In the COVID-19 Pandemic Response, the U.S. Has a Health Disparities Data Problem

One year into the COVID-19 pandemic, the Biden Administration doesn’t have access to the data needed to develop a pandemic response that is equitable across intersections of race, ethnicity, socioeconomic status, age, and geographic location. 

Dr. Marcella Nunez-Smith, head of President Biden’s COVID-19 Health Equity Task Force, has acknowledged the need for better health disparities data related to testing, tracing, quarantine/isolation, treatment, and vaccination. Substantial investment in our public health data infrastructure is needed, yet was largely absent from the recently passed American Rescue Plan. Fixing this data problem is critical to successfully responding to the COVID-19 pandemic, and to proactively addressing future health threats.  

Why quality data are key to addressing the COVID-19 pandemic

Health disparities in the U.S. are not new. Neither are the barriers to accessing high-quality data. COVID-19 has exacerbated long-standing racial, ethnic, socioeconomic, and geographic disparities in death, disability, and access to care. Yet, we are likely underestimating the magnitude of the disparities that have emerged in the wake of the pandemic. This is because much of what we know is from publicly available sources with insufficient or inconsistent breakdowns by race, ethnicity, socioeconomic status, age, and geographic location. Existing publicly available data are designed for public consumption or to conduct simple statistics, but are insufficient for rigorous demographic analyses of health disparities, which often require record-level data. Even the more comprehensive disparities analyses conducted to date suffer from severe limitations. For example, The New York Times’ comprehensive report on racial inequality amidst coronavirus included a disclaimer that over 45% of county-level data were missing from their analysis.

We need accurate, complete, and real-time data to tailor pandemic response policies, resources, and funding to the communities that need them most. Without that data,  it was difficult to quickly identify populations that were disproportionately contracting and dying from COVID-19 early in the pandemic. History is now repeating itself with egregious inequities in vaccine roll-out (e.g. Florida,  New York and Philadelphia). Population health scientists know that people with more money, time, power, social connections, prestige, and freedom tend to be the first to learn about new health threats, and first in line to get help. Without quality health disparities data, these inequities can worsen long-standing mistrust of the government and of healthcare that already exists in marginalized communities, making it even harder to respond to emerging health threats over time.

 Three data problems that need fixing

Our research team at the Carolina Population Center at the University of North Carolina at Chapel Hill has been trying to collect the data needed to rigorously analyze racial, ethnic, socioeconomic, and geographic disparities in COVID-19 testing and deaths across the US since June 2020. State-to-state variation in data systems and a lack of national coordination have stymied our efforts. With over 60,000 new COVID-19 cases per day and the implementation of vaccination programs implemented across the U.S., many states don’t have the capacity to help us get the data we need

Problem 1: State-level differences in data processes: Navigating over 50 different systems, each with its own red tape, has made rigorous, national-level data analysis nearly impossible. Even finding contact information or forms to submit a data request within state health websites is a major barrier. Calling health departments often means getting transferred to various people, only to leave a voice message that isn’t returned. When the appropriate contact is found and requests are finally submitted, underfunded and understaffed state health departments are often unable to dedicate resources to respond to data requests. Some departments noted that they could not provide any data beyond what was available on their state dashboards, while others reported that they would be unable to provide 2020 data until late 2021.

Problem 2: Poor national-level coordination: Early in the pandemic, there was no national coordination of COVID-19 data. Dr. Fauci even acknowledged the need for “collaboration, cooperation and synergy between the federal government and the states.” As such, the quality of data and reporting requirements varied across hospitals, counties, and states.  Ethnicity and race were frequently left out of COVID-19 death data, meaning that racial/ethnic disparities in COVID-19 deaths are likely larger than have been reported. This makes it difficult to compare “apples to apples” and identify which areas were struggling the most. This will hopefully get better under Biden Administration, given this executive order. But these improvements will take some time.

 Problem 3: The under-appreciated National Center for Health Statistics: The CDC’s National Center for Health Statistics would be the best home for this type of population-level of data, but decades of under-investment in this center have made it difficult for over-burdened staff to quickly compile the granular data necessary for sophisticated and high-quality analyses. Much of their data are incomplete and often delayed due to state-level reporting differences. As a result, no studies to date have analyzed COVID-19 mortality data with the rigorous methods needed to paint a full picture of the disparities at the national level; the granular data needed are too difficult to obtain.

How to fix the problems

As health disparities researchers, we would like to see a coordinated, national effort around accurate and real-time data in order to tailor policies, resources, and funding to the communities that need them the most. Local health departments and national agencies must standardize the types of COVID-19 data collected so that all cases and deaths include data on race, ethnicity, age, county, and sex or gender.

 If we seize this opportunity to improve our data systems for COVID-19, we can also improve access to data for other public health issues. Investing in our data systems now will facilitate a more targeted response to current and future public health crises, with  less waste of time, money, and other limited resources. In the long run, we will also improve the U.S.’s ability to keep a comprehensive pulse of the health of our population, which has larger economic and security implications for our country and the world.