To further investigate...

            Although mean temperature and precipitation do explain to a large degree the variability in malaria prevalence; not all of the variability is explained.  There are many other factors such as minimum and maximum temperature, relative humidity, wind, changes in types and amount of irrigation, rates of deforestation, direction and rates of migration, urbanization rates, and amount and distribution of malaria control efforts, that, if taken into consideration, could help explain more of the variability in malaria rates in Karnataka, India, especially in districts that had lower R² values for climate and malaria alone.

            This study should be seen as a preliminary assessment of malaria and climate in Karnataka that can be used to stimulate further research into the relationship between climate and malaria there with more robust data sets, at smaller resolution, and with more data on non-climatic factors so that better prediction can be done with those results.  Malaria prevalence, as well as mean temperature and precipitation, varies quite a bit within districts, which are political and not ecological or physical constructs (Rajagopalan et al., 1990).  Studying at the block or village level would improve the specificity of the results of this study and the ability to predict where climate is likely to impact malaria rates the most.  The best geographical area at which to do the study would be by malaria vector/parasite ecology boundaries, although getting identifying these areas and getting data for these areas would likely be difficult. 

The results from this study, however, can be used to influence policy on climate change and malaria control practices to stimulate action towards limiting the furthering of anthropogenically-caused climate change.