The National Cooperative Dialysis Study attempted to determine adequacy of hemodialysis based on kinetic modeling of urea. Based on this study, it has been recommended that a dimensionless term quantitating the amount of dialysis delivered (KT/V) be greater than 1.0 to avoid adverse outcomes. With the declining duration of dialysis treatments in the United States, there has been concern that a significant proportion of patients may be receiving inadequate therapy. The purpose of this study was to survey hemodialysis practices and treatment outcomes in our metropolitan area. Sixteen area nephrologists volunteered to study their outpatient hemodialysis patients (N = 617). Demographic data and urea kinetic modeling results were then analyzed at the lead center. The mean length of dialysis was 3.2 +/- 0.4 (SD) hours with dialysis blood flow rates of 333 +/- 74 ml/min. The mean KT/V was 1.03 +/- 0.25 with nearly half of patients failing to attain a KT/V of 1.0. In 55% of patients the reason for a low KT/V was the prescription of an insufficient amount of dialysis treatment. In the remainder, insufficient delivery of prescribed dialysis contributed to the low KT/V. Only 1 of 33 patients undergoing dialysis twice a week achieved the recommended quantity of treatment on a weekly basis. Patients undergoing dialysis in non-profit units had a higher KT/V than those treated in proprietary units (1.1 +/- 0.26 vs. 0.92 +/- 0.22, P less than 0.001). In addition, patients dialyzed in units that performed urea kinetic modeling on all or selected patients had a higher KT/V compared to those in units where urea kinetics were not done (1.12 +/- 0.25 vs. 0.95 +/- 0.23, P less than 0.001). If these findings reflect practices elsewhere in the United States, many hemodialysis patients fail to receive the current recommended quantity of treatment.