The Quality and Management of Penicillin Allergy Labels in Pediatric Primary Care

To determine the quality and subsequent management of pediatric patients labeled as having a penicillin allergy.

Five hundred patients randomly selected from a dual-center birth cohort of 18 015 children were included in the study. All patients were born from January 1, 2010 to June 30, 2020 and were labeled as being penicillin-allergic (allergy to penicillin, amoxicillin, amoxicillin-clavulanate). Demographics from the selected participants were similar to all labeled participants in the birth cohort. Participants were followed at primary care and specialty clinics associated with either Texas Children’s Hospital or Children’s Hospital of Philadelphia.

A retrospective chart review was performed using EpicCare electronic medical records (EMR). Based on data extracted from the allergy tab and other healthcare notes and encounters, reactions to penicillin were risk classified into not allergy, low risk, moderate or high-risk, severe cutaneous risk, or unable to classify groups. Referral to an allergist as well as outcomes of skin testing and penicillin re-exposure were also documented.

Fifty percent (250 of 500) of patients had their penicillin allergy labeled as “unable to classify”. Of these, 141 (56.4%) were either “low risk” or “not allergy” based on review of the healthcare notes. Only 16.8% (84 of 500) of patients were referred to an allergist, with subsequent evaluations occurring in 10.8% (54 or 500). Negative penicillin skin testing was found in all 25 of 500 patients tested and 23 of 24 (96%) tolerated oral amoxicillin in a subsequent follow-up visit. Penicillin allergy label removal overall was uncommon (69 of 500, 13.8%) and occurred more often after evaluation by an allergist (26 of 54, 48%). Of the children who had penicillin allergy labels removed in the EMR, 48 of 69 (69.6%) subsequently received a penicillin derivative that was tolerated in 94% of patients. Children delabeled by primary care pediatricians were just as likely to tolerate penicillin when compared with a patient delabeled by an allergist (94% versus 93%, P = .87).

Electronic medical record documentation of penicillin allergy is often uninformative and few patients labeled as being penicillin-allergic were referred to allergists.

Penicillin remains the most commonly reported drug allergy and a focus of many antibiotic stewardship programs. As this study demonstrates, because of inadequate documentation, most penicillin allergy labels documented in the EMR were unable to be risk classified. When these charts were further evaluated, over half were reclassified as “low risk” or “not allergy.” Clear documentation is imperative to avoid unnecessary labeling and/or prompt allergy referrals. This study also found that there was not a significant difference in tolerance of penicillin between patients delabeled by primary care providers compared with allergists. In cases where the clinical history is not consistent with an IgE-mediated reaction and/or there is not a contraindication to future courses of penicillins (such as a severe cutaneous adverse reaction, for example), the primary care provider should feel empowered to remove the penicillin allergy label or ensure referral to an allergist for further testing if needed.