ONS death figures not averaged correctly

4 February 2021
What was claimed

ONS data shows huge falls in non-Covid causes of death in December 2020.

Our verdict

The data was wrong. It claimed to show a huge fall in 2020 against the previous five-year average. But that five-year “average” hadn’t in fact been divided by five, meaning it was five times too high.

Data from the Office for National Statistics (ONS) has been shared on social media, which apparently shows the number of deaths from various non-Covid causes was far lower in December 2020 than in previous Decembers. 

This has been used to suggest that deaths which are now being classed as Covid-19 deaths are actually just miscategorised deaths from other causes.

But this is not the case. In fact, the data from the ONS was just produced incorrectly. There were falls in the number deaths from some non-Covid conditions in December 2020, but nothing like what the tables initially suggested. 

What was wrong with the data?

The data, originally published on 18 January, was meant to show the number of deaths from the leading causes, averaged over the previous five Decembers (2015-2019), but there were some clear indications that there was something wrong.

It appeared as if the “five-year average” figures for the number of deaths from the listed causes had been added together, but not actually been divided by five in order to create an average.

Firstly, the data claimed to show that there were 114,399 deaths in England from the leading nine causes of death, averaged over the previous five Decembers. But other data within the release showed that the highest number of deaths from all causes in any single December over the previous five years was just 44,286 (in 2019). 

Also, while the five-year average number of deaths from these leading causes were all far higher than the levels in 2020, the rates of death per 100,000 people all appeared to be more in line with the levels in 2020. 

After Full Fact, and others including Italian fact checkers Facta, contacted the ONS about the error, it confirmed this was the mistake that had been made and issued new tables. 

What does the correct data show?

The new, correct, data shows that the number of deaths in December 2020 from some non-Covid causes were slightly lower than the previous five-year average.

The biggest falls were deaths from influenza and pneumonia, and “chronic lower respiratory diseases.”

As we’ve written before, the UK is seeing a very mild flu season, in part due to the fact that flu vaccinations are at their highest ever level and steps the country is taking to reduce the spread of Covid-19 may also be reducing the transmission of influenza.

This article is part of our work fact checking potentially false pictures, videos and stories on Facebook. You can read more about this—and find out how to report Facebook content—here. For the purposes of that scheme, we’ve rated this claim as partly false because this data was produced by the ONS in error. The “five-year average” for 2015-2019 was, in fact, not averaged over the five years.

Full Fact fights bad information

Bad information ruins lives. It promotes hate, damages people’s health, and hurts democracy. You deserve better.