365 research outputs found
Epidemiological and ecological evaluation of the impact of housing quality on malaria incidence in Lilongwe, Malawi
No full textAcross sub-Saharan Africa, malaria remains a significant cause of death across socioeconomic and demographic contexts, despite significant international investment in malaria prevention methods. Current interventions to reduce malaria incidence, including vaccination, are insufficient for malaria eradication, prompting further research into strategies to minimize malaria exposure. Recent vaccination efforts have shown some promising results, though vaccine efficacy is heterogeneous, potentially due to factors in the natural and built environment. The built environment and housing quality can have significant implications on mosquito entry to homes, particularly at night when mosquitoes are most active. Utilizing a series of descriptive analyses and Poisson regression models, this study explores how housing characteristics effect malaria incidence, and how housing modifies the efficacy of the malaria vaccine. Results from these analyses indicate that no windows, plastic/paper/carton windows, open windows, grass roofs, and visible holes are associated with increased malaria incidence compared to the current best practice, whereas cement walls are associated with decreased malaria incidence. We also found that no housing characteristic had a significant effect on vaccine efficacy, except for cement outer walls in the 4-dose treatment group. We can thus conclude that housing improvements offer a promising approach to malaria control and prevention, supporting existing findings in the literature.Bachelor of Art
Spatial distribution and disease ecology of gastric cancer in western Honduras
No full textGastric cancer, etiologically linked to infection with Helicobacter pylori, is the leading infectious-related cancer and the second most common cause of cancer mortality worldwide. Previous research has shown that gastric cancer rates are higher at high altitudes; however causal factors remain poorly understood. This research examines the relationship between altitude and gastric cancer risk, and explores potential explanatory covariates related to human behavior that may help explain the spatial patterns of gastric cancer incidence. Using a case control study of gastric cancer cases in western Honduras from 2002-2012, clusters of high-incidence areas are identified. Binomial multilevel likelihood models are constructed to better understand how altitude affects gastric cancer risk and to explore how individual-level behaviors drive disease incidence patterns. While simple models often assume all individuals are identical, multilevel models incorporate individual and group-level heterogeneity in characteristics that may be related to disease dynamics. Results indicate that age-standardized rates (n=594) are twice as high for males than females (15.07 for males and 6.59 for females), and that high rates are significantly clustered at the municipio (local administrative unit) level. Altitude was an insignificant predictor of gastric cancer when measured both as a continuous (p=0.197) and categorical variable (high/low; p=0.192). The results of the multilevel modeling of individual-level behaviors reveal that use of refrigeration as an adult is associated with a decrease in gastric cancer risk (β = -0.9883, p=6.51e-08). The finding that altitude does not affect gastric cancer risk within the study area suggests the possibility that the study area does not contain enough altitudinal heterogeneity to accurately characterize the relationship between altitude and gastric cancer rates. The finding that use of refrigeration as an adult is protective against gastric cancer suggests that access to refrigeration may decrease dependency on salted and preserved meats and increase access to fruits and vegetables, two established factors related to gastric cancer risk. During the past two decades it has been well-established that infection with H. pylori is linked to increased gastric cancer risk. However, the finding that individual-level behavior impacts disease risk supports the theory that to understand disease dynamics, host-pathogen interactions must be considered within the context of their disease ecology.Master of Art
Cholera Transmission in Bangladesh: Social Networks and Neighborhoods
No full textTransmission of infectious pathogens across networks is well-documented, yet remains primarily focused on diseases spread by sexual contact. Such analytical tools, however, may also facilitate understanding of how other types of health outcomes are related to physical and social contacts. This research examines the relationship between cholera incidence and the social network that links households in rural Bangladesh. Using twenty-one years of longitudinal demographic and health data, clustering of similar disease rates in the network was measured and compared to spatial autocorrelation of cholera at the neighborhood level. Results indicate that rates are significantly concentrated amongst households within the same local environment, and that social clustering is only evident during certain years examined. These outcomes suggest that intervention efforts should place priority on identifying local-level environmental factors, but also consider the potential of networks as they assist transmission, as well as their role in interactions within a defined neighborhood
Diarrheal Diseases in Rural Bangladesh: Spatial-Temporal Patterns, Risk Factors and Pathogen Detection
No full textDiarrheal diseases are still a leading cause of child mortality in less developed countries. In the past three decades, in an effort to reduce the transmission of diarrheal diseases, millions of tubewells have been installed as a way to provide safe drinking water in Bangladesh. However, this effort may have been counterproductive since widespread arsenic contamination has been found in groundwater. Thus, there is a reason to rethink the use of tubewells and to assess risk factors related to diarrheal disease in Bangladesh. This study primarily focused on 142 villages of Matlab, a rural area in Bangladesh, using datasets collected through a local health surveillance system to explore the spatiotemporal patterns of diarrheal disease and its relevant risk factors. First, a geographic information system (GIS) and spatial statistics were used to illustrate the occurrence and spatial-temporal clusters of diarrhea (including community childhood diarrhea data and hospital data on diarrhea caused by rotavirus and Shigella). Second, the study determined the relationship between diarrheal disease among children under five and identified several important risk factors, such as tubewell access, depth and arsenic levels. Additionally, simple and rapid concentration methods were developed and evaluated to detect adenovirus, a common etiologic pathogen of diarrhea in water. The study attempted to answer the following questions: What are the trends and spatial patterns of diarrheal diseases? Are tubewells protective against diarrheal diseases? Does arsenic mitigation by well switching raise the risk of diarrheal disease among children? The results obtained from this study provide some useful information to help policy-makers implement relevant scientific measures for diarrhea reduction and arsenic mitigation. The concentration methods developed in this study are applicable to monitor pathogens in water in Bangladesh and worldwide
Space-time differentiation of drivers of and barriers to H5N1 avian influenza evolution in Vietnam
No full textThe emergence and re-emergence of human pathogens resistant to traditional medical treatment will present a challenge to the international public health community in the coming decades. Geography is uniquely positioned to examine the progressive evolution of pathogens across space and through time, and to link molecular change to interactions between population and environmental drivers. The widespread outbreak of H5N1 avian influenza across Asia in 2003, and its continued circulation within both poultry and human populations, presents an opportunity for the integration of traditional disease ecology with the emergent field of landscape genetics. Combining spatial statistical methods with genetic analytic techniques, geographic space is used to explore genetic evolution of H5N1 highly pathogenic avian influenza viruses (HPAIV) at the sub-national scale in Vietnam. This dissertation investigates the following topics: differences in genetic characteristics by species of isolation, location and timing of barriers to gene flow, and population-environment characteristics associated with increased viral evolution in Vietnam from 2003 to 2007. A variety of methods are used, including cluster analysis, multidimensional scaling, analysis of variance, and linear regression. Results indicate that genetic differentiation of these viruses varies significantly according to both their host species and the isolation time, but has a complex relationship with the geographic location of virus isolation. The effect of geographic space, and underlying landscape differentiation, does not appear to create boundaries to gene exchange across Vietnam. Taking these indicators of the influence of species, temporal characteristics and geographic space into account, the drivers of molecular evolution of H5N1 HPAIV in Vietnam are as predicted by a disease ecology framework, a combination of both population and environmental characteristics. These findings indicate that there are significant spatial and temporal effects on the evolution of H5N1 HPAIVs, and that local-level conditions can affect viral genetic evolution. Given that areas of rapid genetic evolution are more likely to produce a highly pathogenic virus capable of sustained human-to-human transmission, further exploration of spatial variation in molecular change is needed
Energy Access and Reforestation Efforts for Ultra-Poor Households in Southern Malawi
No full textInvestigations into energy access in Sub-Saharan Africa often focus on modern energy transitions and electrification projects. However, these studies fail to consider the household level differences in access to energy sources and lack of opportunity to transition to alternative modern fuels. This study uses household-level data to explore household level reforestation efforts as a strategy to improve access to energy sources and improve environmental resilience on a community level. Specifically: Are reforestation efforts in Southern Malawi clustered in space, and do the surrounding land use land cover change classifications or household characteristics influence these efforts? The study, are conducted in southern Malawi with ultra- poor households that receive social cash transfer payments. Therefore, the focus of this study is on the most vulnerable, lowest income households in the community. It is expected that households with limited surrounding forest cover, and those who have received information on agroforestry or sustainable practices would be most likely to participate in reforestation efforts in the form of tree planting. There is observable spatial clustering of village clusters that have been provided information on agroforestry or sustainable practices and household-level tree planting efforts in village clusters, but the two are not found to be spatially correlated. We find that the total land owned by individual households is strongly correlated with tree planting efforts, especially in areas where wood is not primarily collected from plantations. Contrary to the expected result, reforestation efforts are not found to be linked to a current lack of access to energy sources, but are linked to land ownership. This study concludes that participation in un- aided reforestation efforts in southern Malawi may not be a mechanism for households to reduce vulnerability, but are a result of household characteristics like land ownership that enable the ability to plant trees. This paper suggests that promotion efforts should consider other factors that are associated with the decision to reforest to be most effective in promoting sustainable practice.Bachelor of Art
Relationships between flood control and cholera in Matlab, Bangladesh
No full textImplementation of flood control strategies has been empirically associated with rises in disease rates in the developing world. This research examines the impact of flood protection measures on cholera incidence among a rural Bangladeshi population. Using longitudinal health and demographic data collected over 21 years, analysis of clustering patterns and statistical relationships between cholera incidence and environmental factors was conducted for timeframes prior to and following the introduction of flood control in Matlab, Bangladesh. Results indicate that alteration of normal flooding patterns both temporally and spatially shifted cholera occurrence within the study area, and that these shifts demonstrate further differentiation when information on cholera strain is included in the analysis. These outcomes suggest that introducing flood protection to rural Bangladesh will have significant but complex effects on cholera incidence patterns
The ecology of birth defects: socio-economic and environmental determinants of gastroschisis in North Carolina
No full textGastroschisis is a serious birth defect that has increased in prevalence in North Carolina over the past decade. The causes of the defect, and the reasons for this increase, are largely unknown. This study uses the disease ecology framework and spatial methodologies - spatial statistics, Geographic Information Systems, and hydrological modeling - to explore the geographic distribution of gastroschisis in North Carolina and suggest possible socioeconomic and environmental factors that may contribute to the disease. Specific questions addressed in this study include: 1) Do significant geographic clusters of gastroschisis exist in North Carolina? 2) Do clusters suggest the presence of point-source environmental pollutants? 3) What area-level socioeconomic characteristics are related to gastroschisis outcomes? 4) What can this tell us about possible causes of the disease? Using data from a population-based birth defects registry, this study uses Kulldorff's spatial scan statistic to identify the location and extent of clusters of gastroschisis births in North Carolina between 1999 and 2004. Spatial clusters are controlled for four major risk factors (maternal age, race, prior births and Medicaid status) to ensure that the clusters are not an artifact of the population composition of the State. The relationship between neighborhood socioeconomic characteristics (e.g., race, poverty, education and unemployment) and gastroschisis outcomes are examined using logistic regression models, which combine individual-level and neighborhood-level variables. Finally, simple hydrological models are used to determine if exposure to upstream textile mill effluent increases the risk for a gastroschisis affected pregnancy. Results indicate the presence of a localized cluster of gastroschisis in the rural southern Piedmont of North Carolina. In addition, both individual-level (Medicaid status) and neighborhood-level (poverty and unemployment) socioeconomic factors appear to contribute to the risk of a gastroschisis affected pregnancy, suggesting that neighborhood-level socioeconomic factors exert an independent causal effect on gastroschisis. Despite the localized nature of the cluster, which often suggests the presence of an environmental contaminant, there is no evidence to support this hypothesis. These results may help understanding the myriad social, economic and environmental factors that combine and interact to influence gastroschisis outcomes
The Geography of Groundwater Quality and Childhood Diarrheal Disease in Bangladesh
No full textChildhood diarrhea persists in Bangladesh despite efforts to shift from surface water to groundwater for drinking. It is unknown whether shallow aquifer groundwater extracted through tubewells is a significant source of disease or if other sources such as surface water and local sanitation are driving transmission. Using the disease ecology framework, this study explores the influence of poor sanitation on diarrheal disease transmission. Specific questions addressed in this study include: 1) Does poor sanitation influence shallow tubewell water quality? 2) Does fecal contamination of tubewells influence diarrheal disease? 3) Does the neighborhood water and sanitation infrastructure affect childhood diarrheal disease incidence above and beyond household factors? 4) Does poor sanitation influence diarrheal disease via bathing ponds? 5) Does obtaining drinking water from deep tubewells have a protective effect against childhood diarrhea incidence? This study integrates groundwater microbial data, health and demographic surveillance data, and detailed spatial data of the water and sanitation infrastructure in six villages in Matlab, Bangladesh. The relationship between groundwater quality and poor sanitation is measured at multiple scales using geographic analysis tools. Direct and indirect sanitation influences on childhood diarrheal disease (2002-2006) are explored using neighborhood latrine metrics, and bathing pond latrine metrics. A deep tubewell arsenic mitigation intervention is also examined to determine whether children drinking from deep tubewells experience less diarrhea than children drinking from shallow wells. Results suggest that poor sanitation is predictive of both groundwater contamination and diarrheal disease. Children living in neighborhoods with insufficient access to septic latrines experience higher diarrhea incidence. Additionally, children living near bathing ponds surrounded by latrines leaking effluent also have a higher incidence. While deep tubewells were installed for arsenic mitigation, they are also protective against diarrheal disease. These results shed light on the importance of integrating population and environment data to identify particular circumstances in which groundwater is compromised and children are at risk of contracting diarrheal diseases. These results suggest that poor sanitation diminishes the effect of improved drinking water sources and improvements to the built sanitation infrastructure are needed to reduce diarrheal disease incidence.Doctor of Philosoph
Disease Ecology in the Democratic Republic of the Congo: Integration of Spatial Analysis and Population Surveillance
No full textIn countries like the Democratic Republic of the Congo (DRC) that have limited public health infrastructures, only educated guesses have been made about the spatial distribution of important diseases. This research estimates the spatial distribution of HIV, malaria and anemia prevalence in the DRC, and determines the population, environmental and behavioral drivers underlying these distributions. Using molecular diagnostics from dried blood spots from a 2007 Demographic and Health Survey (DHS) and demographic data available from this survey, the primary research aims are addressed via spatial analysis and multilevel modeling. The creation of an extensive Geographic Information Systems (GIS) database and selection of individual questionnaire responses is informed by disease ecology theory. In addition to discerning patterns and drivers of disease prevalence in the DRC, this research demonstrates how well population-representative surveillance data can be used to improve understanding of disease transmission in other developing countries. While older people were at greater risk for HIV and anemia, younger people were at greater risk for malaria. Individual wealth increased HIV risk, while it protected against malaria. Increased risk for anemia was found in certain cultural groups. Living near urban areas increased risk for HIV and decreased risk for malaria. Certain types of agriculture were protective against anemia. Greater density of nearby conflict since 1994 decreased malaria risk and proximity to a refugee camp was protective against anemia in women. Certain population characteristics and behaviors were equally or more important at the community level as at the individual level. Greater individual wealth was protective against malaria along with the average wealth of the community in which one lived. This research extends beyond the scope of what would have been possible with the DHS dataset alone. The molecular results for malaria parasitaemia as well as habitat data from a variety of sources contributed to the creation of a complex database which enabled all aspects of disease ecology to be explored
- …
