Cortisol is a biological hormone that’s elicited during certain physiological responses, most notably for acute and chronic stress. In some individuals, it can impact decision-making and behavioral outcomes, whereas others are more resilient to its effects.

Two University of Minnesota Medical School faculty, Alexander Herman, MD, PhD, assistant professor in the Department of Psychiatry and Behavioral Sciences, and David Darrow, MD, MPH, assistant professor in the Department of Neurosurgery, are studying hair samples from frontline healthcare workers during the COVID-19 pandemic to determine how their cortisol levels might correlate with their responses on a cognitive assessment. The project was originally motivated from an idea in the lab proposed by a medical student, Erika Kaske, and funded by a CO:VID (Collaborative Outcomes: Visionary Innovation & Discovery) grant.

The COVID-19 pandemic has forced healthcare workers to endure a uniquely stressful combination of social isolation and quarantine along with exposure to the direct effects of the virus at work. Initially, with the advent of COVID-19, Dr. Darrow asked, “What can we do to learn additional information about how the pandemic is affecting particular healthcare workers?” 

Drs. Herman and Darrow are analyzing across populations and cross-sections of people who have gone through stressful situations, rather than how an individual changes over time.

“One of our interests is being able to do these complex behavioral tasks remotely,” Dr. Darrow said. “A lot of important psychiatric work is longitudinal, so we need to be able to do our work at a distance after they leave. A large emphasis of our lab is taking tasks and models from computational psychiatry and studying the fine-grained neurophysiology, taking tasks out and moving them to a population level.”

Both researchers maintain an interest in the “explore versus exploit” tradeoff in human behavior, so they are testing healthcare worker’s cognitive abilities through a “multi-armed bandit task.” These tasks typically offer a limited number of resources that must be allocated between competing choices to maximize a reward. Imagine someone standing in front of four casino slot machines having only $20 and trying to decide which ones to play first or at all. Do you stick with one machine until you win? Or, do you explore all four to potentially increase your odds of winning? Mathematical models of these tasks help understand how humans and animals make decisions about how they pursue rewards in their environments. There is already literature on induced stress, but this study will expand that knowledge by capturing how the pandemic affects people's motivation levels and cognitive exertion. 

“Some people are more resilient than others, which is shown through psychological functioning and task analysis. So, the next question is, why? What enables them to do that? And that’s what we want to engage with. New therapy tools and strategies for people to improve their ability to be resilient,” Dr. Herman said.

The researchers will measure patterns of decision-making during each trial and then compile the data. The final results will be a quantitative assessment of cognition and will have potential to translate into clinical utility in the future. 

“Our first goal is understanding. The results dictate if certain psychotherapy tools might augment important processes for those who show a high level of resilience,” Dr. Herman said.

“We’re excited about the potential of this. It could show a change in the product of all the individual decisions being made —that they are becoming more stress-based,” Dr. Darrow said.

They expect to see some kind of distribution of cortisol but are still hypothesizing about what that could mean. For now, they’re hoping to establish a strong cross-section of data with assessments for each data set. If the duration of the COVID-19 pandemic is prolonged, then the study might be replicable in the future. They can also study healthcare workers when daily work is “back to normal” and compare and contrast data averages or fluctuations during and after the pandemic.