1. Redesigning Kidney Disease Care to Improve Value Delivery
- Author
-
Sandeep Palakodeti, Valerie Reese, Todd M. Zeiger, Bradley Patton, Brandi N. Dobbs, Brayden Dunn, Justin J. Coran, Titte R. Srinivas, Esther J. Thatcher, Nagaraju Sarabu, Patrick Runnels, and Peter J. Pronovost
- Subjects
Nephrology ,medicine.medical_specialty ,Quality management ,Primary Health Care ,Leadership and Management ,business.industry ,Health Policy ,Public Health, Environmental and Occupational Health ,Primary care ,medicine.disease ,Internal medicine ,Electronic Health Records ,Humans ,Medicine ,Renal Insufficiency, Chronic ,business ,Intensive care medicine ,Referral and Consultation ,Value (mathematics) ,Kidney disease - Abstract
This article describes the articulation, development, and deployment of a machine learning (ML) model-driven value solution for chronic kidney disease (CKD) in a health system. The ML model activated an electronic medical record (EMR) trigger that alerted CKD patients to seek primary care. Simultaneously, primary care physicians (PCPs) received an alert that a CKD patient needed an appointment. Using structured checklists, PCPs addressed and controlled comorbid conditions, reconciled drug dosing and choice to CKD stage, and ordered prespecified laboratory and imaging tests pertinent to CKD. After completion of checklist prescribed tasks, PCPs referred patients to nephrology. CKD patients had multiple comorbidities and ML recognition of CKD provided a facile insight into comorbid burden. Operational results of this program have exceeded expectations and the program is being expanded to the entire health system. This paradigm of ML-driven, checklist-enabled care can be used agnostic of EMR platform to deliver value in CKD through structured engagement of complexity in health systems.
- Published
- 2022