An Investigation of Errors in the Preparation of Injectable Medicines in the Pharmacy Environment and On Hospital Wards

Abstract

Introduction: The preparation of injectable medicines involves a sequence of several phases, and an error at any stage of the preparation process could cause potential or actual danger to the patient. Few investigative studies have collected data concerning the incidence, type, severity and contributory factors associated with errors in the preparation of injectable medicines in pharmacy aseptic units and on hospital wards. Aims: To determine the incidence, types and severity of errors arising during the preparation of injectable medicines within the pharmacy environment and in clinical areas of hospital across the UK; to explore pharmacy staff and nurses’ opinions of the factors contributing to preparation errors; and to propose strategies to reduce these errors. Methods: A mixed methods approach was used, comprising three stages. Stage one: direct observation of the preparation of injectable medicine in three pharmacy aseptic units (two were licensed and one unlicensed) and four hospital wards. Data were then analysed using descriptive statistics (One-way ANOVA test) to compare the findings. Stage two: a selfcompletion questionnaire was distributed to a panel of two consultant physicians, two senior pharmacists and one senior nurse. Each respondent was provided with a description of the errors previously observed in stage one and asked to independently score the severity of each on a scale from 0 (no harm) to 10 (death). Mean severity scores were mapped to consequence descriptors as follows: mean severity scores of <0.5 = negligible; 0.5-3.5 = minor; 3.5-6.5 = moderate; 6.5-9.5 = major; and >9.5 = catastrophic. Each of these consequence descriptors was then associated with a consequence score ranging from 1 (negligible) to 5 (catastrophic). The error frequency data was mapped to the NPSA likelihood grades (1 to 5) using the NPSA timeframe descriptors of frequency. A risk score was calculated for each of the types of medication errors observed, and the consequence score multiplied by the likelihood score. Stage three: semi-structured interviews (Face to Face) were undertaken to explore the opinions of pharmacy staff and nurses concerning factors contributing to injectable medicines preparation errors in pharmacy aseptic units and hospital wards. A questionnaire survey was also distributed to nursing staff working on the four hospital wards to confirm their perceptions regarding the factors contributing to injectable medicines preparation errors. A thematic analysis was then applied to the qualitative data, employing the theoretical framework outlined in Reason’s (1990) accident causation model. Results: The overall error rate for internal errors for the three different pharmacy units was 4.6% and the external error was 0.09% in the large licensed unit (A). Wrong batch numbers for starting materials on the worksheets and wrong doses were the most common errors noted. Failure to record syringe volumes on the worksheet was also commonplace at the unlicensed unit (C). The majority of these errors were judged to have a minor to moderate severity. However, after taking likelihood into account and calculating the risk score, two types of errors were graded as extreme risk, and seven types of errors were graded as high risk. Lack of staff experience, lack of training, use of look-alike/sound vi alike medicines, loss of concentration and distractions/interruptions inside the units were the factors most likely to result in an error. Poor layout of storage areas was stated as factor at the large licensed unit (A). Poor design of pharmacy computer systems was specified as a factor at the small licensed unit (B), while the heavy workload and low number of staff were specified as factors at the unlicensed unit (C). The following strategies were recommended in order of priority to minimise injectable medicines preparation errors in the three different pharmacy aseptic units: (1) effective use of computer alert systems (unit (A) & (B)); (2) improving the systems supporting the management of safe medicines (unit A); and (3) additional training of pharmacy staff at the (unit C). The overall rate of errors in the preparation of injectable medicines on the four wards was 32.4%. Disregard for a clean/uncluttered treatment room, breach of aseptic non-touch technique (ANTT), wrong addition/mixing of drug, unused gloves, and failing to double check the final product were the most common preparation errors at both hospitals. Faulty labelling and filter needles not being used as specified were common in one hospital (Wards (H) and (B)). Products being prepared in an unsuitable location (e.g. nursing station) was also common in one of these wards (Ward B). Disregard for a clean/uncluttered treatment room was specified as a factor at ward (S), while no double check for the final product was reported as factors at ward (C). The majority of the errors reported were ranked as of moderate to major severity for patients. However, after accounting for error frequency, twelve types of errors were graded as posing extreme risk. High workload with staff shortages, lack of knowledge or experience, lack of training, lack of concentration, forgetting to complete tasks, and distractions/ interruptions while in the IV treatment room were the most common contributory factors cited. Poor design/layout of the IV treatment room, lack of equipment and materials and lack of commitment or adherence to NHS Trust guidelines and policy processes were especially apparent on wards (H) and (B), while inadequate staff education were specified factors on wards (S) and (C). The following strategies were recommended in order of priority to minimise injectable medicines preparation errors in the four hospital wards: (1) improving training and education programmes (ward (S) and (C)); (2) preventing distractions/interruptions (ward (H) and (B)); (3) creating a commitment to guidelines and policies (ward (H) and (B)); (4) reporting and identifying errors (ward (H) and (B)); (5) systemising workflow (ward (C) and (B)); and (6) offering staff sufficient breaks during each shift (ward B). Conclusion: This is one of the first empirical studies to explore preparation errors in injectable medicines at three different aseptic pharmacy units and four hospital wards. The aim and objectives of the research were achieved. The results confirm injectable medicines preparation errors are prevalent in pharmacy and hospital environments and may cause severe harm to patients. Future work is essential to implement the recommended strategies and evaluate their success in practice.

Date of Award	5 Sept 2018
Original language	English
Awarding Institution	University of Bath
Supervisor	Julie Letchford (Supervisor), Matthew Jones (Supervisor) & Neil McHugh (Supervisor)