Antibiotic allergy labels-the impact of taking a clinical history

Many patients admitted to hospital have an antibiotic allergy (AAL) documented in their medical record. In many of these, the reaction is not a hypersensitivity reaction or may no longer be relevant. Despite this, the label adversely affects patient care directly in terms of antibiotic selection, and indirectly in terms of patient costs and the development of antimicrobial resistance.

To estimate the prevalence of AALs in a cohort of hospitalised patients, to investigate the feasibility of de-labelling through re-challenge based solely upon clinical grounds.

This is a cross-sectional study conducted over a 6-month period on adult inpatients. An allergy history was taken from each patient and compared with medical record data regarding allergy. Antibiotic selection data were collected (if relevant). It was then determined whether immediate de-labelling was appropriate, if direct provocation test (DPT) could be relatively safely performed, and if antibiotic selection was appropriate.

Three thousand eight hundred and fifty five patients were screened, 553 (14.35%) had an AAL, and 352 were interviewed. There were 426 AALs; 276 (64.8%) towards a penicillin. After taking a detailed clinical history of the type of reaction, approximately 20% could be immediately de-labelled and educated (non-allergic, non-severe reactions) and another 38% with either a definite or vague history of mild cutaneous reaction would be suitable for an attempt at clinical de-labelling DPT.

These simple measures to 'de-label' patients appropriately, would increase the quality of care of this group known to have higher costs, infection with more resistant bacteria and worse health outcomes that 'non-labelled' patients.