Article claiming studies show Covid-19 vaccines increase infection risk is false and misleading

An article on The Exposé website reports on three studies claiming they show that Covid-19 vaccines do not work to reduce transmission. The article states: “The evidence is now clear. According to three individual studies published by the US Centers for Disease Control [CDC], the UK Government, and Oxford University, Covid-19 vaccines are harmful and ineffective.”

None of the authors involved drew such conclusions, nor does our analysis of the findings support the article’s claims.

The article was published in January 2023 but references studies from 2021. The evidence on Covid vaccine efficacy is continually changing in the face of new variants, and indeed new vaccines intended to protect against different strains are now being used.

The NHS states that Covid vaccines are safe and effective, and have been shown to reduce the risk of serious illness and death from the virus, as well as reducing the risk of catching and spreading the disease. The UK vaccination program is set to move towards targeting only those at higher risk, rather than the general population, over the course of 2023.

CDC study

The CDC study is part of a Morbidity and Mortality Weekly Report from August 2021. The CDC releases these reports as “the agency’s primary vehicle for scientific publication of timely, reliable, authoritative, accurate, objective, and useful public health information and recommendations”.

The data is from a 14-day period in July 2021 and covered 469 Covid positive cases identified among residents in Barnstable County, Massachusetts, of which 74% of the cases were fully vaccinated. 

The Exposé claims this “reveals that vaccinated individuals are more likely to contract and spread Covid-19 than unvaccinated individuals”. 

It doesn’t specify the logical basis for this claim. It may be as simple as the fact that the data shows that more of the Covid cases were among vaccinated people than unvaccinated people. 

Another interpretation could be that the proportion of cases who were fully vaccinated (74%) was higher than the proportion of the state’s population that were fully vaccinated (69%), suggesting that vaccinated people were more likely to become infected than unvaccinated people. This comparison isn’t made by the Exposé directly. 

But regardless, the report itself states “data from this report are insufficient to draw conclusions about the effectiveness of COVID-19 vaccines.”

A simple comparison of prevalence does not take into account inherent differences that may exist between the groups. 

For example, the UK Health Security Agency identifies differences in risk, behaviour and testing as factors which might affect Covid-19 rates in vaccinated and unvaccinated populations. 

UK Government study

The paper funded by the Department for Health and Social Care (DHSC) cited by the Exposé is a preprint (an article published before going through the review process). The reviewed article is also available and it is this version that we’ve considered. 

Like the CDC study, the article also looked at infections, on this occasion recorded as part of the Covid-19 Infection Survey (CIS) which continues to send swabs and blood tests to thousands of households in the UK to estimate how many people across the country are infected on a week-to-week basis. 

The study looked at a period from early December 2020 to May 2021 when the Alpha variant was dominant, and a period from May 2021 to August 2021 when the Delta variant was dominant. The Alpha variant was first documented in September 2020 in the UK, with Delta being found in India in October 2020 and becoming the dominant strain in the UK by mid-May 2021.

The Exposé claims that “82% of the positive PCR results in the study were in fully vaccinated individuals”.

The study does not actually provide the percentage of positive results in vaccinated and unvaccinated people so it is unclear where the 82% has come from. However, it doesn’t appear to be accurate. 

Across the entire time frame covered by the study 1,913 of 19,658 total Covid-19 cases (10%) occurred in fully-vaccinated individuals (those who had received two doses of vaccine).  

This isn’t particularly surprising given that most cases during this period would have occurred in the winter 2020/21 wave during which very few people were fully-vaccinated.

But even just looking at the May to August 2021 period when Delta was dominant, 1,816 (58%) of the total 3,120 cases occurred in fully-vaccinated individuals.

If you look at individuals who had received any dose of vaccine, not just those fully-vaccinated as the Exposé claims, then they account for 2,693 (86%) of the total 3,210 cases in the later period. 

Full Fact contacted the Exposé regarding this figure but had not received a response at time of writing.

But in any case, as mentioned, comparing the prevalence of Covid-19 cases in vaccinated and unvaccinated populations is not a good way to measure vaccine effectiveness.The study found the Pfizer and AstraZeneca vaccines were between 10% and 16% less effective against the Delta variant of Covid-19 compared to Alpha but that both vaccines remained effective.

Oxford University study

The Oxford University study linked in the article is also a preprint. The reviewed article is available online and it is this version that Full Fact reviewed.

The Exposé article states the study “reveals that a Covid-19 vaccine can actually increase the likelihood of contracting and spreading the virus”. The study does not actually show this at all.

The researchers looked at numbers of vaccinated healthcare workers in an infectious disease hospital in Vietnam who tested positive for Covid over a 14-day period in June 2021. 

As all participants were vaccinated, the study didn’t make an assessment of vaccine effectiveness. 

The Exposé says that “Shockingly, the peak viral loads among the infected, vaccinated group were found to be 251 times higher than those of the unvaccinated staff in March-April 2020”. 

The paper does show this. The researchers calculated a viral load (the amount of virus in a given volume of a body fluid like blood, or in this case mucus in the nose) using PCR nasal swabs and tried to control for factors like symptomatic status, age, gender and medical comorbidities. 

The researchers found that overall the median average peak viral load was 251 times higher than viral loads from a group of cases the researchers took from March and April 2020, all of whom were unvaccinated. 

The Exposé uses this finding to make the argument that vaccines are ‘harmful and ineffective’ but this is not a fair comparison given the Covid-19 variants in circulation in March 2020 were different to those in circulation in June 2021. 

A study in the Lancet found viral loads in vaccinated and unvaccinated people infected with the Delta variant were similar, but vaccinated people were less likely to become infected in the first place, and if they did, saw a faster reduction in viral load than unvaccinated people.