Integrating Molecular Typing into Routine Tuberculosis Surveillance: An Assessment of the Strengths and Limitations of Current Approaches.

Abstract

Molecular typing is increasingly integral to tuberculosis (TB) control programs, providing public health practitioners with a tool to characterize transmission patterns, track the emergence and spread of strains of particular medical and public health importance, and identify transmission venues that contribute to the persistence of M. tuberculosis in populations. While molecular typing is already used extensively as a tool for TB control in many diverse populations across the globe, the sensitivity of molecular typing-based measures to characteristics of both the host and microbial populations is not well understood. To better characterize the relationship between key host and microbial factors and the validity of molecular typing measures, this dissertation work employs a multi-disciplinary research strategy which integrates molecular, epidemiologic, and computer-simulation data. In the rural, stable population of Arkansas, we found that a declining incidence of TB between 1996 and 2003 resulted primarily from a declining incidence of TB due to the reactivation of remotely acquired infection, rather than recently acquired infection. This work suggested the influence of a strong cohort effect on disease patterns in this population. A validation study of molecular typing in this same population, in which extensive epidemiologic interview data were compared to molecular typing results, identified a number of host and microbial factors associated with the validity of typing results. This study also suggested the presence of a regionally endemic strain family which was associated with false positive molecular typing results. Using an agent-based model of TB transmission, we conducted the first quantitative assessment of the importance of the diversity and stability of molecular typing markers, as well as historic and demographic characteristics of the host population, to the validity of typing results. The results of these investigations contribute to an improved understanding of the dynamics of TB transmission in rural populations of the United States, and also highlight key factors that should be considered in the interpretation of molecular typing results in all populations. Additionally, these results may inform the development of more rational approaches to the design of molecular typing systems used in TB control.