Skip to content


How Many Cases of a "Breakthrough" COVID-19 Infection Should We Expect? A Lot.


René F. Najera, DrPH

April 23, 2021

You've probably seen in the news that people who are considered to be fully vaccinated are catching COVID-19. Many of these cases are being called "vaccine breakthrough" cases, where the person is thought of as being protected and the virus still infects them and makes them ill. (Fully vaccinated means two weeks post the second dose of COVID-19 vaccine if receiving the mRNA vaccine or the one dose of the Johnson & Johnson/Janssen vaccine.) As a result of these reports, members of the public are starting to worry that maybe the vaccines are not as good as the 90-something "effectiveness" promised to them by the vaccine manufacturers and the government regulators. So let's break down a few concepts in hopes of alleviating a few worries.

First, let's talk about efficacy and effectiveness. When a vaccine is in clinical trials, the researchers conducting the trials have almost complete control of who gets the vaccine and who gets the placebo. The only thing not in their control is the randomization of the people they select to participate, but participants are self-selected. That is, participants choose to be in the trial and sign up for it. It is then up to a randomization strategy (usually a random number generator) to split the group into those who will receive the vaccine and those who will receive a placebo (or those who receive neither, if that is how the trial is designed).

For example, my wife participated in the trial for the Moderna vaccine. She is a physician assistant with 15 years of experience in family medicine, emergency medicine, mental health, and community health. She also worked at a high-level biosecurity laboratory while in college. That is, she knows her viruses and bacteria, and she knows about infection prevention and infection control. She is also well-educated with two master's degrees, and we live in a single-family home in a suburb of Washington, DC. Essentially, she is not like most of the United States. Her risk profile is different than the risk profile of an immigrant worker living in overcrowded (more than 1 person per room) conditions and lacking an education or an understanding of how viruses spread.

Because the study participants and study design are different from real-world conditions, the difference in risk of infection between the placebo and the vaccination group is called "efficacy." While the conditions are not perfect, they're not what the public is facing. This is by design because you want to eliminate as many variables as possible, so that the final risk of disease comes down to vaccinated or not. (Admittedly, it is difficult to do. For that, epidemiologists and biostatisticians work together to properly analyze the data in the full context of the trial.)

Once the vaccine is approved/authorized/licensed for use in the general public, the difference in risk between the vaccinated and the unvaccinated is called effectiveness. As you can imagine, effectiveness can be the same or different from efficacy. It can be the same if the trial design was as close to real-life as possible. It can be very different if the real world has more variables to it than the trial could ever replicate. For example, you can end up with much lower effectiveness than efficacy if you end up vaccinating only a very high-risk group of people, like the elderly living in congregate care settings. Not only is their immune system less likely to react to the vaccine, their age and living arrangements make them susceptible to infection because things like social distancing cannot be adequately implemented.

When the vaccine manufacturers submitted their trial results to the Food and Drug Administration (FDA) to get an Emergency Use Authorization (EUA), they submitted efficacy data. Unfortunately, much of the press presented those numbers as "effectiveness," leading to confusion when the true effectiveness -- the risk in the vaccinated versus unvaccinated in the real world -- turned out to be different. So be mindful of the language being used by pharmaceutical companies and regulators -- and researchers -- when they present findings of clinical trials versus findings of epidemiological studies in the real world.

As the companies that created the vaccines and conducted the trials for them have pointed out, efficacy and effectiveness of these vaccines (or any vaccine) are not 100%. The vaccines are not perfect, and there are many factors that could influence their effectiveness, from where the recipients live to how they live and what work they do for a living. For example, my wife is at higher risk of infection because of her work than someone of the same age, same socioeconomic status, and same educational background who works remotely at home. Because it is not cost-effective and logistically near-impossible to know what everyone who takes the vaccine does for a living, where they go to meet people, how they live at home, or even if they wash their hands, then it is very difficult to accurately give a hard number for effectiveness. What we in public health end up giving is an average number for the average person in an average setting.

If you've studied statistics, you know that averages come with variances. That is, there is a variation -- a range -- of what values an average can take. This is the same with effectiveness. For some, the vaccine can be 99.99999% effective because they are protected by the vaccine and their immune system, and they wear their masks properly, and they wash their hands, and they are otherwise healthy. For others, the vaccine may be completely ineffective because they're living in a crowded condition with a bunch of unvaccinated people who are actively sick and not wearing masks. There is a point at which their immune system can't fend off the virus anymore and an infection sets in. Again, the number you hear is the average.

As the United States government attempts to vaccinate every adult in the country (about 260,000,000 people), the fact that effectiveness is not 100% means that there will be people for whom the vaccine doesn't work. That doesn't mean that the vaccine will not work to stop the pandemic. It can, and it will if we vaccinate enough to get to herd immunity all around the world (as it is a pandemic and not an epidemic). It just means that plenty of people, perhaps millions, will get the disease even if they are vaccinated. After all, 1% of all adults equals 2.6 million people. If the vaccine is 99% effective, those 2.6 million would be at risk. At 95% effectiveness, about 13 million adults would be at risk if everyone was vaccinated.

To protect those who are at risk because the vaccine is not effective in them, or those who cannot get vaccinated because of medical reasons, and even those who will not vaccinate due to other reasons, we need to achieve herd immunity. But that is a concept for another post at a later time. For now, the key takeaway is that, yes, there will be plenty of people who will get COVID-19 even if they are fully vaccinated... And that is okay because nothing in the known universe is perfect.

Post featured image by on


  • , 
  • , 
  • , 
  • ,