Instagram post incorrectly interprets Pfizer vaccine study

A Pfizer-commissioned study published in July found the company’s COVID-19 vaccine was 96.7% effective in preventing people from developing severe symptoms or being hospitalized.

The finding was in line with previous clinical trials that have shown the Pfizer vaccine is extremely effective at preventing or reducing the severity of coronavirus infections. 

But an Instagram user claimed the figure doesn’t tell the whole story.

"Pfizer releases results of six-month long study showing their vaccine reduces severe COVID symptoms/hospitalizations by 0.1%," the now-deleted Aug. 4 post read.

In a follow-up post, the user didn’t dispute the 96.7% efficacy rate Pfizer reported, but called it misleading. The user claimed that her way of measuring gives people a better idea of a vaccine’s efficacy, and she puts that number at 0.1%, which sounds like it’s close to nothing.

The post was flagged as part of Facebook’s efforts to combat false news and misinformation on its News Feed. Instagram is owned by Facebook. (Read more about our partnership with Facebook.)

The crux of the claims is the difference between "relative risk reduction" and "absolute risk reduction." Both are metrics for measuring a vaccine’s efficacy, and both can be calculated from the raw data of longer-term studies such as Pfizer’s. 

But the claim errs in arguing that one of the measures is misleading or less accurate than the other. Both metrics can be valuable to researchers and public health officials trying to gauge the effectiveness of a vaccine. 

One key difference in how they affect perceptions: The more commonly used measure, relative risk reduction, tends to sound like a high number for a vaccine that works well, while absolute risk reduction is usually a lower number, because it reflects the baseline risk in the population of getting infected or severely ill. Some scientists argue that drug makers should always report both figures to provide a balanced picture of a particular intervention.

We contacted the Instagram user, who provided an explanation to PolitiFact for the basis of her claim. But her calculation left out a step in arriving at the 0.1% efficacy rate, according to a Pfizer representative and an author of the study.

PolitiFact previously checked a similar claim that conflated absolute risk reduction and relative risk reduction for the different COVID-19 vaccines currently in use around the world. We rated Mostly False a claim that vaccine makers were deceiving the public by reporting one of the two figures.

Both figures are used to describe efficacy, and can be expressed in percentage terms, but they measure different things and are calculated differently. Broadly speaking, one describes the probability of avoiding illness by getting vaccinated. The other represents the proportion of people in the population who avoided illness by getting vaccinated.

Featured Fact-check

Facebook posts

stated on April 9, 2024 in a Facebook post

Notice from a New Jersey doctor’s office proves that COVID-19 vaccines are dangerous for athletes.

In the case of the recent Pfizer study, the 96.7% figure is the relative risk reduction, the metric most commonly used to express a vaccine’s efficacy. It compares two groups — those who got vaccinated vs. those who got a placebo — as a ratio to show how much being vaccinated reduced an individual’s likelihood of developing severe symptoms.

Relative risk reduction provides an estimate of how effective a vaccine is at preventing a disease compared with someone who received a placebo, according to Dr. Stephen Thomas, lead author on the Pfizer study and chief of infectious disease at SUNY Upstate Medical University.

Absolute risk reduction draws on the raw numbers from the trial data to estimate how the vaccine would affect the incidence of serious symptoms in the broader population. It’s calculated as the arithmetic difference between the incidence rates in the two populations — vaccinated vs. unvaccinated. 

This measure accounts for people’s baseline risks of getting seriously sick from COVID-19, and changes across different groups and timeframes. 

"Both are valid measures," said Kit Longley, a Pfizer spokesperson. "They are simply calculated differently and used for different purposes to create a fuller understanding of a vaccine’s protection."

In the follow up post, which has also been deleted, and in messages to PolitiFact, the Instagram user said she calculated the absolute risk reduction for the Pfizer vaccine as 0.1% by looking at the the number of people out of the 22,000 in the placebo group who developed severe symptoms or were hospitalized — 30 divided by 22,000.

Both Thomas and Longley said the Instagram user’s approach was wrong. Her calculation looked only at the placebo group. Calculating absolute risk reduction involves comparing the incidence rate for the placebo group with the incidence rate for the vaccinated group.

An Instagram post said a Pfizer study showed that its vaccine "reduces severe COVID symptoms/hospitalizations by 0.1%." 

That’s not what the study showed, Pfizer and one author of the study said. The study showed the vaccine is 96.7% effective in preventing severe symptoms or hospitalization from COVID-19, compared with a placebo.  

The Instagram user said she arrived at her own efficacy number by calculating the absolute risk reduction instead of relative risk reduction. The two metrics measure different things.

A representative from Pfizer and the study’s lead author both said the 0.1% figure was incorrectly calculated but did not provide further information.

Absolute risk reduction is a valid measure of vaccine efficacy, but the Instagram user used a flawed calculation.

We rate this claim Mostly False.