ONS accused of overestimating excess deaths in first year of pandemic
The Office for National Statistics (ONS) has been accused of overestimating excess deaths in the first year of the Covid pandemic.
The statistics body announced it was updating its methodology this week for calculating excess deaths to include current death rates, as well as the growing and ageing population – a change that many experts said was “long overdue”.
But academics from the University of Oxford have warned the new modelling reveals a major drop in expected deaths in 2020, making it appear that far more people had died than normal during the first year of the pandemic.
The new modelling also revised down excess deaths last year, even though many charities and universities have reported unusually large upswings in mortality rates for conditions like heart disease.
Prof Carl Heneghan, the director of Oxford’s Centre for Evidence-Based Medicine, said: “The ONS estimates make excess deaths look more serious in 2020 and less serious in 2023.
“It is clear to us that something is not right. There have been more deaths than expected since the pandemic.
“Last year, the British Heart Foundation reported that, since the onset of the pandemic, as of June 2023, there have been nearly 100,000 excess deaths in England involving ischaemic heart disease or other cardiovascular diseases.
“The number of premature deaths from heart disease soared to a 14-year high in the wake of the pandemic. Yet the Government appears to lack any appetite for investigating the root causes.”
The concerns have arisen because of perceived anomalies in the number of deaths that the ONS calculated were likely in a given year.
For 2020, the body estimated there would be around 26,000 fewer deaths than 2019, primarily because 2019 saw significantly fewer deaths than expected.
However, this was largely because there was very little flu circulating that winter and even before Covid, many experts warned that 2020 would see a spike in deaths because old and vulnerable people who would have died in 2019 were still alive.
Instead of increasing the expected number of deaths in 2020, the ONS model continued the downward trend of 2019, estimating around 26,000 fewer people would die than had been expected the previous year.
It meant that the excess in 2020 is recorded at around 76,000, but Oxford warned it was likely to be far less.
The ONS admitted that its model did not factor in a “displacement effect”.
“Because the mortality rate fell in 2019 after several years of being flat, that generated a negative trend,” it said in explanation.
“This explains why the number of expected deaths in 2020 was lower than in 2019. Our model does not currently account for a displacement effect (the fact that after a large positive number of excess deaths you may expect fewer deaths in the next year, and vice versa), but this is something we are going to investigate.”
The ONS pointed out that the revised methodology had actually lowered the number of excess deaths in 2020 from 84,064 to 76,412 – a drop of 9 per cent.
The new changes this week also saw excess deaths in 2023 revised down from 31,442 to 10,994, a drop of 65 per cent.
But there are claims that it underestimates the true picture today because current death rates are increasing in part because of the excess death problem.
The ONS said that because deaths had been relatively higher in 2022, it “may push the trend upwards” but it said the figures were also based on historical baseline data from the past five years and large population rises.
‘There should be an investigation’
Prof Kevin McConway, the emeritus professor of Applied Statistics at the Open University, said there was “no such thing as a correct number of excess deaths” and urged experts to focus on getting to the bottom of what was causing the increased mortality rates.
“The ONS method gives a measure of how many excess deaths there were, but tells us nothing directly about what was the cause of any excess,” he said
“There should be an investigation. Without more work, by the Government and/or some respectable and reliable researchers, this speculation will go on, and it’s not helpful.”
The ONS defended its new methodology, saying it had worked with many academics and organisations for more than a year to make sure the modelling was robust.
Previously excess deaths were calculated based on the five-year average which ignored the growing and ageing population and current death rates.
An ONS spokesman said: “Our new approach accounts for the growth and ageing of the population and provides a robust method for routine monitoring of excess deaths on an ongoing basis.
“The new and old methods estimate similar numbers of excess deaths during the Covid-19 pandemic. In particular, the two approaches produce similar peaks in estimated excess mortality in the second quarter of 2020 and the winter of 2020-21.”
Prof Sheila Bird, former programme leader at the MRC Biostatistics Unit of Cambridge University said getting excess deaths right was “highly important for both the public and the Covid Inquiry”.