UC GIS Week

Spatial distribution of extended-spectrum beta-lactamase-producing Escherichia coli isolated from patients with community-onset bacteriuria

Introduction: The prevalence of infections caused by extended-spectrum beta-lactamase producing Escherichia coli (ESBL-E. coli) is increasing worldwide. There is growing evidence that community-onset bacteriuria caused by ESBL-E. coli may occur as outbreaks. With data from a public healthcare system in San Francisco, we found increasing prevalence of ESBL-E. coli causing community-onset bacteriuria over 6 years. We examined spatial distribution of community-onset ESBL-E. coli bacteriuria to identify such outbreaks.

Methods: We collected electronic health record data on patients with E. coli isolated from urine cultures in a public healthcare system from January 2014 to March 2020, including geocoded addresses. We identified community-onset bacteriuria as bacteriuria diagnosed < 48 hours after hospital admission or in individuals seen in primary care or emergency services with no hospitalizations in the past 90 days (N=7,751). We mapped all cases of ESBL-E. coli bacteriuria. We performed global and local Moran’s I to identify significant proximity between cases of ESBL-E. coli bacteriuria and to identify spatial clusters and places with unexpectedly high ESBL-E. coli incidence, compared to the incidence of non-ESBL-E. coli.

Results: Patients with community-onset bacteriuria were predominantly women (86%) and mean age was 46 years. The study population was multiethnic, with 42% Latina, 19% Asian, 16% White, and 13% Black patients. Preliminary results did not show spatial clusters of ESBL-E. coli compared to non-ESBL-E. coli. Conclusion: Preliminary maps show similar spatial distribution of community-onset ESBL-E. coli and non-ESBL-E. coli bacteriuria and no spatial clusters of ESBL-E. coli. We will be stratifying analyses by race and ethnicity as prior studies have found differential ESBL-E. coli burden in specific racial and ethnic groups.

Ethical Considerations in the Collection, Analysis and Reporting of Spatial Data

No Abstract

Evaluating health outcome metrics and their connections to air pollution and socioeconomic vulnerability in the Coachella Valley

We use two different measures of health outcomes at the census tract level – one from randomly sampled telephone surveys and the other from emergency room visitations – to investigate the degree to which health outcomes may correlate with criteria air pollutants including fine particulate matter (PM2.5) and ozone, and socioeconomic metrics such as education level, linguistic isolation, poverty, health insurance coverage, and racial demographics in the Coachella Valley. We further investigate biases and differences between different health outcome metrics themselves.