Improving Patient Safety by Reducing Medication Errors

Project Details


DESCRIPTION: (PROVIDED BY APPLICANT) Medication errors are among the mostcommon and potentially preventable types of medical errors, accounting formore deaths than motor vehicle accidents, breast cancer, or HIV, and an annualcost of $17 to 29 billion. The elderly are particularly at risk for suchcomplications. The highest risk drugs include anticoagulants,anticonvulsants, antimicrobials, and digoxin. The most frequent seriousadverse outcomes include bleeding and acute renal failure. Medication errorscan occur anywhere in the medication use process, including diagnosis,prescribing, dispensing, administering, ingesting, and monitoring. Sources ofmedication errors are quite varied. Among health care professionals, factorssuch as work stress, distractions, interruptions, inadequate training,fragmented information, or information overload may increase the risk ofcommitting errors, such as prescribing the wrong drug or dose, or omittingneeded action in the course of delivery of care, such as failing to properlymonitor the use of nephrotoxic drugs or anticoagulants. Among patients,factors such as advanced age, frailty, cultural or literacy barriers, mentalillness or incapacity, or lack of adequate social support may contribute topoor adherence with a specified therapeutic regimen. Poor adherence accountsfor almost a quarter of all hospital admissions attributed to drugs, and cantake the form of overuse, underuse, or erratic use of the drug.Building on its 20 years of experience studying adverse drug reactions andother medication safety problems, the University of Pennsylvania proposes aCenter of Excellence for Patient Safety Research and Practice. The proposedCenter will re-focus this large past experience on the expansion of thispatient safety knowledge base through multidisciplinary research and educationprograms that are designed to identify and implement systems approaches toreducing error in the use of medications. In particular, we propose a programthat will combine investigators in pharmacoepidemiology, health servicesresearch, biostatistics, occupational medicine, sociology, psychology, andeconomics to address a theme of "Improving Patient Safety Through Reduction ofErrors in the Medication Use Process." The program will be composed of fourprojects and four cores, based at the University of Pennsylvania and linked tothe government of the State of Pennsylvania and to the network of Centers forEducation and Research in Therapeutics. Each of the four cores will serve thefour projects, in such a way as to maximize quality and efficiencysimultaneously. Project 1 will study patient and system factors that arepredictive of hospitalizations due to dose-related medication errors amongelderly individuals taking specific high-risk drugs (warfarin, phenytoin, anddigoxin), using a State-run population-based pharmaceutical benefit program.Project 2 will study human and medical practice factors as predictors of pooradherence to warfarin therapy in an anticoagulation clinic setting created asa systems approach to prevent medication errors. Project 3 will studymedication errors as causes of preventable acute renal failure in theinpatient setting, despite the existence of a pharmacokinetic monitoringservice created to prevent such problems. Project 4 will study conditionsthat lead to medication errors among physicians, with emphasis on workconditions that increase stress, such as workload, schedules, workorganization, shifts, and patient/staff ratios.Core A will be an administrative core, responsible for coordination. Core Bwill be a data collection core, responsible for all field activities. Core Cwill be a biostatistics and data management core, responsible for data entry,management, and analysis. Core D will be a dissemination core, responsiblefor an extensive dissemination program, as the results of the program emerge.

Effective start/end date9/1/048/31/05


  • Agency for Healthcare Research and Quality


Explore the research topics touched on by this project. These labels are generated based on the underlying awards/grants. Together they form a unique fingerprint.