No abstract

Spatial modelling was conducted to examine community factors associated with cholera incidence rates in Morogoro Municipality. The study employed both secondary (cholera cases) and primary (geographic coordinates of community risk factors) data. Spatial lag model was applied in examining association between the variables. All wards of Morogoro Municipality were considered in the study to capture their variations because cholera cases have a tendency to be clustered. Results indicated that market density, distance to the market and distance to the dumpster are significant factors associated with cholera incidence rates in the wards (p < 0.05). Geographically weighted Poisson model was used to show the variations of those factors between the wards in Municipality. A statistically significant positive association of cholera incidence rates; and market density was only found in Mazimbu ward (p < 0.05) and distance from the community to the dumpster was found in Kihonda, Kingolwira, Bigwa, Kichangani, Kilakala and Boma wards (p < 0.001) and some wards at the centre of the municipality which are Mji Mkuu and Kingo (p < 0.05). A statistically significant negative association of cholera incidence rates and distance from the centre of the community to the market was found in Kihonda, Kingolwira and Kichangani (p < 0.001) and Bigwa wards (p < 0.05). Therefore, measures taken to control and prevent cholera disease should base on the variations of the risk factors found in the Municipal wards. Keywords:    Cholera incidence rate; Spatial lag regression model; Community risk factor; Geographically weighted Poisson model