We all know the images of the coronavirus pandemic: Ambulances arriving at hospitals, health workers in space suits, overflowing hospital corridors, and EMTs pushing patients on gurneys.
“When infections hit the health care system, and especially emerging infectious diseases, that tends to be when things start to go really badly,” said Eric Lofgren, assistant professor in the Paul G. Allen School for Global Animal Health.
Now with support from a National Science Foundation grant specifically aimed at rapid response research, he and Assefaw Gebremedhin, associate professor in the School of Electrical Engineering and Computer Science, are seeking to better understand the relationship between hospitals and their communities in infection transmission with the hope of improving the resilience of healthcare systems in a pandemic.
Emerging disease outbreaks, whether the COVID-19 pandemic, the SARS or Middle East Respiratory Syndrome (MERS) epidemics of the early 2000s, or the Ebola outbreak of 2014, follow a different pattern than normal hospital infections, Lofgren says. Normally, healthcare-associated infections, such as MRSA or C. Difficile, only threaten the patients, affecting people with weakened immune systems who are already frail. But, in an emerging epidemic, health care workers are also very much at risk. So, the early, big transmission events often take place in health care settings.
As an outbreak occurs, hospital administrators are suddenly grappling with complicated management questions, and the interaction between the community and health care workers becomes a key factor in disease spread.
“We start having to think about how to protect patients and health care workers at the same time,” said Lofgren. “You get this amplifying step within hospitals, and it turns into an acute crisis pretty quickly, especially in traditionally under-resourced settings.”
The researchers will be using the $200,000 in grant funding to study how the interaction between hospital or nursing home facilities and the community affects the spread of infectious diseases.
“One of things we’re learning is that a lot of very granular decisions at the level of the individual hospital or nursing home end up having large impacts for the community as a whole,” he said. “We’re trying to understand that interaction, so we can understand at what level we need to think about those policies.”
To get a better picture of how sickness moves between communities and hospitals, the researchers are bringing together two computer models of disease transmission. One, developed at WSU, is a detailed model of transmissions within hospitals. Lofgren’s lab has developed very granular models of transmission that might involve 20 patients on a minute-by-minute basis in an intensive care unit. The other model that the researchers will use is one of the entire state of North Carolina that looks at how people get infected in their communities and end up at the hospital. Such models of large-scale populations tend to simplify the workings of hospitals.
“The computational question is how to combine these two models, so that we don’t have to keep track of 300 million people and what they’re doing every minute, but where you can make computational steps to increase that granularity,” he said. “We’ve been working on a single hospital or a single ICU, but we want to take a step back and take a look at how that interacts with the rest of the community.
“Bringing the two models together is interesting from a computational perspective and is a complicated space to deal with,” he added, “but it’s also interesting from a public health perspective.”
So, for instance, one scenario that the researchers would want to consider is when nurses can’t or won’t work in a very dangerous epidemic.
“We have to consider the risk of what they face against the care that they are trained to provide,” says Gebremedhin. “That is an equity issue that people struggle with all the time. It’s a question of more than just modeling and computation but other dimensions that really, really should be factored in.”
The researchers hope the work will someday lead to a more proactive rather than reactive response to epidemics.
“Are there simple things we can do, like how we schedule nurses or their interactions between patients, to make the hospital more resilient to large scale spread?” said Lofgren.
The researchers also hope their work leads to quicker answers for healthcare decision makers when they face a public health emergency.
“’We’ll have an answer for this in a couple years,’ isn’t going to do anybody any good,” Lofgren said. “It’s a very difficult computational challenge, and that’s not going to work in the middle of an epidemic.”
The world has changed dramatically in the past year, especially for an epidemiologist like Lofgren, who has studied infectious disease throughout his career. A year ago, when he said he was an epidemiologist, people often thought that he studied bugs or skin.
“Now everyone wants to talk about R naught (a mathematical term to indicate how contagious a disease is),” he said. “It’s been a strange experience.”
For his part, Gebremedhin said he is interested in looking at such a challenging real-world problem and at some of the hard-to-quantify factors that affect disease transmission.
“Part of my excitement is that when things are so real, affecting so many people, and there are so many levels of worry, how would incorporate that into your model?” he said. “That’s challenging, and I’m excited about what sorts of things we can do.”