Article Text

Download PDFPDF

GRP-058 Establishment of a Programme to Detect Drug Interactions Complementary to Electronic Prescriptions
  1. J Bilbao Aguirregomezcorta,
  2. C Floristán Imízcoz
  1. Hospital San Eloy, pharmacy, Bilbao, Spain


Background Ours is a 110-bed regional hospital with electronic prescribing throughout. The electronic prescription programme offers allergy alert systems and the need for dosage adjustment in renal and hepatic impairment. However, no warning of potential drug interactions is included.

Purpose To establish a drug interaction screening system to complement the electronic prescribing programme.

Materials and Methods Prospective one-year study (Sept 2011–Sept 2012). Computerized drug interaction alerts can improve the safety, quality and efficiency of care processes and reduce the rate of medicines errors, but sending mass alerts can generate what is known as ‘alert fatigue’. For this reason we selected those drugs categorised in Micromedex® 2.0 as contraindicated for simultaneous and significant interactions (the interaction can cause death and/or require medical intervention to minimise or avoid serious side effects); drugs classified as D in Lexi-Comp OnlineTM (consider modifying the treatment and X (avoid combination); those classified as clinically important by the Arizona CERT. The interaction was detected by crossing data in Access 2003 with the pharmacotherapeutic profile of patients obtained in the electronic prescribing programme (eOsabide) and a proprietary database which contains a total of 3,133 pairs of interactions.

The report was written in the patient’s medical history (Osabide-Global) and acceptance was verified in 24–48 hours.

Results We detected a total of 1996 interactions and 25% of them were reported, 27% of which led to changes in medical treatment. The main cause of non-notification (36%) was that one drug was prescribed if needed.

Conclusions The project was very well accepted among medical professionals and has improved the quality of prescribing. The biggest drawback is the delay in detecting the interaction; it would be helpful if the system generated the warning at the time of prescription.

No conflict of interest.

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.