|
John Wilson is a semi-retired pharmacist based
in Arnold, Nottinghamshire, who works part-time as a writer and
locum pharmacist
|
Recently I have been contemplating the problems of prescribing errors and the responsibility of the pharmacist to intercept any such errors before they reach the patient and possibly cause harm. Dr Bryony Dean
Franklin and her colleagues at the London University School of Pharmacy
have done much excellent work in this field as far as
errors in prescribing by hospital doctors is concerned.1–4 Others
have looked at the impact of clinical pharmacists on prescribing errors
on medical wards.5 The extent of prescription errors in general practice
has also been surveyed.
A master’s degree project at Nottingham University, supervised
by the late Dr Mo Aslam, found that prescribing errors on
general practice prescriptions were common6 and there was a wide range
of different types of error. This study showed that the error rate varied
significantly between prescribers in different practices and was relatively
high on handwritten prescriptions. The latter phrase is, I believe, significant.
It implies that comp-uter-generated prescriptions are less error-prone,
presumably because of the drug interaction software included in practice
computer systems.
However, general practitioners’ views on the use of computerised
drug interaction alerts vary.7 Of the GPs surveyed, some 22 per cent
admitted to “frequently or very frequently over-riding drug interaction
alerts without properly checking them”. The investigators found
that the type of computer system used may make it more or less likely
that the GP may over-ride an alert. A commentary in the same issue of
that journal8 suggests that mechanisms should be introduced to prevent
computers from issuing potentially dangerous drug combinations. Alternatively, “perhaps
the solution is to couch drug alerts less in terms of a ‘hazard
signal’ or warning of unspecified dangers ahead, but more in terms
of clinical consequences”. So, doctors may sometimes, even “very
frequently”, ignore the warnings of a drug interaction and issue
the prescription anyway.
What happens when the prescription reaches the community pharmacy? Chen
et al examined the problem of prescriptions with potentially dangerous
drug combinations which arrived for dispensing.9 They pointed out that
community pharmacy computer systems varied in their ability to detect
potentially hazardous drug combinations, and that pharmacists do not
always act on such warnings, even if the drug combination is potentially
hazardous. Although drug interactions with immediate serious consequences
were relatively rare in their study, many of the interactions that they
observed would require monitoring.
Real life is, however, not that simple. In some pharmacies in which I
have worked as a locum since retirement, it seems that the mean number
of prescription items per patient goes into double figures. The dispensary
computer pings endlessly with alerts. Therefore, having read the papers
that I refer to above, I decided to carry out a small study of my own
in one pharmacy. For a number of prescriptions dispensed, I noted the
interactions that were brought up on the screen. Some of them were well
known, but others were rather bizarre, to say the least. In some cases,
an interaction alert seems to be applied to a whole class of drugs whereas
a quick check in the BNF will show that only one or two members of that
class of drugs is involved in the interaction.
All the interactions arose from long-established prescriptions that had
been dispensed regularly for months. If I had attempted to contact the
prescriber about even a few of the interactions, the day’s work
would not have been completed, and I suspect that I would have received
a less than welcome response from some of the GPs.
How would we know if harm had befallen any of these patients as a result
of the interactions? This particular pharmacy was, admittedly, something
of a prescription factory. A significant number of the prescriptions
were collected by one of the pharmacy staff from the local surgery, the
items dispensed and the medicines then delivered to patients by a woman
with a small van. There was little or no contact between the pharmacy
and patients, or their carers or relations. This scenario is, I suspect,
the reality in many community pharmacies. It is likely that the situation
with regard to ignoring interaction warnings by community pharmacists
is similar to that identified by Magnus et al7 for doctors.
What is to be done? I suggest that someone with considerable expertise
in drug interactions should go through the databases used in both GP
and pharmacy computers, highlight those that are important or dangerous
and get rid of the rest.
I suspect that the designers of the software for both GP and pharmacy
computer systems are covering their backs by including every conceivable
interaction in the database. However, if this means that even insignificant
interactions are brought up on to the screen, this will lead to users
ignoring all of them — a fine example of the law of unintended
consequences.
References
1 Dean B, Barber N, Schachter M et al. Prescribing
errors in hospital inpatients: why do they occur? Pharm J 2000;265(Suppl):R17.
2 Dean B, Schachter M, Vincent C et al. Causes of prescribing errors
in hospital inpatients: a prospective study. Lancet 2002;359:1373–8.
3 Dean B. Learning from prescribing errors. Quality and Safety in
Health Care 2002;11:258–60.
4 Dean B, Schachter M, Vincent C et al. Prescribing errors in hospital
inpatients: their incidence and clinical significance. Quality and
Safety in Health Care 2002;11:340–4.
5 Dale A, Copeland R, Barton R. Prescribing errors on medical wards
and the impact of clinical pharmacists. Int J Pharm Pract 2003;11:
19–24.
6 Shah SNH, Aslam M. Avery AJ. A survey of prescription errors in
general practice. Pharm J 2001;267: 860–2 (PDF 107K)
7 Magnus D, Rodgers S, Avery AJ et al. GPs’ views on computerised
drug interaction alerts: questionnaire survey. J Clin Pharm Ther
2002;27:377–82.
8 Ashworth M. Commentary. J Clin Pharm Ther 2002;27:311–2.
9 Chen Y-F, Neil KE, Avery AJ et al. Prescriptions with potentially
hazardous/contraindicated drug combinations presented to community
pharmacies. Int J Pharm Pract 2002; 10(Suppl):R29 (PDF 45K) |
|
