End-stage renal disease in the United States creates a large burden for both individuals and society as a whole. Efforts to prevent the condition require an understanding of modifiable risk factors. We assessed the development of end-stage renal disease through 1990 in 332,544 men, 35 to 57 years of age, who were screened between 1973 and 1975 for entry into the Multiple Risk Factor Intervention Trial (MRFIT). We used data from the national registry for treated end-stage renal disease of the Health Care Financing Administration and from records on death from renal disease from the National Death Index and the Social Security Administration. During an average of 16 years of follow-up, 814 subjects either died of end-stage renal disease or were treated for that condition (15.6 cases per 100,000 person-years of observation). A strong, graded relation between both systolic and diastolic blood pressure and end-stage renal disease was identified, independent of associations between the disease and age, race, income, use of medication for diabetes mellitus, history of myocardial infarction, serum cholesterol concentration, and cigarette smoking. As compared with men with an optimal level of blood pressure (systolic pressure or = 210 mm Hg or diastolic pressure > or = 120 mm Hg) was 22.1 (P < 0.001). These relations were not due to end-stage renal disease that occurred soon after screening and, in the 12,866 screened men who entered the MRFIT study, were not changed by taking into account the base-line serum creatinine concentration and urinary protein excretion. The estimated risk of end-stage renal disease associated with elevations of systolic pressure was greater than that linked with elevations of diastolic pressure when both variables were considered together. Elevations of blood pressure are a strong independent risk factor for end-stage renal disease; interventions to prevent the disease need to emphasize the prevention and control of both high-normal and high blood pressure.