PURPOSE OF REVIEW: This review aims to describe current best practice and recent advances in the use of risk stratification as a tool for drug provocation testing (DPT). In particular, we focus on the testing of unsubstantiated penicillin allergy labels.

RECENT FINDINGS: The inherent risks of DPT are mitigated through careful selection of patients. A detailed history will elicit features of a potentially severe index reaction, as well as significant patient comorbidities which may increase the risks associated with DPT. Such patients require skin testing and/or in vitro testing prior to consideration for a DPT. However, there is increasing evidence that patients without these features may be able to proceed directly to DPT. This has been demonstrated extensively with unsubstantiated penicillin allergy labels, and a variety of risk stratification models have been employed to identify the 'low-risk' patient. Improved outcomes and reduced cost have been demonstrated with such models without compromising patient safety.

SUMMARY: Risk stratification tools may enable well-tolerated and effective 'delabelling' of low-risk patients, with less demand on already scarce resources.