Cambridge, Mass., June 10, 2020—Why does the coronavirus kill some Americans, while leaving others relatively unscathed?
A new study by researchers at the MIT Sloan School of Management sheds light on that question. The study, by Christopher R. Knittel, the George P. Shultz Professor of Applied Economics at MIT Sloan and Bora Ozaltun, a Graduate Research Assistant in the Center for Energy and Environmental Policy Research (CEEPR) lab, correlates Covid-19 death ratesin the U.S. states with a variety of factors, including patients’ race, age, health and socioeconomic status, as well as their local climate, exposure to air pollution, and commuting patterns.
The findings have important implications for determining who is most at risk of dying from the virus and for how policymakers respond to the pandemic.
Using linear regression and negative binomial mixed models, the researchers analyzed daily county-level Covid-19 death rates from April 4 to May 27 of this year. Similar to prior studies, they found that African Americans and elderly people are more likely to die from the infection relative to Caucasians and people under the age of 65. Importantly, they did not find any correlation between obesity rates, ICU beds per capita, or poverty rates.
“Identifying these relationships is key to helping leaders understand both what’s causing the correlation and also how to formulate policies that address it,” says Prof. Knittel.
“Why, for instance, are African Americans more likely to die from the virus than other races? Our study controls for patients’ income, weight, diabetic status, and whether or not they’re smokers. So, whatever is causing this correlation, it’s none of those things. We must examine other possibilities, such as systemic racism that impacts African Americans’ quality of insurance, hospitals, and healthcare, or other underlying health conditions that are not in the model, and then urge policymakers to look at other ways to solve the problem.”
The study, which has been released as a Center for Energy and Environmental Policy working paper and is in the process of being released as a working paper on medRxiv, a preprint server for health sciences, contains additional insights about what does, and does not, correlate with Covid-19 death rates. For instance, the researchers did not find a correlation between exposure to air pollution. This finding contradicts earlier studies that indicated that coronavirus patients living in areas with high levels of air pollution before the pandemic were more likely to die from the infection than patients in cleaner parts of the country.
According to Prof. Knittel, the “statistical significance of air pollution and mortality from Covid-19 is likely spurious.”
The researchers did, however, find that patients who commute via public transportation are more likely to die from the disease relative to those who telecommute. They also find that a higher share of people not working at all, and thus not commuting, have higher death rates.
“The sheer magnitude of the correlation between public transit and mortality is huge, and at this point, we can only speculate on the reasons it increases vulnerability to experiencing the most severe Covid-19 outcomes,” says Prof. Knittel. “But at a time when many U.S. states are reopening and employees are heading back to work, thereby increasing ridership on public transportation, it is critical that public health officials zero in on the reason.”
The proportion of Americans who have died from Covid-19 varies dramatically from state to state. The statistical models that Knittel and Ozaltun created yield estimates of the relative death rates across states, after controlling for all of the factors in their model. Death rates in the Northeast are substantially higher compared to other states. Death rates are also significantly higher in Michigan, Louisiana, Iowa, Indiana, and Colorado. California’s death rate is the lowest across all states.
Curiously, the study found that patients who live in U.S. counties with higher home values, higher summer temperatures, and lower winter temperatures are more likely to die from the illness than patients in counties with lower home values, cooler summer weather, and warmer winter weather. This implies that social distancing policies will continue to be necessary in places with hotter summers and colder winters, according to the researchers.
“Some of these correlations are baffling and deserve further study, but regardless, our findings can help guide policymakers through this challenging time,” says Ozaltun. “It’s clear that there are important and statistically significant difference in death rates across states. We need to investigate what’s driving those differences and see if we can understand how we might do things differently.”
Further Reading: CEEPR WP 2020-009
About the Authors:
|Christopher R. Knittel is the George P. Shultz Professor of Applied Economics at the Sloan School of Management, Director of the Center for Energy and Environmental Policy Research, and Co-Director of the MITEI Low-Carbon Energy Center for Electric Power Systems Research at the Massachusetts Institute of Technology. He joined the faculty at MIT in 2011, having taught previously at UC Davis and Boston University. Professor Knittel received his B.A. in economics and political science from the California State University, Stanislaus in 1994 (summa cum laude), an M.A. in economics from UC Davis in 1996, and a Ph.D. in economics from UC Berkeley in 1999. His research focuses on environmental economics, industrial organization, and applied econometrics.|
|Bora Ozaltun recently earned his M.S. in the Technology and Policy Program at MIT and worked as a Graduate Research Assistant for Professor Knittel at the Center for Energy and Environmental Policy Research.|
This article originally appeared on the MIT Sloan Press Room, linked here: https://mitsloan.mit.edu/press/research-mit-sloan-explores-correlations-coronavirus-death-rates-a-variety-factors-including-patients-race-age-socioeconomic-status-and-local-climate