Although needleless connectors (NCs) are widely used in clinical practice, they carry significant risk of bloodstream infection (BSI). In this study, we quantified differences in bacterial transfer and biofilm formation between various NCs. Prospective, clinically simulated in vitro experimental study. We tested 20 NCs in a 5-day clinical simulation of Staphylococcus aureus inoculations onto NC septum surfaces, which were then flushed with saline and cultured for bacterial transfer. Biofilm formation was measured through destructive sampling of the connector-catheter system. Moreover, 8 NC design factors were evaluated for their influence on bacterial transfer and biofilm formation. This study was designed without a disinfection protocol to ascertain the intrinsic risk of each NC. Clave Neutron and MicroClave had the lowest overall mean log density of bacteria in the flush compared to other NCs (P < .05), except there were no statistically significant differences between Clave Neutron, Microclave, SafeTouch, and SafeAccess (P ≥ .05). The amount of biofilm in the NC was positively associated with bacteria in the flush (P < .0005). Among 8 design factors, flow path was most important, with the internal cannula associated with a statistically significant 1 log reduction (LR) in bacteria in the flush (R2 = 49%) and 0.5-2 LR in the connector (R2 = 34%). All factors together best explained bacteria in the flush (R2 = 65%) and biofilm in the connector (R2 = 48%). Bacterial transfer and biofilm formation in the connector-catheter system varied statistically significantly between the 20 NCs, suggesting that NC choice can lower the risk of developing catheter-related BSIs.