UCLA Health hospital data reveals COVID-19 risk factors specific to Latino population
The project demonstrates the power of UCLA’s Data Discovery Repository for identifying trends.
A new UCLA Health study shows that patients of Hispanic/Latino ethnicity who contracted COVID-19 had more severe disease and a higher rate of hospitalization than non-Hispanic/Latino whites.
The study identifies certain medical risk factors that predict a more severe course of disease specifically among the Hispanic/Latino population, and also suggests that disparities in medical care cause members of that population to come to the hospital later in the course of disease.
Finally, the study demonstrates the power of the UCLA Data Discovery Repository (DDR), a new resource developed by the UCLA Health Informatics Team and the Institute for Precision Health for analyzing trends from collections of medical records.
A report on the study, from the UCLA Precision Health Data Discovery Repository Working Group, was published in the March 19 edition of iScience.
When the pandemic began to spread across the United States in 2020, doctors and researchers scrambled to identify patterns among those sickened by the virus. Some people became very ill, others never showed a symptom.
UCLA neurologists Timothy Chang, MD, PhD, and Daniel Geschwind, MD, PhD, turned their expertise in precision medicine to the question, in conjunction with their colleagues Manish Butte MD, PhD, Pediatric Immunology and Bogdan Pasaniuc, PhD, Computational Medicine, who co-led the study with Dr. Chang.
The Data Discovery Repository is a collection of patient health records with the identifying information removed. It includes medical information such as lab test results and vital signs, but no personal information that could link that data back to an individual.
From nearly 2,000 patient records from people who tested positive for COVID-19, the researchers successfully extracted patterns of disease severity and risk factor among different ethnic groups.
“COVID-19 is an example of something where being able to do the research quickly is important,” explained Dr. Geschwind. “Because we’ve built this computational infrastructure that allows us to look at de-identified patient records, it enabled Tim and colleagues to go in immediately and do the study without having to create everything from scratch.”
Because 20% of patients served by UCLA Health hospitals self-identify as Hispanic/Latino, the researchers had access to a large sample size from which to draw correlations. They found that while some risk factors contributed to COVID-19 disease severity regardless of ethnic group, certain pre-existing conditions posed a particular risk to Hispanic/Latino patients.
One such factor is a heart condition called mitral valve disease, which affects the way blood flows through the chambers of the heart.
“The DDR lets us identify these risk factors that you wouldn’t think about,” Dr. Chang said. “We found things like this mitral valve disorder which none of us really expected, nor can we even explain now. We need to do additional studies to see why that’s linked to increased hospitalizations.”
In other words, doctors don’t have to take a guess at what might be a relevant risk factor and then go looking for it. The DDR helps uncover these relationships in an unbiased, data-driven manner.
In addition to finding risk factors, the analysis also revealed a telling indicator of systemic health care disparities in the community.
“Latino patients are coming in sicker, and the reason we know that is that when you look at their laboratory tests, they have more inflammatory markers than the non-Hispanic/Latino patients who are COVID-positive,” Dr. Geschwind said.
He points out that patients who get treated later, after the disease has gotten worse, are more likely to be hospitalized and less likely to survive. Doctors can act on these results by making a particular effort to get their Hispanic/Latino patients who test positive for COVID-19 into treatment sooner.
In the future, the DDR could help identify risk factors for other health conditions that specifically affect people of certain ethnicities. Knowing about such risk factors can help target public health messaging around prevention and awareness.
Learn more about Health Equity, Diversity and Inclusion at UCLA Health.
Caroline Seydel is the author of this article.