In an article published Feb. 7, The Exposé reported Pfizer’s COVID-19 vaccine “increases children’s risk of death 5100%,” based on data recorded by the UK’s Office for National Statistics (ONS).

The ONS is the UK’s largest independent producer of official statistics that include COVID vaccine uptake and mortality by age group.

The ONS data cited by The Exposé is found in table 9 of the dataset, labeled “Deaths occurring between 1 January 2021 and 31 October 2021 edition of this dataset.” It can be downloaded directly from the ONS website here.

The Exposé accurately plotted some of the ONS data in its article:

mortality rate in doubly vaccinated

As the above table shows, the all-cause mortality rate in the doubly vaccinated is in fact 238.7/4.58 = 52 times larger, or 5,100% higher than in the unvaccinated.

However, before drawing any conclusions it is imperative to examine the actual ONS data more closely.

First, we must consider the total number of deaths in each population. There were 96 total deaths among nearly 2.1 million unvaccinated children in the 10- to 14-year-old age group compared to four deaths among 1,678 children in that age group who had received their second dose 21 days or more prior to their demise.

Children who received only one dose or who were inside the 21-day window were not included in the subgroup with the highest mortality.

The Exposé chose to compare the most extreme subset of the vaccinated to support its provocative headline. Moreover, although the specific cause of death is not given (the ONS only separates cause of death into COVID or non-COVID-related), none of the four deaths were related to COVID.

This poses important questions. What if the four perished in an automobile accident? What if they were all suffering from a debilitating chronic condition that led to a predictable fatality?

We arrive at a major confounder: These two populations of children are not necessarily matched.

It is likely parents with children with significant comorbidities would have more readily sought the vaccine for their kids compared to parents with healthy children.

According to actuarial tables from the U.S. Social Security Administration, the risk of death in a 10- to 14-year-old is about 1 to 1.7 in 10,000 in a given year.

These numbers are 2 to 4 times higher than the mortality rate in the unvaccinated group in the ONS dataset, further indicating that this group is healthier than the average.

Nevertheless, taking all children who had received at least one dose of the vaccine as a whole, we find that 11 deaths occurred in 27,359 person-years. This mortality rate is still significantly higher (nearly 9 times) than the unvaccinated.

Based on the Fisher Exact Test — a standard statistical test used to determine if the proportions of categories in two group variables significantly differ from each other — there is less than a 1 in 100,000 probability this difference could have happened at random.