RAPID: Curtailing Nosocomial Amplification of COVID-19

One of the populations hardest hit during an emerging epidemic are frontline healthcare workers. Infections in healthcare workers and in healthcare settings produce two separate but related challenges. The first, and most obvious, is that sick or dying healthcare workers cannot care for patients, causing labor shortages right at the moment when demand on a healthcare system is likely increasing due to the epidemic. The second problem is that, because of how difficult it is to control infections within hospitals, the epidemic itself can accelerate once it reaches the healthcare system, causing a rapid increase in the number of cases. Examples of this phenomena, which we call 'nosocomial amplification', are common, including both previous major coronavirus epidemics before COVID-19 (SARS and MERS) as well as Ebola. This project will seek to model and understand what factors within a hospital can prevent this from happening, to increase the resilience of healthcare systems. Outcomes from this effort will be informing behavioral guidelines for healthcare systems. Other broader impacts from this project include training and professional development opportunities for students.

The researchers will adapt an existing model of within-hospital infection transmission to COVID-19 and combine this model with a spatially explicit agent-based model of the healthcare facilities in the state of North Carolina. This approach allows for the representation both of the healthcare environment, as well as the community where initial cases are seeded, where healthcare workers can infect – and be infected by – their community, etc. It also allows for the modeling of facility-level impacts of state-level decisions and allows us to address the question of how to best protect the health of both patients and healthcare workers in an environment where both are at significant risk of infection and critical supplies such as PPE are not necessarily unlimited. Simultaneously, they will collect data from hospitals in the SHEA Research Network on the changes brought on to staffing and clinical practice from COVID-19, such as whether or not the ratio of nurses to patients has changed, as well as ascertaining to what extent modeling has been used in hospital decision making, and whether or not it has been useful in that role. This project will thus have both a robust and sophisticated model for the interaction between hospitals at a granular level and the surrounding community, as well as timely parameter estimates from a diverse array of hospitals on how COVID-19 has changed their practices, as well as how modeling might be better tailored to inform the control of both COVID-19 and future epidemics.

This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.