Drug interaction decision support


Drug interaction decision support in software can help clinicians decide whether certain drug combinations can be used safely, but evidence suggests that information provided in the alerts is not always relevant or helpful. An NPS study conducted in 2008 identified shortcomings in the quality of drug interaction decision support in commonly used prescribing and dispensing systems. This study builds on previous NPS work to seek health professionals’ opinions on drug interaction alerts in their software.


To explore Australian GPs and pharmacists’ preferences in relation to the content, format and usability of drug interaction alerts in prescribing and dispensing software.

Design and methods

Surveys were mailed to a random sample of 1000 GPs and 1000 community pharmacists in June 2010 to seek their opinions on the following: (1) usefulness of seven components of drug interaction information; (2) preferred format of drug interaction alerts; (3) extent of agreement for which features would be valuable in practice; and (4) aspects of drug interaction decision support users would most like to change. Descriptive statistics and thematic analysis were used to analyse responses.


219 GPs and 170 pharmacists responded. Most GPs and pharmacists considered the severity of the interaction, clinical effects and management advice to be ‘mostly useful’ in drug interaction alerts. The most popular drug interaction alert format was one with headings and 1-2 succinct bullet points beneath each (see figure 1). GPs and pharmacists wanted to be able to differentiate drug interaction alerts by severity. A majority of respondents agreed that it should be made more difficult to override alerts for severe interactions and mandatory to provide a reason for doing so.


Transforming drug interaction information into a suitable format for decision support is challenging—the information must be relevant, useful, concise and easy to read and comprehend.

NPS has made recommendations to improve drug interaction decision support based on this study’s findings, and has communicated with software vendors and drug interaction knowledge base providers about these recommendations.


  • Yu K, Sweidan M, Williamson M. Drug interaction alerts – necessary but need improving? Australian Pharmacist 2012; 13(1)
  • Yu K, Sweidan M, Williamson M, Fraser A. Drug interaction alerts in software – what do general practitioners and pharmacists want?. Medical Journal of Australia 2011; 195 (11/12): 676-680.(Abstract)
  • Yu KH, Sweidan M, Williamson W, et al. Drug interaction alerts in clinical software: What do GPs want? GP11 Conference; 2011 Oct 6-8; Hobart, Australia.(Presentation)
  • Yu KH, Sweidan M, Williamson W, et al. Improving drug interaction decision support in clinical software: GPs and pharmacists tell us what they want. Primary Health Care Research Conference; 2011 July 13-15; Brisbane.(Abstract/Presentation)