A Data-Driven Approach Can Enhance COVID-19 Vaccine Rollout
Detailed data of vaccine distribution in near real time could have improved the government’s ability to identify hotspots and reach inaccessible populations
We are reaching a plateau in vaccine administration much earlier than anticipated1. As of May 18, 2021, there are 112.7 million Americans fully vaccinated, which still leaves over 62% of Americans requiring at least one more COVID-19 vaccine dose. The vaccination rollout could have more effectively utilized prior data, and collected new data, to better inform distribution strategies and aid in this “last leg” of vaccination – identifying and reaching out to the unvaccinated.
It has been made clear throughout the COVID-19 pandemic that the U.S. public health data infrastructure is lacking at best. Since COVID-19 was first detected in the United States in January 2020, testing, positivity, death, and hospitalization data have been nebulous, missing, and/or incompatible between states. The vaccination rollout that began in December 2020 provided an opportunity for the U.S. to learn from those prior missteps and pursue clear, uniform data collection standards. Unfortunately, that has not been the case.
In defense of the CDC and public health and political leaders that developed and implemented vaccination strategies, the pandemic, vaccine development, and vaccine deployment have progressed at an unprecedented pace, making it difficult to plan. That is not to say that we could not and did not prepare for this. In 2018 the CDC’s Advisory Committee on Immunization Practices (ACIP) released Allocating and Targeting Pandemic Influenza Vaccine During an Influenza Pandemic2. This document laid out the plans for distribution of a new vaccine, including who should be prioritized, during a novel influenza pandemic.
While COVID-19 is not the flu, these guidelines are an excellent framework for vaccine preparedness and the establishment of data infrastructure. Additionally, during the COVID-19 pandemic, multiple non-government organizations including the Johns Hopkins Bloomberg School of Public Health3, produced guidelines for how to ethically and effectively distribute a new COVID-19 vaccine. Those detailed plans offered many contingencies for what were unknowns at the time, including who was most affected by COVID-19, how many vaccines would be available, and how many doses would be required per person.
The foundation for an extraordinary vaccine rollout existed, so why are we now struggling to vaccinate the remainder of the country, and what could data do to help us?
Given that you are currently reading this on a device that is connected to the internet, it may be difficult to conceive of the significant population of Americans that cannot access a computer due to either financial or personal reasons, do not have broadband internet connectivity due to their rural location, or do not have the technological literacy necessary to utilize vaccine registration systems. Even if people can access a computer, they may not have a personal vehicle or reliable public transportation to get to a vaccination appointment. These obstacles unequally impact Americans experiencing financial hardship, which is particularly disconcerting when it has been shown that COVID-19 has disproportionately affected those in poverty4.
While President Biden proposes to address broadband internet access as part of his administration’s American Jobs Plan5, 23% of Americans do not currently have access to broadband internet at home as of February 20216, which is a necessity for vaccine registration in almost every state. Pockets of people without broadband could be generally located through public data on regional internet access, as well as individual computer ownership7. Computer literacy is more difficult to gauge with data, although census data regarding age could at least identify the locations of populations more likely to have difficulties using computers. With this information, states could have designed vaccine registration systems accessible by phone and/or created mobile vaccine registration units to help people in areas with no broadband access register for vaccination. Data on public transportation and personal vehicle registration from state departments of motor vehicles could have helped plan vaccination site locations and mobile vaccine unit deployment to reach this impoverished, inaccessible population that would have been less likely to pursue vaccination.
That brings us to where we are today: a significant proportion of Americans still need a COVID-19 vaccine, but existing data cannot pinpoint who they are or where they are because the data are not centralized, accessible, or detailed at the local level. Most states do not record vaccination data for people within their borders who have been vaccinated at federally-operated FEMA sites or who have received shots in other states.
Politicians have been tasked with protecting their citizens and getting vaccinations to everyone who needs one, but they are not equipped with the complete datasets they desperately need to make critical decisions on continued vaccine distribution. As of now, local leaders can only design targeted outreach strategies by guessing where hotspots of unvaccinated citizens exist through comparisons of incomplete demographic data on people vaccinated with census-level population data. Mobile vaccine units, if available, could be deployed to fully vaccinate portions of counties since no data identifying the unvaccinated exists. It’s no better than throwing a dart at a map of their regions to find unvaccinated citizens.
While these improvements have not been adopted so far, there is still hope. We do not have access to the data required to finally end this pandemic, and even if we could access it, many questions remain. Can we access granular data without infringing on personal privacy? Can data help the U.S. push towards a goal of 100% vaccination and get our lives back to normal? Can we use the knowledge from our vaccine distribution failures and successes to help international vaccination efforts?
We will address these questions and more as the Pandemic Data Initiative continues to highlight public health data concerns that have been exposed by the COVID-19 pandemic. Check back later this week for an interview with Johns Hopkins Coronavirus Resource Center vaccination expert, Dr. William Moss, where we will discuss vaccination data issues in more detail.
1. E. Findell, Covid-19 Vaccine Supply Set to Outpace Demand, The Wall Street Journal.
2. Allocating and Targeting Pandemic Influenza Vaccine During an Influenza Pandemic, Centers for Disease Control and Prevention, 2018.
3. A.B. Eric Toner, Carleigh Krubiner, Justin Bernstein, Lois Privor-Dumm, Mathew Watson, Elena Martin, Christina Potter, Divya Hosangadi, Nancy Connell, Crystal Watson, Monica Schoch-Spana, Tener Goodwin Veenema, Diane Meyer, E. Lee Daugherty Biddison, Alan Regenberg, Tom Inglesby, Anita Cicero, Interim Framework for COVID-19 Vaccine Allocation and Distribution in the United States, Johns Hopkins Bloomberg School of Public Health Johns Hopkins Center for Health Security, 2020.
4. S. Adhikari, N.P. Pantaleo, J.M. Feldman, O. Ogedegbe, L. Thorpe, A.B. Troxel, Assessment of Community-Level Disparities in Coronavirus Disease 2019 (COVID-19) Infections and Deaths in Large US Metropolitan Areas, JAMA Network Open 3(7) (2020) e2016938.
5. FACT SHEET: The American Jobs Plan, The White House, 2021.
6. Internet/Broadband Fact Sheet, Pew Research Center, 2021.
7. C. Ryan, Computer and Internet Use in the United States: 2016, American Community Survey Reports, U.S. Census Bureau, 2017.