The medication use process involves a complex interplay between many factors – people, tasks, technologies, organisation and environment – and the impact of interventions is not always clear, according to Mary Tully (Reader in Pharmacy Practice, Division of Pharmacy and Optometry, University of Manchester, UK).
Errors in the medication use process occur predominantly at the prescribing and administration stages, with only 11% of errors occurring during dispensing and preparation, she continued.
Awareness of one’s own errors can be a useful learning tool, but all too often junior doctors are unaware of their own errors and the opportunity for learning is lost. The Equip study showed an overall prescribing error rate of about 9% but this is not the perception of most junior doctors.
Audit and feedback is an effective way of improving performance. Work in Manchester that involved giving personal feedback to junior doctors and then holding group discussions showed that it was possible to normalise sub-optimal prescribing by this method. Participating doctors commented that it was useful to see that they were not alone and that “there were others in the same boat”.
Dr Tully said that there is widespread misunderstanding among nurses and pharmacists about what is meant by an ‘independent double check’ – often the importance of independence is overlooked – and this is critical. Moreover, only one controlled study has shown that it makes a difference.
People use workarounds in order to do a good job. For example, if a wristband barcode is damaged or not accessible, nurses will often stick barcodes on to bedheads so as not to impede the medication administration process. Knowing about workarounds can help planners to understand where the system is failing and could help to develop a safer system, said Dr Tully.
Tools and technology can improve safety; for example, automated dispensing systems reduce the rate of dispensing errors and smart pumps significantly reduce serious error rates.
Prescribing stationery is different in every hospital in England but poor design can increase prescribing or administration errors. One study examined effects of redesigning the prescription chart using ‘nudge’ technology and this appeared to have a positive impact, she said.
Interruptions and distractions are both associated with dispensing errors and administration errors. However, a systematic review showed that there is only weak evidence of the effectiveness of interventions to reduce interruptions and very limited evidence of their effectiveness in reducing medication administration errors.
Identifying risk points
Pharmacists need to understand the processes for medicines’ use in hospitals because it is often the processes that lead to errors, according to Matthew Grissinger (Director, Error Reporting Programs, Institute for Safe Medication Practices, US). He described a case in which a 16-year-old woman in labour had been given an epidural infusion containing fentanyl and bupivacaine by the intravenous route instead of penicillin G, and died.
At first sight this seemed to be a ‘look-alike’ packaging error but on further investigation, multiple problems were uncovered. Nurses were routinely expected to prepare medicines before the orders were written, the barcode checking system was not working and the nurse was tired and distracted.
The documentation of allergies is critical because important decisions are made on the basis of this information. Often, minor events such as gastrointestinal disturbance after antibiotic therapy are incorrectly recorded as ‘allergy’ and so it is important to have a description of the reaction that occurred, he said.
Another common problem in the US is recording the patient’s weight in pounds – sometimes on the chart in the box labelled ‘kilograms’. On occasions this value has been used to calculate a dose using the mg per kilogram figure, with consequent over-dosage.
The dangers of failing to label IV drugs that have been drawn up were illustrated by a case where vecuronium had been drawn up but was not used during the operation. Because it was not labelled, it was assumed to be a saline flush and was later used to flush an intravenous line in a three-year-old child, with fatal results.
Regarding drug storage, Mr Grissinger said, “Take a look for yourself – and always think of the process behind poor storage”.
Error reporting provides useful qualitative information but it is also important to identify areas of risk. Mr Grissinger recommended performing failure mode and effect analysis (FMEA), hospital and departmental ‘walkarounds’ to see what is happening and asking staff, “What keeps you awake at night when you go home from the hospital?” The responses to such questions can be eye-opening, he said.
In 2000, the National Patient Safety Agency said that 80% of errors are attributable to human factors at individual level, organisational level or, commonly both – little has changed since then, Annmarie Brown (Consultant in Emergency Medicine, Royal Liverpool Hospital, UK) told the audience.
Errors and failures are inevitable in any complex system and so awareness of the risk is critical. Some systems predispose to errors, for example, 60-page guidelines that make it impossible to find information quickly and adverse environmental conditions, such as when care has to be delivered in hospital corridors because there is no room in the Emergency Department. Unsafe acts can be divided into unintended actions, such as slips and lapses and intended actions such as violations and transgressions. Slips and lapses occur when people are hungry, angry, late or tired – known as HALT, as an aide memoire.
Most NHS staff are affected by these conditions for much of the time, commented Dr Brown. Violations and transgressions are often workarounds that improve efficiency and so it is important to understand why they are used. Corrective actions that make it difficult to do the wrong thing are strong (effective) interventions whereas retraining and the issue of new policies or procedures are weak (often poorly effective) interventions.
Patsy Tipene (Critical Care Senior Nurse, the Queen Elizabeth Hospital, Kings Lynn, UK) described how after a critical report from the Care Quality Commission, a simple device – the DruGuardian – had been installed in her hospital to improve the security of medicines. The device incorporates a light sensor, a camera and an alarm; it records the time when, for example, the controlled drug (CD) cupboard is opened and an image of the individual concerned.
If the cupboard is open and unattended for more than two minutes, an alarm is triggered. At her hospital, DruGuardians were installed in all CD cupboards in the operating theatres. Over a two-week period, the CD cupboard was accessed without authority on five occasions and each time a picture of the person concerned was captured. She concluded that DruGuardian offers a simple but effective way to detect and deter tampering and drug theft in hospitals.
Advanced therapy medicinal products
“In 10 years’ time advanced therapy medicinal products (ATMPs) could be routine in pharmacy”, Anne Black (Assistant Director of Pharmacy – Quality Assurance, Newcastle-upon-Tyne Hospitals, UK) told the audience. Three types of ATMP are used at present: gene therapy products; somatic cell therapy products; and tissue-engineered products.
All of these are technically medicines but because they have not come through the traditional route for medicines, they raise many questions for pharmacists and appropriate handling will be important, she explained. Three ATMPs already have marketing authorisations in Europe and it is likely that pharmacists will encounter others as advanced therapy investigational products (ATIMPs).
ATMPs are governed by a number of regulations, including blood products and tissue culture regulations in addition to medicines’ regulations. For example, the starting material may be a patient’s blood that is collected, sent to the manufacturer for processing and then returned for administration. It will be important to have robust governance and logistics arrangements in place to ensure safe handling of these products, many of which are cryopreserved in liquid nitrogen and have very short half-lives once thawed.
In future, hospitals may have dedicated cellular therapy units with expert staff who are trained to handle an administer ATMPs, said Ms Black. The chief pharmacist is accountable for the use of ATMPs and so pharmacists should take steps to find out if ATMPs are in use in their hospitals, she advised.
Dose-banding for chemotherapy was started at the Royal Marsden Hospital as part of the electronic prescribing project, Richard Nuttall (Electronic Prescribing Pharmacist, Royal Marsden NHS Foundation Trust, UK) told the audience. One method does not suit all drugs and a scheme was eventually developed that provides volume-based doses for inexpensive drugs, but doses based on both volume and vial size for expensive drugs. ‘Pick and mix’ syringes then allow the dose to be made up from pre-mixed units – typically only two or three are needed, he explained.
Dosing according to body surface area (BSA) is a “deeply flawed” concept, said Mr Nuttall. It was originally assumed that BSA was a good predictor of drug clearance but in one study of 33 drugs, BSA-based dosing reduced variation in clearance in only five drugs, and even then by only 15–35%. It will take time to move to a more rational approach, such as dosing based on lean body mass, because the use of BSA-based dosing is so deeply rooted, he acknowledged.
At the Royal Marsden Hospital, it was decided to allow variation of ±6% of the calculated dose for chemotherapy drugs and ±10% for monoclonal antibodies. A recent study with 5-fluorouracil showed that interpatient variability in drug handling has a large effect and would easily mask any differences that dose-banding might introduce.
A study with pembrolizumab showed that in spite of mg/kg dosing there was wide variation in the areas under the curve (AUCs – a reflection of the dose to which the body is exposed). However, the AUC made no difference to the overall response rate. In addition, mg/kg dosing of nivolumab gave similar AUCs to banded dosing.
Dose-banding is cost-efficient for the Pharmacy department, reduces the potential for occupational exposure for all staff and reduces waiting times for patients. One hospital had reported using 99 different doses of paclitaxel before dose-banding but after the introduction of dose-banding, seven (out of 28) doses accounted for 85% of all doses given.
A nationally agreed dose-banding scheme for adult systemic anti-cancer therapy has now been agreed (see www.england.nhs.uk/) (for England). A patient information leaflet explaining dose-banding has also been compiled.
In future, the NHS needs to leverage its buying power and agree on presentations for commercially manufactured banded doses of injectable chemotherapy agents, concluded Mr Nuttall.
An incident in which a patient became upset when he realised part-way through his chemotherapy that it was not going to be curative was one of several episodes that prompted the establishment of a multidisciplinary chemotherapy consent clinic, Carl Booth (Lead Pharmacist for Cancer Services and Clinical Trials, Airedale NHS Foundation Trust, UK) told the audience.
At the clinic, patients have the opportunity to discuss all aspects of the planned treatment with a nurse, a doctor and a pharmacist, and baseline checks and measurements are made. Mr Booth’s role is to prescribe the treatment in accordance with the oncologist’s plan.
The consent clinic prepares patients for their first visit to the oncology day unit – and it reduces the likelihood of patients having last-minute second thoughts and drugs being wasted as a result, he explained.
Another part of his work is in pharmacist-led chemotherapy review clinics, following a model developed in Nottingham.
Clinics have been established for prostate cancer patients receiving abiraterone and enzalutamide and also for patients receiving adjuvant treatment for breast cancer. In the first year it was estimated that 159 hours of medical time were saved, saving nearly £29,000 for commissioners and generating income of more than £67,000 for the Trust. The feedback from patients and colleagues has been very positive, said Mr Booth
The B Braun Medical Changing practice to improve safety symposium was held at the International Conference Centre in Birmingham, UK, on 12 October 2016