Fig. 3

Kaplan–Meier plots for re infection by RANTES INT1.1 genotype (TT, TC or CC). Cumulative re-infections are plotted against weeks to the next infection. The study participants were actively followed-up with visits once every 2 weeks at their homes, to obtain information about re-infection. Using a Cox proportion hazard regression model, adjusting for age, malaria history and ITN use, the predictors of length of time to first re-infection for TT (wild- type); TC (heterozygous) and CC (homozygous) were not statistically different with p-value of 0.28 and 0.67, respectively