PJ Online: Broad Spectrum
Is self-checking (by anyone) a risk too far?
The publication by the Department of Health of the document An organisation with a memory was a key moment in British health care, when the culture of no blame and an appreciation of the importance of systems failure analysis gained governmental recognition. The more recent Building a safer NHS describes the plan for the establishment of the National Patient Safety Agency, the need to develop a learning culture from errors, the need for further research, and improved systems of investigating adverse events. One of the key targets for action is to reduce by 40 per cent the number of serious errors in the use of medicines. Against this background, it is a welcome development that the Royal Pharmaceutical Society is consulting its membership about standard operating procedures for dispensing.1 More specifically, the Society is asking if the current recommendation that non-pharmacists should not carry out self-checking of the accuracy of their dispensing should be revoked.
It has been reported that community pharmacists spend 37 per cent of their time on routine dispensing and many hospital pharmacies have had to restrict service developments because of the workforce crisis. The advent of robotic dispensing systems and electronic prescribing or transmission of prescriptions might free pharmacists in both sectors, as well as reduce the opportunity for some types of error. However, to use the skills of pharmacists for direct patient care, technical staff will have to take on some of the roles and associated responsibilities currently occupied by pharmacists.
In recent years, large numbers of technicians have been trained and accredited as checkers, and, although little UK research exists, North American studies have shown technicians make no more or even fewer errors than pharmacists.2 Therefore the argument that the Societys guidance on technician self-checking should be relaxed is persuasive. However, no matter how well trained an individual is, he or she will still make or fail to detect errors. So, whether or not the final accuracy check is made by a technician or a pharmacist is perhaps less important than the question, Is self-checking a safe system of working?.
James Reason, professor of psychology, University of Manchester, describes the system approach to human error emphasising that defences, barriers, and safeguards occupy a key position in the system approach. High technology systems have many defensive layers: some are engineered (alarms, physical barriers, automatic shutdowns, etc), others rely on people (surgeons, anaesthetists, pilots, control room operators, etc), and yet others depend on procedures and administrative controls. Their function is to protect potential victims and assets from local hazards. Mostly they do this very effectively, but there are always weaknesses.3
The analogy he uses is that each layer of defence should be thought of as a slice of Swiss cheese with its associated holes. Any given system will have a number of slices of cheese with holes in varying positions. An error has to find a trajectory through the holes in each layer of cheese to reach an undesirable outcome, such as a patient taking the wrong medicine. The more layers of cheese, then the more barriers an error has to overcome, and the less likely is an adverse outcome.
Independent checks do act as a defence against errors. Research in the UK has shown that hospital pharmacies without an independent check on pharmacist dispensing have a statistically significant higher rate of errors (35 per 100,000) than those departments with a second check on all work (11.5 per 100,000).4 There is no equivalent study for community pharmacy, but a small study in four community pharmacies showed an internal error rate of 1 per cent (50 errors, 5,004 items).5 Nine of these errors (0.18 per cent) were considered to be serious with the potential to require significant medical intervention, although none left the pharmacy. It is not clear from this study if an independent check was in place, but it does demonstrate the potential for serious adverse events, which might be prevented by the additional defence of an independent check.
There is little work on the cognitive processes involved in self-checking or the best system to reduce errors when self-checking. A small qualitative study on the process of checking did uncover concerns that pharmacists had about self-checking of their dispensing6 Self-checking is more difficult; it is harder to reread objectively; The essentials of own work checking are bound to be fraught; It differs whether I am checking myself or others. A system of dispensing that incorporates an independent accuracy check may be less likely to break down under the pressure of work than an attempt to separate dispensing and checking internally by an individual practitioner. Also, a self-checker will have already made the error that he or she has to detect, which may blind them to its presence. The brains potential to see what it expects to see and fill in missing information, known as confirmation bias, is also an important consideration.
The National Co-ordinating Council for Medication Error Reporting and Prevention, an independent body formed in 1995 in the US, recommends that a series of checks be established to assess the accuracy of the dispensing process. Whenever possible, it recommends an independent check by a second individual. The Societys proposed guidance, also states there should be an accuracy check within the procedure and that wherever possible the check should be undertaken by a second person.
Higher error rates?
Self-checking is a necessary way of working when single-handed, but this should not be used as an argument for the expansion of self-checking within normal working hours. The benefit of supplying an urgent drug on-call may negate the additional risk of self-checking, but during normal working hours should self-checking be avoided? It has been suggested that the re-engineering of the hospital service requires the use of technicians as self-checkers. Assuming technicians are as good, or as bad, as pharmacists, then in hospitals there is the possibility that a higher rate of errors might have to borne if increased amounts of self-checking are undertaken.
A parallel can be drawn with patient group directions. The Crown report recommended that most patients would continue to receive medicines on an individual patient-specific basis, but that new groups of professionals would be able to apply for authority to prescribe in specific clinical areas, where this would improve patient care and patient safety could be assured. Perhaps for self-checking we could say: The majority of patients should continue to receive medicines which have received a independent check, but that registered technicians and pharmacists will be able to self-check in specific clinical areas, where this will improve patient care and patient safety can be assured.
1. Royal Pharmaceutical Society of Great Britain. Consultation on SOPs for dispensing. Pharm J 2001; 266:616?9.
Anthony Cox is a senior pharmacist at City Hospital NHS Trust, Birmingham, and ADR pharmacist at the West Midlands Centre for Adverse Drug Reaction Reporting, Birmingham. John Marriott is senior lecturer in pharmacy practice at Aston University, Birmingham
Citation: The Pharmaceutical Journal URI: 20004446
Recommended from Pharmaceutical Press
Pharmaceutical Press is the publishing division of the Royal Pharmaceutical Society, and is a leading provider of authoritative pharmaceutical information used throughout the world.Visit rpharms.com