The full picture of the racial impact of the Covid-19 pandemic is clouded by systemic failures to collect race and ethnicity data, even in states that are leaders in promoting health equity.
In California, for example, a key benchmark for reopening and allocating vaccines doesn’t fully incorporate race and ethnicity data. Meanwhile, nationally there remain glaring absences in testing and hospitalization data by race and ethnicity a full year after it was shown Covid-19 had a disproportionate effect on Black, Latino and Native American communities.
Black Americans have the highest Covid-19 death rate nationally and the Centers for Disease Control and Prevention (CDC) estimates that Native Americans, Latinos and Black people are two times more likely than white people to die of the disease.
The nation’s top infectious disease expert, Dr Anthony Fauci, said in May that as society returns to “some form of normality”, people should remember that “the undeniable effects of racism in our society” has created unacceptable disparities in Covid-19 cases, hospitalizations and deaths.
“Covid-19 has shone a bright light on our own society’s failings,” Fauci said during a graduation ceremony in Atlanta.
Several public health experts who focus on health disparities told the Guardian gaps in reporting on race and ethnicity during the pandemic, and more broadly in health, have limited the deployment of resources in vulnerable communities and masked the true scope of Covid-19’s toll.
The Cares Act requires people administering Covid-19 tests to collect data on race, but the federal rule is not enforced. As of 14 July, race and ethnicity data was not known for 40% of US Covid-19 cases, according to the CDC.
Nationwide, race and ethnicity data was missing for 32% of fully vaccinated people as of 14 July, according to the CDC.
In California, a national leader in promoting health equity, race and ethnicity data is missing from a standard it uses to set benchmarks for reopening. California reserved 40% of its vaccine supply for census tracts in the lowest quartile of this standard, known as the Healthy Places Index (HPI).
The HPI brings together data on 25 factors such as income and education to show which areas are most socially and economically vulnerable during the pandemic. But race and ethnicity data are separated from the combined factors in a “complementary” data set, a legacy of the state ban on affirmative action.
A critic of the index, University of California Los Angeles professor Dr Vickie Mays, said that requiring additional layers of data analysis to understand where race and ethnicity fit in the index creates dangerous delays in the government’s response.
“Those populations are undiscovered until somebody does additional analyses,” Mays said. “That is what inequity looks like.”
In a statement, California’s department of public health did not dispute the absence of race and ethnicity data in the HPI standard.
The agency said it “is committed to prioritizing equity” and highlighted its equity dashboard, which shows Covid-19 data broken down by race and ethnicity. It also includes case data based on income and access to health insurance and notes that those issues, which can increase the risk of poor health outcomes, are “often the result of structural racism”.
The Public Health Alliance of Southern California developed the HPI and repeatedly highlights on its website and in briefs the important role race and ethnicity play in shaping health outcomes and in Covid-19.
The website also notes that race and ethnicity data are separate because of a 1996 California ban on affirmative action, known as Prop 209. California is one of eight states that blocks the government from giving jobs, educational opportunities or contracts based on race. In November, California voters rejected a measure that would repeal the ban on affirmative action.
“In order to ensure the Healthy Places Index can be used in a variety of applications, including informing policy decisions, the Alliance has made additional, informational layers about Race/Ethnicity available on our online map,” the HPI website said.
More integrated data sets for California do exist.
Mays and other UCLA researchers published a predictive model for Los Angeles county in November that includes definitive information on which racial and ethnic groups are most vulnerable. Like the HPI, it also includes other socio-economic factors, such as healthcare access and environmental risk.
Los Angeles county was the center of the US outbreak in the winter and the model accurately predicted that the most vulnerable communities, and therefore those that should be prioritized for vaccinations, were in areas where there are large groups of racial and ethnic minorities, low-income households and unmet social needs.
The model was endorsed by Dr Edward Sondik, former director of the CDC’s National Center for Health Statistics. Sondik said if this model was replicated across the US, “these insights could be extremely helpful for bringing the pandemic under control”.