Bacteremia is associated with high morbidity and mortality, but the utility and optimal timing of follow-up blood cultures (FUBCs) remain undefined. To assess the optimal timing of FUBCs related to appropriate antibiotic therapy (AAT), adults with community-onset bacteremia and FUBCs after bacteremia onset were retrospectively studied during the 6-year period in two hospitals. Based on the time gap between the initiation of AAT and FUBC sampling, 1,247 eligible patients were categorized as FUBCs prior to AAT (65 patients, 5.2%), 0-3 days (202, 16.2%), 3.1-6 days (470, 37.7%), 6.1-9 days (299, 24.0%), and ≥9 days (211, 16.9%) after AAT. The prognostic impact of the growth of the same bacteria in FUBCs on 30-day mortality was evidenced only in patients with FUBCs at 3.1-6 days after AAT (adjusted odds ratio [AOR], 3.75; P < 0.001), not in those with FUBCs prior to AAT (AOR, 2.86; P = 0.25), 0-3 days (AOR, 0.39; P = 0.08), 6.1-9 days (AOR, 2.19; P = 0.32), and ≥9 days (AOR, 0.41; P = 0.41) of AAT, after adjusting independent factors of 30-day mortality recognized by the multivariable regression in each category. Conclusively, persistent bacteremia in FUBCs added prognostic significance in the management of adults with community-onset bacteremia after 3.1-6 days of AAT.