A mixed method study of the merits of e-prescribing drug alerts in primary care.

Objectives: The objective of this paper was to describe primary care prescribers' perspectives on electronic prescribing drug alerts at the point of prescribing.

Design: We used a mixed-method study which included clinician surveys (web-based and paper) and focus groups with prescribers and staff.

Participants: Prescribers (n = 157) working in one of 64 practices using 1 of 6 e-prescribing technologies in 6 US states completed the quantitative survey and 276 prescribers and staff participated in focus groups.

Measurements: The study measures self-reported frequency of overriding of drug alerts; open-ended responses to: "What do you think of the drug alerts your software generates for you?"

Results: More than 40% of prescribers indicated they override drug-drug interactions most of the time or always (range by e-prescribing system, 25% to 50%). Participants indicated that the software and the interaction alerts were beneficial to patient safety and valued seeing drug-drug interactions for medications prescribed by others. However, they noted that alerts are too sensitive and often unnecessary. Participant suggestions included: (1) run drug alerts on an active medication list and (2) allow prescribers to set the threshold for severity of alerts.

Conclusions: Primary care prescribers recognize the patient safety value of drug prescribing alerts embedded within electronic prescribing software. Improvements to increase specificity and reduce alert overload are needed.