Nevirapine (NVP) is the primary antiretroviral (ART) available in Zimbabwe. However, its use as first line ART is complicated by interactions with rifampicin (RIF), the primary drug available for tuberculosis (TB). In some cases ART treatment is delayed or avoided in TB co-infected patients. Previous studies of the effect of RIF on NVP pharmacokinetic (PK) exposure has been inconclusive, failing to generate specific recommendations regarding co-treatment. Moreover, PK and safety data for NVP with RIF have not been generated in our population in Zimbabwe. This study will compare the PK, safety and treatment outcomes of NVP 200 mg BID versus 300 mg BID during RIF co-administration in HIV-TB co-infected Zimbabwean patients. Thirty-two HIVTB patients stabilized on RIF based TB therapy will initiate NVP 200 mg BID for 14 days followed by randomization to either 400 mg or 600 mg daily NVP-based ART. Intensive PK for NVP will be estimated at 2 and 4 weeks following NVP initiation with sparse samples collected at 16 and 24 weeks. Patients will be followed for 24 weeks for safety and clinical and or laboratory outcomes. Sixteen control patients (HIV only, no TB) will initiate NVP 200 mg QD for 14 days, escalated to 200 mg BID. They will undergo identical procedures. Plasma NVP will be quantitated using LC tandem MS. PK results will be generated using non-compartmental methods with WinNonlin®5.2.1 including AUC0-12h, Cssave; Cmax and Cmin. Comparisons of the three study groups will be analyzed using analysis of variance (Student t test). Descriptive statistics will evaluate safety and tolerability of NVP at varying doses with RIF. CD4 cell counts and HIV RNA will be measured at baseline and monitored at weeks 4, 16 and 24 to assess treatment outcomes in the intervention arms.