Simulation Modeling

We have used both agent-based and compartmental models to study transmission dynamics and the impact of interventions to reduce transmission across a variety of population settings and scales. In collaboration with CDC partners, we developed a regional simulation representing a regional network of hospitals, nursing homes and long-term acute care hospitals (LTACHs) to investigate the impact of coordinated vs facility-independent responses to an outbreak of antibiotic-resistance, finding a clear benefit coming from healthcare facilities sharing information about where outbreaks are occurring and responding through implementation of interventions to reduce transmission. 

This regional simulation model was extended to look at the impact of an intervention that included active surveillance for Carbapenem-resistant Enterobacteriaceae (CRE) carriers as well as enhanced isolation of identified carriers targeted specifically at ‘high-risk’ LTACHs on the spread of CRE across all of the other facilities finding that focusing intervention efforts on LTACHs is potentially a highly efficient strategy for reducing CRE transmissions broadly. Additionally we have used simulations to study the impact of a vaccine in development for symptomatic Clostridioides difficile (C. diff) infection (CDI) on overall transmission across the region. We found potential for a vaccine against CDI to reduce transmissions in healthcare facilities, even with no direct effect on carriage susceptibility. 

We have also investigated the impact of antibiotics on infection with antibiotic-resistant bacteria and C. diff, including a simulation C. diff transmission and infection, including components such as 1) patients and health care workers, and their interactions; 2) room contamination via C. difficile shedding; 3) C. difficile hand carriage and removal via hand hygiene; 4) patient acquisition of C. difficile via contact with contaminated rooms or health care workers; 5) and patient antimicrobial use. We then introduced six interventions, simulated alone and "bundled" together: aggressive C. diff testing; empiric isolation and treatment of symptomatic patients; improved adherence to hand hygiene and contact precautions; improved use of soap and water for hand hygiene; and improved environmental cleaning. All interventions were tested using parameter values representing base-case, typical intervention, and optimal intervention scenarios.