Social Networks and the Spread of Influenza and other Nosocomial Infections

点击翻译按钮获取中文摘要

DESCRIPTION (provided by applicant): In this K01 application, Philip Polgreen, MD, seeks to gain expertise in graph theory pertaining to social networks and in mathematical simulations for developing more effective interventions to minimize the spread of nosocomial infections (specifically influenza, MRSA, and C. difficile). Dr. Polgreen has assembled a group of extremely strong mentors and interdisciplinary collaborators who are all highly committed to his success. Healthcare associated infections affect about 2 million patients in U.S. hospitals each year. Furthermore, hospitals serve as amplifiers for the spread of infectious pathogens: patients who are infected or colonized with transmissible pathogens are often in close proximity to uninfected patients with compromised immune systems. For example, SARS did not spread much in the community but spread widely in hospitals. MRSA and C. difficile historically spread first in healthcare facilities and later in the community. Also, many infectious disease experts are concerned that the spread of H5N1 influenza or strains of vancomycin-resistant S. aureus could be magnified in hospitals. Vaccination and hand hygiene are the most effective measures for preventing the spread of hospital-acquired infections. However, no data or theoretical framework exist to identify the healthcare workers who are most likely to acquire and transmit infectious agents (i.e., those who should have the highest priority in influenza vaccine campaigns or should be the focus of programs to increase adherence with hand hygiene). In addition, the mathematical models previously used to study spread of infections in hospitals are based on the assumption that the mixing of healthcare workers and patients is homogeneous. These models ignore the social networks (clustering and small world properties) that define the true nature of the interactions between healthcare workers and patients and thus can yield misleading results. Dr. Polgreen's hypothesis is that some groups of healthcare workers are substantially more likely to spread nosocomial pathogens than are other groups. To test this hypothesis, he will collect information [from direct observation, keyboard login-data, Radio Frequency Identification Badge (RFID) data, and sensor mote data] on the contacts between healthcare workers and patients at University of Iowa Hospitals and Clinics. This information will allow him to develop graph (social network) models that describe the spread of pathogens that are transmitted by the respiratory route (contact within 3 feet) and those spread by direct contact. He will then estimate, through mathematical simulations, how different vaccination and hand hygiene strategies in various groups of healthcare workers affect the spread of nosocomial pathogens. This work will lead to more effective infection control interventions and strategies.