In this study, we compared the outcomes of forearm arteriovenous grafts (AVGs) and upper arm AVGs in a large, prospectively collected data set, which represents real-world experience with upper extremity prosthetic dialysis access, to determine if there are clinically significant differences in the upper arm and forearm positions. We identified 2,063 patients who received upper extremity AVGs within the Vascular Quality Initiative data set (2010-2018). Axillary to axillary upper arm AVGs were excluded (n=394) from the analysis. The main outcome measures were primary and secondary patency rates at 12months. Other outcomes were 6-month wound infection, steal syndrome, and arm swelling. The log-rank test was used to evaluate patency loss using a Kaplan-Meier analysis. Cox proportional hazards models were used to examine adjusted association between locations (forearm and upper arm) and outcomes. There were 1,160 forearm AVGs and 509 upper arm brachial artery AVGs in the study cohort. Patients with forearm AVGs were more likely to have a body mass index>30 (45% vs. 38%, P=0.013), no history of previous access (73% vs. 63%, P<0.001), and underwent local-regional anesthesia (56% vs. 43%, P<0.001). The 12-month primary patency (51.5% vs. 62.9%, P<0.001) and secondary patency (76.4% vs. 89.1%, P<0.001) were significantly lower for forearm AVGs. Wound infection, steal syndrome, and arm swelling were similar between forearm AVGs and upper arm AVGs at the 6-month follow-up. In multivariable analysis, the primary patency loss (adjusted hazard ratio (aHR) 1.66, 95% confidence interval (CI) 1.33-2.01, P<0.001) and 12-month secondary patency loss (aHR 2.71, 95% CI 1.84-3.98, P<0.001) were significantly higher for forearm AVGs at 12months. From this observational study of the Vascular Quality Initiative data set, the primary and secondary patency rates were superior for upper arm brachial artery AVGs compared with forearm AVGs.