A year into the coronavirus pandemic in the United States, we still lack a complete understanding of who is getting sick, and where, and when. Demographic data from many states are astonishingly incomplete, and even widely collected information, such as the age of patients at the time of diagnosis or death, is so inconsistently presented that it has been impossible to assemble into a clear national picture. The federal government is now making more demographic data available, but the information continues to emerge at a snail’s pace.
This has left government outsiders to try to assemble the data—groups like us, the COVID Tracking Project, which is housed at The Atlantic. For more than nine months, we’ve compiled data from states to create a composite national picture of the pandemic. Time and again, we have seen that a lack of federal support has left overburdened state public-health authorities to fend for themselves, resulting in incomplete reporting, incompatible data definitions, and inconsistent data pipelines.
With vaccine data, the United States has the opportunity for a do-over. The national vaccination effort itself is fragmented and inconsistent, guided by state and county policies in the absence of a comprehensive federal system of support—but the data about vaccinations need not mirror this incoherence. Tracking the distribution of vaccines and the pace of vaccination can provide meaningful insights into the volume of future cases, hospitalizations, and deaths. And particularly given the well-established racial and ethnic disparities we see in COVID-19 cases and deaths, we must have access to data that would reveal whether these disparities are being remediated—or intensified—by our national vaccination effort.
As the CDC states, “A strong, nationally coordinated approach is critical to collect, track, and analyze vaccination data, especially in the early phases of vaccine administration.” The agency requires the uniform and timely reporting of complete demographic data about each vaccination. Yet almost a month after the first doses of COVID-19 vaccine were administered in the U.S., we have very little public information on who is being vaccinated. The CDC is publishing data on vaccine distribution and first doses administered—which is a good starting point —but it has not yet released any demographic data.
The vaccine data that individual states are publishing replicate the patchwork nature of the other state-level COVID-19 data our teams have been compiling since March. Our researchers have found 17 states currently reporting some data about the race and ethnicity of the people receiving vaccines. As with other demographic information about COVID-19, in the absence of public reporting standards, states have chosen to report data in widely varying ways. Several states combine respondents who describe themselves as multiracial and as “other”; Massachusetts combines four categories into one; Mississippi identifies respondents only as Black, white, or Asian; Oregon reports people in every category they choose, meaning that vaccine recipients who identify themselves in more than one category are counted more than once. These inconsistencies make accurate comparisons among states—or any national understanding of the vaccination campaign—impossible.
Without complete, public federal public-health data or uniform standards for state-level reporting, we will continue to have an incomplete and imprecise understanding of which Americans are getting vaccinated, and where, and when. Without this information, it will be impossible to tell whether vaccination efforts throughout the country are achieving their stated aims—and it will be impossible to identify and fix efforts that aren’t. No amount of nongovernmental effort can remedy this problem. Federal public-health agencies must publish all non–personally identifiable vaccine data—not in two or three or six months, but now.
Charlotte Minsky and Kara Schechtman of the COVID Tracking Project contributed research and reporting.