Clinical Data Intelligence & Advanced Analytics to Reduce Drug Diversion across the Care Delivery Cycle and Drug Supply Chain in Health Systems

This SBIR project will research mechanisms to detect when Health Care WorkersHCWsin hospitals steal ordivertlegal drugs either to abuse themselves or to illegally sell to othersWe focus on HCWs in hospitals because of the alarming rates of substance abuse and diversion in hospitalswith multiple studies finding roughlyof our nation s nursesanesthesiologistsand pharmacists are currently diverting drugs in their workplacesHCWs are becoming addicteddestroying their careersjeopardizing their patientssafetyand increasingly dying from drug diversion overdosesDiversion continues even though most hospitals already lock p addictive drugs in Automated Dispensing MachinesADMsand run monthlyanomalous usagecomputer reports to try to detect diversionHospitals broadly agree these current methods have two main weaknessesData in the ADM only show part of the equationthe dispensing of the drug from the locked cabinetignoring drug administration data in the Electronic Medical RecordEMRas well as other data available in other existing hospital computer systemsMotivated diverters can game the system with falsified data entries to avoid detectionThis SBIR project will conduct research to address these two problems by building a computer system withaautomated data feeds from multiple existing hospital computer systems andbadvancedanalytics to flag potential diversion for investigationWe will test the following four hypothesesData Consolidation hypotheses and experimentation planPhaseIf we consolidate data from two systemsEMR andampADMthen we can detect diversion that would have been undetected using data only from the ADMHypothesisPhaseIf we consolidate data from five systemsEMRADMPurchasing SystemsInternal Inventory System sand Employee Time Clocksthen we can detect diversion that would have been undetected using only EMR andampADM dataHypothesisData Analytics hypotheses and experimentation planPhaseIf we create and test algorithms on blindedconsolidatedhistorical data from EMR ADMthen we can detect known cases of drug diversion that that current methods do not detectwith fewer Type II errorsfalse negativesHypothesisPhaseIf we refine and test additional algorithms using near real timeconsolidated data from the five computer systems abovethen we can detect drug diversion that current methods do not detectfasterwith fewer Type I errorsfalse positivesand fewer Type II errorsHypothesisThis SBIR Phaseproject will research mechanisms to detect when Health Care WorkersHCWsin hospitals steal ordivertlegal drugs either to abuse themselves or toillegally sell to othersDrug diversion is considered a very important and visible public health issue because of its impact on patient safetyIn additionthis research hasthe potential to reduce injuries to HCWs who are becoming addicteddestroying their careersjeopardizing their patientssafetyand increasingly dying from drug diversionoverdoses