75% of medication alert overrides pose harm to patients

Alert fatigue for physicians has the potential to cause negative outcomes—but so does overriding those alerts. In a study published in Journal of American Medical Informatics Association, researchers evaluate the effects of overriding clinical decision support alerts.

Led by first author Karen Nanji, PhD, the study collected data on types of inpatient clinical decision support alert—including overall numbers, frequency, reasons for override and appropriateness of override—to identify rates of alerts and the effects of overriding them. The study included data covering a 3-year period from a 793-bed tertiary-care teaching institution.

Results showed that, while the rate of overrides varied by type of alert, 73.3 percent of patient allergy, drug-drug interaction and duplicate drug alerts were overridden. Of these overrides, 60 percent were appropriate. Further broken down, researchers concluded 2.2 percent of renal, 26.4 percent of mediation substations, 98 percent of duplicate drug, 96.5 percent of patient allergy and 82.5 percent of formulary substation alerts were appropriately overridden. However, researchers also noted that 75 percent of alert overrides in other areas were inappropriate and could have posed significant harm to patients.

“Almost three-quarters of alerts were overridden, and 40 percent of the overrides were not appropriate,” concluded Nanji and colleagues. “The vast majority of duplicate drug, patient allergy, and formulary substitution alerts were appropriate, suggesting that these categories of alerts might be good targets for refinement to reduce alert fatigue. Future research should optimize alert types and frequencies to increase their clinical relevance, reducing alert fatigue so that important alerts are not inappropriately overridden.”