Variation by Race in Antibiotics Prescribed for Hospitalized Patients With Skin and Soft Tissue Infections

This cohort study examines antibiotics prescribed and variations by race among hospitalized patients with skin and soft tissue infections (SSTIs).


Introduction
Antibiotic prescriptions for skin and soft tissue infection (SSTI) are increasing. 1 SSTI treatment guidelines consider a β-lactam, such as cefazolin, as first-line therapy to target Staphylococcus aureus and Streptococcus species; alternative SSTI treatments, such as clindamycin, are inferior. 2 Racial differences in treatment exist for several medical conditions. However, little is known about disparities in SSTI treatment. 3 We hypothesized that racial differences in the management of SSTI exist.

Methods
This cohort study conducted a subanalysis of multisite, cross-sectional data collected through a national survey of Acute Care Hospital Groups within Vizient Inc (October 16, 2018, to January 13, 2019) (eMethods in the Supplement), considering adult inpatients treated for SSTI. The indicator of interest was race determined from electronic health record and categorized as Asian, Black, American Indian or Alaska Native, White, or other (which also included those who declined). The outcome was antibiotic use. Mass General Brigham and Tufts institutional review boards reviewed the study and determined it exempt from review and the requirement for informed consent because it was not human participant research. This study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline. Because the number of individuals identified as Asian (n = 13), American Indian or Alaska Native (n = 5), and other races (n = 146) was small, we only compared the antibiotic use between Black (the referent group) and White individuals. To account for clustering by hospital, we used generalized estimating equations to obtain unadjusted and adjusted odds ratios (aOR) with 95% CI. All P values were 2-sided, and P Յ .05 was considered statistically significant. Statistical analysis was performed with SAS software version 9.4 (SAS Institute) in September 2021.

+ Supplemental content
Author affiliations and article information are listed at the end of this article.

Discussion
This study found that race was associated with differential management of SSTI. Black inpatients were less likely to receive cefazolin and more likely to receive clindamycin compared with White inpatients. Cefazolin is one of the first-line SSTI treatments. Clindamycin is not recommended given frequent dosing and high potential for adverse effects including Clostridioides difficile infection (CDI).  h Includes doxycycline (n = 61) and minocycline (n = 6).
i Includes ciprofloxacin (n = 36), levofloxacin (n = 29), and moxifloxacin (n = 3). Although prior studies of antibiotic choice for infections have revealed practice inconsistent with guidelines, 4,5 racial differences have not been investigated.

JAMA Network Open | Infectious Diseases
Black race was associated with increased clindamycin use, even after controlling for methicillinresistant Staphylococcus aureus (MRSA) colonization, infection and penicillin allergy. Although penicillin allergy is described as more prevalent in White patients, 6 we observed an increased prevalence of penicillin allergy in Black inpatients compared with White inpatients treated for SSTI.
Given that historical penicillin allergies are associated with increased clindamycin use and risk of CDI, but are often disproved with formal testing, 6 racial disparities in penicillin allergy documentation and assessment requires additional study.
Although our data came from a large sample of hospitals, these data may not be nationally representative. Race was extracted from the medical record and may have been incorrectly assigned.
Other race was not specified, and Hispanic ethnicity was not captured. Data on structural vulnerability such as income, employment, and education level, were not collected, limiting our evaluation of the impact of structural racism. Cross-sectional data did not permit determination of cumulative antibiotic utilization metrics.
We detected a potential racial disparity in antibiotic choice in this study. Future work should evaluate the determinants of this observed difference and devise interventions to achieve pharmacoequity.