Ethnographic study of incidence and severity of intravenous drug errors

On the subject of bolus doses from IMED infusion devices, if Phillip
is referring to the Gemini models then the secondary infusion rate is not
a bolus dose system at all and such confusion can lead to further
problems.

The use of the secondary infusion rate assumes both a separate fluid
container and giving set attached to the primary administration set. In
the absence of a second giving set the fluid will be taken from the
primary container reducing the volume available to be infused without
decrementing the volume to be infused display.

Multiple bolus doses can significantly reduce the actual volume without
reducing the amount the pump will attempt to infuse so that the container
can empty (apparently) prematurely leading to problems such as air in the
line and discontinuity of infusion while a new fluid container and giving
set are prepared.

Such delays with critical infusions were identified in the original
article as potentially serious.

Of the volumetric infusion pumps in use at this hospital none are able to
deliver a true bolus though several have secondary infusion rates that are
used as such by staff. I have no information about how many errors have
occurred as a result of this but I feel that only if there were an extreme
deterioration of the patient’s condition would anything be reported.

Competing interests:  
Trying to teach too many pumps to too many nurses

Editor,

Can there be a “system” of drug administration independent of
cultural considerations?

I read the study cited[1] in the response by Anton et al., and though
it’s not mentioned in the response, I wasn’t surprised to find the authors
of the cited study noting the higher rates of arithmetic error in drug
calculations and drug mixing among their study participants in clinical
settings, compared to industrial arithmetic error rates with “self-
checking”.

Administrators should endorse attitudes which allow nurses to do
their work without the excess and paranoia of checking each others every
move and utterance. Studying nurses who are already busy studying each
other, is madness. Educating nurses to “tertiary” standard, only to
systematically deskill them in an out of date hospital culture of
cloistered “sisterhood” which seems preoccupied with bedpan rinsing, and
call bell answering, and baby-feeding, is counterproductive.

Phillip J. Colquitt, New Farm, Qld., Australia

PS. Please feel free to comment on errors.

[1] Ferner RE, Langford NJ, Anton C, Hutchings A, Bateman DN,
Routledge PA. Random and systematic medication errors in routine clinical
practice: a multicentre study of infusions, using acetylcysteine as an
example. British Journal of Clinical Pharmacology 2001; 52: 573-7.

Competing interests:  
Education

As researchers who specialise in qualitative methodologies we read
with great interest the recent study by Katja Taxis and Nick Barber (BMJ,
326, pp684-687). Whilst we appreciate the attempts of the authors to shed
some light on the nature of intravenous drug errors, and an often
neglected area of clinical practice, we were left somewhat perplexed as to
the authors’ understanding of ‘ethnography’. Whilst we are aware that the
use of qualitative methods in medical related research is relatively new,
we are concerned, as are others, about the quality of research that is
being published in this area. As published, this study is not an
ethnographic study and is barely qualitative.

The authors have conducted an observational study, one it seems with
roots in quantitative not qualitative methodology. This is most obvious
from the outcome measures used where the authors set out to document the
‘number, type and clinical importance of errors’. One clue to reviewers
would be the use of terms like ‘number’. Also later we find that the
researchers have used ‘a validated scale to assess the clinical importance
of intravenous drug errors’. Ethnographic studies do not rely on the use
of quantitative scale assessments. The use of ‘confidence intervals’ to
analyse supposedly ‘ethnographic data’ is again somewhat worrying. Lack of
time here precludes us from expansion, but we refer interested parties to
the following sources for an exploration of the meaning of ‘ethnography’
and ethnographic methods:

Agar, M. 1980 The Professional Stranger: An Informal Introduction to
Ethnography.
Spradley, J.P. 1979 The Ethnographic Interview.
Hammersley, M. and Atkinson P. 1995 Ethnography: Principles and
Practices.

In a broader sense, this paper raises concerns for us about the
understanding of qualitative methods in medicine. The misrepresentation of
this study as ‘ethnography’ can unfortunately do nothing but undermine the
work of qualitative researchers. Ethnographers and other qualitative
researchers are constantly called upon to defend the integrity and rigour
of their methodologies against quantitative ones and we feel it our duty
to defend proper qualitative research against methodological drift. There
already exists an abundance of misunderstanding about qualitative research
without prestigious journals like the BMJ contributing to the situation.

Competing interests:  
None declared

This study confirms what most of us informally are aware of - when
people are under pressure performance suffers.

Whatever one knows or can do (i.e. one's competence) is not always
translated into day to day activity (performance). Who else has not taken
a short cut today in one or more clinical or administrative activities,
because they had five other important, or urgent (or both!), things to get
done before the end of the day?

There would appear to be two logical potential solutions:

a) Use infusors rather than require nurses to spend time
administering i.v. medication

b) Staff the wards sufficiently to allow nurses to have the time to
devote to this function, in addition to the multifarious other tasks
accruing in their portfolios.

The latter option would allow time for the nurses actually to have
human contact with the patient for a limited but relatively undisturbed
period of time (as when specialist oncology nurses administer cytotoxic
drugs). Given the extra cost this would represent to a resource limited
system, I think I can predict which option would be the more attractive to
someone designing a service with cost control as a (the?) significant
driver.

Competing interests:  
The amount of work I would like to finish to a satisfactory standard, and the amount of time I am prepared to devote to my working life