Home' Australian Pharmacist : Australian Pharmacist January 2012 Contents 70 Australian Pharmacist January 2012 I ©Pharmaceutical Society of Australia Ltd.
Do you sometimes feel bombarded with
too many drug interaction alerts? Can they
be annoying and irrelevant or unhelpful?
Feedback from pharmacists and GPs suggests
that many answer 'yes' to these questions,
and yet they find sufficient value in these
alerts not to want them to be switched off.
We ask pharmacists and GPs how this form of
decision support can be improved.
Drug interaction (DI) alerts are commonly
implemented in prescribing and dispensing
systems, but evidence suggests that
information provided in the alerts is not
always relevant or helpful.1--3 Previous NPS
research identified shortcomings in the
quality of DI decision support in Australian
prescribing and dispensing systems.4,5
A number of recommendations on the
content and format of DI decision support
resulted from this work.5
NPS recently conducted a survey study
to find out what Australian community
pharmacists and GPs' preferences are in
relation to the content, format and usability
of DI alerts in their software. Results of this
survey study were recently published in the
Medical Journal of Australia6 (Yu KH, Sweidan
M, Williamson M, et al. Drug interaction alerts
in software -- what do general practitioners
and pharmacists want? MJA 2011;195:676--
80.) -- here, we will provide a summary of the
results and discuss the practical implications.
What we did
Surveys were mailed to a random sample
of 1,000 community pharmacists and 1,000
GPs to seek their opinion on:
Drug interaction alerts
-- necessary but need
By Kitty Yu, Michelle Sweidan, Margaret Williamson
All three authors are staff of NPS: Better choices,
Better Health. Kitty Yu is a Program Officer and
Michelle Sweidan is the Team Leader of the e-Health
and Decision Support Team. Margaret Williamson is
the Manager of Research and Development.
• The usefulness of seven components
of information in DI alerts (e.g. clinical
effects, management advice).
• Which DI alert format they preferred.
• Whether they would find six usability
features desirable or valuable in their
practice (e.g. differentiate DI alerts by
severity, ability to customise alerts).
• The three aspects of DI decision support
they would most like to change in their
Descriptive statistics and thematic analysis
were used to analyse responses.
What we found
Surveys were returned by 170 pharmacists
and 219 GPs, 138 and 190 surveys were
analysed, respectively, after excluding
surveys from ineligible respondents.
Content of drug interaction alerts
Information on the severity of the
interaction, clinical effects, management
advice, and -- to a lesser extent -- timeframe
and mechanism of an interaction, were
considered useful by the majority of
pharmacists and GPs (see Figure 1).
Pharmacists and GPs want DI alerts to be
more relevant to practice. Respondents gave
examples of irrelevant or unhelpful alerts,
including: theoretical interactions; dosage
forms that usually don't interact (e.g. creams,
inhalers); individual drugs within a class that
do not interact (e.g. not all calcium channel
blockers have the same interactions); drug
combinations routinely used together;
medicines no longer being taken by the
patient; and duplicate alerts (e.g. when
different strengths of warfarin are prescribed).
Of the three DI alert formats, the most
popular -- preferred by 83% of pharmacists
and 69% of GPs -- was the one with
headings and 1--2 succinct bullet points
of information beneath each (see Figure
2). Supporters said this format was clear,
concise, and easy to scan and navigate.
More than 80% of respondents believed
that alerts should be differentiated by type
and severity, and that the software should
make it harder for users to override alerts for
severe interactions. In all 61% of pharmacists
and GPs thought it should be mandatory to
provide a reason to override severe alerts,
and around half of the pharmacists and GPs
favoured user customisation of alerts.
Figure 1. Proportion of pharmacists who considered seven components
of information in DI alerts 'mostly useful' or 'sometimes useful'. Note: GP
responses were similar.
Links Archive Australian Pharmacist February 2012 December 2011 Navigation Previous Page Next Page