Rapid Responses to:
|
|
Rapid Responses: Rapid Responses published:
-
![[Read Rapid Response]](/icons/misc/sectionDOWN.gif)
Color coded labels of urgent care medications are useful
- Eckehard A E Stuth, Milwaukee, Wisconsin 53226 (19 January 2002)
-
Reading the label!
- Subrata K. Basu, PO6 3LY (20 January 2002)
-
Remember Aviation
- Edward S Cooper (21 January 2002)
-
Why not use our eyes?
- Alon Duby, Bromley, Kent, BR2 9AJ (21 January 2002)
-
Reporting adverse incidents is not whingeing
- Miles D Witham (22 January 2002)
-
A WOLF IN SHEEP'S CLOTHING
- SIMON BIRCH (22 January 2002)
-
Correcting a misperception
- Anthony E. Pickering, Chris R Monk (24 January 2002)
-
Humans are fallible machines
- Anthony R Cox (28 January 2002)
-
Re: Multicolured labels of urgent care medications are useful
- John R Davies (30 January 2002)
-
Unhelpful Dogma
- Craig S. Webster (16 April 2002)
-
Re: Re: Multicolured labels of urgent care medications are useful
- Craig S Webster (18 April 2002)
-
Colour blindness, presbyopia, and bad lighting in dark dorms.
- Phillip J. Colquitt (19 April 2002)
-
Don't think of zebras
- Craig S Webster (27 April 2002)
|
|
|||
|
Eckehard A E Stuth, Associate Professor of Anesthesiology Medical College of Wisconsin, Milwaukee, Wisconsin 53226
Send response to journal:
|
We read with interest professor Wildsmith's letter concerning medication labels. While it seems obvious that in a perfect world everybody will carefully read the labels on the drugs about to be injected, the recurrent theme of medication errors especially in urgent care environments suggest that system improvements are necessary. To our knowledge color coded syringe labels for all commonly used anesthesia medications are standard in most, if not all US operating rooms, and in our experience this practice can significantly decrease the incidence of medication errors. The color coding is by substance class, i.e. muscle relaxants have one uniform orange color label, similarly inotropes (purple), induction agents (yellow), reversal agents, etc. In summary, at least in the anesthesia environment color coding makes sense, is cheap and protects the patient. |
|||
|
|
|||
|
Subrata K. Basu, Staff Grade Anaesthetist Q.A.Hospital,Cosham,Portsmouth,, PO6 3LY
Send response to journal:
|
Dear Sir, There is no doubt that every doctor should read labels of the drug ampoule/vial/container before using it.Respected Dr.Wildsmith has highlighted this universal rule.But does the history of development and practice of Anaesthesia follow this rule? To mention a few: Why the shape of the knob of the oxygen flowmeter in the anaesthetic machine is different compared to the fellow knobs? Why the anaesthetic vaporisers and cylinders are colour coded? Shape and colour definitely helps in alerting and identifying especially when they are in accordance with a universal standard.They should probably be seen as an adjunct rather than an alternative to "reading the label". In this regard,some important drugs e.g.suxamethonium, should probably have a standard universal labelling used nationally (! may be internationally ),to reduce the chances of error/confusion.Human error could probably only be reduced to a minimum,not possibly be eliminated altogether. Thanking you, Dr.Basu. |
|||
|
|
|||
|
Edward S Cooper, Consultant Pediatrician Newham General Hospital
Send response to journal:
|
Professor Wildsmith is right that we should read the label. But still, there is a stage beyond that sort of remark where pilots have already gone but the doctors have not reached yet. For example, in the Staines Trident ("Papa India") crash of many years ago, there was quite a long chain of causation: the captain was in a foul temper and had undiagnosed atherosclerosis, this combination led to a coronary occlusion at the moment of take-off, the rather inexperienced and average-skilled No. 2 pilot suddenly had to take over; he was not quite ready; he retracted the leading-edge flaps prematurely; the stall warning went off; it was ignored because it was just one more signal in the confusion and difficult to process cognitively; the aircraft stalled and fell. Professor Wildsmith might well say that a pilot should not retract the flaps at insufficient airspeed: quite right, but it did happen and the aeroplane did crash. The outcome of the investigation was not to blame the pilot but to modify the controls so that it was no longer possible to retract the flaps prematurely. Pilots did not lose skill through this "nanny" manoeuvre, they just became a fraction safer. And so on, and so on, and so should we be. |
|||
|
|
|||
|
Alon Duby, SpR A&E Bromley Hospital, Bromley, Kent, BR2 9AJ
Send response to journal:
|
Professor Wildsmith’s letter (1) is an unfortunate example of how we sometime fail to grasp an opportunity to improve when one is offered. Of course, it is true that medical personnel should always read the label. But as in just about every other aspect of life, using visual information in the form of shape and colour helps the human brain grasp information more efficiently. This is why road signs have colour patterns which alert us as to which are more important, even though we still have to read them. And I am sure professor Wildsmith would take longer to find his car in the parking lot if all cars were the same shape and colour, despite all having clearly readable number plates. From my limited experience in anaesthetics and in other aspects of practice, when I open the drug cupboard to look for a box of something, my eyes settle on the box that looks the shape and colour of what I remember it to be. THEN I check what it says on the box, but some time has already been saved. If I were not to read the label, identical looking boxes would not help my patient. As for increasing the costs of manufacture, which in my opinion is a lesser issue than safety anyway, he is mistaken again. The labels mentioned by Nduka & Leff (2) are white with yellow and black print. Without adding any other colour, the manufacturer could have added a black & yellow stripe pattern, yellow print on a black background, black on yellow, white on black, etc. And, although not every drug must be visibly different from all others, logic dictates that drugs which (as in this case) are stored together, are visibly different. Alon Duby
(1) Wildsmith JAW. Doctors must read drug labels, not whinge about them (letter). BMJ 2002;324:170 (19 January) (2) Nduka C, Leff D. Medical mishaps: mistaken identity. BMJ 2001;323:615. (15 September) |
|||
|
|
|||
|
Miles D Witham, Clinical Research Fellow Section of Ageing and Health, Ninewells Hospital and Medical School, Dundee DD1 9SY
Send response to journal:
|
EDITOR – I was disappointed to read of Wildsmith’s attitude to the problem of errors involving lignocaine and saline ampoules[1]. I do not believe that his denial of a labelling problem, or his suggestion that clinicians are absolving themselves of responsibility by drawing attention to this problem, are likely to improve error rates. Three points stand out: Firstly, it is abundantly clear that the current system for preventing the administration of the wrong substance from an ampoule is inadequate. We know this because these errors happen[2,3]. Whilst ambiguous labelling is unlikely to be the only problem in this system, it is clearly part of the problem Secondly, simply telling people to read a label is unlikely to change their behaviour, regardless of the consequences of not doing so for the patient or for the individual concerned[4]. Indeed, continually remonstrating with a group of people who are trying to do their best and are not being wilfully negligent is likely to reduce morale and increase stress. Thirdly, having established that telling people to do something is not a useful way to proceed, we need to find ways of changing the systems in which we all work. Changes to labelling, storage and handling procedures have been shown to reduce error rates in similar situations, e.g. changes to the packaging and storage of strong potassium chloride[5]. Such changes do not have to be limited to colour coding. Wildsmith notes that the responsibility will always lie with the user to read the label before use. I agree with this, but as the user is often the most fallible component of a system, we should be aiming to give the user a helping hand to avoid error, rather than whingeing about the fallibility of humans. Wildsmith’s anaesthesiology counterparts in Australia have shown the usefulness of adverse incident reporting in anaesthesia[6], and we should promote wider adverse incident reporting rather than castigate those who report such incidents. Reference List 1. Wildsmith JAW. Doctors must read drug labels, not whinge about them. BMJ 2002;324:170 2. Nduka C,.Leff D. Medical mishaps - Mistaken identity. BMJ 2001;323:615 3. Malhotra A, Matson M, Chan O. ...and packaging of lignocaine must be changed. BMJ 2001;322:548 4. Reason J. Human error: models and management. BMJ 2000;320:768-70 5. Cohen MR. Drug product characteristics that foster drug-use-system errors. Am.J.Health Syst.Pharm. 1995;52:395-9 6. Webb RK, Currie M, Morgan CA, Williamson JA, Mackay P, Russell WJ et al. The Australian Incident Monitoring Study: an analysis of 2000 incident reports. Anaesth.Intensive Care 1993;21:520-8 |
|||
|
|
|||
|
SIMON BIRCH, SpR PAEDIATRICS ROYAL HAMPSHIRE COUNTY HOSPITAL, WINCHESTER, SO22 5DG
Send response to journal:
|
Professor Wildsmith is obviously correct that doctors should read drug labels - just as they should not administer intrathecal Vincristine or anoxic gas mixtures. Surely, thinking that errors can be simply avoided will only make them more likely. Often tragically, instant visual cues have been fundamental to recognition and classification throughout history. A single colour can advertise a specific cigarette brand. Perhaps we could package all drugs identically (other than the label text) or in certain colour codes according to class. What we have with lignocaine, however, is a dangerous drug packaged in a way many would associate with hundreds of simple 'flushes' they have given. Journal articles provide can provide a forum for raising such safety concerns without being alarmist or allocating individual blame. If this is regarded as a 'whinge', then we will perpetuate the 'club culture' which will force external bodies to increasingly dictate our practice |
|||
|
|
|||
|
Anthony E. Pickering, Lecturer Sir Humphry Davy Department of Anaesthesia, United Bristol Healthcare Trust, Bristol, BS2 8HW., Chris R Monk
Send response to journal:
|
Editor - Wildsmith in his recent letter (1) referring to the risk of mistaking ampoules of lignocaine for saline (2-5) takes the opportunity to remind us that it is vital for clinicians to read the label before administering drugs. This is self evidently excellent advice. The lignocaine/saline swap is a recurrent error as illustrated by a recent study in which we documented five instances over the last year where anaesthetists in the South-West administered lignocaine instead of saline as a flush (summary report available at http://www.ubht.nhs.uk/anaesthesia/Clinical_Audit/drug_error/sld001.htm). It is not however reasonable to conclude that this type of drug error occurs simply because the doctors concerned "did not read the label". Rather it should be made clear that doctors are at risk of "incorrectly reading the label". They and many others (ourselves included) have fallen victim to misperception. Reading is not a simple and infallible process. What we perceive is enormously influenced by what we expect to see. When reading we do not scan in individual pixels, assemble first letters, then words and subsequently interpret meaning. Rather we recognise patterns and word shapes and our interpretation depends crucially on context. FOR EXAMPLE IT IS KNOWN THAT IT IS MORE DIFFICULT TO READ TEXT IN CAPITALS BECAUSE WORDS LOOSE THEIR DISTINCTIVE SHAPES. Furthermore there are a number of famous literary illusions that expose the process of language. E.g. I Love Paris in the the springtime. We typically establish context with our evolutionally ancient skills of object recognition to synthesise a coherent internal model of the world. The more recent development of language is superimposed onto this model. In the specific case of ampoules the drug is selected and brought into close visual proximity because of its location, shape, colour and associated features. So we often establish a preconception about the identity of the drug before any reading has occurred. With this in mind it is vital that we give ourselves the best chance to avoid misperceptions in drug administration. A quick glance round any anaesthetic room will reveal many potent drugs packaged in small ampoules (requiring dilution) that have a similar appearance (e.g. adrenaline, atropine, morphine, ephedrine). The labels of such ampoules can have the key information (generic drug name and strength) printed in text of six point or less (half the size of that in this letter). Leaving aside the difficulties experienced by presbyopic colleagues, a six point font is the size of the fine print in legally binding contracts (not intended to be read!). The obvious similarities in the names of these drugs contrasts with their wildly different pharmacological actions. The risk of misperception is clear and is further emphasised by our finding that ampoule swaps between these four drugs alone resulted in 16 anaesthetists giving the wrong drug in our region last year. They did not make the mistake because "they did not read the label" rather they read it and saw what they anticipated. The issue of using colour coding and shapes to assist drug recognition is controversial with polar views being expressed. However it is worth noting that the pharmaceutical industry is united in the value it places on branding with distinctive colour, shape, labelling and packaging being deployed at considerable expense to encourage product loyalty. There is a clear conflict here as there are many instances where the similar manufacturer's branding of dissimilar drugs has resulted in drug errors. Surely we should either condemn all use of distinctive visual cues in drug packaging (and retreat to a monochrome world stripped of variety) or we should usefully codify and standardise these visual cues to introduce hazard warnings and perhaps drug class recognition markings. Our current halfway house position is lamentable and even dangerous. Unfortunately the recent deliberations at the Medicines Control Agency about drug packaging (http://www.mca.gov.uk/inforesources/publications/mlx275report.pdf) do not seem likely to result in a radical shift in the guidance issued to the pharmaceutical industry, perhaps an opportunity is being missed. As clinicians we must and do take considerable personal responsibility for the safety of drug administration. It is to be hoped that the introduction of the National Patient Safety Agency will help us to collate "whinging" critical incident and near miss reports to enable the recognition of the common sources of clinical error. Hence facilitate the design of solutions to reduce the systematic factors that currently underlie many mistakes. We fear the response of Wildsmith perpetuates the culture of blaming the individual for poor practice rather than identifying them as the unwitting victim of a well-documented system flaw in product presentation. A.E. Pickering
C.R. Monk
Sir Humphry Davy Department of Anaesthesia, United Bristol Healthcare Trust, Bristol, BS2 8HW. 1. Wildsmith JAW. Doctors must read drug labels, not whinge about them. BMJ 2002;324:170. 2. Dawes R, Vanner R. Minerva picture. BMJ 2001;322:308. 3. Ferner RE. Misleading drug packaging. BMJ 1995;311:514. 4. Malhotra A, Matson M, Chan O. ... and packaging of lignocaine should be changed. BJA 2001;322:548. 5. Nduka C, Leff D. Medical Mishaps: mistaken identity. BMJ 2001;323:615. |
|||
|
|
|||
|
Anthony R Cox, Adverse Drug Reaction Pharmacist West Midlands Centre for Adverse Drug Reaction Reporting, Birmingham, B18 7QH
Send response to journal:
|
Professor Fieldman is correct in stating that professionals have a responsibility to read drug labels carefully, but this should be coupled with attempts to make labelling clearer and packaging of pharmaceuticals distinctive. Humans are fallible machines, and will occasionally make wrong responses to information presented. The incorrect reading of a medicine label by health professionals cannot be avoided merely by telling people to concentrate more on the job in hand. Slips and lapses, which account for many medication errors, are neither fully amenable to training or threats, since they occur at an unconscious level.(1) Several thought processes conspire to make human errors inevitable. People make sense of the world by matching inputs to their existing knowledge – ‘similarity matching’.(2) If a similar looking product is used, it may be matched to a previously used product. More frequently occurring inputs in a particular circumstance may be incorrectly chosen - ‘Frequency gambling’. For example, perceiving a similar looking lignocaine ampoule as the expected Sodium Chloride 0.9% ampoule when administering an intravenous flush. Decisions based on limited data, such as the initial glance at pharmaceutical packaging, may interfere with the later interpretation of improved information upon reading the label – ‘confirmation bias’.(2) Once the mind has incorrectly perceived the lignocaine as sodium chloride 0.9%, it does not wish to discount this likelihood. Similarities in packaging give signals re-enforcing the initial incorrect identification of product. Professor Wildsmith points out the paucity of shapes that can be used to hold fluids, but does not take into account that properties of the fundamental shape, such as texture and dimensions can be used to make ampoules distinct from one another. Although colour coding has deficiencies, the careful and judicious use of colour to help differentiate between products, as the recent Committee on the Safety of Medicines Working Group on Labelling and Packaging of Medicines (WGLPM) suggested,(3) may provide visual cues to prompt detection of an error. The CSM working group also recommended the use of the largest font size possible for critical information on medicine labels and that text should be black on a white background to improve legibility. Existing text does not always make best use of the space available and may have coloured background printing. Product packaging and labelling should serve only one purpose, the clear unambiguous identification of the drug. By all means, encourage professionals to read labels, but let us encourage manufacturers to make it easier for them to do so safely. Anthony R Cox Competing interests: Mr Cox was a member of the Working Group on Labelling and Packaging of Medicines. 1. Ferner RE. The Primrose path – errors in prescribing and giving medicines. Drug Information Journal 2001;35:633-638 2. Reason J. Human Error. New York: Cambridge University Press, 1990. 3. MLX 275: Report to the Committee on Safety of Medicines from the Working Group on Labelling and Packaging of Medicines. Committee on Safety of Medicines. London. 2001. |
|||
|
|
|||
|
John R Davies, Consultant Anaesthetist Royal Lancaster Infirmary
Send response to journal:
|
Sir, Of the labels in my anaesthetic room, relaxants are blue, inotropes are red, induction agents are white and opiates are orange. Of the drugs, co -amoxoclav goes a lovely pink then fades to nothing, meloxicam is a violent yellow and as for methylene blue... The gases have no colour, and although they have coloured labels and anaesthesia only uses four, the colours will be different if I go to Europe, let alone North America. Human society communicates by language and writing, not colours. Please READ the label. |
|||
|
|
|||
|
Craig S. Webster, Research Fellow Green Lane Hospital, Auckland 1003, New Zealand
Send response to journal:
|
It astounds me that Prof. Wildsmith can apparently believe that the best way to improve rates of drug administration error in medicine is to simply exhort doctors to read the label. His opinion not only flies in the face of an overwhelming body of evidence from human factors and safety theory in many other high-risk industries, but is also incompatible with recent major reports on safety in medicine itself ("To Err is Human" by the US Institute of Medicine and "An Organisation with a Memory" from the NHS). In discussing the present state of safety in medicine, the Institute of Medicine report clearly states: "The status quo is not acceptable and cannot be tolerated any longer", and "Health care is a decade or more behind other high-risk industries in its attention to ensuring basic safety". The vast majority of doctors who make mistakes are attempting to carry out their jobs in a conscientious manner. They do no make mistakes because they are careless. A recent multi-centre study estimates that the average anaesthetist makes approximately 7 drug administration errors a year (1). It is therefore very likely indeed that Prof. Wildsmith has also made at least this kind of error in his own practice (presuming he administers any kind of drug). The fact of the matter is that error is a statistically inevitable concomitant of being a human being. Redesigning the existing error-prone systems within which doctors work is the most effective known way to reduce error below current levels and to increase patient safety. Exhortation to be more careful is, ironically, the best way to ensure that errors keep occurring, because it directs attention away from the faulty and error-prone hospital systems which badly require improvement. It is time medicine got over its denial of the safety problem and did something about it. (1) Webster CS, Merry AF, Larsson L, McGrath KA, Weller J. The frequency and nature of drug administration error in anaesthesia. Anaesth Intensive Care 2001; 29: 494-500. Craig Webster, Research Fellow, Green Lane Hospital, Auckland, |
|||
|
|
|||
|
Craig S Webster, Research Fellow Green Lane Hospital, Auckland 1003, New Zealand
Send response to journal:
|
A consistent colour standard for anaesthetic drugs exists in Australia, New Zealand, the US and Canada (1). The colour coding on labels in Britain remains in confusion, with three different colour coding schemes in use throughout the country, and even within some individual hospitals! Clearly, this is an accident waiting to happen (1). The simplest solution would be for Britain to adopt the international standard. While society may communicate with language and writing, these are all recent inventions. Our brain evolved to recognise shape and colour long before writing or modern society came about. We can recognise a colour without conscious effort - the same cannot be said of writing. Colour is therefore a psychologically powerful cue that should be used in a consistent and meaningful fashion, as an adjunct to writing, to aid drug identification (2). There has been a great deal of discussion about the difficulties of employing colour in this way, but none of the difficulties are insurmountable (3). Neither will such an approach necessarily be prohibitively expensive (4). The current level of safety in medicine will not do, neither will repeating "read the label" ad nauseam. More constructive and informed approaches need to be adopted and a great deal of work has already been done in this area (5). (1) Webster CS, Merry AF. British syringe label "standards" are an accident waiting to happen. Anaesthesia 2000; 55: 618. (2) Webster CS. Human psychology applies to doctors too. Anaesthesia 2000; 55: 929-30. (3) Webster CS, Mathew DJ, Merry AF. Effective labelling is difficult, but safety really does matter. Anaesthesia 2002; 57: 201-2. (4) Webster CS, Merry AF, Ducat CM. Safety, cost and pre-drawn emergency drugs. Anaesthesia 2001; 56: 818-20. (5) Merry AF, Webster CS, Mathew DJ. A new safety-orientated, integrated drug administration and automated anesthesia record system. A |
|||
|
|
|||
|
Phillip J. Colquitt, Independent Technical Advisor New Farm, Qld. Australia, 4005
Send response to journal:
|
Sir, Colour coding of drug ampoules to assist prevention of drug administration errors, is suggested[1,2,3,4] in response to Wildsmith's claim[5] that "Doctors must read drug labels, not whinge about them." It follows that increasing reliance on ampoule colour, and the implied increase in drug administration speed and efficiency, will disadvantage colour blind doctors, which doctors, according to this journal, are known to exist[6]. Further, older doctors may be disadvantaged by age related presbyopia, as emphasis already needed on using larger printing fonts for vital information on ampoules[7], is possibly sacrificed in favour of colour coding. Unforgettable the occasion when I picked up a plastic ampoule of "Water" to mix with Amoxil prescribed for IV route, based on the fact that the ampoule was in a container normally used exclusively for water ampoules, did appear the same colour as the water ampoules, and was "identical" to the water ampoules; on reading the ampoule label, I was shocked to read the word "Lignocaine". I concluded, "always read the label, because you can't rely on colours." I have often lamented the poor working conditions of doctors and nurses administering drugs, often in semi-emergency situations, and very often with poor lighting both generally and at the bedside, presumably in deference to promoting a "calm environment" for patients at night. I would prefer that all patients had a single room, in which lighting and privacy could be quickly controlled as one walked in, and could also be directed as appropriate for the task. I once took a lux meter with me to work at hospital, and concluded that as things are, hospitals are well behind other industries, as the staff blunder around in dark dorms. Phillip J. Colquitt
[1] Color coded labels of urgent care medications are useful Eckehard A E Stuth, et al. bmj.com, 18 Jan 2002 [2] Reading the label! Subrata K. Basu, et al. bmj.com, 19 Jan 2002 [3] Why not use our eyes? Alon Duby, et al. bmj.com, 20 Jan 2002 [4] Re: Re: Multicolured labels of urgent care medications are useful Craig S Webster bmj.com, 16 Apr 2002 [5] J A W Wildsmith. Doctors must read drug labels, not whinge about them. BMJ 2002; 324: 170. [6] C J M Poole, D J Hill, J L Christie, and J Birch Deficient colour vision and interpretation of histopathology slides: cross sectional Study. BMJ 1997; 315: 1279-1281. [7] Humans are fallible machines Anthony R Cox bmj.com, 28 Jan 2002 |
|||
|
|
|||
|
Craig S Webster, Research Fellow Green Lane Hospital, Auckland 1003, New Zealand
Send response to journal:
|
Thank you to Phillip Colquitt for yet another letter reminding us to read the label. Mr Colquitt apparently has no problem with improving hospital working conditions in terms of better lighting, but balks at the idea of colour-coded labels because they may disadvantage colour-blind doctors! I would be delighted to hear from any colour-blind doctors regarding the problems they have with colour-coding, because in all my years of working in safety research in anaesthetics, I believe I have yet to meet one. I have heard the "but-what-about-colour-blind-anaesthetists" objection to colour-coding many times and it seems ironic to me that doctors are so quick to think of zebras when it comes to safety improvements despite the axiomatic advice in the rest of medicine to first think of horses. Even if some colour-blind doctors were disadvantaged by colour-coding there is no reason to expect that this would offset the net safety benefits of better labelling. Indeed, haphazard colour-coding already exists on drug packs throughout the world, so there is even less reason to believe that consistent colour-coding standards would be any more of a disadvantage than the status quo. Safety is a probabilistic endeavour - even a small improvement will lead to fewer cases of error at the national and international levels (1). In addition, colour-coding is only one safety improvement strategy (albeit perhaps one of the more pressing, given the currently appalling state of labelling, particularly in Britain). Safety experts from every field of high-risk industry tell us that multifactorial approaches must be employed throughout entire organisations to achieve lasting and effective improvements in safety (1,2). Such an approach is why it is so safe to fly, even in the face of terrorist attacks. What makes clinicians think that they are so special and different that medicine is an exception to these safety lessons? Clearly doctors and nurses do conscientiously read and re-read labels before giving drugs, and clearly (as evidenced by on-going drug error, year after year) this is inadequate to ensure appropriate patient safety. Clinicians are doing their best in terms of avoiding error, but would do better if working in a less inherently error-prone system (1,2). (1) Merry AF, Webster CS, Mathew DJ. A new, safety-orientated, integrated drug administration and automated anesthesia record system. Anesthesia and Analgesia 2001; 93: 385-90. (2) Anderson DJ, Webster CS. A systems approach to the reduction of medication error on the hospital ward. Journal of Advanced Nursing 2001; 35: 34-41. |
|||