INTRODUCTION: Single center work demonstrated a safe reduction in unnecessary blood culture use in critically ill children. Our objective was to develop and implement a customizable quality improvement framework to reduce unnecessary blood culture testing in critically ill children across diverse clinical settings and various institutions. METHODS: Three pediatric intensive care units (14 bed medical/cardiac; 28 bed medical; 22 bed cardiac) in 2 institutions adapted and implemented a 5-part Blood Culture Improvement Framework, supported by a coordinating multidisciplinary team. Blood culture rates were compared for 24 months preimplementation to 24 months postimplementation. RESULTS: Blood culture rates decreased from 13.3, 13.5, and 11.5 cultures per 100 patient-days preimplementation to 6.4, 9.1, and 8.3 cultures per 100 patient-days postimplementation for Unit A, B, and C, respectively; a decrease of 32% (95% confidence interval, 25-43%; P < 0.001) for the 3 units combined. Postimplementation, the proportion of total blood cultures drawn from central venous catheters decreased by 51% for the 3 units combined (95% confidence interval, 29-66%; P < 0.001). Notable difference between units included the identity and involvement of the project champion, adaptions of the clinical tools, and staff monitoring and communication of project progress. Qualitative data also revealed a core set of barriers and facilitators to behavior change around pediatric intensive care unit blood culture practices. CONCLUSIONS: Three pediatric intensive units adapted a novel 5-part improvement framework and successfully reduced blood culture use in critically ill children, demonstrating that different providers and practice environments can adapt diagnostic stewardship programs.