Drug Allergy in Hospitalized Patients: The Contribution of Allergy Consultation and a Structured Questionnaire

Abstract

Background: Hospitalized patients with an alleged history of drug allergy pose medical and economic concerns when selecting medications for treatment, possibly leading to deviations from standards of care and the use of expensive agents. Accurate history taking and clear documentation of drug allergy are essential for preventing subsequent administration of the offending drug and overdiagnosis of drug allergy. We aimed to evaluate drug allergy-related history taking by internists compared to allergists and to prospectively assess the effect of a simple, structured questionnaire on the accuracy of drug allergy diagnosis. Methods: Consenting patients with an alleged drug allergy who were able to give a coherent history were recruited from two internal medicine wards. In both wards, the internists’ drug allergy diagnosis was initially compared to that of the allergists. In the second part, in the intervention ward, after the same procedure, the internists completed the structured questionnaire. Their diagnostic conclusions with and without the questionnaire were compared. Results: 202 patients labeled with a medication allergy were enrolled. In the control and intervention wards, 54 and 58% of the patients, respectively, labeled by the internists as allergic, were found not to be allergic by the allergist. In the intervention ward, after using the questionnaire, the percentage of patients tagged by the internists as allergic dropped initially by 31% and finally by 59%. Conclusions: Discrepancies between drug allergy diagnosis of internists and allergists are common. Allergist consultation or use of a simple structured questionnaire may be beneficial for accurate diagnosis of drug allergies.

β‐Lactam allergy de‐labelling

“…In one study, 617 skin-prick test positive or negative patients with reported non-immediate reactions to penicillins were challenged orally with one-tenth of the normal dose of the antibiotic, followed by the full dose for 5 days. 1 Nine had an immediate mild rash without systemic symptoms and 608 (98.5%) had no reaction. Skinprick testing results did not predict reactions.…”