With 95% of low-risk patients labelled penicillin allergic tolerating the medicine on direct oral challenge, every penicillin allergy label should be questioned, Professor Jason Trubiano told the National Medicines Symposium today (11 November).
Before accepting a label, ask:
- When did the reaction happen?
- How severe was it?
- Did it need treatment?
The answers can identify patients at low risk of a true allergy – and for most of them, a single supervised dose of penicillin is all it takes to remove the label, Prof Trubiano, an infectious diseases physician at Austin Health, told delegates at the virtual event, themed ‘Medicines stewardship: safer use starts with all of us’.
‘We’ve got evidence now for assessment, direct oral challenge, and impact on AMS [antimicrobial stewardship] and AMR [antimicrobial resistance] in our Australian hospitals,’ he said. ‘So let’s make it the standard of care.’
How common are penicillin allergy labels?
Very. In a National Antimicrobial Prescribing Survey (NAPS) audit at a Melbourne tertiary hospital, 10% of inpatients carried a penicillin allergy or adverse reaction label, and 25% carried a label to at least one antimicrobial. A Perth hospital audit found antibiotic allergy labels in 18% of inpatients, and 71% of those were penicillin labels.
Not every label needs a challenge. A 2024 study across five Victorian community pharmacies found 18.4% of patients with a penicillin allergy label had a Type A or non-immune-mediated reaction history – the kind of label that can often be removed on history alone.
Why does a wrong label matter?
It pushes prescribing towards broader-spectrum alternatives. In the Melbourne NAPS audit, patients with an antimicrobial allergy label received antimicrobials for longer than those without (median 6 versus 4 days).
A United Kingdom cohort of more than 300,000 adults linked a documented penicillin allergy to a 69% higher risk of methicillin-resistant Staphylococcus aureus and a 26% higher risk of Clostridioides difficile (C. difficile) infection, driven largely by greater use of beta-lactam alternatives such as macrolides, clindamycin and fluoroquinolones.
Penicillin delabelling is therefore a way of ‘restoring penicillin narrow-spectrum antimicrobial usage’, Prof Trubiano said.
Who can assess a penicillin allergy?
Any trained clinician. ‘Non-allergists can safely perform penicillin allergy assessment, and we have a range of antibiotic allergy assessment tools available to us,’ Prof Trubiano said. ‘We’ve also got some great traffic light systems – green being low risk, somebody that’s unlikely to have a penicillin allergy.’
PEN-FAST is the best-known phenotyping tool, which Prof Trubiano said has now been used in more than 21 countries and validated in more than 15 settings – including emergency departments, inpatient care, paediatrics and low- to middle-income countries.
‘It really is a tool that anybody can use to risk-assess somebody’s penicillin allergy, and we know it can be used by an allergist and a non-allergist from our own validated data,’ he said.
How does PEN-FAST work?
In the original validation study, led by Prof Trubiano, PEN-FAST allocates points based on the following criteria:
- allergic reaction 5 or fewer years ago (2 points)
- anaphylaxis, angioedema or severe cutaneous adverse reaction (2 points)
- treatment required for an allergic reaction (1 point).
A score below 3 classifies the allergy as low risk, with a negative predictive value of 96.3% for a true penicillin allergy.
‘When somebody is assessed as low risk by those assessment tools, we know they can go on to receive a direct oral challenge,’ Prof Trubiano said.
Is direct oral challenge safe for low-risk patients?
Yes. A randomised controlled trial compared direct oral challenge (a simple supervised test dose) with skin testing followed by a test dose in patients with a PEN-FAST score below 3. Overall, 95% of patients were negative on direct oral challenge, with Prof Trubiano noting meta-analysis data show that figure holds globally.
‘People always want more evidence, so we’ve got more,’ he said, pointing to five further randomised controlled trials showing the same safety data.
‘We now have level 1A evidence that penicillin allergy delabelling by direct oral challenge is safe. But it’s not enough to show something safe. We have to implement it into healthcare, and we have to demonstrate AMS benefit.’
What happens when hospitals roll out delabelling?
Prescribing shifts fast, according to the International Network of Antibiotic Allergy Nations (iNAAN) study, which ran across 40 hospitals in eight countries between November 2022 and May 2025.
Of 5,121 patients assessed, 1,573 had a direct oral challenge and 1,502 (95.5%) of those patients were delabelled. Serious adverse events were rare, at 0.4%.
Importantly, Prof Trubiano said the process was ‘safe irrespective of the discipline leading it – nursing, pharmacy, medical clinician’.
‘We saw an immediate impact on prescribing,’ he said. Patients who had a challenge were 13 times more likely to be prescribed penicillin at 90 days, and less likely to receive World Health Organization Watch or Reserve antibiotics.
‘The impact went beyond prescribing; it went to AMR,’ he said, pointing to reductions in multidrug-resistant Gram-negative infections and C. difficile infections. Length of stay was unaffected.
Compliance was also high (90% across the study period). ‘We saw it adopted, and this is key. And we saw it implemented,’ Prof Trubiano added.
Have pharmacists already led delabelling?
Yes. A pharmacist-led penicillin allergy ward round at Austin Health ran for more than 3 years, with pharmacists using PEN-FAST and a validated allergy assessment tool to work through inpatient labels.
The study found 23.2% had a non-immune-mediated label and were delabelled directly, without a challenge – with 95.2% of low-risk patients delabelled after direct oral challenge.
Antimicrobial prescribing during admission and after discharge also improved.
Now, delabelling is heading beyond hospitals. ‘The next steps for us are looking beyond healthcare, looking to community practice, First Nations practice, to look at children,’ Prof Trubiano said.
