| Part1Study of distribution and spatial clustering of oncomelania in Hubei ProvinceObjective:To investigate the distribution, geographic regularity of spatial distribution and spatial attributes of the oncomelania in Hubei Province, so as to provide a scientific basis for the development of the oncomelania-controlling plan for schistosomiasis control. Methods:This study was based on the database of eliminating oncomelania that was kept by Hubei provincial schistosomiasis control administration and the GIS geospatial database. We retrieved the two databases from the years1980to2009. The two databases were associated to establish the Hubei geographic information database of oncomelania. Dynamic changes about the oncomelania eleminating, area and density indicators were observed using descriptive analysis method. Spatial clustering analysis was conducted to analyze the oncomelania density. The global and local Moran’s Index were calculated. Results: The oncomelania density at intervals of five years from the years1980to2009in Hubei Province showed a downward trend. The oncomelania density was higher from1980s to1990s. After the year of1990, the oncomelania density decreased and was at the lowest level in2009. Oncomelania area changed (increased or decreased) significantly in Hubei Province in1980s. And in1990s, the oncomelania area remained unchanging or slightly reduced. In2009, the oncomelania density was higher in Jiangling, Shashi, Gongan and Tianmen. Form the global Moran’s I index, the results showed that the oncomelania density in Hubei had significant spatial autocorrelation. Hotspot analysis indicated that Jiangling, Yangxin and Gongan were the significantly higher oncomelania density areas in2009. Conclusion:The oncomelania density showed a downward trend in the past30years, which indicated that the control for oncomelania in Hubei was effective. Oncomelania in Hubei was distributed mainly along the middle reaches of the Yangtze River, the Han River, and the lake floodplains between the two rivers. The oncomelania was zonary-distributed along the two rivers mentioned above. The oncomelania density in Hubei showed a strong spatial clustering over time, but a trend of more gathering was not shown. Part2Analysis of the oncomelania density using mixed-effects model and Bayesian space-time modelObjective:We explored the factors that may affect the oncomelania density distribution in Hubei Province, and studied the changes of oncomelania density over time and space. Our findings can provide some theoretical basis for early prediction and control of oncomelania in Hubei Province. Methods:The study was based on the Hubei geographic information database. Environmental factors including climatic factors, and remote sensing factor were collected, associated with the oncomelania spatial database. Data from2009and1990-2009, respectively, were used to construct the no-time-effect mixed model and time mixed-effects model. AIC, AICC and BIC rules were employed to compare different error distributions and the connection function combination of generalized linear mixed models. The main factors that affecting oncomelania distribution were analyzed by a multi-level analysis, and the parameter were estimated. Bayesian space-time model was adopted to analyze the oncomelania density. Results:Results from the univariate analysis showed that air humidity (r=0.320, p<0.05), NDVI (r=0.384, p<0.05), LST (r=0.318, p<0.05), and the distance between the water and the village (r=-0.383, p<0.05correlation) were significantly associated with oncomelania density. The best generalized linear mixed model (GLMM) was Poisson-distributed error, log link function and variance function mean. GLMM results showed that oncomelania density was affected by LST (p=0.020), the distance between the sampled counties and the Yangtze (p=0.020), and whether oncomelania control or not (p=0.001), NDVI (p=0.003), air humidity (p=0.001) and time (p=0.001). Results from the Bayesian model showed a gradually downward trend. Spatial random variability was statistically significant, and each observation point of the spatial variability was different. Bayesian space-time model was the best according to the principles of the smaller the DIC, the better the model. Parameter estimates both using the Bayesian space-time model and generalized linear models were close. Conclusion:The best generalized linear mixed model (GLMM) was Poisson-distributed error, log link function and variance function mean using the data of oncomelania density in Hubei. Bayesian space-time model indicated that the oncomelania density had a gradually downward trend, and the spatial random variability had statistically significant. When building models, it is necessary to consider the effects of time, space and their interaction. Humidity, NDVI, LST, from the distance between the Yangtze River and village and whether oncomelania control or not were significantly associated with the distribution of the oncomelania. Findings from this study can provide the basis for the density monitoring and prevention and control of the oncomelania. Part3A Bayesian Approach to Estimate the Prevalence of Schistosomiasis Japonicum Infection in the lake region, Hubei Province, ChinaObjective:A Bayesian technology was introduced to estimate community prevalence of schistosomiasis japonicum infection based on the data of a large-scale survey of schistosomiasis japonicum in lake-regions in Hubei Province. Methods:A multistage cluster random sampling approach was applied to the endemic villages in lake-regions of Hubei province in2011. IHA test and Kato-Katz test were applied for the detection of the S. japonicum infection for sampled population. Expert knowledge on sensitivities and specificities of IHA test and Kato-Katz test were collected based on a two-round interview. Prevalence of S. japonicum infection was estimated by Bayesian Hierarchical model in two different situations. Results:In situation1, Bayesian estimation used both IHA test data and Kato-Katz test data to estimate the prevalence of S. japonicum. In situation2, only IHA test data was used for Bayesian estimation. Finally14cities and46villages including50980residents were sampled from lake-regions of Hubei province. Sensitivity and specificity for IHA test ranged from80%to90%and70%to80%, respectively. And for the Kato-Katz test, sensitivity and specificity were from20%to70%and90%to100%, respectively. Similar estimated prevalence was obtained in the two situations. Estimated prevalence among sampled villages was almost below13%in both situations and varied from0.95%to12.26%when using data of IHA test only. Conclusion:The study indicated that it is feasible to apply IHA test only combining with Bayesian method to estimate the prevalence of S.japonicum infection in large-scale surveys. |
PDF Full Text Request