New Clinical Decision Rule Aims to Limit Unnecessary Penicillin-Allergy Labels

In a health care landscape where many physicians are cautious about over-prescribing antibiotics, there’s 1 antibiotic category in which caution leads physicians to under-prescribe: penicillin.

That’s because many patients are labeled as having penicillin allergies despite being at a low risk of allergic reaction, outgrowing the allergy, or even being mis-labeled as allergic in the first place.

The challenge for clinicians is being able to identify patients who could safely be prescribed penicillin. That’s why investigators have been working to find ways to safely identify and “de-label” people who could take the antibiotic but who have been classified as allergic.

In a new report published this month in JAMA Internal Medicine, Jason A. Trubiano, MBBS, PhD, of Austin Health and the University of Melbourne, in Australia, proposes a solution.

Trubiano and colleagues developed and tested a penicillin allergy clinical decision rule called PEN-FAST, and found it to be highly effective at identifying low-risk penicillin allergies.

Trubiano told Contagion® the potential impact of such a tool is high because over-labeling is a huge problem.

“Penicillin allergy is reported in 1 in 10 individuals yet up to 10% of these will actually have a true penicillin allergy,” he said. “Having a penicillin allergy leads to inferior hospital outcomes, excess mortality and use of inappropriate antibiotics, driving antimicrobial resistance such as MRSA and C difficile.”

Trubiano and colleagues used data from 622 allergy-tested patients from 2 tertiary care sites in order to develop their model based on characteristics that were highly predictive of allergies. External validation of the resulting model took place among 945 patients at clinics in Australia and the United States.

Features found to be linked with actual penicillin allergies included being diagnosed with the allergy 5 or fewer years ago, anaphylaxis/angioedema, severe cutaneous adverse reaction (SCAR), and treatment being required as a result of the allergy episode. In the model, 2 points were assigned for each of the first 3 categories, and 1 point was assigned for the latter. Patients with scores of less than 3 were found to be “low risk,” according to the model, as only 3.7% of patients with scores under 3 tested positive for a penicillin allergy.

Trubiano said a tool like PEN-FAST is necessary in part because the problem of over-labeling with penicillin allergies has not gone away despite increased awareness of the issue.

“I believe the problem is staying the same in regard to minor or low risk reactions,” he said. “We have a huge cohort of elderly patients that have historical persistent labels and new cohorts coming through acquiring labels in childhood (eg viral rash) or inappropriately in adulthood (e.g. mild rash or drug side effect rather than true allergy—nausea).”

He said there’s been some progress, but much more work is required.

Asked if part of the problem might be patient resistance to removing labels, Trubiano noted that that question was not directly addressed in his study. However, he said other research has been conducted on this question.

“I believe the level of post-testing compliance with the revised allergy labels is proportional to the level of post testing documentation and explanation doctors/pharmacists provide to both the patient and the primary caregiver,” he said.