Obstetrics and gynecology care as an opportunity for penicillin allergy testing: Results from a novel survey

Avani R. Patel: Writing – original draft, Investigation, Data curation, Conceptualization. Carmen L. Vazquez: Writing – review & editing, Data curation. William H. Kobak: Supervision, Conceptualization. Quetzal A. Class: Writing – original draft, Software, Resources, Project administration, Methodology, Formal analysis, Data curation.

This study was approved by the University of Illinois, Chicago Institutional Review Board (#2023-1408).

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

A total of 189 eligible pregnant patients met the inclusion criteria and were approached for participation via in-person QR code card at their routine clinic visit ( n  = 134, 71%) or MyChart message ( n  = 55, 29%) if the potential participant missed their scheduled in-person visit. A total of 39 survey responses were collected, all included in the final analysis, for a response rate of 21%. These included 35 responses from in-office responses and 4 via MyChart messaging. One of the 39 (2.6%) surveys was completed in Spanish, a percentage that reflects the population we serve. Contamination of survey responses was avoided because informational cards or messaging were not presented until either the end of the visit or the end of the completed linked survey (when completed from MyChart link). As presented in Table 1 , responders self-reported being primarily 30+ years old ( n  = 29, 78%), Black ( n  = 26, 68%), non-Hispanic ( n  = 31, 81.6%), and using public insurance ( n  = 21, 55%). Most respondents indicated that high school/general educational development (GED) certificate was their highest level of education ( n  = 19, 51%). Again, this sample follows the same demographic breakdown of our general population served. We noted no significant differences across primary and secondary survey outcomes across demographic factors. Table 1 Description of respondent demographics, allergy history, knowledge of penicillin allergy course, and allergy testing. Factor n (%) Total 39 Age (years)  25–29 6 (15.4)  30–34 10 (25.6)  35–40 8 (20.5)  41 or older 11 (28.2)  Unknown 4 (10.2) Race  Black 26 (66.7)  White 7 (17.9)  Asian 1 (2.6)  Other 5 (12.8) Ethnicity  Non-Hispanic 31 (79.5)  Hispanic 8 (20.5) Insurance  Public 21 (53.8)  Private 14 (35.9)  Other 4 (10.3) Completed education  High school or General Educational Development Cert. 19 (48.7)  Bachelor's degree 11 (28.2)  Graduate school or technical certificate 7 (17.9)  Unknown 2 (5.1) Age that allergy was discovered (years)  2 or younger 10 (25.6)  3–15 12 (30.8)  16–25 10 (25.6)  26–35 6 (15.4)  36 or older 1 (2.6) How allergy was discovered  Family report 14 (35.9)  Personal experience 24 (61.5)  Unknown 1 (2.5) Use of alternate antibiotic due to allergy  Yes 29 (74.4)  No 4 (10.3)  Unsure 6 (15.4) Knowledge that allergy testing exists  Yes 25 (64.1)  No 14 (35.2) Knowledge of how to obtain allergy testing  Yes 9 (23.1)  No 30 (76.9) Ob/Gyn has counseled you on allergy testing  Yes 6 (15.4)  No 33 (84.6) Received counseling that allergy testing is safe during pregnancy  Yes 1 (2.6)  No 38 (97.4) Would you undergo allergy testing while pregnant?  Yes 14 (35.9)  No 24 (61.5)  Unknown 1 (2.6) Would you undergo allergy testing while not pregnant?  Yes 30 (76.9)  No 8 (20.5)  Unknown 1 (2.6) Description of respondent demographics, allergy history, knowledge of penicillin allergy course, and allergy testing. Table 1 also presents that most participants stated last penicillin use was greater than 11 years ago ( n  = 26, 67%) and that penicillin allergy was discovered through personal experience of allergic reaction ( n  = 24, 62%). A majority ( n  = 29, 76%) experienced a need for penicillin but had to receive a different antibiotic because of their allergy. Only 63% (n = 24) had heard of allergy testing though they expressed a lack of knowledge in how to obtain allergy testing ( n  = 30, 76%). We found patients reported a lack of counseling from Ob/Gyn providers regarding allergy testing ( n  = 33, 85%). Only one respondent reported being told by their Ob/Gyn provider that allergy testing is safe during pregnancy ( n  = 1, 3%). Notably, Fisher's Exact test showed that significantly more patients reported considering allergy testing during pregnancy if their Ob/Gyn provider discussed allergy testing with them during a prior session [ n  = 5 (83.3%) versus n  = 9 (28.1%), OR = 12.78, Phi coefficient = 0.41, p  = .018]. However, a self-reported discussion with their provider was not associated with increased patient knowledge regarding how to obtain testing [ n  = 7 (21.2%) versus n  = 2 (33.3%), OR = 1.86, Phi coefficient = 0.10, p  = .61]. Table 2 presents that patients have a good basis of knowledge on penicillin allergy, with 26 (67%) participants correctly scoring 80% or higher on the knowledge quiz that contained 5 true/false questions covering knowledge on allergy course and testing accessibility (range = 40–100% correct, median percent correct = 80%, IQR = 25%). Table 2 Penicillin allergy knowledge assessment with correct answers in bold. Factor n (%) answer Allergies found in childhood may be resolved over time  True 31 (79.5)  False 8 (20.5) There are many reasons why penicillin is used in pregnancy  True 31 (79.5)  False 8 (20.5) Being pregnant can affect the results of an allergy test  True 12 (30.8)  False 27 (69.2) Allergy testing is safe to do during pregnancy  True 22 (56.4)  False 17 (43.6) Allergy testing is available through UI Health  True 39 (100.0)  False 0 (0.0) Penicillin allergy knowledge assessment with correct answers in bold.

We created a novel, online, anonymous survey to assess pregnant patients' knowledge of their penicillin allergy and allergy testing during and outside of pregnancy. The survey was developed by the authors and underwent two revisions after editorial review by faculty from the University of Illinois, Chicago Department of Ob/Gyn. A native Spanish speaker translated our survey into Spanish, and we back-translated it to English through the university's translation services. The survey was piloted on our non-patient medical student author (C.L.V.) and other faculty within the department. Qualtrics was used to host the survey. Both the English and Spanish versions are presented in Supplemental Materials Tables A.1 and A.2. Our project was approved by our university's Institutional Review Board as an exempt study (#2023–1408). Potential participants met inclusion if they had a penicillin allergy listed in their electronic medical record, being 18 years old or older, and being pregnant with an Ob/Gyn appointment of any kind during the study period. Author A.P. placed an electronic EPIC “flag” in the patient electronic medical record on an ongoing basis during the study period. EPIC flags are customizable alerts that appear in the patient banner of their chart. Upon clinic check-in, these flagged patients were provided a study information sheet with a QR code that linked to our anonymous online survey by check-in staff. The patient then had the opportunity to voluntarily complete the online anonymous survey while they waited for their provider. All patients who completed the survey did so prior to their clinic visit. At the conclusion of the clinic appointment, the nurse collected the patient's other paperwork and provided the patient with an information card describing the efficacy, safety, and benefits of penicillin allergy testing, along with contact information for scheduling a testing appointment. If the potential patient did not attend their scheduled in-person appointment, they were sent a MyChart message containing study information, the survey QR code, and a link to the anonymous survey. The same information provided at the conclusion of the in-person visit was included at the end of the survey for those who completed the survey via the MyChart link. Survey responses were collected from January 16, 2024 through March 31, 2024. Our primary outcomes included personal experiences with penicillin allergy (e.g., duration, use of alternative antibiotics). Secondary outcomes included the percent correct on the 5-question true/false quiz at the end of our survey. The quiz assessed allergy course knowledge and the accessibility of allergy testing. Planned analyses included examining whether primary and secondary outcomes differed across self-reported demographic factors and self-reported conversations with their Ob/Gyn provider about their penicillin allergy. Descriptive and Fisher's Exact Test were used in analyses using SPSS 30.

We assessed openness to penicillin allergy testing and knowledge of penicillin allergy among pregnant minority women attending an academic training hospital in an urban setting. We found that most patients became aware of their allergy during childhood and that they had to alter their antibiotic prescriptions due to their perceived allergy. In general, patients showed good basic knowledge about penicillin allergy on a 5-question true/false quiz. Patients were less aware of changes in allergies over time and how to obtain allergy testing. Importantly, patients reported being open to allergy testing during pregnancy if they also reportedly had a prior discussion about their allergy with their provider. This finding supports previous research showing that uptake of penicillin allergy testing is moderate in outpatient care clinics during pregnancy [9] , though we did not study uptake. Most (76.9%) of our patients reported not knowing how to obtain penicillin allergy testing. We interpret our findings to show that the incorporation of a discussion of penicillin allergy history, the benefit of testing for allergy resolution, and referral to an allergy clinic may benefit patients, especially during pregnancy [3] . Our study has several limitations that should be considered. First, there may be selection bias. Access to the survey was limited to those with a “smart” device capable of scanning a QR code or those with a computer or smart device at home. While we did create both an English and a Spanish version of the survey, those with the least health literacy may have been left out of our sample due to a lack of an internet-connected device or a preference for another language. Second, we had limited power due to a small sample size, as we approached this project as exploratory and descriptive. We were therefore unable to control for confounders using multivariable analyses, and generalizability may be limited. Response rates were lower for MyChart responders than for in-person connected-device responders, as is expected by the format differences. Part of this response rate difference may have also been driven by health behavior differences, as those who received MyChart invites had already missed their scheduled in-person visit. Our results are obviously not generalizable to people who are not pregnant and future work may consider promoting penicillin allergy knowledge and testing attitudes in males. Third, there may have been a “patient-effect,” in which the reason for the patient-provider conversation was the patient's interest rather than the provider's initiative. This may have also driven the patient to be more open to penicillin allergy testing in the first place. More women believe that they have a penicillin allergy than truly do have an allergy [1] . This assumption may prevent them from receiving the best line of antibiotics for many issues, especially during pregnancy, and has implications for antibiotic resistance as a population [3] , [4] . Our work adds that the option to test for a persistent penicillin allergy may be promoted by the Ob/Gyn provider and supports the need to incorporate counseling into Ob/Gyn care. If the infrastructure is in place at a given institution, creating a streamlined scheduling process would likely increase patient compliance with testing. Voicing openness on a survey is different from engaging in the test, however. One previous study showed that 74% of pregnant women who were offered penicillin allergy testing accepted the option but only 36% followed-through with completing the testing [9] .Ob/Gyn and Allergy and Immunology departments can work together to promote the health and well-being of women during and beyond the prenatal period. Further research needs to focus on the contents of provider counseling and on the action of how outpatient care centers seamlessly integrate penicillin testing into prenatal, and annual well-woman, visits [3] .

No generative AI or AI-assisted technology was used in any way during any part of this project or in the manuscript preparation process.

Allergy to penicillin, a beta-lactam antibiotic, is one of the most commonly reported antibiotic allergies [1] . While 1 out of 10 women report penicillin allergy, 9 out of 10 pregnant women who obtain allergy testing are not found to be allergic [2] . As immunoglobulin-E (IgE) - mediated penicillin allergy wanes over time, approximately 80% of patients become tolerant after a decade [2] . However, most reproductive-age women do not undergo penicillin allergy verification testing, leading to individuals being unaware of their true allergy status [3] . Unverified penicillin allergy is associated with increased maternal morbidity, increased likelihood of postpartum endometriosis, use of less effective antibiotic regimens, and increased antibiotic-resistant microorganisms [1] , [4] , [5] . Further, neonates born to women with unverified penicillin allergy have longer hospital stays than do those without an allergy [5] . These risks can be diminished with increased awareness of an individual's penicillin allergy status through allergy testing as recommended by primary care providers. The reason behind a lack of penicillin allergy verification testing, however, is not readily understood. Importantly, verifying a penicillin allergy can help increase antibiotic stewardship and provide long-term benefit to patients. Penicillin and its adjacent antibiotics are used widely throughout the field of Obstetrics and Gynecology (Ob/Gyn) [3] . Pregnancy is a crucial time that benefits from penicillin in management for Group B Streptococcus (GBS) prophylaxis, intra-amniotic infection and inflammation, preterm labor rupture of membranes, and surgical prophylaxis [4] .Verification of penicillin allergy is therefore crucial. Ob/Gyn providers do not routinely recommend penicillin allergy testing during pregnancy despite it being a safe and recommended procedure [2] , [3] . Indeed, the American College of Obstetricians and Gynecologists (ACOG) states that penicillin is the first-line treatment for GBS infection and that penicillin allergy testing in pregnancy is recommended for benefit to the mother and baby [6] . Further, several studies have shown the need for testing within the pregnant population [3] , [7] , [8] . Previous research suggests that streamlined outpatient penicillin allergy testing can be conducted in a pregnant population [7] , [9] . These studies show 93–99% of patients do not have a penicillin allergy, and no patients experienced an “adverse event” from the allergy testing [7] , [9] . Less is known, however, about knowledge and openness to penicillin allergy testing during pregnancy. In the current exploratory and primarily descriptive study, we gather patient attitudes on penicillin allergy testing during and outside of pregnancy in pregnant minority women living in an urban setting who have a penicillin allergy listed in their electronic medical record. We use a novel, anonymous, online survey to examine knowledge regarding common allergy course and allergy testing opportunities. Spanish- and English-language surveys were completed during in-person Ob/Gyn visits or via MyChart messaging at home. Our study adds to the understanding of patients' preferences and knowledge regarding penicillin allergy, with implications for practice changes.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.