ABSTRACT
To facilitate healthcare quality improvement initiatives, we previously developed an algorithm to identify diabetes mellitus (DM) patients using only electronically available administrative data. In this study, we have validated our prediction model, screening over 28,000 admissions and determining factors associated with false positive assignment. These factors will be incorporated into a revised algorithm.
Subject(s)
Algorithms , Diabetes Mellitus , Adult , Hospitalization , Humans , Mathematical Computing , Patient Admission , Prospective StudiesABSTRACT
Using a previously described computer algorithm to prospectively identify diabetics, we observed the adherence to published guidelines for treatment of dyslipidemia in diabetics. Despite national guidelines, dyslipidemia remains widely under-treated. Further interventions are needed to improve lipid-lowering treatment in appropriate candidates.
Subject(s)
Diabetes Complications/drug therapy , Dyslipidemias/drug therapy , Guideline Adherence , Hypolipidemic Agents/therapeutic use , Algorithms , Diabetes Mellitus/diagnosis , Female , Hospitalization , Humans , Male , Middle Aged , Practice Guidelines as TopicABSTRACT
Commercial rule bases can be implemented to identify medication orders that fall outside recommended dosage ranges, but they are likely to produce an excessive number of nuisance and clinically insignificant alerts. Strategies for customizing commercial dosing rules can be implemented to minimize this problem. This paper describes specific strategies implemented in a dose checking application necessary for achieving a clinically acceptable alert rate.
Subject(s)
Drug Therapy, Computer-Assisted , Medication Systems, Hospital , Pharmaceutical Preparations/administration & dosage , Reminder Systems , Clinical Pharmacy Information Systems , Creatinine/metabolism , Decision Support Systems, Clinical , Hospitals, Community , Hospitals, University , Humans , Medication Errors/prevention & control , Retrospective StudiesABSTRACT
Using an automated method to prospectively identify diabetic patients, we measured the impact of an administrative policy to perform LDL-cholesterol (LDL-c) testing on all diabetics not having the test performed within a specified time period. Automatic testing resulted in significant increases in LDL-c testing rate, and identified a greater proportion of patients who were candidates for statins. Further interventions are needed to increase prescriptions for lipid-lowering therapy.
