This study identifies factors that are associated with the risk of access thrombosis in 267 new hemodialysis patients. There are few longitudinal studies evaluating the risk of access thrombosis despite the need for long-term use of the access for maintenance hemodialysis. We used a prospective design following patients from 26 providers in Renal Network Council #12 (Iowa, Missouri, Kansas, and Nebraska) for 1 year who were starting hemodialysis. There were significant increases in access thrombosis relative risk (RR) associated with the placement of a polytetrafluoroethylene graft compared with patients with the arteriovenous fistula (RR 1.98; 95% confidence interval [CI] = 1.3, 3,01). The probability of remaining thrombosis free 90 days after first use was 90.1% (95% CI = 82.9, 94.4) for arteriovenous fistula patients, but only 71.6% (95% CI = 63.5, 78.2) for polytetrafluoroethylene graft patients. In arteriovenous fistula patients with more than 30 days maturity time the risk of thrombosis was significantly lower than in those with less maturity time (RR 0.40; 95% CI = 0.14, 0.84); however, there was no significant difference for maturity time among patients with a polytetrafluoroethylene graft. Reduced thrombosis risk also was observed in patients with dialyzer blood flow rates greater than 300 mL/min (RR 0.66; 95% CI = 0.44, 0.99). Total heparin dose and erythropoietin therapy were not associated with the risk of thrombosis. No differences in risk were found for age, renal diagnosis, or type of dialyzer.