Prediction modeling of Dengue risk based on socio-cultural and environmental factors using GIS : case of Jalor town, Rajasthan, India | |
| Author | Bohra, Alpana |
| Call Number | AIT Diss. no. SR-02-1 |
| Subject(s) | Dengue--India--Rajasthan--Forecasting India--Rajasthan--Social life and customs--Health aspects Environmental health--India--Rajasthan Health risk assessment--India--Rajasthan |
| Note | A dissertation submitted in partial fulfillment of the requirements for the Degree of Doctor of Technical Science, School of Engineering Technology |
| Publisher | Asian Institute of Technology |
| Abstract | Dengue Fever (DF) associated with dengue haemorrhagic fever/dengue shock syndrome (DHF/DSS) has emerged as an important public health problem in India due to its high fatality rate. In India dengue fever has been known since the l 91h century and epidemics have been reported from almost all parts of the country. Until now nearly all research efforts have been focused either on biological, entomological or clinical aspects of DF/DHF/DSS. However, location-specific studies, demonstrating an integrated approach of social, physical and climate information are lacking. This calls for an in-depth study of the interrelationships involved in socio-cultural practices, physical and climatic factors in dengue incidences. The present study identifies the most significant indicators of social, physical and climate factors describing the disease incidences, under the influence of the local conditions of the dengue endemic area of Jalor using statistical, spatial and Geographic Information Systems (GIS) modeling. Statistical techniques and spatial modeling have been carried out to develop a dengue risk map based on socio-cultural factors. The socio-economic and socio-cultural parameters from 77 households belonging to both dengue affected (DA) and unaffected (UA) samples have been collected through structured questionnaire of sixty variables. Out of these, nineteen variables have been found to be significantly correlated with incidence both at 95% and 99% of confidence levels. A step-wise regression analysis has generated eight variables significantly contributing to the dengue incidence with the R2 value of .958 at the 95% significant level. These variables are: frequency of cleaning of water storage containers, pattern of houses, frequency of use of cooler, frequency of cooler cleaning, water storage practices, mosquito protection measures, frequency of water supply, and frequency of waste disposal. GIS has been used to link the spatial and significant socio-cultural indicators with the disease data. Using factorial discriminant analysis and spatial modeling with these eight sociocultural indicators, five classes of risk categories ranging from "very low" to "very high" have been identified. Validation of these risk categories on individual houses shows that 94.5 % of the houses are correctly classified. The nearest neighborhood method has been used to prepare a spatial extrapolated social risk area map. Amongst the physical parameters, dengue risk associated with settlement patterns, land use and land cover, climate. Considering these physical factors and the flight range of Aedes aegypti mosquitoes three buffer zones with 200m, 400m and 600m have been created around the each dengue affected and unaffected sample houses. Percentage of land use and land cover within these buffers for each house has been calculated. Levene's T-test for equality of variances has been applied to identify the significant land use and land covers in each buffer. Results indicate that water bodies and culturable waste land have significant effect on dengue incidence. Water bodies have a significant effect at (p ~ 0.01) level on all three buffer zones. Whereas, culturable wasteland have significant effect at (p ~ 0.05) and (p ~ 0.01) for 400m and 600m buffer sizes respectively. Spatial modeling and factorial discriminant analysis are used to determine four classes of risk for buffer sizes of 200m and 600m respectively, and five classes of risk for buffer size of 400m. Database has been prepared in GIS format and the nearest neighborhood spatial modeling method has been used to create spatial extrapolated physical risks area map of Jalor in all buffer sizes, separately. Validation on the buffer sizes 'demonstrates that 600m buffer size provides the best (75.3 % ) classification. The statistical analysis of fifteen years of climate data ( 1985-1999) including temperature (mean monthly maximum and minimum) relative humidity (minimum and maximum) and rainfall (monthly average) have been analyzed and grouped separately for disease and non-disease years. The Leven's T-test for equality of variances has been computed to identify the significant climatic 11 parameters. In the disease group, the minimum related humidity has been found significantly different at (p s; 0.01) confidence level. Analysis of entomological data indicates that Aedes aegypti breeding, associates with climate factors. Occurrence of DF/DHF/DSS in the particular months of March to April is associated with a relatively higher vector density. The mean range of ambient temperature (23.9 and 26.7 °C), relative humidity (48.5 and 47.5 %) and water temperature (16 to 28 °C) is favorable for higher breeding of the vector fauna and disease appearance during the months of March and April respectively. The results of studies on different types of breeding habitats shows that cement tanks lying inside the house had the most preferred breeding place throughout the year with the R2 value of 0 .45. The correlation of adult house index and container index (cement tank) has been found to be significant with the R value of 0.67. Finally, spatial modeling has been used to integrate significant socio-cultural and physical risk indicators for each buffer-zones (200m, 400 and 600m). The three integrated models has been developed which describe different predictive indicators, demonstrating social cultural and physical spatial risk area of Jalor. The integration of social cultural and physical risk map provides a baseline on to monitor the dengue risk. Results provide new information for spatial decision support systems along with direct relevance to the needs of DF/DHF/DSS control. This study would from now assist in the prediction of estimates of spatial outbreaks of DF/DHF/DSS. This research had brought new insights in the application of GIS in understanding of dengue incidences at the spatial, temporal levels. It demonstrates the importance of socio-cultural and physical environments in relation to the occurrence of different levels of dengue incidences in the Jalor, Rajasthan. It may therefore, be concluded that any step to improve any of eight social risk indicators would have favorable effect on reducing the dengue cases. Such analysis provides a valuable information for planning precautionary measures and in controlling the spread of DF/DHF/DSS. It elaborates an integrated use of GIS, statistical and spatial analyses to model the significant socio-cultural and bio-physical environmental indictors affecting the dengue risks levels. Such-indepth-study widens the scope of this study. Spatial extrapolated maps have been derived, classifying the area into the different levels of health risk. |
| Year | 2001 |
| Type | Dissertation |
| School | School of Advanced Technologies (SAT) |
| Department | Other Field of Studies (No Department) |
| Academic Program/FoS | Space Technology Application and Research (SR) |
| Chairperson(s) | Andrianasolo, Haja; |
| Examination Committee(s) | Tripathi, Nitin Kumar ;Athapol Noomhorm ;. Remigio, Amador A. ; |
| Scholarship Donor(s) | French Government ; |
| Degree | Thesis (Ph.D.) - Asian Institute of Technology, 2001 |