Examining spatial patterns in the distribution of low birth weight babies in southern india- the role of maternal, socio-economic and environmental factors

Authors:Mark Rohit Francis, Rakesh PS, Venkata Raghava Mohan, Vinohar Balraj, Kuryan George
Int J Biol Med Res. 2012; 3(1): 1255-1259  |  PDF File


While individual-level maternal risk factors continue to play a significant role in explaining low birth weight (LBW) outcomes, the need to better understand the contribution of other explanatory factors at both the individual and the neighbourhood level are vital to proposing future prevention strategies and interventions. Prevalence of LBW babies were calculated for all the villages in the Kaniyambadi block of Vellore district using the data from the Health Information System of the Department of Community Health, Christian Medical College, temporal trends in the distribution of LBW babies were mapped over a 20 year period (1991 to 2010). Spatial analysis was performed on a total of 7,058 births during 2006-2010 to identify statistically significant spatial clusters of high values of LBW deliveries using the Getis-Ord Gi* hotspot analysis tool in ArcGIS 10. At the village level, the LBW deliveries during the same period were modelled using global Ordinary Least Squares (OLS) linear regression and Geographically Weighted Regression (GWR) looking at various maternal risk factors, socio-economic indicators, and community-level environmental factors. The overall prevalence of LBW for the 20 year period was found to be 16.8%. Significantly high spatial clustering of LBW was observed in the region with hotspots being noted in 10 villages. The OLS regression for all term LBW births between 2006 and 2010 revealed that mothers with anaemia and under-educated members were significant predictors of LBW in the region (AICC = 442.4, adjusted R2 = 0.843). The GWR model provided a better fit, with an AICC = 436.4 and an adjusted R2 = 0.867. Spatial autocorrelation using the Global Morans I method revealed no statistically significant spatial clustering among the residuals of the GWR. After adjusting for spatial nonstationarity, the important risk factors predicting the burden of LBW babies in the region are maternal under-education and anaemic mothers. GIS tools provide a powerful way of exploring spatial phenomenon and trends, aiding a better understanding of public health concerns.