BACKGROUND: Cardiopulmonary resuscitation (CPR) for out-of-hospital cardiac arrest (OHCA) is associated with higher survival. The association between time to bystander CPR at different time thresholds, compared with those with no bystander CPR, is less clear. METHODS: Within the Cardiac Arrest Registry to Enhance Survival, we identified 194,807 witnessed OHCAs during 2013-2023. Multivariable hierarchical logistic regression was used to evaluate the association between each time interval for initiation of bystander CPR (0-1, 2-3, 4-5, 6-7, 8-9, 10 + minutes), compared with no bystander CPR, for survival to discharge and favorable neurological survival (i.e. without severe neurological deficits). RESULTS: The mean age was 64.4 ± 15.9 years, and 33.8% were female. Bystander CPR was provided in 48.4% of cases, with a median initiation time of 2 min (IQR: 1-5). Overall, 15.3% survived to discharge, and 12.9% had favorable neurological survival. Compared with no bystander CPR, survival to discharge was higher for patients with bystander CPR initiated at 0-1 min (OR 1.78 [95% CI: 1.73-1.84]), 2-3 min (OR 1.57 [1.51-1.64]), 4-5 min (OR 1.23 [1.17-1.30]), 6-7 min (OR 1.25 [1.15-1.35]), and 8-9 min (OR 1.13 [1.03-1.25]), but no survival association was seen at ≥ 10 min (OR 0.80 [0.74-0.86]). A similar pattern was observed for neurological survival. CONCLUSIONS: Compared with no bystander CPR, bystander CPR was associated with improved survival even when started at 8 to 9 min. Given that there is a graded, inverse relationship between time to bystander CPR and survival, these findings underscore the urgency of immediate bystander CPR initiation to optimize OHCA survival.