In 2004, the World Health Organization performed a survey to assess transmitted drug resistance in Mexico City among drug-naive persons with newly diagnosed human immunodeficiency virus (HIV) infection and likely to be recently infected who were attending 3 voluntary counseling and testing sites. A parallel study comparing 2 alternative methods of enrolling survey participant was conducted in 9 voluntary counseling and testing sites in central Mexico. In study arm 1, subject information, consent and blood specimens were obtained during the HIV diagnostic testing visit. In study arm 2, consent and blood specimens were obtained at the return visit, only from those who were HIV infected. This survey classified nonnucleoside reverse-transcriptase inhibitor and nucleoside reverse-transcriptase inhibitor transmitted drug resistance as <5% and 5%-15%, respectively. Arm 2 yielded major advantages in cost and workload, with no evidence of increased sampling bias.