Penicillin-allergy notes in patient files are widely recognized as containing many false flags, and delabeling those has long been the responsibility of allergy specialists in the outpatient setting.
Efforts to relocate this function to the hospital setting – and create a more straightforward process – are leading Vanderbilt University Medical Center allergist Cosby Stone, M.D. and colleagues to develop a protocol and EHR tools that signal when and how unconfirmed penicillin allergies should be investigated.
Whether applied mistakenly due to co-occurring but unrelated symptoms – or in an attempt to avoid a family member’s adverse reaction – the percentage of incorrect listings of penicillin allergies in patient files exceeds that of valid labels.
“For people who have low-risk symptoms, especially those that are from a long time ago, those patients can be tested in a wide variety of settings,” said Stone, an assistant professor of medicine. “I think we should be doing that as much as possible.”
And with so many cases to address, the task requires broad support.
“There are too many people reporting too many untested penicillin allergies for allergy doctors in the U.S. to handle this on their own. We need an all-hands-on-deck approach,” Stone said.
To begin marshalling these new forces, Stone and colleagues recently introduced the protocol to 12 inpatient units following initial testing in the Vanderbilt ICU. As presented at the 2024 American Academy of Allergy, Asthma and Immunology annual meeting, implementation of the protocol led to a rise in penicillin allergy-risk assessments from 2.7 to 8.4 percent and an increase in delabeling from 3.0 to 4.4 percent.
The protocol is now available across inpatient units throughout Vanderbilt University Medical Center, with pilot programs in the outpatient setting currently underway.
“For people who have low-risk symptoms, especially those that are from a long time ago, those patients can be tested in a wide variety of settings.”
Risk-Stratification in the ICU
Although the medical records for 10 percent of the United States population flag a penicillin allergy, less than 1 percent of the population are truly penicillin allergic. When alternate antibiotics are prescribed, patients may be receiving less effective antimicrobial care and the risk for antimicrobial resistance increases.
“There’s so much evidence now showing that if we aren’t testing penicillin allergies, that patients are getting second-class outcomes,” Stone said.
To guide providers in spotting and evaluating a low-risk penicillin allergy, Stone and Elizabeth Philips, M.D., the John A. Oates Chair in Clinical Research at the Vanderbilt Drug Allergy Clinic, set to work on a penicillin allergy-risk stratification protocol.
[See a flowchart illustrating the protocol here.]
Testing the protocol in the Vanderbilt ICU proved the approach to be 99-percent accurate in identifying patients who are at low risk for a penicillin reaction. Using amoxicillin, the patients are offered an oral challenge and monitored for an hour. Low-risk patients may still display a rash, which occurs in about one in 200 cases, Stone said.
“Four years later, we’re getting close to having removed 400 penicillin allergies at the point-of-care in the ICU.”
Stone and Philips, alongside international colleagues, have shown that it’s safe to offer patients with a low-risk penicillin allergy label an oral amoxicillin test dose without first administering a skin test, an approach that is now recommended by national allergy associations. But Stone stresses that patients who report symptoms that sound like anaphylaxis and other severe reactions still need expert consultations with an allergy specialist.
“Four years later,” says Stone, “we’re getting close to having removed 400 penicillin allergies at the point-of-care in the ICU. Our rate of reported penicillin allergies in ICU patients came down, especially among the really sick patients who were in and out of the hospital a lot.”
Calling for National Guidelines
According to Stone, 85 percent of patients approached in the hospital setting will agree to a challenge to their low-risk penicillin allergy.
“The thing that makes patients more likely to say ‘no’ is if the provider who is explaining it to them doesn’t feel entirely comfortable with it themselves,” Stone said.
This insight points to a need for greater awareness and education – a barrier that has to be overcome.
“We’ve all been trained for decades that penicillin allergies are a hard-stop sign. And now I’m telling you that it’s a stop sign that needs to come down for most patients.”
“We’ve all been trained for decades that penicillin allergies are a hard-stop sign. And now I’m telling you that it’s a stop sign that needs to come down for most patients,” Stone said. “That disconnect with how we were trained and the new evidence that most of these allergies are past their expiration date just makes people feel uncomfortable.”
While national medical societies have published specialty-specific practices for removing penicillin allergy labels, a lack of national guidelines has created a speed bump making it difficult to promote – to both patients and providers – the benefits and safety of penicillin-allergy delabeling.
