On Sept. 9, President Joe Biden announced a new set of pandemic response policies. In the process, he said that 1 in 5,000 vaccinated people become infected by SARS-CoV-2 each day. He was citing David Leonhardt’s New York Times newsletter. Biden and Leonhardt frame the statistic as encouraging: only 1 in 5,000. But it isn’t good news or bad news. It’s essentially meaningless. What’s more important is knowing that when community spread of the coronavirus is high, the risk of breakthrough infection needs to be taken seriously, which means combining vaccination with other public health measures.
Let’s take a look at where the 1-in-5,000 figure comes from. Leonhardt was trying to assess whether breakthrough infections are rare or not. He reviewed daily infection reports from a few jurisdictions that track breakthrough infections including Utah, Virginia, and King County in Washington state. He decided the numbers were trustworthy and could be applied nationally. He also identified other locations that reported only 1 in 10,000 daily infections. Interpreting the data, he writes, “Here’s one way to think about a one-in-10,000 daily chance: It would take more than three months for the combined risk to reach just 1 percent.” If his math is correct, one may indeed argue that a 1 percent risk over three months qualifies as rare. Although it unfortunately seems likely the pandemic, and therefore our exposure to the virus, will last far longer than three more months.
Leonhardt is wrong, however, to take these health department reports at face value. Overall, the Centers for Disease Control and Prevention estimates that less than 25 percent of infections are confirmed through testing. This underreporting problem is likely worse for people who are vaccinated—their symptoms tend to be milder and they are less likely to be tested in a health care setting. Another problem is that health departments struggle to confirm whether an infected person has in fact been vaccinated. If someone has been vaccinated in another state or at a Veterans Affairs facility, they would in most cases be mislabeled as unvaccinated. Finally, the few jurisdictions that Leonhardt reviews do not include states like Florida and Louisiana, which have experienced the highest coronavirus spread in recent weeks.
What is the actual risk that a vaccinated person will experience a breakthrough infection? There is no simple answer. It depends on levels of community spread, whether individuals are exposed at high levels (due to risk levels in their workplaces, schools, and households, for example), and how long the pandemic lasts. We are also still learning about why breakthrough infections occur and whether it involves waning vaccine effectiveness over time. But if we make some realistic assumptions about underdetection, a place that consistently reports a 1-in-5,000 daily breakthrough infection risk could easily translate to 10 percent or more of its vaccinated population becoming infected over three months.
Maybe breakthrough infections aren’t rare, but can they be ignored? It is true that most vaccinated people who become infected will be fine, experiencing no worse than a miserable few days in bed. It is also true that large majorities of the people who are hospitalized or die from COVID-19 are not vaccinated.
But there are three main reasons I think we need to take breakthrough infections seriously. First, vaccinated people can still transmit the virus to people who are unvaccinated, including children who are not yet eligible for the vaccine. Second, there are people whose health conditions put them at high risk of hospitalization and death even if they are vaccinated. This includes people with certain immunocompromised conditions who may not respond as well to vaccines, even with a booster shot. Third, we are still learning about the risks of long COVID among the vaccinated. Even if a small percentage of breakthrough infections lead to persisting symptoms, that adds up to a large number of people when infections are widespread.
The bottom line here is that vaccination is not an individual’s golden ticket out of the pandemic. Despite what CDC Director Rochelle Walensky says, this is not just a “pandemic of the unvaccinated.” Instead, we are all in this together, because we breathe the same virus-filled air. In recent weeks, about 15 percent of COVID deaths have been among the vaccinated. If this trend holds and we allow another 100,000 people to die of COVID, that will include 15,000 deaths of people who followed the government’s advice and got vaccinated.
Back in August of 2020, while the vaccines were still under development, Anthony Fauci believed that the vaccines were likely to offer imperfect protection, and bringing the pandemic under control would therefore require combining vaccination with other public health tools such as universal testing, ventilation improvements, and masking policies. But these public health tools can be costly and require regulation of business, so there has been a reluctance by government to fully pursue them, especially in light of the vaccine’s initial stellar effectiveness. Instead, the Biden administration pushed for a more perfect vaccine that could prevent breakthrough infections; they initially planned to offer booster shots for the entire U.S. adult population. This plan was quashed, however, by an influential Food and Drug Administration advisory committee, and a CDC advisory group ultimately recommended restricting boosters to those at high risk of severe disease (older Americans and people with certain underlying conditions). In an unprecedented move, the Biden administration announced it would overrule the CDC committee and expand eligibility to people with high occupational exposure to the virus. While I don’t know how broadly the government should distribute boosters, I do know that the vaccine-as-silver-bullet approach has not served us well, and that a broader public health approach would prevent many deaths.