Use this URL to cite or link to this record in EThOS: http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.700800
Title: Evaluation of types and causes of dose administration errors
Author: Alqasoumi, Abdulmajeed Abdullah A.
ISNI:       0000 0004 5988 8913
Awarding Body: King's College London
Current Institution: King's College London (University of London)
Date of Award: 2016
Availability of Full Text:
Access through EThOS:
Access through Institution:
Abstract:
Medication incidents were reported by the National Patient Safety Agency in England and Wales as the third most common type of incidents in 2014. Incidents during the administration stage of the medication use process were reported as one of the most frequent types of medication incidents. A retrospective, quantitative study using incidents reports was performed to determine the rate of medication administration incidents (MAI) relative to capacity and bed occupancy, types of incidents, drugs involved and severity of patient harm in medication incidents reported in one UK acute NHS hospital electronic reporting system over 12 months. The results confirmed that administration incidents were the most common incident stage, accounting for 49.6% (937/1889) of incidents in all stages of medication use processes. Most common incident types were drug omission and wrong dose, frequency, or infusion rate. The highest number of incidents was from Children's Services (215/937). However the number of medication administration incidents per 1000 bed-days was highest in the Perioperative, Critical Care and Pain directorate (16.9 MAIs/1000 patient days) followed by Children's Services (3.9 MAIs/1000 patient days), then Surgery (3.8 MAIs/1000 patient days). Morphine was the most common drug involved in MAI followed by enoxaparin. Weighting MAI rates by bed occupancy may be more appropriate measure than simple frequency of reported incidents from each clinical directorate and it is a better method of prioritising interventions for improvement. Determining the causes and contributing factors of MAIs is essential to plan intervention to reduce these incidents. A semi-structure interview-based study was undertaken to investigate the views of nurses’ and midwives about factors contributing to MAIs in hospitals. Twenty-five nurses with differing years of experience and from a range of clinical directorates were included in the study. Interviews were transcribed and thematically analysed using NVIVO software. Different factors related to the work environment, task, team, personal, organisational, and patients related factors were identified. Common work environment factors reported were staffing level, skill-mix, workload, night shifts, and frequent interruption/distraction during the preparation and administration of medicines. The main task factors included inappropriate checking and non-availability of laboratory results required for medication administration. Heparin, insulin, and drugs requiring dose adjustment based on blood tests were reported as high risk medications. Patient-specific factors identified were patients with complex clinical conditions, confused, unconscious or patients with dementia. Team factors such as poor communication within teams or with doctors were also considered risks. Organisational factors identified included inadequate training, lack of feedback on medication incidents, and unclear policies /guidelines. Finally, personal factors reported included nurses’ fatigue and sleepiness particularly during long and night shifts, and inadequate knowledge, or skills about medicines. The nurse interview study identified that fatigue / sleepiness were considered by nurses to significantly contribute to MAI particularly during long shifts and night shifts. Results from an American study which aimed to assess the impact of extended shifts and fatigue on medication safety supported a possible relationship between extended work hours and the likelihood of error by nurses. However, this study relied on self-reported errors by nurses using logbooks. Therefore, the third study aimed to investigate whether nurses’ fatigue and sleepiness during long shifts (12 hours shift) have an impact on the error rate and type during IV preparation and administration. This study was based on a direct observation method using simulated tasks of IV preparation and administration. Participants completed a baseline fatigue scale (OFER) scale for acute fatigue, chronic fatigue, and inter-shift recovery) at the start of their shift and their acute sleepiness was measured at different testing points: at start of shift, 8 hours and at end of the shift (12h) using Karolinska Sleepiness Scale (KSS). Both previously validated scales were short (8 to 15 questions). Nurses also prepared an injectable medication of similar level of complexity at the start and end of their shift. Nurses were presented with a prescription in the format routinely used on the study wards. The simulated task consisted of preparation of a medicine, including, as required: selection of materials, calculations, manipulations, labelling and set up for infusion. The simulated preparations were performed in ward medication preparation areas. Deviations from hospital standards for drug preparation, errors and preparation time were recorded. Using SPSS 22, the internal consistency of all subscales in the fatigue scale was assessed using Cronbach’s alpha. The number and mean of deviations from standards and errors for each period of observation (pre and post shift) were calculated. Paired t-tests and Wilcoxon’s signed-rank test were used to determine differences in fatigue levels, sleepiness scores, deviations from standards, means of errors and time needed between periods of observation (pre and post shift). 39 nurses were observed from medical wards (n=14), surgery wards (n=12), and intensive care units (n=13). Sleepiness scores were significantly higher at the end of shift compared to the beginning of the shift. The mean number of deviations from standards (6.33) at the end of the shift was significantly higher from the mean (4.18) at the beginning of the shift. However, the number of observed errors, and time needed did not significantly differ. McNemar's test showed a significant association between being sleepy and making errors at the end of the shift. However, this association was not significant at the beginning of the shift.
Supervisor: Whittlesea, Catherine Margaret Cecilia Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID: uk.bl.ethos.700800  DOI: Not available
Share: