Clinical Observation
Health Care Reform
September 28, 2009

Clinicians' Assessments of Electronic Medication Safety Alerts in Ambulatory Care

Author Affiliations

Author Affiliations: Center for Patient Safety, Dana-Farber Cancer Institute, Boston, Massachusetts (Drs Weingart and Isaac, Messrs Simchowitz, Shiman, and Spencer, and Mss Brouillard and Morway); Harvard Medical School, Boston (Drs Weingart, Davis, Isaac, Sands, and Weissman); Blue Cross Blue Shield of Massachusetts, Boston (Ms Cyrulik); Division of General Medicine and Primary Care, Beth Israel Deaconess Medical Center, Boston (Drs Davis, Isaac, and Sands); PatientsLikeMe, Cambridge, Massachusetts (Dr Massagli); Cisco, San Jose, California (Dr Sands); Institute for Health Policy, Massachusetts General Hospital, Boston (Dr Weissman); Executive Office of Health and Human Services, Commonwealth of Massachusetts, Boston (Dr Weissman); and University of Massachusetts Medical School, Worcester (Dr Weissman).


Copyright 2009 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2009

Arch Intern Med. 2009;169(17):1627-1632. doi:10.1001/archinternmed.2009.300

Background  While electronic prescribing (e-prescribing) systems with drug interaction and allergy alerts promise to improve medication safety in ambulatory care, clinicians often override these safety features. We undertook a study of respondents' satisfaction with e-prescribing systems, their perceptions of alerts, and their perceptions of behavior changes resulting from alerts.

Methods  Random sample survey of 300 Massachusetts ambulatory care clinicians who used a commercial e-prescribing system.

Results  A total of 184 respondents completed the survey (61%). Respondents indicated that e-prescribing improved the quality of care delivered (78%), prevented medical errors (83%), and enhanced patient satisfaction (71%) and clinician efficiency (75%). In addition, 35% of prescribers said that electronic alerts caused them to modify a potentially dangerous prescription in the last 30 days. They suggested that alerts also led to other changes in clinical care: counseling patients about potential reactions (49% of respondents), looking up information in medical references (44%), and changing the way a patient was monitored (33%). Altogether, 63% of clinicians reported taking action other than discontinuing or modifying an alerted prescription in the previous month in response to alerts. Despite these benefits, fewer than half of respondents were satisfied with drug interaction and allergy alerts (47%). Problems included alerts triggered by discontinued medications (58%), alerts that failed to account for appropriate drug combinations (46%), and excessive volume of alerts (37%).

Conclusion  Although clinicians were critical of the quality of e-prescribing alerts, alerts may lead to clinically significant modifications in patient management not readily apparent based on “acceptance” rates.