-30T13:14:00Zen
dc.date.issued2020-04-15en
dc.description.abstractProblem: The ambulatory care management team at Carilion Clinic lacked the necessary tools to demonstrate readmission risk reduction for patients undergoing care transitions. Purpose: This quality improvement project aimed to determine if implementing a real-time workflow management system which supported the prioritization, intervention tracking, and coordination of transitions of care, would result in readmission avoidance through risk reduction. Background: The Accountable Care Strategies team implemented an electronic Transition Tracking Tool (T3), as one aspect of Carilion’s readmission reduction program. Evidence from the literature: Approximately 20% of Medicare beneficiaries are readmitted within 30 days following hospital or facility-based care (Fischer et al., 2014). Many health systems across the country have developed strategies to reduce hospital readmissions after the passage of the Patient Protection and Affordable Care Act and its requirement for the implementation of a Hospital Readmissions Reduction Program (ACA, 2010). While there are a variety of readmission risk stratification tools used to identify patients, the predictive performance of these tools, according to Kansagara et al., (2011), has been marginal due in part to the complex factors contributing to a readmission. These researchers recommend incorporating a larger data set to include social determinants of health (Kansagara et al., 2011). Patient’s social determinants have a significant impact on their readmission risk, thus ambulatory programs which address these factors are essential (Calvillo-King et al., 2013). EBP Question: (1) Is there an impact on readmission for a patient who undergoes risk reduction strategies by a nurse using an automated patient prioritization tool with predictive interventions? Methods: The ambulatory care management team uses a relationship-based model, partnering with patients in self-care which is grounded in Dorothea Orem’s Theory of Self-Care (Petiprin, 2016). The aim is to support personal agency in the achievement of effective self-management. A tool was needed to replace a manual system, which could identify and prioritize at risk patients and track interventions and readmissions. A real-time data system was implemented called T3, it aggregates patients from both in and out of network hospitals. T3 also ingests information from Jvion, a machine learning platform that provides a readmission risk scoring and associated interventions. A dashboard displays patients and their risk scores, along with recommended interventions. Ambulatory nurses working remotely select a patient for outreach, review machine-recommended interventions and use nursing judgement for a patient-centric approach. Readmissions prevented are recorded using specific criteria. Outcomes: On average 2200 patient were managed each month and received risk reduction interventions. Over 11 months 212 patients had a readmission prevented. With the average cost of a hospital stay at $11,200.00, these 212 prevented readmission would have cost well over 2 million dollars. Most importantly the team saved patients from sustaining additional health complications due to a readmission. Implications for practice: Health systems focusing on readmission reduction need to consider using a predictive tool which incorporates social determinants of health and recommends targeted interventions. Prioritizing discharged patients, managing and tracking interventions, and recording readmissions prevented by ambulatory nurses will demonstrate improved quality of care transitions. References: (avail)en
dc.identifier.urihttps://hdl.handle.net/10919/117729en
dc.language.isoenen
dc.rightsIn Copyrighten
dc.rights.urihttp://rightsstatements.org/vocab/InC/1.0/en
dc.subjectHospital dischargeen
dc.subjectTransition of careen
dc.subjectReadmission risken
dc.titleImproving Care Transitions Through Risk Reduction with Machine Learning Supporten
dc.typePosteren
dc.type.dcmitypeTexten
pubs.organisational-group/Virginia Techen
pubs.organisational-group/Virginia Tech/VT Carilion School of Medicineen
pubs.organisational-group/Virginia Tech/VT Carilion School of Medicine/Family and Community Medicineen
pubs.organisational-group/Virginia Tech/VT Carilion School of Medicine/Family and Community Medicine/Family and Community Medicineen

Files

Original bundle
Now showing 1 - 1 of 1
Name:
Improving Care Transitions Through Risk Reduction with Machine Learning Support _Carver presentation.pptx
Size:
60.24 KB
Format:
Microsoft Powerpoint XML
Description:
Presentation
Download
License bundle
Now showing 1 - 1 of 1
Name:
license.txt
Size:
1.5 KB
Format:
Plain Text
Description:
Download

Collections

All Faculty Deposits
Scholarly Works, Virginia Tech Carilion School of Medicine (VTCSOM)