Integrated geographical tools can enable interventions to control risk of malaria transmission in Dar es Salaam, Tanzania
Date
2017
Authors
Chuma, Victoria Mathew Mwakalinga
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
Rationale: Although public health and urban planning share the common goal of preventing urban outbreaks of infectious diseases, there is little operational overlap between these disciplines today. The separation of the fields has contributed to uncoordinated efforts to address the health of urban populations. Recently, infectious diseases like malaria are posing a great health hazard, particularly to developing countries. On one hand, almost all current research focuses on tackling this issue through public health programmes with little direct influence on the root causes of risk. On the other hand, urban planning is guiding infrastructure development/upgrading and delivery of adequate housing programmes which essentially can provide the basis of complementary interventions for malaria if appropriately and selectively targeted to high-risk areas. Integration of such public health considerations into this larger development agenda is often neglected, but could help target the areas of highest risk of malaria transmission with far more effective mitigation and intervention measures than are possible through the health sector alone. However, a greater challenge remains in how to identify areas for spatial targeting of interventions at finer geographic scales, because in cities malaria risk is highly heterogeneous, dynamic and unpredictable at finer spatial resolution.
Aim: Firstly, to identify and characterize risk of malaria at finer scales; by calibrating a comprehensive range of high resolution topographic predictors of risk using entomological and epidemiological field surveys. Secondly, to provide evidence-based guidelines for optimal design and spatial targeting of interventions that can be integrated into public health, infrastructure and settlement development programmes to effectively prevent urban malaria transmission.
Methods: The study was conducted in Dar es Salaam city (Tanzania), located along the shores of Indian Ocean; with an area of 148km2 and 3.6 million residents. Longitudinal and household cross-sectional surveys; were conducted to monitor entomological (adult mosquitoes using tent traps) and epidemiological (positivity obtained using Rapid Diagnostic Tests (RDTs)) indicators in 2,013 and 430 survey locations respectively, between March 2010 to February 2013. Indicators of city’s resilience were obtained from existing maps of land uses and infrastructure. Topographical predictors of local water accumulation potential were extracted from a 20m grid of a Digital Elevation Model (DEM). The spatial locations of study sites were collected using global positioning system (GPS) receivers at 5m accuracy. Spatial autocorrelations in the outcome variables were assessed using Global Moran’s I statistics (MI) and Variance Inflation Factor (VIF) was used to assess multi-collinearity among geographic predictors. Proximity analysis was used to compute the distance metrics from households to various infrastructures. Hotspots based on entomological and parasitological indicators were detected using Flexible spatial scan statistics. Boosted regression trees (BRT) accounting for spatial effects and generalized linear mixed models (GLMMs) were separately fitted to evaluate the influence of resilience indicators upon malaria transmission risk. Multivariable - Conditional Autoregressive Models (M-CARs) using Markov chain Monte Carlo (MCMC) simulation were used to predict hazard (M-CAR-Poisson) and risk (M-CAR-Binomial) of malaria using local topographic predictors of water accumulation potential and factors for population resilience.
Description
A thesis submitted to the School of Public Health, Faculty of Health Sciences, University of the Witwatersrand, Johannesburg, in fulfilment of the requirements for the degree of Doctor of Philosophy
Johannesburg, 2017
Keywords
Malaria Control, Urban Planning