The objective of this field trial was to reduce bovine leukemia virus (BLV) transmission and prevalence in commercial dairy herds using proviral load (PVL) and lymphocyte count (LC) measurements as indicators of the most infectious animals for culling or segregation. Bovine leukemia virus causes lymphoma in <5% of infected cattle, and increased lymphocyte counts (lymphocytosis) in about one-third. Recent research has shown that dairy cows infected with BLV have altered immune function associated with decreases in milk production and lifespan. Recent findings show that a minority of infected cattle have PVL concentrations in blood and other body fluids of over 1,000 times that of other infected cattle. In combination with a high LC, these animals are thought to be responsible for most transmission of BLV in a herd. Milk or blood samples from adult cows in our 3 Midwestern dairy farm field trials were tested semiannually with ELISA for BLV antibodies, and ELISA-positive cattle were then retested using a blood LC and a quantitative PCR test for PVL to identify the animals presumed to be most infectious. Herd managers were encouraged to consider PVL and LC status when making cull decisions, and to segregate cows with the highest PVL and LC from their BLV ELISA-negative herd mates where possible. After 2 to 2.5 yr of this intervention, the incidence risk of new infections decreased in all 3 herds combined, from 13.8 to 2.2, and the overall herd prevalence decreased in all 3 herds combined from 62.0 to 20.7%, suggesting that this approach can efficiently reduce BLV transmission as well as prevalence. This is encouraging, because a very low prevalence of BLV infection would make it economically feasible to cull the remaining ELISA-positive cattle, as was achieved in national eradication programs in other countries decades ago.