Tuberculosis in Fulton County, Georgia, 2008-2014: Risk Markers for Isoniazid Monoresistance and a Pilot Study of Novel Spatial Methods Open Access

Worrell, Mary Claire (2015)

Permanent URL:


Background: Tuberculosis (TB) continues to cause morbidity and mortality in the United States (US), particularly in poor and transient populations. Disease transmission, particularly in homeless populations, can occur in multiple locations, and practical approaches are needed to understand the geographic characteristics of transmission.

Objective: (1) Identify the effect of homelessness on isoniazid (INH) monoresistant TB in Fulton County, Georgia. (2) Pilot novel spatial analysis techniques for disease surveillance that includes transient populations.

Methods: In Fulton County, Georgia, two separate TB outbreaks were recorded since 2008. A random sample of active TB cases diagnosed between 2008 and 2014 was selected from Fulton County Department of Health records. Disease and demographic indicators and all addresses and locations of interest were abstracted from charts. Logistic regression was used to understand the relationship between INH resistance and homeless status and various covariates. Kernel density methods were used to characterize the distribution of reported cases when using a single residential address compared to using multiple addresses, adding additional residences, work, school, and hangout spots. Activity space analysis evaluated the intersection of cases with particular traits.

Results: Homelessness was highly related to INH resistance; the odds of INH resistance in homeless cases was 3.3 times higher than the odds of INH drug resistance in the non-homeless population when controlling for year of diagnosis, excessive alcohol use, HIV status, history of incarceration, age, and sex (CI: 1.22-9.14). Greater dispersion of cases was found when utilizing all addresses for each case for kernel density interpolation compared to using a single address. Activity spaces of homeless and INH resistant cases overlapped with other homeless and INH resistant cases respectively more than non-homeless and INH susceptible cases (p<0.0001 and p<0.0001 respectively).

Conclusion: INH resistant TB remains a major problem in Fulton County; 50% of this study's TB cases diagnosed in 2014 were INH resistant. Alternative spatial methods offer insight into the spatial context of TB and provide information for cases without permanent addresses. Activity space analysis, prominent in exposure science and chronic disease, can provide insight into the investigation of infectious disease and should be utilized alongside standard epidemiologic methodologies.

Table of Contents

Chapter 1: Extended Background and Literature Review. 1

Tuberculosis. 1

Microbiology and Transmission. 1

Active v. Latent TB Infection. 2

Symptoms. 3

Testing. 3

Treatment. 5

INH Monoresistant, Multi-drug Resistant, and Extensively-drug Resistant TB. 6

Epidemiology. 8

Tuberculosis in the United States. 8

Tuberculosis in Georgia. 8

Geospatial Analysis Methods. 11

Chapter 2: Manuscript. 14

Background. 14

Methods. 18

Results. 26

Discussion. 31

Conclusion. 35

Tables, Figures, and Maps. 37

Tables. 37

Figures. 45

Maps. 46

Chapter 3: Public Health Implications. 51

References. 53

Appendix. 60

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.
Subfield / Discipline
  • English
Research Field
Committee Chair / Thesis Advisor
Committee Members
Partnering Agencies
Last modified

Primary PDF

Supplemental Files