Antibiotic allergy labels strongly associated with disease burden in hospitalization
Click Here to Manage Email Alerts
Antibiotic allergy labels appeared strongly associated with disease burden but not with hospital length of stay, according to a study published in Annals of Allergy, Asthma & Immunology.
“While implementing an inpatient penicillin delabeling program at our institution, we were evaluating the potential economic impact of such a program,” Derek Lanoue, MD, MEcon, a PGY3 internal medicine resident at University of Ottawa, told Healio.
Lanoue and his colleagues reviewed studies showing increased lengths of hospital stays among patients with penicillin allergies. Based on their experience, he continued, patients with antibiotic labels tend to be sicker and more complicated than those without.
“We wanted to determine if the association between antibiotic allergy and hospital length of stay persisted after adjusting for overall patient sickness. We thought that the economic incentive for delabeling would be even stronger if the association persisted,” he said.
Lanoue and Carl van Walraven, MD, MSc, FRCPC, professor in the department of epidemiology and community medicine at University of Ottawa, examined data from 111,611 hospitalizations of 76,460 patients (mean age, 57.3 years ± 21.3; 58.2% women) at The Ottawa Hospital between Jan. 1, 2012, and Dec. 31, 2015. More than half of the patients had at least one comorbidity. The median length of stay was 4 days.
The researchers documented antibiotic allergy labels (AALs) in 14.8% of these hospitalizations, including penicillin (8.1%), other beta-lactams (2.4%) and non-beta-lactams (8.1%). Also, 549 hospitalizations involved patients with AALs for all three types of antibiotics.
Hospitalized patients with an AAL were notably sicker, the researchers wrote, and were older (mean age, 61.8 years vs. 56.6 years) and more likely to be women (69.5% vs. 56.3%) than those without an AAL.
Additionally, patients with an AAL had a greater prevalence of chronic disease (Charlson score 3: 34.4% vs. 24.2%), greater ED utilization ( 2 annual visits, 40.2% vs. 22%) and a more than doubled 1-year death risk (15.7% vs. 7.1%). They were more likely to be admitted from the ED as well (68.3% vs. 58.7%).
The hospital administered antibiotics during 42.1% of these admissions, including 20.6% who had an AAL, with 53.1% of them receiving antibiotics that were included on their AAL.
The AAL was identified after the offending antibiotic was administered to 21.5% of these patients. Administration of antibiotics lasted less than 24 hours in 52% of the other patients.
Treating physicians deemed AALs to be clinically unimportant, based on the prolonged administration of potentially offending antibiotics, in 20% of the AAL patients who received antibiotic treatment during their admission.
Further, there was no association between AAL and daily discharge likelihood, the researchers wrote.
For example, there was no significant change in discharge likelihood without antibiotics for patients with AALs for penicillin (adjusted OR = 0.99; 95% CI, 0.95-1.02), or nonpenicillin beta-lactam or non-beta-lactam antibiotics (aOR = 0.99; 95% CI, 0.93-1.05 for both).
Also, none of the interaction terms between any AAL and administration of antibiotics approached significance. According to the researchers, this indicates that daily discharge likelihood for patients who received antibiotics did not change based on their AAL status.
Hospitalizations with any AAL without antibiotic (aOR = 0.99; 95% CI, 0.93-1.05) and with antibiotic (aOR = 1.03; 95% CI, 0.97-1.09) or all three AAL subtypes without antibiotic (aOR = 0.99; 95% CI, 0.91-1.08) and with antibiotic (aOR = 1.01; 95% CI, 0.92-1.1) saw similar results, the researchers continued.
Even when they limited their analysis to patients who had very low disease burden, such as those admitted electively to the obstetrics service, the researchers said there was no association between AAL and decreased likelihood of daily discharge.
“Consistent with prior studies, we found in a univariate analysis that patients with penicillin allergy had increased lengths of hospital stay,” Lanoue said.
“However, antibiotic allergy labels were also strongly associated with a greater disease burden. After adjusting that disease burden, we found no significant association between antibiotic allergy label and hospital length of stay,” he added.
These results, along with the strong association between AAL and overall disease burden, should be considered in projecting the potential cost implications of delabeling antibiotic allergies among hospitalized patients, according to the researchers.
“Physicians should recognize that an antibiotic allergy label is strongly associated with a greater disease burden. That burden, rather than the allergy label itself, likely accounts for these patients’ increased hospital stay,” Lanoue said. “An antibiotic allergy label is a marker of greater health care interaction and should be questioned wherever it is appropriate.”
For more information:
Derek Lanoue, MD, MEcon, can be reached at delanoue@toh.ca.
Perspective
Back to Top
These findings are not particularly surprising, because we all recognize that previous associations between antibiotic allergies and length of stay do not equal causality. I also am not surprised that sicker patients have more antibiotic allergy labels.
This is a very statistics-heavy paper, so in this particular model, antibiotic allergy is not associated with an increased length of stay. But this study also did not look at anything outside of that particular admission, such as readmission, infection recurrence, other antibiotic adverse effects or antimicrobial resistance, all of which are other reasons to delabel antibiotic allergies.
Other studies support the importance of delabeling antibiotic allergy for reasons other than length of stay, but this is an important light with which to view the length-of-stay data. A lot of times, patients are referred for antibiotic allergy delabeling so they may be treated with the most appropriate antibiotic, and this is still the most important goal.
Innovative ways to better target appropriate patients for delabeling is an important step, such as developing models to predict the best patients to target. Further, innovating how delabeling takes place is also important, such as using telemedicine, e-consults or pharmacist-driven initiatives.
Collapse
Read more about
Click Here to Manage Email Alerts