Objective: Prospectively examine the use of a hemodialysis prognostic nutrition index (HD-PNI) as a predictor for morbidity and mortality in hemodialysis patients and its correlation to adequacy of dialysis. Design: Prospective randomized collaborative study group. Setting: There were 211 chronic hemodialysis centers; 202 from 43 United States, 9 from Canada. Patients: There were 1527 hemodialysis patients undergoing treatment a minimum of 3 months and at least 18 years of age. Sample mirrored United States Renal Data system data for age, sex, race, and etiology of renal failure. Interventions: None; routinely collected demographic, biochemical, and clinical data for 8-month baseline and 3-month predictive phases. Methods: HD-PNI calculated from baseline data as linear mathematical equation using level of serum albumin, level of serum creatinine, and number of days and times hospitalized; HD-PNI risk defined as ≥0.8. Adequacy of dialysis calculated as urea reduction ratio (URR) from baseline data; adequacy risk defined as URR of ≤65%. Main outcome measures: Number of times and days hospitalized, mortality. Results: For this research, 1167 patients completed the study (76%) with 360 (24%) dropped due to death, incomplete data, transfer, or change in modality. Patients completing study with HD-PNI risk (n = 208, 18%) compared with patients who had no HD-PNI risk (n = 959, 82%) were hospitalized more often (57.2% v 28.5%, P < .01), hospitalized with infection more often (14.6% v 4.6%, P < .01), and had greater mortality (7.7% v 2.5%, P < .01). Stratification of HD-PNI risk by URR of ≥65% did not significantly improve prediction. Conclusions: Use of HD-PNI is an effective screening tool to identify hemodialysis patients at risk for morbidity and mortality. No correlation was found between URR and HD-PNI.