ABSTRACT
SETTING: Two rural districts in Malawi: Thyolo, where voluntary counselling and human immunodeficiency virus (HIV) testing (VCT) is offered to all tuberculosis (TB) patients and adjunctive cotrimoxazole to HIV positives, and Mulanje, where no such interventions are offered. OBJECTIVES: For all TB patients registered in 2001: 1) to determine the uptake of VCT and cotrimoxazole in Thyolo, and 2) to compare treatment outcomes between Thyolo and Mulanje. DESIGN: A cohort study using routinely collected programme data. RESULTS: There were 1239 TB patients in Mulanje and 1103 in Thyolo. In Thylo, 1064 (97%) patients consented to VCT, 1006 were HIV tested (91%) and 761 (69%) were started on cotrimoxazole a median of 4 days from registration; 77% of patients tested in Thyolo were HIV-positive. For all TB patients, in Thyolo and Mulanje, treatment success was respectively 75% and 61% (P < 0.001); death was 21% and 25% (P = 0.026); and other outcomes were 4% and 14% (P < 0.001). The adjusted relative risks of treatment success (1.23), death (0.84) and other outcomes (0.26) in Thyolo were significantly different from those in Mulanje (P < 0.001). CONCLUSION: VCT and adjunctive cotrimoxazole is well accepted by TB patients in Thyolo and, with other HIV care and support services, is associated with good treatment outcome indicators for the National Tuberculosis Programme. This intervention is being expanded to other districts in Malawi, and other African countries should consider a similar approach to the dual HIV-TB epidemic.