Background: It is unknown whether in malaria-endemic areas, cotrimoxazole (CTX) prophylaxis can be stopped in HIV-infected patients on antiretroviral therapy (ART) who have regained immune competence. The aims of this thesis were to: review the literature on the effect of CTX on malaria; investigate the effect of stopping CTX on malaria incidence in a randomised trial; and assess the effect of CD4 count and ART regimen on malaria. Methods: (i) The literature was systematically searched for relevant papers. (ii) Data from the recently completed COSTOP trial were used to examine the effect of stopping CTX on malaria incidence among HIV-infected Ugandan adults on ART. Participants with CD4 count ≥250 cells/μl were randomised (1:1) to continued CTX or placebo. CD4 counts were determined at ART initiation, enrolment and during follow-up. Malaria was defined as fever with parasitaemia. Incidence and rate ratios (RR) were estimated using random effects Poisson regression, accounting for multiple episodes. Results: (i) Six studies were identified. All reported an increase in malaria following CTX discontinuation. However, all studies were subject to bias and/or confounding. (ii) In COSTOP, 2180 participants were followed for a median of 2.5 years. They experienced 453 malaria episodes. Malaria incidence was 3.5 (95%CI=2.7-4.4) times higher on placebo than CTX. Few cases of severe malaria occurred, and no increase in malaria mortality. CD4 count had no effect on incidence. Partcipants on a protease inhibitor-based regimen experienced malaria significantly less often than those on other regimens. Conclusion: Among participants with CD4 of ≥250 cells/μl, malaria incidence increased when CTX was stopped. This effect was lower than shown in previous (unblinded) trials. Malaria mortality did not increase. CD4 count had no effect on malaria incidence. These results support current WHO guidelines on CTX use in malaria-endemic areas, but there may be subgroups who benefit from CTX discontinuation.