Preventing antimicrobial resistance and infections in hospitalized neonates in low resource settings

点击翻译按钮获取中文摘要

Project Summary/Abstract Annually, 2.5 million babies die within the first four weeks of life, nearly a quarter due to infectious causes. Newborns admitted to the Neonatal Intensive Care Unit (NICU) are especially vulnerable, due to such factors as prematurity, an immature immune system, and need for life-sustaining invasive procedures and devices. In low and middle income countries (LMIC), an increasing number of NICUs care for premature and critically ill newborns. Healthcare-associated bloodstream infections (HA-BSI) in LMIC are more common due to inadequate infection prevention and control (IPC) and more difficult to treat due to high rates of antimicrobial resistance (AMR). Previous research in this setting focuses primarily on outbreak investigations and does not adequately describe risk factors for HA-BSI. Healthcare facilities lack effective tools to assess maternal and neonatal IPC and create improvement strategies. Preliminary data from the applicant's ongoing prospective cohort study that has enrolled over 6600 neonates in three NICUs in Pune, India, reinforces the high incidence of HA-BSI in this setting with a rate of 7.6 per 1000 patient-days, as well as high rates of AMR. Among Klebsiella pneumoniae isolates, the most common BSI pathogen, 96% are resistant to third-generation cephalosporins and 38% to carbapenems. Among neonates with BSI, mortality is 22%. Within the framework of this study, the following are proposed: (1) To identify modifiable risk factors for HA-BSI in the NICU; (2) To develop a model for predicting infection with carbapenem-resistant organisms (CRO); and (3) To develop and pilot a novel tool to assess IPC practices in the NICU and Labor & Delivery. Identifying risk factors for HA-BSI in the NICU will promote development of targeted IPC strategies. Creation of a prediction model using a decision tree algorithm will help identify babies at highest risk of CRO infections. Such a model can support NICU clinicians in selecting the right antibiotics when infection is suspected, reducing time to appropriate therapy and decreasing unnecessary use of last resort antibiotics such as colistin. Development of an IPC assessment tool that incorporates human factors engineering (HFE) principles will enable healthcare facilities to optimize IPC and reduce risk of hospital-acquired infections and associated mortality. This mentored research will train the applicant in advanced epidemiologic methods and application of IPC in LMIC. The applicant is a neonatologist at Johns Hopkins University committed to patient-oriented research in resource-limited settings. Her long-term goals are to become a leader in neonatal IPC in low resource settings and devise interventions to reduce global burden of HA-BSI and associated mortality. This K23 will facilitate skill development in longitudinal data analysis, prediction models, survey development, HFE, and qualitative data analysis. Training will include formal coursework, supervised data analysis, and mentorship by a team with expertise in infectious diseases, IPC, biostatistics, epidemiology, patient safety, and HFE. Collectively, the activities of this K23 will provide a pathway to an independent career as a clinical investigator with expertise in healthcare epidemiology and IPC in low resource settings.