P-2011. Evaluating the Impact of Penicillin Allergy Cross-Reactivity Alert Removal on Antibiotic Prescribing and Utilization

TL;DR

Removing a computer alert about penicillin allergy and cephalosporin cross-reactivity led to more use of cephalosporins without causing more allergic reactions.

Contribution

This study shows that removing an EHR alert increases safe use of beta-lactam antibiotics in penicillin-allergic patients.

Findings

Cephalosporin orders increased by 10% after alert removal in penicillin-allergic patients.

Only one mild allergic reaction occurred among thousands of cephalosporin doses given post-intervention.

Use of alternative antibiotics like quinolones and aztreonam decreased significantly.

Abstract

Patients with a documented penicillin (PCN) allergy are often prescribed second-line antibiotics due to electronic Health Records (EHRs) warnings about cross-reactivity with cephalosporins, despite the low risk. This PCN allergy-cephalosporin cross-reactivity alert can discourage beta-lactam use, leading to negative outcomes. In June 2024, the Montefiore Health System removed this alert to improve beta-lactams use. This study evaluates the impact of this removal on antibiotic prescribing and utilization.Figure 1:Percentage of antibiotic orders that are cephalosporins in patients with reported penicillin allergiesFigure 2:MMC Facility wide inpatient AU rate by antibiotic class

Percentage of antibiotic orders that are cephalosporins in patients with reported penicillin allergies

MMC Facility wide inpatient AU rate by antibiotic class

This retrospective, pre-post study included patients admitted to Montefiore Medical Center who had an order for antibiotics. Antibiotic prescribing patterns and administrations were compared in patients with documented PCN allergies before (Jan-Jun 2024) and after (Jun-Apr 2025) the removal of allergy alert, and in non-PCN allergic patients as a reference group. Data collected included antibiotic class orders, antibiotic utilization rate (AU, measured as days of therapy/1000 days present), incidence of new cephalosporin allergies, and allergy severity.Figure 3:Monthly percentages of antibiotic orders by class among penicillin allergic patients before and after alert removal compared to non-penicillin allergic reference group

Monthly percentages of antibiotic orders by class among penicillin allergic patients before and after alert removal compared to non-penicillin allergic reference group

A total of 9,863 patients were included, 6,744 (68%) of who were women. The mean (SD) age was 64 (17) years. Among patients with documented PCN allergies, cephalosporin orders increased by 10% after the intervention (27.8% vs. 37.8%; p < 0.0001) (Figure 1). Cephalosporin AU rate increased from 142.1 to 156.3, p< 0.001, figure 2. During the post-intervention period, only one mild allergic reaction to cephalosporin was reported among 4,388 PCN-allergic patients who received a total of 24,643 cephalosporin doses. Orders for quinolones, clindamycin, and aztreonam declined by 4.8% (p < 0.0001),1.8% (p=0.06), and 1.3% (p < 0.0001), respectively (Figure 3).

Removal of EHR alerts for PCN allergy-cephalosporin cross-reactivity is associated with a significant increase in cephalosporin use in PCN allergic patients without a rise in allergic reactions.

All Authors: No reported disclosures

Figures