Trends in spatio-temporal dynamics of visceral leishmaniasis cases in a highly-endemic focus of Bihar, India: an investigation based on GIS tools

• DOI: 10.1186/s13071-018-2707-x
• Publication Date: 2018-04-04T04:34:13Z
• Authors: Rakesh Mandal; Shreekant Kesari; Vijay Kumar; Pradeep Das
• Keywords: Male; Spatial Analysis; Kala-azar; Endemic Diseases; Research; Incidence; India; Infectious and parasitic diseases; RC109-216; Spatio-temporal analysis; 3. Good health; 03 medical and health sciences; 0302 clinical medicine; Geographical Information Systems; Geographic Information Systems; Animals; Humans; Leishmaniasis, Visceral; Female; Disease Eradication; Spatial autocorrelation; Software

Visceral leishmaniasis (VL) in Bihar State (India) continues to be endemic, despite the existence of effective treatment and a vector control program to control disease morbidity. A clear understanding of spatio-temporal distribution of VL may improve surveillance and control implementation. This study explored the trends in spatio-temporal dynamics of VL endemicity at a meso-scale level in Vaishali District, based on geographical information systems (GIS) tools and spatial statistical analysis.A GIS database was used to integrate the VL case data from the study area between 2009 and 2014. All cases were spatially linked at a meso-scale level. Geospatial techniques, such as GIS-layer overlaying and mapping, were employed to visualize and detect the spatio-temporal patterns of a VL endemic outbreak across the district. The spatial statistic Moran's I Index (Moran's I) was used to simultaneously evaluate spatial-correlation between endemic villages and the spatial distribution patterns based on both the village location and the case incidence rate (CIR). Descriptive statistics such as mean, standard error, confidence intervals and percentages were used to summarize the VL case data.There were 624 endemic villages with 2719 (average 906 cases/year) VL cases during 2012-2014. The Moran's I revealed a cluster pattern (P < 0.05) of CIR distribution at the meso-scale level. On average, 68 villages were newly-endemic each year. Of which 93.1% of villages' endemicity were found to have occurred on the peripheries of the previous year endemic villages. The mean CIR of the endemic villages that were peripheral to the following year newly-endemic villages, compared to all endemic villages of the same year, was higher (P < 0.05).The results show that the VL endemicity of new villages tends to occur on the periphery of villages endemic in the previous year. High-CIR plays a major role in the spatial dispersion of the VL cases between non-endemic and endemic villages. This information can help achieve VL elimination throughout the Indian subcontinent by improving vector control design and implementation in highly-endemic district.