Socio-demographic and Clinical Risk Factors Associated with Tuberculosis Mortality in the U.S., 2009-2013 Open Access

Hannah, Haylea Ariel (2015)

Permanent URL:


In 2009, the United States National TB Surveillance System (NTSS) was updated to include death attributable to TB. The purpose of our study is to examine factors related to TB mortality and whether they vary between TB-specific mortality and all-cause mortality. The three outcomes of interest were death from any cause (all-cause mortality), TB-specific death, and no death, during January 2009-December 2013. "No death" was used as a control throughout. We used multinomial logistic regression to adjust for risk factors associated with mortality. Stepwise, backwards elimination, and forward selection model techniques were used to determine the best predictive model of the outcome. Statistical significance was assessed at an alpha=0.001 level. Of the 52,714 TB cases, 2009-2013, 4,331 TB cases resulted in death: 1,404 (32%) TB-specific, 2,413 (56%) all-cause, and 514 (12%) missing cause of death. Following a sensitivity analysis concluding that cases with a missing cause of death did not significantly impact the effect estimates, those 514 cases were excluded. There were 48,358 (92%) TB cases reported as alive. All risk factors were independently associated with TB outcomes in bivariate analyses except for previous episode of TB, homelessness within the past year, and MDR-TB diagnosis. The survival time between TB mortality and all-cause mortality was significantly different (p<0.0001). MDR-TB diagnosis, reporting TB symptoms as a reason evaluated, and Hispanic race were all significant risk factors for TB mortality, but not all-cause mortality. Advanced age (45 years and older), male gender, birth in the U.S., being unemployed or retired at the time of diagnosis, the presence of both extra-pulmonary and pulmonary disease, end-stage renal disease, any immunosuppressive risk factors, and reporting an abnormal X-ray as the reason evaluated were all significant risk factors of both TB and all-cause mortality among cases. Being treated with both self-administered therapy (SAT) and directly observed therapy (DOT) was protective against both TB and all-cause mortality. These findings have implications for TB case management and suggest that some high risk groups, such as older TB cases, may need to be followed more closely.

Table of Contents

Chapter 1: Literature Review on Tuberculosis Epidemiology, Control, and Mortality Risk Factors 1

Table of Contents 2

Chapter 2. Socio-demographic and clinical risk factors associated with Tuberculosis mortality in the United States, 2009-2013 26

Introduction 27

Methods 30

Results 32

Table 1. 39

Table 2. 40

Table 3. 41

Figure 1. 42

Figure 2. 43

Discussion 43

Chapter 3: Future Directions and Implications 44

Appendix A: Additional Analyses 49

a. Secular Trends in Reporting 50

b. Kaplan-Meir Survival Curves 51

References 46

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

Primary PDF

Supplemental Files