Utilizing statistical process control tools and highly-detailed simulation for biosurveillance in intensive care units to control healthcare associated infections

Objective: To determine the effectiveness of combining statistical process control and simulation tools to predict and control outbreaks of healthcare-acquired infections in Intensive Care Units. Background: In the United States, 1 in 20 hospitalized patients become infected every year with a healthcare associated infection (HAI), and 20% of HAI patients become infected in intensive care units (ICUs). Some of the complications of HAIs include increased morbidity, increase of drug-resistant infections, and a strain in the resources of healthcare systems. Statistical process control (SPC) tools have been used in healthcare settings to predict outbreaks. Based on these predictions, analysts make recommendations to assist stakeholders on control plans to mitigate the outbreak. Methods: Data has been collected for the patient population from a regional hospital in Southwest Virginia for one year through electronic medical records of patients and direct surveillance of healthcare professionals. A highly detailed synthetic representation of one of the hospital's ICUs was created using patient and healthcare worker data. Different infection scenarios were used to simulate outbreaks within the ICU. SPC control charts were utilized to predict potential outbreaks in the ICUs. Specific types of control charts were more effective in identifying outbreaks on time. The simulated ICU biosurveillance scenarios are analyzed to derive useful inferences, which will assist in improving control plans for HAIs. Public health significance: The use of high performance simulation and biosurveillance of HAIs will assist in improving HAI prevention and control plans, and planning for optimal allocation of limited healthcare resources to reduce morbidity and mortality of patients with HAIs.