Buruli Ulcer Disease and Its Association with Land Cover in Southwestern Ghana.

Background: Buruli ulcer (BU), one of 17 neglected tropical diseases, is a debilitating skin and soft tissue infection caused by Mycobacterium ulcerans. In tropical Africa, changes in land use and proximity to water have been associated with the disease. This study presents the first analysis of BU at the village level in southwestern Ghana, where prevalence rates are among the highest globally, and explores fine and medium-scale associations with land cover by comparing patterns both within BU clusters and surrounding landscapes. Methodology/Principal Findings: We obtained 339 hospital-confirmed BU cases in southwestern Ghana between 2007 and 2010. The clusters of BU were identified using spatial scan statistics and the percentages of six land cover classes were calculated based on Landsat and Rapid Eye imagery for each of 154 villages/towns. The association between BU prevalence and each land cover class was calculated using negative binomial regression models. We found that older people had a significantly higher risk for BU after considering population age structure. BU cases were positively associated with the higher percentage of water and grassland surrounding each village, but negatively associated with the percent of urban. The results also showed that BU was clustered in areas with high percentage of mining activity, suggesting that water and mining play an important and potentially interactive role in BU occurrence. Conclusions/Significance: Our study highlights the importance of multiple land use changes along the Offin River, particularly mining and agriculture, which might be associated with BU disease in southwestern Ghana. Our study is the first to use both medium- and high-resolution imagery to assess these changes. We also show that older populations (≥ 60 y) appear to be at higher risk of BU disease than children, once BU data were weighted by population age structures. Author Summary: We studied the relationship between a neglected tropical disease, Buruli Ulcer (BU), and landscape disturbance. We hypothesized that the increased presence of BU was related to landscape disturbance, especially alluvial mining, in endemic areas of the disease. We characterized recent cases of BU in southwestern Ghana and characterized six types of land cover classes at different spatial extents at the village level. We illustrated that mining and water areas were prevalent in BU clustered areas and may potentially play an interactive role in BU occurrence, a finding that should be explored further. No previous study has quantified the relationship between mining and BU at the village level. In addition, we found that people ≥ 60 years old had a higher prevalence than other age groups, when data was weighted by the population age structure. To our knowledge, these findings have not been reported in Ghana before and might provide new insight in BU disease intervention and control. While it is unclear how people come in contact with the bacterial source, our study shows that the importance of where people live (specifically, what land activities are occurring in the area) seems to play a large role in determining disease risk. [ABSTRACT FROM AUTHOR]