Antibiotic timing in neonates with suspected hospital-acquired infections

There exists general agreement within neonatology that antibiotics should be administered promptly to neonates with possible bacterial sepsis and meningitis. We initiated a series of quality improvement cycles designed to reduce delays in the initiation of antibiotic therapy to less than 2 hours when hospital-acquired infection (HAI) was suspected. All infants in this study were in neonatal intensive care (level II or III) who were started on antibiotics for a suspected HAI (defined as an infection that occurred 72 hours after admission to the NICU) were audited. Through a series of quality improvement cycles, we analyzed sources of delays in the initiation of antibiotic therapy from the time the order was written through administration. In subsequent cycles, we intervened to reduce delays through education, standardize the evaluation process, and develop an online ordering system that streamlined the workflow patterns in the nurseries and pharmacy. Using a prospective cohort design, we compared antibiotic delivery times after each process improvement cycle. Antibiotic delivery time was reduced from a median of 137.5 minutes to 75 minutes and variation of practice was reduced in terms of standard deviation and range (P < .001). The use of computerized physician order entry significantly improved the writing of STAT orders (P < .0001). A systematic analysis of workflow patterns and efficiencies, coupled with improvement cycles targeting delays and development of a computerized physician order entry system, allowed us to improve antibiotic delivery time in neonates with suspected HAI in an intensive care nursery system.