Skip to:

In Public Health, Data Is Our Compass

September 2021

Many of us have heard stories of a fully vaccinated friend, or friend of a friend, who contracted a COVID-19 breakthrough infection. Thankfully such infections do not typically lead to serious disease, yet they are important to track because they could lead to further spread of the virus. And evolution of the virus  could lead to even more infectious variants than the currently predominant one, Delta.

What do we know about current rates of breakthrough infections in the United States? Unfortunately not much because the critical data to provide reliable answers are not being systematically collected in the United States. While the Centers for Disease Control and Prevention (CDC) does track hospitalizations and deaths associated with COVID-19, the CDC does not currently count all breakthrough infections — such as vaccinated people who experience mild symptoms or those who don’t experience any symptoms at all. Because we don’t know how many breakthrough infections exist in our country, we can’t begin to answer the question of whether vaccinated people may be transmitting infection in significant numbers to others, including children. The answer is important for guiding public health policy including masking recommendations for both unvaccinated and vaccinated people. 

A recent study from the Kaiser Family Foundation attempted to determine the rate of breakthrough infections by piecing together data from websites and other available data sources posted by public health agencies from all fifty states and the District of Columbia. They found that only about half of states have some data on breakthrough infections, and only a fraction of those are regularly updating their data.

Controlled randomized clinical trials have provided key scientific data that led the FDA to grant full approval to the Pfizer vaccine. But in addition to controlled clinical trials, we need reliable public health surveillance data in order to conduct studies that track the real-world effectiveness of vaccines in populations. Public health surveillance data helps answer timely questions such as: How stable is vaccine protection over time against the current predominant Delta variant or against new emerging variants across our communities and all segments of our population? Without this information, we jeopardize our efforts to protect the health of our communities.

With vaccine booster shots on the immediate horizon, now is the time to get health information data systems ready to help answer key questions including the real-world effectiveness of boosters. A recent study in England evaluated the real-world effectiveness of vaccines for preventing symptomatic Delta variant infections, and a new nationwide study in Israel demonstrated the real-world effectiveness of boosters against the variant among people over 60 years of age who had been fully vaccinated at least five months previously. These studies were possible because of a network of national, interconnected (linkable) data systems. We can learn from these and other countries about building public health data systems.

Our nation’s public health surveillance data system in the United States is not a single network, but instead more than fifty separate state and local systems that are woefully inadequate and severely understaffed. Accurate and linkable real-time data are the cornerstone for developing effective public health control strategies and sounding the alarm when pivots are necessary to address shifting trends in emerging infectious diseases. The time is long overdue to invest in our public health data infrastructure, beginning with a national blueprint that can serve as a framework for our path forward.

Public health data is our compass — it guides scientists and policy makers — and the public should be able to count on its robust collection to help keep us all healthy and prevent future pandemics.





Dr. Ron Brookmeyer
UCLA Fielding School of Public Health