There has been ongoing debate in the scientific community on the main factors causing the observed exacerbation of epidemic malaria in regions where altitude previously limited transmission. Two leading hypotheses that are typically viewed as alternative are the spread of anti-malarial drug resistance and the increase in transmission intensity due to climate change.  However, as we demonstrate here, these two drivers are not necessarily independent, since at different locations or under different drug treatments, an increase in transmission intensity due to warmer temperatures can act either to increase or decrease the spread of resistance. Our modeling approach is based on the observation that individuals living in areas of endemic P. falciparum develop immunity to malaria with age and exposure; as levels of immunity are higher, there is a decline of parasite density in the blood which is assumed to lead to lower probabilities of transmission, milder clinical symptoms, and possibly an extension in the duration of infections. Our model is constructed as a multi-stage SIS model with different levels of immunity as a function of exposure, and is an extension of a recently proposed two-stage model by Klein and colleagues (2008).

Results/Conclusions

Analytical analysis and simulations demonstrate that an increase in transmission beyond a threshold can lead to a decrease in drug resistance, as previously shown, but that a second threshold can occur, where further increases in transmission lead to the re-establishment of drug resistance. Our results provide an explanation for the ‘valley phenomenon’ described in empirical observations for Chloroquine treatment failure, which was found to be higher in areas of low and high transmission, and at its lowest at intermediate transmission areas.  We discuss the implications of our findings for the argument of drug resistance vs. climate change in regions of epidemic (unstable) malaria.