Treatment of HIV infection in developing countries, particularly those in Africa, is still a major challenge to healthcare systems with limited laboratory resources. While drug costs have fallen to levels where antiretroviral therapy is now possible in these countries, and results suggest that adherence is as good as in developed countries, questions remain regarding the effect of scarce laboratory resources on treatment monitoring and, hence, outcome. The DART trial aimed to measure the effect of laboratory monitoring. This randomized controlled trial evaluated the differences between routine laboratory monitoring and monitoring driven by clinical events, and was conducted between January 2003 and December 2008 at sites in Uganda and Zimbabwe. The results indicate that clinically driven monitoring is likely to be more cost-effective in this resource-limited situation.