Accidents happen, but researching errors and near-misses in pharmacy compounding units will help improve patient safety and quality assure services for the future
PhD, BPharm, PGDPRM(Open), PGCert LTHE, MRPharmS
Reading School of Pharmacy
University of Reading
MSc, BPharm, MRPharmS
East and South East England Specialist
Estimates show that in developed countries as many as 1 in 10 patients is harmed while receiving hospital care. The harm can be caused by a range of factors or adverse events including those from medication errors. There is a plethora of evidence to show that rates of errors in prescribing, preparing and administering injectable medicines are higher than for other forms of medicines. But specialist hospital pharmacy units are considered the safest places for the compounding and preparation of sterile products, certainly compared to the ward and clinical areas. On this basis, UK hospital pharmacies have been involved in the preparation and supply of sterile products such as intravenous injections for many decades but the practice across Europe and even in the UK is variable. For example in 2001, there was strong evidence that high-risk injectable medicines were often being prepared in clinical areas in English hospitals and a later risk assessment study of injectable medicine preparation in the north of England found high-risk products, including cytotoxics, adult parenteral nutrition (PN) bags and intra-ocular injections, being prepared in clinical areas.
In fact, on the basis of such findings, in 2007 the National Health Service (NHS) body responsible for patient safety, the National Patient Safety Agency (NPSA), issued an alert for promoting safer use of injectable medicines, in which it made six recommendations to English and Welsh hospitals. The alert made the recommendations to undertake a risk assessment of injectable medicine procedures and products in all clinical areas to identify high risks, and develop an action plan to minimise them; ensure there are up-to-date protocols and procedures for prescribing, preparing and administering injectable medicines in all clinical areas; ensure essential technical information on injectable medicines is available and accessible to healthcare staff in clinical areas at the point of use; implement a ‘’purchasing for safety’’ policy to promote procurement of injectable medicines with inherent safety features; provide training for, and supervision of, all healthcare staff involved in prescribing, administering and monitoring injectable medicines; and include an audit of medication practice with injectable medicines as part of the annual medicines management audit programme. A very specific recommendation was to “prepare all cytotoxic and total parenteral nutrition (TPN) products, and make all additions to TPN, in the pharmacy department.”
There is an English proverb that to be forewarned is to be forearmed, on the basis that anticipating an event in advance may help with its better management (through planning or preparation) in due course. This is exactly the line of reasoning we took when researching and highlighting the pattern of errors and near-misses in hospital pharmacy sterile compounding units (aseptic units) in the UK. Our aim was not to overstress safety issues with pharmacy units, but we hoped that in researching errors and near misses in UK-wide pharmacy compounding units we could help identify and highlight any patterns of mistakes that do exist to help colleagues in quality assuring services in the future.
Although by their very nature accidents (for example, errors in preparing medicinal products) are unexpected and could happen without prior warning, certainly the NHS in the UK has advocated a ‘’systems approach’’ to error management in its recent history. This means that while a healthcare worker may be involved in making a mistake, rather than focussing on blaming them, it is the wider, more global structure and organisation of the workplace that merit the closer scrutiny and, if needed, modification and change. So the seven steps NHS organisations should take to improve patient safety include: building a safety culture; leading and supporting staff; integrating risk management activity; promoting reporting; involving and communicating with patients and the public; learning and sharing safety lessons; and implementing solutions to prevent harm. Globally too, the World Health Organization advocates a similar approach to improve the safety of patients worldwide.
National Aseptic Error Reporting Scheme
We used data from a scheme that has been operating in the UK since 2003. This scheme, known as the National Aseptic Error Reporting Scheme (NAERS), has been collecting data on pharmacy compounding errors via regional pharmacy quality assurance specialists and returning quarterly summary reports. In fact, we did find that overall the rate of error reported to the scheme was very small indeed when calculated as a percentage of the reported number of products made by the participating units. We found a 0.49% error rate including near-misses, which is in strong contrast to errors rates associated with the preparation of injectables on hospital wards (20%–99%).[7,8] This figure in itself could be an indication that the rigorous application of the principles of Good Manufacturing Practice and the associated Quality Management Systems within pharmacy aseptic units greatly reduce the likelihood of drug preparation errors being made and thereby deliver a true benefit in terms of patient safety. There were seven stages during which errors could have been detected. These were: first check in assembly area; operator check in preparation area; during labelling; final check prior to release; at release stage; in clinical area prior to administration; and in clinical area during or after administration. However, only 24 of the 4691 errors in our study were detected during or after administration to the patient, so the majority of the reports did indeed relate to near-misses.
Adult cytotoxic preparations and parenteral nutrition
A total of 40% of the errors that had been reported to the scheme related to adult cytotoxic preparations. This could simply be explained by high quantities of such products made by the participating units but an equally valid explanation is the sheer complexity of cytotoxic preparations that often require the handling of variable data especially for patient-specific doses. We are certain that there is a need for uniform and robust final checks to ensure that if they occur, errors with cytotoxic product do not leave compounding units. The preparation of PN solutions is also complex, and even where standardised multicompartment bags are used, compounding still involves adding a number of extra components according to patient needs. When we looked at errors reported with PN solutions it seemed that staffing levels being below establishment and workload being above planned capacity were issues thought to have had an effect on errors with these products, much more than with any other type of injecteable being prepared. Certainly in the UK many aseptic units prepare PN solutions in the last work session of the day and this could be a contributory factor.
Paediatric PN was more likely to be associated with more serious adverse outcomes compared to any other product and while this might be anticipated with any paediatric preparation, still the complexity of the compounding process and the large number of ingredients used in the preparation of such products must merit a call for ensuring robust compounding, checking and release procedures with these and indeed paediatric cytotoxic preparations. Whereas technology such as use of automated compounding systems has the potential to reduce or remove many of the risks associated with manual preparation of solutions, these systems can in themselves introduce other risks and so it is essential to ensure validation processes are robust and adequately resourced. In-process and end-product controls and testing are valuable tools in assuring quality of PN solutions, but it must be seen that there is no single check or test that can assure product quality. Instead it is the design and validation of the whole process of product compounding, checking and release that can give us a high level of assurance in the quality and fitness-for-purpose of the final product. Errors with both types of paediatric product were in fact more likely to have been detected in the latter stages of the drug use process. Indeed errors with paediatric PN preparations, although few, were detected and reported as accidents that had reached the patient more than other products were.
Types of errors
These data can also provide some guidance for directing risk management resources towards specific parts of the aseptic preparation process. It is perhaps not surprising that labelling errors were by far the largest single category in terms of error type (34.2% of all reported errors), but this clearly highlights the need to ensure robust risk management and checking controls are in place for labelling activities. The next largest category of error was associated with the transcription of information (11.1%) and there were also reports of errors with expiry dates (7.5%), calculations (5.5%), and selection of the correct drug (4.2%), strength (5.2%) or diluent (4.3%). The errors associated with selection can of course highlight the need for good segregation and checking practice in aseptic units, but also deliver a message to the manufacturers of the products handled in aseptic units to ensure packaging is well designed and that similar drugs and those with multiple strengths are clearly differentiated.
Overall, our findings could be seen as a starting point from which better error detection and reporting systems can be developed and further monitored in due course. The findings are also helpful because they set a benchmark against which future comparative studies can be conducted, in the UK and across Europe. At the time of writing this article, the future of the NPSA was under threat as the new UK government was looking at ways to make financial savings. While the NPSA has been running its own National Reporting and Learning Service, at this time of uncertainty it is a comfort to the authors that at least the NAERS, a system run voluntarily by participating pharmacists and quality assurance officers, is seemingly unaffected at least by the budgetary considerations and we would encourage colleagues across Europe to consider setting up a similar scheme for more locally relevant data. Indeed we would be interested to hear from anyone willing to share their own experience of such a scheme in another European country.
1. World Health Organization. Facts on patient safety. Available from http://www.who.int/features/factfiles/patient_safety/patient_safety_facts/en/index.html (Last accessed 18 August 2010)
2. Audit Commission. A spoonful of sugar: medicine management in NHS hospitals. London: Audit Commission. 2001
3. Hardy L, Mellor L. Hosp Pharm 2007;14:58–64.
4. National Patient Safety Agency. Patient safety alert number 20; promoting safer use of injectable medicines. 2007. Available from www.nrls.npsa.nhs.uk/ (Last accessed 18 August 2010)
5. Bateman R, Donyai P. Qual Saf Health Care. doi:10.1136/qshc.2009.034751
6. National Patient Safety Agency. Seven steps to patient safety. 2004. Available from http://www.nrls.npsa.nhs.uk/resources/collections/seven-steps-to-patient-safety/ (Last accessed 18 August 2010)
7. Taxis K, Barber N. BMJ 2003;326:684–7.
8. Cousins DH, et al. Qual Saf Health Care 2005;14:190–5.