2023
Using Multi-Modal Electronic Health Record Data for the Development and Validation of Risk Prediction Models for Long COVID Using the Super Learner Algorithm
Jin W, Hao W, Shi X, Fritsche L, Salvatore M, Admon A, Friese C, Mukherjee B. Using Multi-Modal Electronic Health Record Data for the Development and Validation of Risk Prediction Models for Long COVID Using the Super Learner Algorithm. Journal Of Clinical Medicine 2023, 12: 7313. PMID: 38068365, PMCID: PMC10707399, DOI: 10.3390/jcm12237313.Peer-Reviewed Original ResearchComposite risk scoreRisk scoreElectronic health recordsAnalyses identified several factorsValidation of risk prediction modelsModerate discriminatory abilityRisk prediction modelPost-acute sequelae of COVID-19Health recordsCombined risk scorePost-acuteIdentification of individualsPrevention effortsSuper Learner algorithmMedical recordsHealthcare challengesPublic healthMedical phenotypesCOVID-19Increased riskPredictive factorsCOVID-19 infectionRecord DataPost-acute sequelaeHigh risk
2021
A Phenome-Wide Association Study (PheWAS) of COVID-19 Outcomes by Race Using the Electronic Health Records Data in Michigan Medicine
Salvatore M, Gu T, Mack J, Sankar S, Patil S, Valley T, Singh K, Nallamothu B, Kheterpal S, Lisabeth L, Fritsche L, Mukherjee B. A Phenome-Wide Association Study (PheWAS) of COVID-19 Outcomes by Race Using the Electronic Health Records Data in Michigan Medicine. Journal Of Clinical Medicine 2021, 10: 1351. PMID: 33805886, PMCID: PMC8037108, DOI: 10.3390/jcm10071351.Peer-Reviewed Original ResearchPhenome-wide association studyCOVID-19 outcomesIntensive care unitAssociation studiesNon-Hispanic blacksNon-Hispanic whitesAcademic medical centerAssociated with hospitalizationHealthcare deliveryAssociated with mortalityMedicine backgroundPre-existing conditionsMedical phenomeDisease preventionVulnerable populationsPulmonary heart diseaseTargeted screeningMental disordersCOVID-19Associated with intensive care unitMedical CenterRecord DataCare unitGenitourinary conditionsHeart disease