Objective: Malnutrition is present in a significant proportion of patients commencing dialysis. However, the prevalence and prognostic significance of malnutrition within the chronic renal insufficiency (CRI) population before the initiation of dialysis is poorly characterized. The aim of this study was to determine the prevalence and prognostic significance of malnutrition in an unselected group of patients with CRI. Design: Cohort analytic study. Setting: Ambulatory care practice of a university teaching hospital. Patients: Fifty patients with CRI (serum creatinine concentration ≥ 1.7 mg/dL) were enrolled. Patients with a recent acute illness, nephrotic syndrome, intercurrent steroid therapy, gastrointestinal disease, or other severe organ failure that may have independently influenced nutritional status were excluded. Intervention: At baseline, patients had a nutritional assessment consisting of subjective global assessment (SGA), measurement of body mass index (BMI), midarm circumference (MAC), serum albumin concentration, total lymphocyte count, and single frequency bioelectrical impedance analysis. Patients received standard medical care and were followed prospectively at quarterly intervals for 12 months. Results: At baseline assessment, 28% of patients had evidence of malnutrition by SGA criteria. The malnourished group of patients had a significantly lower creatinine clearance (18.9 ± 9.8 v 36.5 ± 14.0 mL/min/1.73 m 2, mean ± SD, P < .001), glomerular filtration rate (20.7 ± 10.9 v 28.5 ± 12.5 mL/min/1.73 m 2, P = .04), BMI (22.7 ± 2.9 v 29.0 ± 5.0 kg/m 2), and MAC (24.3 ± 4.9 v 30.7 ± 4.8 cm, P < .001), but there were no differences in serum albumin concentration or total lymphocyte count between the groups. At the 12-month follow-up, there was significantly increased mortality (21% v 3%, P = .04), composite endpoint of death or dialysis (50% v 11%, P = .02), and likelihood of acute hospitalization (78% v 23%, P = .001) in the malnourished group. A significant association was observed between baseline nutritional status and subsequent admission to hospital and baseline glomerular filtration rate and progression to end-stage renal failure. Conclusion: These data suggest that SGA provides a useful means of assessing nutritional status and is helpful in identifying patients with increased risk of morbidity and mortality in the setting of CRI.