When experts say COVID-19 vaccines are 95% effective it doesn't mean you'd have a 5% chance of getting sick

Hilary Brueck
·6 min read
covid vaccine trials south africa
A volunteer receives an injection for a potential vaccine against COVID-19 at the Baragwanath Hospital on June 28, 2020 in Soweto, South Africa. Felix Dlangamandla/Beeld/Gallo Images via Getty Images
  • Vaccine efficacy measures how well shots prevent disease among vaccinated people in clinical studies, when compared to unvaccinated control groups.

  • Most vaccines are not 100% effective at preventing symptomatic infections.

  • But, if enough people get shots, they can prevent the spread of disease near-perfectly anyway.

  • Visit Business Insider's homepage for more stories.

95%. 94.5%. 66%.

Lots of numbers are being thrown around, touting how effective different vaccines are at preventing symptomatic coronavirus infections.

These calculations are based on vaccine trials, which have been performed on tens of thousands of volunteer shot-takers around the world.

But what do these percentages really mean?

Vaccine efficacy measures how well a vaccine worked at preventing disease during a well-managed clinical trial. The number is an estimate, designed to give a general sense of how good shot protection may be.

Watch: Doctors discuss Johnson & Johnson vaccine efficacy

To reach it, researchers tally up all the people in a vaccine trial who got sick during the study period. Then they compare how many of those participants were vaccinated, and how many were not. Both Moderna and Pfizer said that roughly 95% of the people who got sick with COVID-19 during their trials had not received vaccines, while 5% had.

That does not mean that a vaccine is going to be 95% effective at stamping out all disease, even the asymptomatic kind. It does not mean that 5% of vaccinated people globally will get sick. And it does not mean that, once you are fully vaccinated, you have a 5% risk of getting COVID-19.

At best, it provides a temporary snapshot, a vaccine maker's rough estimate as to how well their vaccine will perform at curbing symptomatic infections in the general public, based on just a few months worth of trial data.

Vaccine efficacy rates are a rough guess at how the shots will work in practice

Coronavirus vaccine trial UK
Kirsty O'Connor/PA Images via Getty Images

In Pfizer's Phase 3 vaccine trial, just eight vaccinated people got sick with COVID-19 - mostly mild infections - out of more than 18,000 who got both doses of the vaccine.

One might be tempted to think, then, that the odds of getting COVID-19 for people vaccinated with Pfizer's shots could be calculated as 8/18,000, or just 0.04%.

But Yale immunobiologist and leading virus expert Akiko Iwasaki says that's a terribly misleading way to think about this trial data.

"If you were to ensure that the 18,000 people all got exposed to the virus, then yes, the vaccine efficacy would be incredible, right? But they're not," she told Insider. "That's why we have to compare the vaccine arm with the placebo arm, because the placebo arm is what tells us how much exposure and infection there would have been, if they weren't immunized with the vaccine."

In Pfizer's control group, consisting of 18,325 unvaccinated study participants, 162 people got sick with COVID-19.

That means even many unvaccinated people in the same study managed to evade the coronavirus, likely through a combination of factors - varying exposure risk, mitigation measures like mask-wearing and hand-washing, and perhaps some immunity from previous infections.

So, it's more helpful to compare the eight people who got sick with vaccines to the 162 people who got sick without them at the same time. "That's the better comparison," Iwasaki said, "and that's where the 95% comes from."

In Moderna's Phase 3 coronavirus vaccine trial, 11 people out of nearly 14,000 fully vaccinated individuals got sick, while 183 of the unvaccinated people in the similarly-sized control group fell ill with COVID-19 during the trial.

Watch: Dr. Fauci discusses how variants can affect vaccine efficacy

Vaccine effectiveness is a different metric

The efficacy rates from Pfizer and Moderna would have likely been different if the study had continued for a longer period of time, giving both the vaccinated and unvaccinated participants more and more chances to become infected.

Both trials only went on for a matter of months before this data was collected. They were also done during a time when there weren't as many worrisome virus variants circulating, either. They reflect only the lowered risk of infection vaccinated people were exposed to at that specific time and place.

Vaccine effectiveness, a different metric, measures how well shots work on the general public. We don't have those numbers for coronavirus vaccines yet, because none of them have been around long enough to collect robust data on how well these new vaccines protect people from future infections outside of a controlled environment.

It remains to be seen whether the COVID-19 vaccines will be as near-perfect as, say, polio vaccines, which are more than 99% effective at preventing polio infection, or more like annual flu shots, whic are not as great at completely halting that disease (ranging from 40% to 60% effective).

The plus side of even the less effective shots is that they can make a case of the flu, or another disease like the coronavirus, milder, even if they don't completely prevent infection.

The vaccines will only be truly effective if they reach most people in the world, and that could take a while

Schott glass vials
Glass vials. Schott

Herd immunity from diseases can only be achieved when large numbers of people are immune to a pathogenic threat, either from a prior infection or a vaccine.

A highly effective vaccine doesn't work very well if it is not widely used. On the other hand, if enough people get an imperfect vaccine, it can provide near-perfect disease protection.

"What I would like to see is the overwhelming majority of people get vaccinated so we can essentially really crush this outbreak," Fauci told the New York Times last November. "As we get into the fall [2021] we could be quite close to some degree of normality, certainly from the standpoint of the economy: of getting businesses open, of getting sports events being attended to, that is feasible."

There are still some unanswered questions about how well these new vaccines work.

It's possible that coronavirus vaccines may not provide people with full-on sterilizing immunity, meaning that vaccinated people could still get the virus to replicate in their bodies, and pass it along to others, asymptomatically.

But, even if that were to be the case, having such highly-effective vaccines widely available will greatly reduce the burden of death and disease wrought by this virus so far, and improve pandemic conditions around the country, and the world.

Update: This story was originally published on November 24, 2020, when vaccine trial results were first starting to trickle out. It has been updated.

Read the original article on Business Insider