A medical student at the University of Kansas discusses a research project she started in her first year of school—2020—looking at whether a racial bias in pulse oximetry was present in patients with acute hypoxemic respiratory failure.
Julia Balmaceda began her medical education at the University of Kansas Medical Center in 2020. In this interview, she describes the backdrop for the abstract that was published at CHEST 2022 that looked at the direct arterial blood gas and pulse oximeter readings of White and Black patients experiencing acute hypoxemic respiratory failure in the intensive care unit.
What prompted you and your co-authors to do this study?
So as a med student, we’re allowed to do some research. And at the time, when I started this, it was my first year of medical school, so that would have been in 2020. And during that time, obviously, it was very tumultuous for the entire country. COVID-19 was going on, you know, the tragedies of George Floyd and Breonna Taylor had occurred. And so, it was really striking home with a lot of us in the health care system. You know, we're definitely not isolated from things that are going on socially in the country. So, as a med student, I had to ask myself, what can I do? And one of the things that we could do is do research.
So, I decided to really focus my research on health care disparities. Part of our med school curriculum is talking about the historical and structural racism that occurs in health care. And we learned about racism that's built into things like osteoporosis risk calculations and things like that. And so pulse oximetry, it's not like a new discovery that there's a race-based bias built into it. But honestly, it was mainly me trying to see what's going on at the institution that I'm going to school at...to see what was happening in the [intensive care unit] here at KU. So that's kind of where it stemmed in. And looking at the results now, we see that we reproduced something that other institutions have seen. And thankfully, we have, I think, a larger data set that we're going to look at, and hopefully that'll help us elucidate some of the more practical implications of what this bias is.