Major disparities in all-cause mortality during first months of COVID-19 pandemic seen in new study

Deaths from all causes combined increased dramatically early in the COVID-19 pandemic for certain demographic groups in the U.S.—particularly for Black and Hispanic people, even when accounting for socioeconomic factors, according to new University of Michigan research.

In fact, the study—which focused on groups of adults under age 65—found the wealthiest non-Hispanic Black residents had a higher increase in mortality than the poorest non-Hispanic white people. Hispanic residents also tended to have larger mortality increases than those who are non-Hispanic, although the differences were smaller.

Co-authored by U-M Ross School of Business researcher Sarah Miller, Ph.D., the paper has been published online by the journal Health Affairs. Miller noted that although other studies have found racial and ethnic disparities in the impact of COVID-19 itself, the new paper is noteworthy both for the size of the gaps found in all reasons for dying, as well as for showing how the gaps transcend other effects.

“We had gone in thinking that social and economic factors, such as income and insurance coverage, were going to be the most dominant effects that we would see, and that these might mediate the effects of race and ethnicity,” said Miller, assistant professor of business economics and public policy. “But what we actually found was that race and ethnicity swamped the differences in incomes or insurance status. Uninsured white people had far lower increases in mortality than insured Black people.”

While the research did not directly study reasons behind these gaps, it notes past work has suggested factors like “racial segregation, living in poor neighborhoods, environmental exposures, lack of access to high-quality health care and chronic exposure to racial discrimination” as possible explanations.

In another key finding, Miller and colleagues found a huge spike in mortality among residents of residential health care facilities, even though they focused on adults younger than 65. The increase was more than 100 times larger than the increase among those not living in group settings. People living in correctional facilities also experienced large increases in mortality—more than five times the increase for those not living in group quarters.

Overall, people without health insurance, those with incomes below the poverty line, and those in occupations with limited work-from-home options also experienced significant increases in mortality, the researchers found.

Taken together, the results show groups at an economic or social disadvantage experienced large mortality increases during the early months of the pandemic. They argue that this pattern demonstrates the need for targeting COVID-19 relief efforts to these populations most likely to be affected.

“To me, these results suggest that extra measures to target these communities—to improve vaccination rates or to provide more support to help bridge the gap to a post-COVID world—would be appropriate, given that they’ve been hit especially hard,” Miller said. “Ours is not the only paper to show that, but this research definitely shows that it cuts across socioeconomic factors in a way that’s very striking.”

Unlike past studies, which tended to rely on data from smaller samples, the new paper analyzes national numbers from the U.S. Census Bureau. The data provides a more detailed and in-depth look at how different factors such as ethnicity, occupation, living situation, income and insurance status can interact to affect death rates. The researchers studied people aged 18-65, comparing mortality figures from the second quarter of 2020—April-June, when the pandemic’s first wave was in high gear—to the second quarter of 2019.

The researchers found significant increases in mortality among all the subgroups they studied, but the sizes of the increase varied considerably. Among the findings:

Across all socioeconomic groups, non-Hispanic Black adults under 65 experienced significantly higher increases in mortality than non-Hispanic white people in the same age range. For example, non-Hispanic Black people in the highest income group studied had an increase in mortality more than 3.5 times larger than the increase in mortality experienced by the poorest non-Hispanic white people (46.5 deaths per 100,000 people compared to 12.9).

Hispanic residents tended to see significantly higher mortality increases than non-Hispanic white residents. This held true among people with insurance, those not living in group quarters, lower-income families, people with work-from-home options, and those in essential industries.
People under 65 living in nursing facilities or other health group quarters experienced an increase in mortality of 1,619 deaths per 100,000 residents. Those in correctional facilities experienced an increase of 83.3 deaths per 100,000. This compared to an increase of 16.1 deaths per 100,000 among people not living in group quarters.

Uninsured people had an increase of 27.6 deaths per 100,000, compared to an increase of 15.5 deaths among the insured.

People with family income at or below the poverty line saw an increase in mortality of 27.1 deaths per 100,000, compared to 13.3 among those with incomes four times the poverty level or higher.

Mortality rates increased by 19.1 deaths per 100,000 among people in occupations without work-from-home options, compared to 10.9 among those able to work from home.

Among specific occupations, mortality increases were highest among installation, maintenance and repair workers; production workers; legal professions; and transportation workers. In contrast to previous studies, the researchers did not find exceptionally high death rates among health care practitioners or food-service workers.

Miller said the new results underline questions about long-established health inequities in the U.S.

“Even before the pandemic, people in low-income households, for example, had really high mortality rates, and then the pandemic made this worse,” she said. “If we had smaller disparities to begin with, if or when there’s another pandemic, we might not see such stark differences in its impact. Why do poor people do so poorly in terms of their health in the United States, and is there anything that policy can do to improve on that?”

Miller’s co-authors are Laura Wherry, assistant professor of economics and public service at New York University’s Wagner Graduate School of Public Service, and Bhashkar Mazumder, senior economist in the Federal Reserve Bank of Chicago’s Economic Research Department.