Complex system modeling for hospital-associated infection transmission in pediatric ICUs

Hospital-associated infections are a major problem that not only affect the health of the patient but also result in increased cost of treatment and strains healthcare resources. This problem persists in spite of the fact that the best practices to reduce the occurrence for common hospital-associated infections are well-documented in the literature. This research proposes a complex system modeling framework to tackle this problem.;The framework enables one to look beyond the traditional infection control practices of concentrating on a single infection affecting a limited patient population. Instead, this research considers the 5 most common types of hospital-associated infections affecting the entire patient population in a pediatric ICU. This framework encompasses variety of engineering techniques such as simulation modeling, risk analysis, and statistical analysis. Simulation modeling is utilized to model an ICU work environment from the perspective of infection transmission. The model depicts patient population, invasive procedures performed on the patients and the main functions of staff. Risk analysis is then utilized to evaluate the effect of different patient physiological factors and the invasive procedures performed on the patients to determine a 'risk of infection' model. Using the simulation and the risk model, different interventions are then modeled and their effectiveness in reducing the risk of infection is evaluated and compared. Experimental design is then used to identify the factors that contribute the most, towards reducing infections.;The findings show that the combination interventions are more effective in reducing the risk of HA infections as compared to the effectiveness of individual interventions. They also show that interventions oriented towards reducing the risk of HA infection due to CVCs, mechanical ventilation, PICCs, Foley catheters, patient length of stay, hemodialysis/CRRT and broviac lines are more likely to succeed in reducing the overall risk of HA infections.;While it is inconceivable to try these interventions in a real ICU to test their effectiveness, the complex system modeling framework allows one to do just that. The framework is flexible and can accommodate additional factors that may be discovered in future as those affecting the risk of HA infections.