RT Journal
A1 Rosenberg SN, Sullivan M, Juster IA, Jacques J
T1 OVerrides of medication alerts in ambulatory care
JF Archives of Internal Medicine
JO Archives of Internal Medicine
YR 2009
FD July 27
VO 169
IS 14
SP 1336
OP 1340
DO 10.1001/archinternmed.2009.224
UL http://dx.doi.org/10.1001/archinternmed.2009.224
AB 
Isaac et al,1 writing in the February 9, 2009, issue of the Archives, describe an important and disturbing situation in which the promise of electronic prescribing systems to protect patient safety has been only partially realized. We have encountered a similar situation, which encompasses both electronic and paper prescriptions, in a system that generates clinical alerts using pharmacy claims data for prescriptions that have recently been filled. We agree with the authors' point (and the article by Shah et al2 that they cite) that such systems can achieve better physician acceptance if they limit alerts to those with the highest clinical importance, thereby avoiding the initiation of “alert fatigue.” We have also found that adding other sources of information—medical and laboratory claims, test results, feedback from physicians, and self-reported data from patients who are enrolled in disease management or complete health risk assessments—appears to greatly increase the specificity and credibility of clinical alerts, and the number of clinicians who respond by discontinuing the use of potentially dangerous medications.3- 4