Thrice-weekly hemodialysis is the most commonly used form of renal replacement therapy, yet it is associated with unacceptably high morbidity and mortality. Attempts to improve outcomes for hemodialysis patients by increasing their per-session dose of dialysis have recently proven unsatisfactory in the multicentered Hemodialysis (HEMO) study. Interest has thus turned to increasing dialysis frequency. Short daily and long nocturnal dialysis, which are typically performed 6 days per week, are gaining acceptance and are associated with significant improvements in secondary outcomes, including nutrition, left ventricular hypertrophy, hypertension, anemia, and calcium-phosphorus balance. Studies to date have not been adequately powered to detect the survival benefits that these changes may confer. Large-scale randomized studies are planned, but will likely not answer the survival question for several years. Until this issue is resolved, funding policies are unlikely to change, confining current dialysis patients to potentially suboptimal therapy. By capturing data from current and future daily dialysis patients using an international registry, a survival benefit might be demonstrated more quickly. Such a project will soon be undertaken by the London Daily/Nocturnal Study Group with endorsement from the International Society for Hemodialysis and the U.S. National Institutes of Health. This database will also provide useful descriptive data that will help develop methodologies in this growing field. Historically the interpretation of dialysis registry data has been plagued with various methodological problems. These are briefly reviewed, and some potential solutions and necessary precautions are discussed.