Date of Award

January 2016

Document Type

Open Access Thesis

Degree Name

Master of Public Health (MPH)


School of Public Health

First Advisor

Mayur Desai

Second Advisor

James Hadler


INTRODUCTION: Tuberculosis (TB) transmission continues to be problematic in the United States. Contact investigations are essential for identifying new infections and must prioritize limited resources by pursuing more infectious cases. Current infectivity measures like sputum smear grade, chest x-rays, and nucleic acid amplification tests (NAAT) are sub-optimal. Time-to-detection (TTD) of Mycobacterium tuberculosis in liquid culture has shown promising results.

OBJECTIVES: Our objective is to examine the relationship between TTD and TB transmission.

METHODS: This study examined a retrospective cohort of pulmonary TB cases with TTD data from 2009-2014 from the King County Laboratory, Seattle, WA. The primary outcome was transmission indicated by previously undiagnosed latent tuberculosis infection (LTBI) in household contacts. The proportion of contacts positive for LTBI was modeled using Poisson regression. TTD, highest smear grade, NAAT, and chest x-ray were assessed for their association with LTBI in contacts.

RESULTS: Of 390 pulmonary TB cases, 92 had TTD and household transmission data. Overall, 76.1% of them had evidence of household transmission. Among all household contacts, 36.5% (187/512) had previously undiagnosed LTBI, including 33% of contacts related to cases with long TTD and 30% with negative smears. Short (0-7 days) TTD was significantly associated with increased LTBI risk compared to long TTD (15+ days) (RR: 1.39; MH X2 p-value: 0.025) and medium TTD (RR: 1.36; MH X2 p-value: 0.023). However Poisson modeling of transmission rates did not demonstrate a significant relationship. Multivariate analyses adjusted for highest smear, NAAT, and chest x-ray were also non-significant. All cases with short TTD were smear positive, but positive smears identified 60% more transmission events.

CONCLUSIONS: In this limited study, short TTD was not a promising addition to current infectivity measures. Observational studies of TB transmission are difficult when there are high background rates of LTBI. Any future studies should take place in previously TB-naïve households.


This is an Open Access Thesis.