The Effect of Neighborhood-Level Median Household Income on Emory Healthcare Outpatients’ Risk of Antibiotic Non-Susceptible Uropathogens, 2022-2023 Open Access

Mwansa, Chanda (Spring 2024)

Permanent URL: https://etd.library.emory.edu/concern/etds/br86b533t?locale=en
Published

Abstract

Introduction: Urinary tract infections (UTIs) are a common outpatient complaint, most frequently caused by gram-negative bacteria such as Escherichia coli and Klebsiella pneumoniae. Continued emergence of antibiotic resistance increases the risk of harboring resistant bacteria in the urinary tract which can result in an antibiotic resistant UTI. However, data on how neighborhood-level socioeconomic status impacts this risk is scarce. We examine the effect of census block group-level median household income (MHI) on the risk of having a uropathogen resistant to first-line antibiotics,i.e trimethoprim/sulfamethoxazole (TMP/SMX) and nitrofurantoin, among patients attending Emory Healthcare outpatient facilities.

Methods: We combined electronic health record data and antibiotic susceptibility test results for urine E. coliand K. pneumoniae isolates collected at Emory Healthcare outpatient facilities in the Greater Atlanta Area of Georgia between October 2022 and September 2023. We used 2017-2021 US Census data to determine MHI for each patient’s block-group of residence. We performed a logistic regression with a priori risk factors using a generalized estimating equation. A sub-group analysis was performed for patients with diabetes, who are at higher risk of antibiotic-resistant infections.

Results: The analyses included 9,327 urine E. coli and K. pneumoniae isolates. K. pneumoniae had a higher prevalence of non-susceptibility to first-line antibiotics (34.2%, vs 13.5% for E. coli; p <0.001). K. pneumoniaewas more likely to be non-susceptible to nitrofurantoin (27.4%, vs. 1.8% for E. coli; p <0.001) and less likely to be non-susceptible to TMP/SMX (6.8% vs 11.6% for E. coli; p <0.001). The risk of having a non-susceptible urine isolate generally decreased as block group-level MHI increased after controlling for patient age, sex and race/ethnicity, along with neighborhood-level educational attainment, language and race/ethnicity. Compared to the lowest MHI quintile, patients living in block groups with the highest MHI quintile had 0.78 the odds of harboring a uropathogen non-susceptible to first-line antibiotics (95%CI: 0.64, 0.95). There was no association among people with diabetes.

Conclusion: We found that higher neighborhood-level MHI was associated with lower individual risk of first-line antibiotic-non-susceptible uropathogen. Antibiotic prescribing should be guided by local resistance patterns and individual antibiotic susceptibility results.

Table of Contents

Introduction 1 

Literature Review 3 

Methods 12 

Figure 1: Diagram showing exclusion process for urine E. coli & K. pneumoniae isolates from Emory Healthcare outpatients 14 

Figure 2: Directed acyclic graph showing potential confounders of the relationship between block group-level median household income and patient’s risk of non-susceptible urine isolate 16 

Results 18 

Table 1: Sociodemographic Characteristics of 3,867 Emory Healthcare Outpatients with 9,325 Urine Culture Isolates, Stratified by First-Line Antibiotic Susceptibility, 2022-2023 19 

Table 2: Sociodemographic Characteristics of 3,867 Emory Healthcare Outpatients with 9,325 Urine Culture Isolates, Stratified by Organism, 2022-2023 20 

Table 3: Patterns of Non-Susceptibility to First Line Antibiotics for 9,325 Urine Culture Isolates from 3,867 Emory Healthcare Outpatients, Stratified by Organism, 2022-2023 21 

Figure 3: Effect of median household income in a patient’s neighborhood on their risk of first-line antibiotic non-susceptible uropathogen, stratified by type 2 diabetes mellitus 22 

Discussion 23 

Public Health Implications & Recommendations 28 

References 30 

Supplementary Table 1: American Community Survey (2017-2021) Variables Used to Derive Block Group-Level Median Household Income and Other Covariates of Interest 47 

About this Master's Thesis

Rights statement
  • Permission granted by the author to include this thesis or dissertation in this repository. All rights reserved by the author. Please contact the author for information regarding the reproduction and use of this thesis or dissertation.
School
Department
Degree
Submission
Language
  • English
Research Field
Keyword
Committee Chair / Thesis Advisor
Last modified

Primary PDF

Supplemental Files