The World Health Organization (WHO) treated at least 350 million persons with antiparasitic medicines globally, in an effort to control and eliminate onchocerciasis (oncho), lymphatic filariasis (LF), and other Neglected Tropical Diseases (NTDs). Unintended side effects, however, occurred for people in regions endemic with loiasis, with thousands of Serious Adverse Events (SAEs) that occurred solely as a result of the WHO's strategy, and 85% reported in Cameroon. To mitigate these undesired consequences, Schluter, et al. (2016) developed a probability model to gauge a village's average intensity of loiasis by calculating the population's average microfilariae count within a milliliter of blood (mf/ml), and generating a recommendation output whether or not to administer antiparasitics. This project aims to reduce SAEs by making this statistical tool more accessible through a graphical user interface (GUI) and aid global policymakers in their decisions about dispersing the antiparasitic tablet, ivermectin. Data were collected in September 2016 in Cameroon and captured demographic information, as well as test results for oncho and LF prevalence, loiasis mf/ml intensity, and the Schluter, et al., (2016) probabilities. 2,700 persons were tested within 27 villages. 93% of the villages were found to have loiasis with a mean of 4,009 mf/ml, as well as an 89% prevalence on oncho, and a 30% prevalence of LF. Approximately half of these villages, according the Schluter, et al. (2016) test, were found to have too high of a loiasis intensity to merit mass treatment of ivermectin. These findings were then programmed into a GUI that depended on the RStudio Shiny application, which provides source code to make statistical tools accessible as a user interface. This application generated a successful GUI that allows anyone with a Microsoft Excel file (CSV) to upload a spreadsheet, and instantly view the statistical model's prediction with a useful graphic ('Thumbs up' for treating the village with ivermectin sans pre-testing; 'Thumbs Down' for not treating). With this innovation, steps towards approximating loiasis in Cameroon, and other African counties, can be made more accessible to a wider array of global health stakeholders.
Table of Contents
CHAPTER 1. INTRODUCTION 4 CHAPTER 2. LITERATURE REVIEW 5 Literature Review Part I: NTD Elimination and Loiasis 5 Literature Review Part II: eHealth History and Implementation 13 CHAPTER 3. PROJECT CONTENT 21 Methods 21 Results and the GUI Prototype 25 CHAPTER 4. DISCUSSION 27 CHAPTER 5. ADDITIONAL PAGES 33 Public Health Implications 33 Tables 34 Figures 36 References 45
About this Master's Thesis
|Committee Chair / Thesis Advisor|
|Gooey Predictions: Utilizing a Graphical User Interface (GUI) to Increase Access to a Complex Predictive Statistics Model for Neglected Tropical Disease Management in Cameroon and Loiasis Endemic Regions ()||2018-08-28||