Research Activities, January 2011: Health Information Technology: A customized computer drug alert requiring active provider response is no more effective than a passive alert

Health Information Technology
A customized computer drug alert requiring active provider response is no more effective than a passive alert

The effectiveness of drug alerts in computerized decision support systems in preventing drug-drug interactions has been modest because of low response by providers to the automated alerts. Studies show, for example, that these alerts are overridden as much as 90 percent of the time. A new study has found that a customized alert that required a physician response was no more effective than a passive alert in preventing the prescribing of drugs that interact in unsafe ways. A research team led by Brian L. Strom, M.D., M.P.H., of the University of Pennsylvania, examined the effectiveness of the two types of alerts within an inpatient computerized physician order entry (CPOE) system.

One alert required an affirmative response from the provider and the other alert was a commercially available passive alert that was already part of the CPOE system. The test case used was coprescribing of a nonsteroidal anti-inflammatory drug (NSAID) with the anticoagulant warfarin, a practice generally advised against. The researchers found that a customized CPOE alert that required a provider response had no effect in reducing concomitant prescribing of NSAIDs and warfarin beyond that of the passive CPOE alert. The randomized trial included 1,963 clinicians in 2 urban hospitals, who were assigned to receive 1 of the 2 alerts. Over the 17-month study period, there were 1,024 alerts: 464 in the active-alert group and 560 in the passive-alert group. The proportion of desired ordering responses (not reordering the alert-triggering drug within 10 minutes of an alert) was lower in the active-alert group than the passive alert group (25 vs. 28 percent).

The researchers concluded that CPOE alerts cannot be assumed to be effective in improving prescribing, and need to be formally evaluated. This study was partly supported by the Agency for Healthcare Research and Quality (HS16946).

See "Randomized clinical trial of a customized electronic alert requiring an affirmative response compared to a control group receiving a commercial passive CPOE alert: NSAID-warfarin co-prescribing as a test case," by Dr. Strom, Rita Schinnar, M.P.A., Warren Bilker, Ph.D., and others in the Journal of the Medical Informatics Association 17, pp. 411-415, 2010.
Research Activities, January 2011: Health Information Technology: A customized computer drug alert requiring active provider response is no more effective than a passive alert