Study: Commercial e-prescribing can cut hospital med errors

Commercial electronic prescribing systems can significantly reduce medication errors in the hospital setting, according to a study published in PLoS Medicine in January.

The study from Australia begins to answer questions about the effectiveness of commercial e-prescribing systems posed in 2011 by the U.S. Agency for Healthcare Research and Quality. Most previous studies of electronic prescribing and inpatient medication-error rates have used customized systems developed by health systems for their own use.

Researchers studied 2 major teaching hospitals in Sydney, Australia. One of the 2 institutions used Cerner Millennium, the second institution used iSoft MedChart. The before and after e-prescribing study design used 1 intervention ward and 3 control wards in the Cerner hospital and 2 intervention wards in the iSoft hospital.

E-prescribing cut total medication errors by 57.5%, 60.5%, and 66.1% in the 3 intervention wards for an overall decrease of 55% across the study population. The drop in total errors was driven largely by a 90% decrease in procedural errors involving unclear, incomplete, or inappropriate medication orders.

Serious clinical errors fell by 44% in the 3 intervention wards compared to a 16.7% decline in control wards. The overall rate of clinical errors, a combination of all errors ranging from the insignificant to the probably fatal, showed no significant change.

Researchers found that both e-prescribing systems introduced new system-related errors, such as checking the wrong medication on a drop-down list. Each of the 2 e-prescribing systems produced different system-related errors, but these new types of errors accounted for 35% of all medication errors after e-prescribing was introduced.

“The results highlight the need to continually monitor and refine the design of these systems to increase their capacity to improve both the safety and appropriateness of medication use in hospitals,” the authors said. “System-related errors require close attention, as they are frequent, but are potentially remediable by system redesign and user training.”