The inequitable distribution of COVID-19’s terrible burden has been well documented. There are notable disparities by race and ethnicity. COVID-19 rates, including incidence and fatality rates, are higher among Black, Indigenous, and other Persons of Color (BIPOC). This is structural racism at its worst, in which BIPOC individuals are disproportionately represented among essential workers and have higher rates of underlying health conditions, and thus suffer extraordinary health consequences during a global pandemic.
COVID-19 has also deeply affected rural residents, who have more underlying risk factors for COVID-19 mortality. Rural residents are older, with greater numbers of underlying health conditions, and more limited access to care. This includes greater distances to care as rural hospitals close and ongoing barriers to accessing primary care, as well as much more limited access to specialty care, ICU beds, and ventilators. Too often we describe these issues in isolation, falsely implying that health risks of racism and rurality are mutually exclusive. Yet, the reach of structural racism and its impact on health outcomes extends to rural places.
Lack of data on the intersection of race and rurality
Throughout the pandemic, there has been limited attention to the intersection of racism and rurality in COVID-19 risk. This opaqueness is related, in part, to the availability of data. Many federal and state websites release data by race and ethnicity and by geography. But, data are not commonly available across both dimensions.
For example, the Centers for Disease Control and Prevention (CDC) releases data on provisional COVID-19 death rates by race, ethnicity, and county. But, the publicly available data only includes counties with more than 100 deaths. Across the U.S., 1,335 counties are classified as rural, non-core [pdf] in the 2013 National Center for Health Statistics Urban-Rural County Classification Scheme (a widely-used measure of rurality). These are less densely populated US counties, where there is not a town with more than 10,000 residents. This is also the most common type of county, making up more than 42% of all counties across the US. As of March 17, 2021, CDC data on COVID-19 deaths by race and county were available for only six rural, non-core counties. That is, nationally, COVID-19 fatality data by race and ethnicity are available in less than one-half of one percent of rural, non-core counties.
Race and rurality data at the state level
As rural health researchers focused on health equity, we care deeply about COVID-19 risk among rural BIPOC residents. To understand the intersection of race and rurality here in Minnesota, we had to request restricted data from the state. However, everyone from clinical health care delivery systems to non-profit and community-based organizations to local public health should have access to the data needed to appropriately target programs and resources to reach those most vulnerable to COVID-19. In this blog, we present such data and give examples of how they could be used to address the dual inequities caused by racism and structural urbanism. Structural urbanism is a bias toward large population centers that disadvantages rural residents in terms of resources, research, and information.
Through a special request, we were able to obtain restricted data from the Minnesota Department of Health (MDH) on numbers of COVID cases and fatalities by racial and ethnic group within each of Minnesota’s 87 counties, from the onset of the pandemic through February 10, 2021. We merged these data with Urban Influence Codes (another common measure of rurality) to classify each county as metropolitan (urban) or non-metropolitan (rural). Then, we merged on 2014-2018 American Community Survey data to obtain the total county population for each racial and ethnic group. We matched these groupings to the classifications used by MDH to examine county-level COVID-19 case and fatality rates for Black, Indigenous, Asian, Hispanic, and white residents of each county.
Methodological challenges for small populations
One challenge in examining these data for counties with relatively small populations was that MDH marks case and death numbers fewer than 5 as “<5” in the data to prevent disclosure of small cell sizes. To analyze those, we converted instances of <5 to 4. We also conducted sensitivity analyses converting those to 1 (instead of 4), leading us to largely the same conclusions. However, this issue highlights the challenges posed by conducting analysis among small populations or in sparsely populated areas. A tension exists between data privacy and the urgent need to highlight inequities, especially for smaller populations.
Highest COVID-19 burden in Minnesota
Figure 1: Minnesota County-Level COVID-19 Case Rates by Race, Ethnicity, and Rurality (click for full-size figure)
The results of our analyses highlight two clear findings. BIPOC residents have faced disproportionate burden from COVID-19 in Minnesota. And, rural residents have faced a greater burden than urban residents within each racial and ethnic group. For example, the COVID-19 case rate was highest for rural Asian adults (18%), followed by rural Black and rural Indigenous people (14.5% and 14.3%), respectively (Figure 1).
The story was largely the same for county-level COVID-19 case fatality rates. Again, in every group, the case fatality rate was higher in rural counties than in urban counties (Figure 2). These data show extremely high fatality rates among rural Indigenous people (8.4%), followed by rural Asian people (2.8%) in Minnesota.
Figure 2: Minnesota County-Level COVID-19 Case Fatality Rates by Race, Ethnicity, and Rurality (click for full-size figure)
Our findings are limited to one state (Minnesota). Still, they clearly highlight the dual disparity caused by racism and structural urbanism, in which BIPOC rural residents bear the greatest burden from COVID-19. Other research indicates that the story is similar across the U.S., and also that these findings are not isolated to the current pandemic. The reach of structural racism into rural areas is long and its deleterious impact on health is lasting.
Going forward: data transparency and antiracist policies
The essential contribution of granular data is to point to the specific place where the health burden is greatest. The necessary next step is to target prevention and treatment efforts to mitigate this burden. We should do this by building equitable systems and enacting antiracist policies to sustainably improve population health and reduce disparities by race and geography.
More data and public reporting on the intersection of race and geography should inform such efforts. We were able to access these data through a request to MDH. But, this information should be readily available for community and policy leaders across the country to assess who is at greatest risk of COVID-19 and other health outcomes. Leaders urgently need this information to prioritize policy efforts.
Making data available will require resources and attention to data privacy, balancing a need for evidence-based and equity-minded policy. States may consider legislative authorization for appropriate data sharing with protection built-in for individual privacy. At a minimum, releasing data on health outcomes by race and rurality would begin to show where disparities are deepest. Such data requirements must be accompanied by sufficient federal or state funding for data collection, infrastructure, and analysis.
Data on their own are not enough, however. The inequities faced by rural BIPOC individuals are already clear. As we rebuild from the pandemic, including current efforts to administer the vaccine and longer-term efforts to bolster our economy and health care system, rural BIPOC residents should be prioritized. This includes vaccine administration efforts that reach hard-to-reach rural residents who may otherwise have transportation barriers, limited access to broadband internet to find and make appointments, or who may have both historical and contemporary experiences with health care that have caused harm.
Need for immediate and long-term actions
The data we analyzed in Minnesota indicate a particular need for attention to rural Indigenous and Asian communities. Successful, swift vaccination efforts in tribal communities in rural Minnesota offer an instructive example. These efforts show how resources and decision-making authority in the hands of local (sovereign) leaders can rapidly produce positive change. In Minnesota, Mahnomen and Cook counties, the traditional and reservation lands of the White Earth Ojibwe and Grand Portage Band of Lake Superior Chippewa, reached the highest adult vaccination rates in the state rather quickly, owing to extraordinary efforts by these tribes partnering with county public health departments. Still, prioritization of rural Asian Minnesotans has not been a focus of statewide efforts or of data reporting.
Attention to public health infrastructure is essential to support health equity across race and place. This requires more robust and granular data reporting, including in the current vaccination phase of the pandemic. The lessons learned through this pandemic have come at a terrible cost in terms of human lives and suffering. Moving forward, we need data-informed efforts to recover and rebuild the nation’s public health, no matter one’s race or location.
Carrie Henning-Smith
Latest posts by Carrie Henning-Smith (see all)
Latest posts by Mariana Tuttle (see all)
Katy Kozhimannil