Low frequency of allergy referral for penicillin allergy evaluation in an urban Boston primary care setting.

Background: A key strategy to combat the public health crisis of antimicrobial resistance is to use appropriate antibiotics, which is difficult in patients with a penicillin allergy label. Objective: Our aim was to investigate racial and ethnic differences related to penicillin allergy labeling and referral to allergy/immunology in primary care. Methods: This was a retrospective study of Tufts Medical Center's Boston-based primary care patients in 2019. Univariable and multivariable logistic regression models were used to examine demographic associations with (1) penicillin allergy label and (2) allergist referral. Results: Of 21,918 primary care patients, 2,391 (11%) had a penicillin allergy label; of these, 249 (10%) had an allergist referral. In multivariable logistic regression models, older age (adjusted odds ratio [aOR] = 1.06 [95% CI = 1.04-1.09]) and female sex (aOR = 1.58 [95% CI = 1.44-1.74]) were associated with higher odds of penicillin allergy label carriage. Black race (aOR = 0.77 [95% CI = 0.69-0.87]) and Asian race (aOR = 0.47 [95% CI = 0.41-0.53]) were associated with lower odds of penicillin allergy label carriage. In multivariable regression, allergist referral was associated with female sex (aOR = 1.52 [95% CI = 1.10-2.10]) and Black race (aOR = 1.74 [95% CI = 1.25-2.45]). Of 93 patients (37%) who completed their allergy visit, 26 (28%) had received penicillin allergy evaluation or were scheduled to receive a penicillin allergy evaluation at a future visit. Conclusions: There were racial differences in penicillin allergy labeling and referral. Allergy referral for penicillin allergy assessment was rare. Larger studies are needed to assess penicillin allergy labeling and delabeling with an equity focus on optimizing patient health outcomes.