The complexity of late medication errors
- UNCG Author/Contributor (non-UNCG co-authors, if there are any, appear on document)
- Carey J. Estes (Creator)
- Institution
- The University of North Carolina at Greensboro (UNCG )
- Web Site: http://library.uncg.edu/
- Advisor
- Nancy Hoffart
Abstract: The medication administration process is complex and frequently leads to errors. Medication errors have a global impact of over $42 billion annually, and in the US an impact of over $21 billion annually. Medication errors have been researched for over 20 years following the Institute of Medicine’s landmark study, To Err is Human, but they continue to increase. The purpose of this study was to evaluate contributing factors to late medication administrations (LMAs). Complexity theory guided the study design and data analysis, supporting the wide array of factors that have been shown, individually, to contribute to medication errors and the inter-reliant system structure of the medication administration process. A six-hospital system was the setting for the study. Descriptive statistics and multilevel Negative Binomial regression modeling was performed to model relationships among variables. Three levels of nested predictor variables were tested in the modeling: shift characteristics were nested within nurse characteristics, which were nested within unit characteristics. Shift characteristics were time of shift (day or night) and presence of a permanent charge nurse. Nurse characteristics were years of experience, highest degree obtained, full-time equivalent status, and specialty certification. Unit characteristics were patient population, unit size, nurse manager years of experience, and nurse manager specialty certification. Results showed that registered nurses working on units with intensive care unit (ICU) patient populations had higher average count of LMAs when compared to nurses working with patient populations on medical-surgical, stepdown or mixed units, after controlling for all other predictors in the model and nurse and unit clustering. Nurses who had earned an associate’s degree were found to have higher average count of LMAs when compared to bachelor’s prepared nurses, controlling for all other predictors in the model and nurse and unit clustering. Shifts that had a permanent charge nurse had a higher average count of LMAs when compared to shifts that were staffed with a relief charge nurse and controlling for all other predictors in the model and nurse and unit clustering. Both individual nurse and unit characteristics appear to influence the occurrence of LMAs on nursing units and the use of multilevel regression modeling mirrors the inter-reliant concept supported through complexity theory and nested structure frequently found in healthcare.
The complexity of late medication errors
PDF (Portable Document Format)
778 KB
Created on 5/1/2022
Views: 169
Additional Information
- Publication
- Dissertation
- Language: English
- Date: 2022
- Keywords
- Late Medications, Medication Errors, Missed, Omitted
- Subjects
- Medication errors
- Nursing errors