A randomized, double-blind, double-dummy controlled, multicenter trial was conducted that involved 554 antiretroviral-naive human immunodeficiency virus-infected adults (plasma HIV type 1 [HIV-1] RNA level, >or=400 copies/mL; CD4(+) cell count, >100 cells/mm(3)) and compared a 300-mg once-daily (q.d.) regimen of lamivudine (3TC) versus a 150-mg twice-daily (b.i.d.) regimen of 3TC, combined with zidovudine (300 mg b.i.d.) and efavirenz (600 mg q.d.), during a 48-week period. Treatments were considered equivalent if the 95% confidence interval (CI) for the difference in proportions of patients achieving an HIV-1 RNA level of <400 copies/mL was within the bound of -12% to 12%. At week 48 of the study, an intent-to-treat analysis in which patients with missing data were considered to have experienced treatment failure showed that the 3TC q.d. and 3TC b.i.d. regimens were equivalent (HIV-1 RNA level <400 copies/mL, 178 [64%] of 278 vs. 174 [63%] of 276; treatment difference, 1% [95% CI, -7.1% to 8.9%]; HIV-1 RNA level <50 copies/mL, 165 [59%] of 278 vs. 168 [61%] of 276; treatment difference, 1.7% [95% CI, -9.7% to 6.6%]). Median increase above baseline in CD4(+) cell count was similar (q.d. group, +144 cells/mm(3); b.i.d. group, +146 cells/mm(3)), and the incidences of adverse events, disease progression, and HIV-associated conditions were comparable.