Computer systems fail to spot harmful interactions
ISMP surveys error alerts
Computer software integrated into hospital pharmacy systems to flag potential medication errors failed by a wide margin to detect a sample of 10 error-types in a field survey of 307 hospital systems conducted by the Institute for Safe Medication Practices. The ISMP used errors that had been previously reported to its Medication Errors Reporting Program as the way to test computer programs, covering aspects such as potentially harmful drug interactions, labeling concerns, and dosage alerts.
Overall the field survey found that only four of the 307 systems tested caught potential problems with all 10 medication orders. The survey found that a variety of computer system vendors are being used in hospital pharmacies, including (in descending order of frequency) HBOC, Meditec, Cerner, SMS, Pharmacon, and Digimedics, as well as drug information providers such as First Data Bank, Medispan, and Medicom. (Overall the survey found no discernible differences among the systems and did not evaluate or rank them as part of the survey.) In other words, all of the systems equally flagged, or failed to flag, the same types or number of potential error alerts.
The survey found that 87% of the systems failed to detect toxic doses of antibiotics for patients with renal impairment; 87% did not detect single or cumulative lethal doses of colchicine; 65% did not detect potentially toxic drug ingredient duplication with acetaminophen and Percocet; 62% did not detect lethal overdoses of cisplatin or vincristine; 61% did not flag an order when an oral suspension was ordered for IV use; and 58% did not link pharmacies and labs.
"It’s frightening to believe that the vast majority of pharmacy computer systems may be incapable of preventing lethal errors," says ISMP president Michael Cohen, who stresses that computer systems cannot be blindly relied upon.
The institute historically has recommended hospitals install systems that allow for routine manual updating of new error alerts, and that systems should be installed that can be tailored to specific types of alerts matched with the medicine being practiced by an institution or to errors commonly unique to an institution.
The survey did find that 54% of the hospitals, which voluntarily agreed to be part of the testing, do integrate drug information updates into their systems at least quarterly.
And not all of the survey results were bad. Of the systems tested, 88% did detect a ketorolac- aspirin cross allergy; 85% detected interactions between ketoconazole and cisapride; and 69% of the systems do provide for updated inclusion of specific medication alerts.
Complete survey results can be found at http://www.ismp.org.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content