LAUSR

Abstract The paper determines climate variability in Delhi represented by changes in weather pattern and makes an empirical assessment of their impact on dengue incidence by identifying the ‘high risk’ weeks of a year. Statistical technique of iterative chi-square which compares daily data of climatic elements for each weeks during the recent years (1999–2011) and previous years (1973–1998) selected for this study has been used for determining the weeks of a year during which there have been change in maximum and minimum temperatures, rainfall and relative humidity. Association of climate variability and dengue have been established by modelling longitudinal (panel) data for weekly dengue incidence, changes in climatic elements and the recorded climatic elements during January 2008 to May 2013 using the most suited Poisson regression applying generalized estimating equation approach. Findings reveal significant changes in weather across the year having significant and positive association with dengue cases at specified lags. Weeks in April and July to October with gaps have been identified as the high risk weeks based on the estimated relative risk. There has been intra-annual expansion in dengue risk period extending beyond monsoon and post-monsoon. The outcome of the study contributes to framing of future adaptation strategies.