ORIGINAL ARTICLE Year : 2023  |  Volume : 9  |  Issue : 2  |  Page : 67-72

The effect of reported penicillin allergy on length of stay for patients admitted to internal medicine services: A retrospective cohort study

Susan Linn Evenhouse1, Jingwen Zhang2, Samuel Owens Schumann1
1 Department of Medicine, Division of General Internal Medicine, Medical University of South Carolina, Charleston, SC, USA
2 Department of Medicine, Section of Health Systems Research and Policy, Medical University of South Carolina, Charleston, SC, USA

Date of Submission03-Jun-2022Date of Acceptance15-Apr-2023Date of Web Publication26-Jun-2023

Correspondence Address:
Dr. Samuel Owens Schumann
Department of Medicine, Division of General Internal Medicine, Medical University of South Carolina, 135 Rutledge Avenue, MSC 591, Suite 1240, Charleston 29425, SC
USA

Source of Support: None, Conflict of Interest: None

Check

DOI: 10.4103/ijam.ijam_48_22

Introduction: Penicillin allergy is the most common medication allergy reported in the United States; however, >90% of these patients can tolerate penicillins. Reported penicillin allergy is associated with increased morbidity and mortality for patients. This retrospective, observational study evaluates the effect of reported penicillin allergy on length of stay (LOS) for adult patients admitted to the internal medicine services of a tertiary care, academic medical center.
Materials and Methods: We evaluated adult patients admitted to internal medicine services from June 2014 to June 2019 comparing those who reported penicillin allergy to those who did not. Patients who did not receive antibiotics were excluded. Univariate analyses were performed on multiple demographic and clinical variables between patients who do and do not report a penicillin allergy. Multivariable linear regression was used to model the effect of reported penicillin allergy on LOS.
Results: 14,144 patients were admitted to internal medicines services during the study period, with 8697 patients receiving an antibiotic during their admission. 10.5% of these patients reported a penicillin allergy. The unadjusted mean LOS for patients who report penicillin allergy was 10.4 days, while the unadjusted mean LOS for patients who do not report penicillin allergy was 9.3 days. Adjusting for covariates, the estimated LOS for patients who report penicillin allergy was 11.1 days, while the estimated LOS for patients who do not report penicillin allergy was 9.5.
Conclusions: A reported penicillin allergy is associated with an increase in estimated LOS of 1.6 days.
The following core competencies are addressed in this article: Systems-based practice, Medical knowledge.

Keywords: Inpatient, length of stay, penicillin allergy

How to cite this article:
Evenhouse SL, Zhang J, Schumann SO. The effect of reported penicillin allergy on length of stay for patients admitted to internal medicine services: A retrospective cohort study. Int J Acad Med 2023;9:67-72
How to cite this URL:
Evenhouse SL, Zhang J, Schumann SO. The effect of reported penicillin allergy on length of stay for patients admitted to internal medicine services: A retrospective cohort study. Int J Acad Med [serial online] 2023 [cited 2023 Jun 26];9:67-72. Available from: https://www.ijam-web.org/text.asp?2023/9/2/67/379347   Introduction 

Penicillin allergy is the most common medication allergy reported in the United States (US).[1] Approximately 10% of the general US and Canadian populations report a beta-lactam or penicillin allergy.[1],[2] The incidence of reported penicillin allergy is even higher (13%–15%) in hospitalized patients.[2],[3],[4] Despite the high prevalence of reported penicillin allergy, 90%–95% of people with a reported penicillin allergy can tolerate penicillin antibiotics.[5] In addition, only 7.78% of patients who report a penicillin allergy have a positive skin test or oral challenge (if the skin test is negative).[6] This discrepancy from reported penicillin allergy to true penicillin allergy has been attributed to multiple causes: The initial reaction to the medication was not a true allergic (IgE) mediated reaction, the patient reports a penicillin allergy due to a family history of penicillin allergy or the patient's symptoms were due to a viral illness instead of the antibiotics.

Reported penicillin allergy is associated with increased morbidity. Hospitalized patients with reported penicillin allergy have higher rates of methicillin-resistant Staphylococcus aureus, Clostridium difficile and vancomycin-resistant Enterococcus infections,[7],[8],[9] higher antibiotic costs,[2],[10],[11] higher rates of intensive care unit (ICU) transfer,[12] and longer length of stay (LOS)[7],[10],[12],[13],[14],[15],[16] compared to patients without a reported penicillin allergy. Data on whether reported penicillin allergy is associated with increased in hospital and all-cause mortality is conflicting with studies showing increased association, decreased association and no association.[8],[13],[14],[16],[17] Patients with a reported penicillin allergy are more likely to be prescribed multiple antibiotics and antibiotics with worse side effect profiles than penicillin antibiotics (fluoroquinolones, clindamycin, and vancomycin).[7],[8] Reported penicillin allergy was associated with an increased time to the administration of the first dose of antibiotics by almost an hour in patients with sepsis, pneumonia, urinary tract infection and bacteremia in one emergency department study.[18] Although reported penicillin allergy is associated with increased health-care costs and worse health-care outcomes, only 0.1% of all patients with a reported penicillin allergy undergo allergy testing.[16],[19]

The aims of this study are to quantify the prevalence of penicillin allergy reported by adult inpatients and examine the health-care outcomes for these patients with a reported penicillin allergy admitted to internal medicine services (ICU, general medicine, and subspecialty medicine teams) at a large academic medical center. This study looks to validate other studies[7],[10],[12],[13],[14],[16] which have showed higher LOS for patients with reported penicillin allergy compared to patients without a reported penicillin allergy.

  Methods 

Study design and exposure

This was a retrospective cohort study of adult patients admitted to internal medicine services at a tertiary care, academic medical center in the southeastern US from June 2014 to June 2019 designed to evaluate the outcomes associated with reported penicillin allergy. This study was conducted using administrative data at the patient level. This study was submitted to our institutions Internal Review Board and determined to be exempt from review. Patients with a reported penicillin allergy in their chart were defined as the exposed cohort. Any reported reaction (hives, serum sickness, anaphylaxis etc.) to a penicillin antibiotic was considered a penicillin allergy. Patients without a reported penicillin allergy on their chart were the comparison cohort. For this study, penicillin allergy includes all antibiotics in the penicillin class, but does not include other beta lactam antibiotics such as cephalosporins.

Patient population

The study included all patients over 18 years of age who were admitted internal medicine services. Admission to an internal medicine service was defined as the patient being discharged from one of the following services: hospitalist, general internal medicine (teaching), cardiology, malignant hematology, hepatology, critical care (pulmonology), or other. Discharging service was used to define the treating team as historically this variable at the administrative level has more accurately identified the treating team an “admitting service” variable. If a patient was admitted multiple times during the study period to internal medicine services, data were only included from their first admission, which is in keeping with other studies on this subject.[13] Patients who did not receive antibiotics during their admission were excluded from the study as penicillin allergy reported in their chart was not believed impact medical decision-making.

Study outcomes and variables

The primary outcome was LOS (days) occurring after the first dose of antibiotics. The secondary outcomes included in hospital mortality and discharge disposition. Discharge disposition is defined by the location or level of care to which a patient is discharged. Discharge dispositions reported in this study include the patient's home, rehabilitation or skilled nursing facilities, hospice (including home hospice or an inpatient hospice facility) or death. To evaluate the characteristics of the study population, demographic and comorbid disease data were collected on all patients included in the study. Comorbidities were identified by International Classification of Diseases - 9/10 codes and included in multivariable regressions as independent dichotomous variables. Co-morbid disease data were used to also calculate the Charleson Co-morbidity Index (CCI) for univariate assessment of aggregated patient morbidity.[20],[21] Limited data were available on social determinants of health; therefore, we used patient zip codes linked to the 2010 Census data to determine the poverty rate of the patient's area of residence. A dichotomous poverty variable was assigned a value of 1 if the patient's zip code had ≥25% of citizens below the federal poverty level (FPL).

Statistical analysis

Univariate analyses were performed using the Pearson's Chi-square for the categorical variables and the Student's t-test for continuous variables, comparing patients who report and do not reported a penicillin allergy. Multivariable linear regression (Proc Genmod with a gamma distribution) was used to model the relationship between penicillin allergy and LOS, adjusting for other variables. Least square means and 95% confidence intervals (CI) are reported. Multicollinearity was assessed and if detected, highly correlated variables were removed based on statistical and clinical relevance. Independent dichotomous comorbidities were included in the models rather than CCI, to better control for the independent effects of comorbid disease. All statistical analyses were performed using SAS 9.4 (SAS Institute Inc., Cary, NC, USA), and significance was determined at the 5% level.

  Results 

45,736 patients were admitted to hospitals during the study period (July 2014–June 2019), with 14,255 of these patients being admitted to internal medicine services at the medical center. 237 patients were excluded because they were <18 years old and 111 patients were excluded because they did not have a discharge disposition listed in the chart. Of these 14,144 patients, 8697 received an antibiotic during their admission and were included in the final analysis [Table 1].

Of the 8697 patients who received an antibiotic, 10.5% reported a penicillin allergy. The mean age of patients in the reported penicillin allergy group was slightly older than the patients who did not report a penicillin allergy (60.9 ± 17.3 vs. 62.5 ± 16.5, P = 0.0132). Patients who reported a penicillin allergy were more likely to be female (59.6%) than male (40.4%) (P < 0.0001) and were more likely to be white (74.4%) than black (23%) (P < 0.0001). Patients who reported a penicillin allergy were more likely to live a further distance from the medical center than those who did not report a penicillin allergy. There was no statistical significance in the percentage of patients living in zip codes with ≥25% of citizens below the FPL.

The CCI was calculated for each patient included in the study to characterize their risk of 1-year all-cause mortality base on associated comorbid disease.[20],[21] Patients included in our study who reported a penicillin allergy had a lower mean and median CCI as compared to patients who did not report a penicillin allergy (4.7 ± 3 vs. 5.0 ± 3.2, P = 0.035). A diagnosis of uncomplicated diabetes was more frequently identified (16.7% vs. 14.0%, P = 0.0274) in patients who reported penicillin allergy and less frequently found in patients with a diagnosis of renal disease (26.4% vs. 29.5%, P = 0.0419) or cancer (17.6% vs. 22.6%, P = 0.0007) as compared to patients who do not report penicillin allergy. Patients in both groups were admitted most frequently to the general internal medicine services, followed by the hospitalist services. Patients who did not report penicillin allergy were more frequently admitted to subspecialty services of cardiology, hematology, and critical care (pulmonology).

Mean LOS was statistically and clinically different for patients in the study who reported penicillin allergy compared to those who did not report penicillin allergy at 10.4 ± 15.4 days versus 9.3 ± 18 days, respectively (P = 0.0001) [Table 2]. Median LOS was also statistically and clinically different for patients who reported penicillin allergy compared to those who did not at 6.0 days versus 5.0 days (P = 0.001).

The secondary outcome was discharge disposition. Patients who reported a penicillin allergy were less likely to die while in the hospital (12.2% vs. 13.9%, P < 0.0001). In addition, patients with a reported penicillin allergy were more likely to be discharged to a postacute care facility, ex. skilled nursing facility or subacute rehabilitation facility, (17.9% vs. 13.3%, P < 0.0001) and less likely to be discharged to hospice (4.8% vs. 8.6%, P < 0.0001).

Multivariable linear regression revealed the estimated LOS for patients who report penicillin allergy was 11.1 days (95% CI: 10.3–12.0), while the estimated LOS for patients who do not report penicillin allergy was 9.5 days (95% CI: 8.9–10.0). Thus, when adjusting for all aforementioned variables patients who report a penicillin allergy are at an increased risk for longer estimated LOS of 1.6 days compared to thosepatients who do not report a penicillin allergy.

  Discussion 

Our study evaluated the impact of reported penicillin allergy on LOS and discharge disposition. Our study validates other studies which found that the LOS for patients with a reported penicillin allergy was statistically and clinically significantly longer compared to patients without a reported penicillin allergy.[7],[10],[12],[13],[14],[16] Other studies have proposed the observed increase in LOS could be due to treatment failure associated with the use of second line antibiotics and increased rate of adverse events due to the administration of nonpenicillin antibiotics.[9],[13] We agree with this hypothesis, however our study was not designed to answer this specific question. Instead, we set out to identify if adult inpatients who report penicillin allergy were more likely to have specific comorbid diagnoses or be disproportionately admitted to specific services. Our goal was to identify the concentrations of patients who report penicillin allergy among disease states or admitted to clinical service lines. We found that inpatients who reported a penicillin allergy less frequently had a comorbid cancer diagnosis and were most frequently treated on the hospitalist and general internal medicine services. Other comorbid diagnoses were either not clinically or statistically significant.

Patients who reported a penicillin allergy had a lower CCI and thus lower likelihood of 1-year mortality than patients who did not report a penicillin allergy. This differs from other studies on adult inpatients who report a penicillin allergy, which showed an increase in CCI and 1-year mortality risk compared to patients who did not report a penicillin allergy.[14],[16] It is unclear why patients admitted to our center reporting penicillin allergy would have a lower CCI. However, these patients in our study were less likely to die in the hospital or be discharged to hospice care, thus the lower CCI appears accurate. This observation also helps to explain the longer LOS for patients reporting penicillin allergy; patients reporting penicillin allergy were more frequently discharged to postacute care facilities. It has been well documented for internal medicine and other inpatients that discharge to a postacute care facility is associated with increased LOS.[22],[23],[24],[25] Conversely, patients who did not report penicillin allergy and were more frequently discharged with hospice services, which is associated with a shorter LOS.[26] Of note our cohort was different from the aforementioned studies by Pérez-Encinas et al.[14] and Sousa-Pinto et al.[15],[16] who studied a Spanish national dataset and a Portuguese national dataset respectively.

Patients with a reported penicillin allergy were less likely to die during their hospitalization compared to patients without a reported penicillin allergy, reflecting the lower CCI score in our study for patients who report a penicillin allergy. This finding of lower odds of inpatient mortality (or no association with inpatient mortality) was noted in multiple other studies, which includes population level studies and single center studies.[9],[14],[16] It is unclear why this finding continues to occur, especially as other studies have shown an increase risk of adverse events and readmissions associated with reported penicillin allergy.[8],[9],[10] Patients with a reported penicillin allergy were more likely to be discharged to a rehabilitation or nursing facility compared to patients without a reported penicillin allergy. Although we did not specifically look at this in our study, perhaps, these patients required facility placement to receive the extended courses of nonpenicillin antibiotics.

Limitations

The prevalence of reported penicillin allergy among the patients included in our study was 10.5%, which is lower than previously reported values (13%–15%) for inpatients in the US.[3],[4] This is also notably higher than studies from other countries evaluating the prevalence of penicillin allergy amongst hospitalized patients.[14],[16] The prevalence of reported penicillin allergy among the patients in our study more closely reflects the estimated prevalence of penicillin allergy for the general population.[1] Although we believe our sample size of >8600 patients over a 5-year period to be appropriate to avoid sample bias, we cannot explain our variance from other US studies which report on inpatient penicillin allergy prevalence, although we believe that reported penicillin allergy can be difficult to measure.

While treatment with antibiotic therapy was an inclusion criterion, we did not report on the percentage of patients who reported penicillin allergy but still received a penicillin antibiotic during their hospitalization. It is possible that while patients had a reported penicillin allergy in their chart, this information was incorrect or did not influence whether a patient received a penicillin antibiotic. In addition, our study did not also evaluate for reported cephalosporin allergy as many patients with nonanaphylactic penicillin allergy can safely receive cephalosporins as an appropriate first line alternative to penicillin.

As with all retrospective observational cohort studies, we are subject to confounding bias. We attempted to mitigate this affect, by accounting for multiple demographic and clinical variables during our univariate analyses and our multivariable linear regression. Of note, however, we did not account for patients who reported multiple antibiotic allergies.

Future directions

Our study did not account for other beta lactam allergies, such as cephalosporin allergies. Future studies could additionally include other beta lactam allergies when evaluating patients admitted to internal medicine services. Our study excluded patients cared for by surgical or obstetrical services. As surgical and obstetrical patients frequently receive beta lactams as first line preprocedural antibiotics, future studies could evaluate LOS and surgical site outcomes for patients on these services who report penicillin allergy. We believe that all adult patients who report a penicillin allergy, especially those at increased risk for hospitalization, should undergo penicillin allergy testing during adulthood. However, due to access and resource limitations, most patients who report penicillin allergy currently cannot undergoing testing. We believe that future studies should identify patients at highest risk for adverse outcomes related to a reported penicillin allergy and later target these patient cohorts for penicillin allergy testing.

  Conclusions 

Our study found that the prevalence of reported penicillin allergy was on par with the reported prevalence of penicillin allergy for the general US population at 10.5%. Patients with a reported penicillin allergy are at an increased risk for longer LOS compared to patients without a reported penicillin allergy. Patients with a reported penicillin allergy who are at increased risk for hospitalization should be strongly considered for penicillin allergy skin testing.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Ethical conduct of research

Appropriate approval was obtained for the study center's Institutional Review Board for Human Research (IRB) with reference number Pro00091624. This study was determined to IRB Exempt. The IRB also recommended approval of the HIPAA Waiver of Authorization, as the criteria of the Privacy Rule were satisfied. This study was done according to the reporting quality, formatting, and reproducibility guidelines set forth by the EQUATOR Network.

 

  References 
1.Macy E. The clinical evaluation of penicillin allergy: What is necessary, sufficient and safe given the materials currently available? Clin Exp Allergy 2011;41:1498-501.  
    2.Picard M, Bégin P, Bouchard H, Cloutier J, Lacombe-Barrios J, Paradis J, et al. Treatment of patients with a history of penicillin allergy in a large tertiary-care academic hospital. J Allergy Clin Immunol Pract 2013;1:252-7.  
    3.Lee CE, Zembower TR, Fotis MA, Postelnick MJ, Greenberger PA, Peterson LR, et al. The incidence of antimicrobial allergies in hospitalized patients: Implications regarding prescribing patterns and emerging bacterial resistance. Arch Intern Med 2000;160:2819-22.  
    4.Baxter M, Bethune C, Powell R, Morgan M. Point prevalence of penicillin allergy in hospital inpatients. J Hosp Infect 2020;106:65-70.  
    5.Sacco KA, Bates A, Brigham TJ, Imam JS, Burton MC. Clinical outcomes following inpatient penicillin allergy testing: A systematic review and meta-analysis. Allergy 2017;72:1288-96.  
    6.Kuniyoshi Y, Tsujimoto Y, Banno M, Taito S, Ariie T, Kubota T, et al. Differences in the prevalence of positive penicillin allergy test results in children and adults. Postgrad Med 2021;133:875-6.  
    7.Macy E, Contreras R. Health care use and serious infection prevalence associated with penicillin “allergy” in hospitalized patients: A cohort study. J Allergy Clin Immunol 2014;133:790-6.  
    8.Blumenthal KG, Lu N, Zhang Y, Li Y, Walensky RP, Choi HK. Risk of meticillin resistant Staphylococcus aureus and Clostridium difficile in patients with a documented penicillin allergy: Population based matched cohort study. BMJ 2018;361:k2400.  
    9.MacFadden DR, LaDelfa A, Leen J, Gold WL, Daneman N, Weber E, et al. Impact of reported beta-lactam allergy on inpatient outcomes: A multicenter prospective cohort study. Clin Infect Dis 2016;63:904-10.  
    10.Powell N, Honeyford K, Sandoe J. Impact of penicillin allergy records on antibiotic costs and length of hospital stay: A Single-Centre observational retrospective cohort. J Hosp Infect 2020;106:35-42.  
    11.Sade K, Holtzer I, Levo Y, Kivity S. The economic burden of antibiotic treatment of penicillin-allergic patients in internal medicine wards of a general tertiary care hospital. Clin Exp Allergy 2003;33:501-6.  
    12.Charneski L, Deshpande G, Smith SW. Impact of an antimicrobial allergy label in the medical record on clinical outcomes in hospitalized patients. Pharmacotherapy 2011;31:742-7.  
    13.Huang KG, Cluzet V, Hamilton K, Fadugba O. The impact of reported beta-lactam allergy in hospitalized patients with hematologic malignancies requiring antibiotics. Clin Infect Dis 2018;67:27-33.  
    14.Pérez-Encinas M, Lorenzo-Martínez S, Losa-García JE, Walter S, Tejedor-Alonso MA. Impact of penicillin allergy label on length of stay and mortality in hospitalized patients through a clinical administrative national dataset. Int Arch Allergy Immunol 2022;183:498-506.  
    15.Sousa-Pinto B, Araújo L, Freitas A, Delgado L. Hospitalizations in children with a penicillin allergy label: An assessment of healthcare impact. Int Arch Allergy Immunol 2018;176:234-8.  
    16.Sousa-Pinto B, Cardoso-Fernandes A, Araújo L, Fonseca JA, Freitas A, Delgado L. Clinical and economic burden of hospitalizations with registration of penicillin allergy. Ann Allergy Asthma Immunol 2018;120:190-4.e2.  
    17.Blumenthal KG, Lu N, Zhang Y, Walensky RP, Choi HK. Recorded penicillin allergy and risk of mortality: A population-based matched cohort study. J Gen Intern Med 2019;34:1685-7.  
    18.Conway EL, Lin K, Sellick JA, Kurtzhalts K, Carbo J, Ott MC, et al. Impact of penicillin allergy on time to first dose of antimicrobial therapy and clinical outcomes. Clin Ther 2017;39:2276-83.  
    19.Joint Task Force on Practice Parameters, American Academy of Allergy, Asthma and Immunology, American College of Allergy, Asthma and Immunology, Joint Council of Allergy, et al. Drug allergy: An updated practice parameter. Ann Allergy Asthma Immunol 2010;105:259-73.  
    20.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: Development and validation. J Chronic Dis 1987;40:373-83.  
    21.Quan H, Sundararajan V, Halfon P, Fong A, Burnand B, Luthi JC, et al. Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care 2005;43:1130-9.  
    22.Barba R, Marco J, Canora J, Plaza S, Juncos SN, Hinojosa J, et al. Prolonged length of stay in hospitalized internal medicine patients. Eur J Intern Med 2015;26:772-5.  
    23.Foer D, Ornstein K, Soriano TA, Kathuria N, Dunn A. Nonmedical factors associated with prolonged hospital length of stay in an urban homebound population. J Hosp Med 2012;7:73-8.  
    24.Anderson ME, Glasheen JJ, Anoff D, Pierce R, Capp R, Jones CD. Understanding predictors of prolonged hospitalizations among general medicine patients: A guide and preliminary analysis. J Hosp Med 2015;10:623-6.  
    25.Siegler JE, Boehme AK, Fowler BD, George AJ, Monlezun DJ, Albright KC, et al. Inpatient rehabilitation centers and concern for increasing volume of ischemic stroke patients requiring rehabilitation. South Med J 2013;106:693-6.  
    26.Weckmann MT, Freund K, Bay C, Broderick A. Medical manuscripts impact of hospice enrollment on cost and length of stay of a terminal admission. Am J Hosp Palliat Care 2013;30:576-8.  
    

 
 

  [Table 1], [Table 2]