Observational studies suggest improvements with frequent hemodialysis (HD), but its true efficacy and safety remain uncertain. The Frequent Hemodialysis Network Trials Group is conducting two multicenter randomized trials of 250 subjects each, comparing conventional three times weekly HD with (1) in-center daily HD and (2) home nocturnal HD. Daily HD will be delivered for 1.5-2.75 h, 6 days/week, with target eK(t)/V(n) > or = 0.9/session, whereas nocturnal HD will be delivered for > or = 6 h, 6 nights/week, with target stdK(t)/V of > or = 4.0/week. Subjects will be followed for 1 year. The composite of mortality with the 12-month change in (i) left ventricular mass index (LVMI) by magnetic resonance imaging, and (ii) SF-36 RAND Physical Health Composite (PHC) are specified as co-primary outcomes. The seven main secondary outcomes are between group comparisons of: change in LVMI, change in PHC, change in Beck Depression Inventory score, change in Trail Making Test B score, change in pre-HD serum albumin, change in pre-HD serum phosphorus, and rates of non-access hospitalization or death. Changes in blood pressure and erythropoiesis will also be assessed. Safety outcomes will focus on vascular access complications and burden of treatment. Data will be obtained on the cost of delivering frequent HD compared to conventional HD. Efforts will be made to reduce bias, including blinding assessment of subjective outcomes. Because no large-scale randomized trials of frequent HD have been previously conducted, the first year has been designated a Vanguard Phase, during which feasibility of randomization, ability to deliver the interventions, and adherence will be evaluated.