Analytics-Based Platform for Diabetic Retinopathy Care Management

ABSTRACT Despite the overefficacy of diabetic retinopathyDRtreatmentit continues to be the leading cause of blindness in working age AmericansOvermillion adults in the US have diabetesOf thesewill develop DR andwill develop vision threatening DRVTDRCurrentlyless than half of all patients with diabetes receive the recommended annual dilated eye exam andof those diagnosed with VTDRonlyundergo timely treatmentNew screening protocols are increasing screening effectivenesshowevereven if every patient is screenedthe majority of those at increased risk do not receive evidence based follow up carea classic public health failure common to population based screeningWith expertise in health systems engineering and ophthalmologythe research team is uniquely suited to eliminate the root causes of this failureresulting in a newcomprehensive care model designed to prevent blindness for themillion Americans with DRRetinal Care IncRCIs focus is on eliminating blindness by applying targeted care coordination for patients at increased risk for VTDRensuring they progress through the care path to treatmentHowevertwo significant deficiencies weaken RCI s ability to deliver effective care coordinationThe first is insufficient capacity due to coordinating care forof all patientsthough less thanactually have VTDRThis derives from the low positive predictive value of the current systembased on a handheld electroretinography and pupillography device deployed in a primary care settingThe first aim is to improve the ability to classify a patient as lowrisk for VTDRthereby reducing the care coordination burden and improving quality and effectivenessAimImprove the ability to accurately identify patients at increased risk for VTDR by including patienthealth and demographic attributes in a machine learning based predictive algorithm for VTDR diagnosisPatient clusters will be identifiedaand used to create sub population specific risk modelsbThe second deficiency is the lack of adherence to follow up and treatment after identification as increased risk for VTDROnce a patient is identified as increased riskthe care coordination of the RCI DR program has averaged over five phone calls per patient to achieve follow up with an eye care providerThe second aim is to identify and eliminate these barriers to coordinated care between patient and providerAimApply text analytics and simulation to improve accesscompliancecostand equity by enhancing care coordinationfrom diagnosis to treatmentfor patients at increased risk for VTDRThe application of data analytics and systems engineering methods to integrate and improve each critical component of the care pathaccess to careaccuracy of diagnosisand adherence to follow up and treatmentwill enable a paradigm shift in DR treatmentRCI s platform is designed to eliminate deficiencies of the current care delivery model by delivering a first of its kindend to end solution for diabetic retinopathy care PROJECT NARRATIVE Diabetic retinopathyDRaffectingmillion Americansis the leading cause of blindness in working age Americansyetblindness is preventable inof the cases with timely detection and treatmentTo reach themillion adults living with diabetes in the USthe Retinal Care DR platform is designed to eliminate deficiencies of the current care delivery model via a first ofits kind end to end solution for DR care implementable in any care settingThis will be accomplished via the application of data analyticsmachine learning and statistical analysisand systems engineeringsentiment analysis and simulationmethods to integrate and improve each of the three critical components of the care pathaccess to careaccuracy of diagnosisand adherence to follow up and treatment