Transmission patterns of extensively drug-resistant tuberculosis in South Africa: a network approach 公开
Nelson, Kristin (Fall 2018)
Published
Abstract
Tuberculosis (TB) is the leading infectious cause of disease worldwide and there were over half a million cases of drug-resistant TB in 2017. Transmission plays a critical role in the spread of extensively drug-resistant (XDR) TB in South Africa and globally. However, an incomplete understanding of the risk factors for XDR TB transmission prevents design of effective interventions to curtail transmission. Further, there has been little focus on the critical role of missing cases in transmission networks, though understanding missingness is essential to ensure accurate estimation of underlying transmission patterns.
We combined bacterial whole genome sequencing to identify transmission events with network analysis to investigate the clinical and behavioral factors driving XDR TB transmission. In Aim 1, we used exponential random graph models (ERGMs) to measure associations between clinical markers of infectiousness and transmission of XDR TB. Cases reporting 2 to 3 months of cough were more highly connected in the network than those reporting no cough and smear-positive cases were more poorly connected than smear-negative cases. In Aim 2, we examined associations between social mixing patterns and transmission. Cases who spent time in urban settings were more highly connected in the network than those who did not, and cases with extended hospital stays were less connected that those who reported shorter hospital stays.
In Aim 3, we assessed the impact of missing XDR TB cases in the transmission network. We found that no single scenario we tested could account for the missingness in the empirical transmission network. However, missingness was unlikely to be random based on our models; the most likely scenarios involved oversampling of low-transmitting cases or omission of a factor strongly related to transmission from our models. Our results were strongly influenced by several key assumptions. This highlights the uncertainties in our transmission model, and about TB transmission broadly, that preclude more exact inference regarding underlying XDR TB transmission patterns.
Through gaining a clearer understanding of XDR TB transmission patterns in settings of high TB incidence, we can directly inform interventions that will halt the spread of drug-resistant TB in countries with the highest burdens of disease.
Table of Contents
Chapter 1. Background......................................................................................................................... 1
Chapter 2. Rationale and Specific Aims............................................................................................... 33
Chapter 3. Data source......................................................................................................................... 38
Chapter 4. Preliminary Work................................................................................................................ 37
Chapter 5. Aims 1 and 2........................................................................................................................ 69
Chapter 6. Aim 3................................................................................................................................... 107
Chapter 7. Public health implications and overall significance............................................................ 160
Chapter 8. References........................................................................................................................... 167
About this Dissertation
| School | |
|---|---|
| Department | |
| Degree | |
| Submission | |
| Language |
|
| Research Field | |
| 关键词 | |
| Committee Chair / Thesis Advisor | |
| Committee Members |
Primary PDF
| Thumbnail | Title | Date Uploaded | Actions |
|---|---|---|---|
|
|
Transmission patterns of extensively drug-resistant tuberculosis in South Africa: a network approach () | 2018-11-06 14:29:32 -0500 |
|
Supplemental Files
| Thumbnail | Title | Date Uploaded | Actions |
|---|