The Pfizer vaccine reduces SARS-CoV-2 infection and hospitalization risk in middle-aged adults

The Pfizer-BioNTech vaccination had a 30% lower risk of COVID-19 infection and hospitalization than the AstraZeneca two-dose immunizations, according to a new medRxiv* preprint study. There were no differences in declining immunity between both vaccines when Delta was the dominant strain for at least six months.

The findings, led by Junqing Xie of the University of Oxford in the United Kingdom, reveal that the Pfizer-BioNTech mRNA COVID-19 vaccinations provide more protection against SARS-CoV-2 than the AstraZeneca vaccines. This may make mRNA vaccines for boosters and mix-and-match vaccines more appealing.

doctor giving her patient injection

Obtaining information

The researchers used data from the UK BioBank, which includes 500,000 people between the ages of 40 and 69, from 2007 to 2010. Sociodemographic, lifestyle, and health-related characteristics were all included in the data. Because vaccination data for Scotland and Wales was not available at the time, the researchers concentrated on people residing in England. They comprised people who received a single dosage of COVID-19 vaccine between January 11, 2021, and February 28, 2021, or a two-dose between March 22, 2021, and May 9, 2021.

The first Pfizer-BioNTech dose was given to about 70,097 persons aged 50 and up. The AstraZeneca COVID-19 vaccine was chosen by 98,551 persons. Pfizer-BioNTech and AstraZeneca each gave two doses to 67,813 and 89,030 patients, respectively.

Pfizer-BioNTech vaccination recipients were slightly older, with an average age of 71.35, compared to AstraZeneca vaccine recipients, who had an average age of 71.06.b The majority of the participants were white men.

When they were vaccinated, and socioeconomic characteristics like income, were the most significant disparities between vaccination groups. 200 patients out of the 14,000 who received one dose of the Pfizer-BioNTech vaccine later tested positive for COVID-19 infection. This translates to a 13.7 per 1,000-person incidence rate. A total of 261 people tested positive for SARS-CoV-2 out of the approximately 20,000 people who received one AstraZeneca dose, yielding a rate of 12.6 per 1,000 people.

However, after more changes (the hazard ratio was changed to 0.72), the Pfizer-BioNTech vaccine was found to provide the best protection against SARS-CoV-2. For both vaccines, the risk of hospitalization was comparable to the risk of infection.


SARS-CoV-2 was found in 1,361 of the 34,991 patients who got two doses of Pfizer-BioNTech. After two doses of the AStraZeneca vaccine, 2,497 patients out of 44,084 tested positive. The Pfizer-BioNTech vaccine was found to be the most effective in preventing infection and hospitalization.

Unadjusted and adjusted HR were almost identical, favouring BNT162b2. The rates of Covid-19 hospitalisation remained low in both cohorts, but higher amongst ChAdOx1 (IR: 4.55 per 1,000 person-years) compared to BNT162b2 recipients (IR: 3.47 per 1,000 person-years), with adjusted HR of 0.71 favouring BNT162b2,” explained the researchers.

The rate of COVID-19 infections after participants received their first doses increased first from January to March 2021, before levelling 14 weeks later.

From June to October 2021, there was a surge in infections 12 weeks following the second dosage. Regardless, results showed that patients who received two Pfizer-BioNTech doses versus two AstraZeneca doses had a decreased probability of COVID-19 infection. For at least 6 months, the rate of infection and hospitalization remained stable.


Several study flaws could have influenced the final results. The observational data could have been impacted by disparities in testing rates between vaccinated and unvaccinated people who were exposed to SARS-CoV-2.

Participants’ personal information was obtained over a decade ago and may have changed since then. The researchers claim that all of the participants were middle-aged or older, and that only minor changes in socioeconomic position and education are expected. Other unexplained variables could also change the link between infection and vaccine effectiveness, thereby underestimating comparative vaccine effectiveness.

*Caution: This is a very important notice.

Preliminary scientific papers published on medRxiv are not peer-reviewed and should not be regarded as conclusive, should not be used to influence clinical practice or health-related behavior, and should not be recognized as established information.

Journal reference

  • xie J, et al. (2021). Comparative effectiveness of the BNT162b2 vs ChAdOx1 vaccine against Covid-19.