Terrifying death graphs used to justify second Covid lockdown ‘proved to be incorrect’
DEATH graphs used to justify a second Covid lockdown in England have been branded "misleading" by experts who say they are "mathematically incorrect".
Heat maps presented to the general public on Saturday evening made for scary viewing as the colour coded charts showed the pandemic was out of control - predicting a worst case scenario of more than 4,000 deaths a day by December.
⚠️ Read our coronavirus live blog for the latest news & updates
Prime Minister Boris Johnson was forced to hold the press conference after details of a second lockdown were leaked to the media.
Chris Whitty and Sir Patrick Vallance presented data, including one graph which showed the death toll predictions from various studies.
Today, speaking at the Science and Technology Committee, the two scientific advisers defended their decision to present the graphs to the general public.
While they admitted that the heat maps and graphics may have been "confusing" Prof Whitty noted that the committee has "repeatedly asked for more data" and claimed that when it is published then it's "too much".
The data presented on Saturday included data from Public Health England (PHE) and Cambridge University, which predicted the most extreme death toll - up to 4,000 fatalities a day.
But, one of the scientists who worked on that study, said the data was presented incorrectly.
Professor Danlela de Angelis said the curve was made to look like a prediction, when in fact it was a model that was a month old and did not take into consideration the tier restrictions that had previously been introduced.
The Cambridge model has since revised its estimates which are around 1,000 deaths by the start of December.
Data was also presented from Imperial College London, the London School of Hygiene and Tropical Medicine and Warwick.
Professor Neil Ferguson, of Imperial College London, said that the 4,000 deaths a day scenarios were "preliminary work" to create a new "reasonable worst case planning scenario".
He added: "SPI-M [Scientific Pandemic Influenza Group on Modelling] undertakes a wide range of modelling for government.
"The 'up to 4,000 deaths a day' scenarios represent preliminary work to generate a new reasonable worst case planning scenario to assist NHS and other Government planning."
He added: "Even allowing for the effects of the current tier system, the most recent SPI-M projections suggest that without further action, the second wave is set to exceed the first wave in hospital demand and deaths."
However, experts failed to plot their own "worst case scenario" model, which the graph above shows to be far less than other estimates.
Sir Patrick this afternoon stood by the decision to show the graphics to the public and said that the intention was "not to frighten people", but was in fact to show how the virus could spread in the future.
'Mathematically incorrect'
The most recent data from the UK revealed that cases plunged by 2,000 on last Monday as 18,950 more contracted the deadly bug and 136 people died.
Speaking on BBC Radio 4's Today Programme, Professor Carl Heneghan said the maps shown during the Downing Street press conference were "incorrect".
He said: "Mathematically it is now proven to be incorrect, particularly the 4,000 estimate of deaths that would occur in December.
"That is because it's already about four weeks out of date and actually Cambridge who are doing it at the MRC unit there have already provided updates to provide lower estimates and there estimates are much nearer to the truth and they have a system called now-casting and forecasting where it's much nearer to the correct data.
"As an example, what they modelled four weeks ago was showing about 1,000 deaths on November 1, and actually only 200 actually occurred."
Heating up?
During the news conference on Saturday night, Prof Whitty also revealed a "hospital heat map".
They presented the charts to show that the areas marked in red now had more Covid patients than at the peak of the virus.
Amber shows that there was more than half as many than the peak, and green shows fewer patients than at the peak.
The experts mapped just 29 NHS hospitals when there is in fact 482.
On October 31, figures showed that there were 9,213 patients in hospital with the disease compared with 17,171 at the peak.
Prof Heneghan, who is the director for the Centre of Evidence-Based Medicine said the graphs that government advisers revealed to the public should "never have been shown".
He said nerve casting and forecasting tools only predict two weeks into the future and after that the tool can become "highly uncertain".
Prof Heneghan explained that this is because cases are "shifting in a way they weren't three weeks ago.
'Cases flatlining'
He added: "They are starting to flat line - admissions are flatlining and in effect deaths are starting to flatline.
"There will be an update I hope on this system that will give us a clear understanding of where we are going.
"Also today the ONS (Office for National Statistics) will be reporting on excess deaths so the next 24 hours will give much more useful information that should inform whether a lockdown occurs or not."
Looking at the full NHS data set from all 482 hospitals and there are 232 hospitals that have no Covid patients at all.
Some of the areas in the UK that have been hit hardest by the virus aren't even using half fn their beds on Covid patients.
On October 27, data from the NHS showed that Covid patients at Liverpool University Hospitals were occupying 450 out of 1,595 beds.
Liverpool was one of the first areas to enter into Tier 3 restrictions as cases continued to rise in the area.
'Significant drop'
Prof Heneghan added that the R value in Liverpool is "well below one at this moment in time".
He said there is a "problem" in Liverpool but that cases in the city have halved and hospital admissions have "stabilised".
"You've got ... these pockets around the country where trusts like Liverpool have got into trouble with over half the patients being Covid patients.
"But let's look at the data, the data in Liverpool is showing cases have come down by about half, admissions have now stabilised, so, yes, there is a problem in Liverpool.
"But, actually, the tier restrictions... the people in Liverpool have dropped cases from about 490 a day down to 260 a day - a significant drop
"The R value is well below one in Liverpool at this moment in time."
Other places in England that have been hit hard by the virus include Nottingham.
Looking at NHS data and on October 27, just 289 Covid patients were taking up 1,512 of the beds at the trust.
In Manchester University Hospital, 207 of 1,368 beds were taken by Covid patients.
National lockdown
While cases of the virus are rising in some areas, the data from NHS shows that this does not translate to hospital admissions in all areas.
In defence of the graphics shown on Saturday - which are based on modelled estimates, Sir Patrick this afternoon added that "they are scenarios not predictions".
He added: "It's difficult to project forward, it's a model, it's not real. This is a disease that is spreading and will affect us in weeks to come.
"An illustration of what may happen in the future is an important part of this."
Prof Whitty added that while the graphs shown are models, it's not unreasonable to think that deaths would reach the heights seen at the peak of the pandemic in April.
From Thursday the government will implement new restrictions in England, temporarily doing away with the tier system that had been implemented in order to curb cases.
A four week lockdown will commence in which pubs and restaurants will close, as well as non-essential shops.
Schools will remain open and workers who cannot do their job from home - such as those who work in the construction industry, will be able to continue to do their job.
At the press briefing on Saturday, Mr Johnson said that there was no alternative.
Most read in Health News
Addressing MPs yesterday he said: "If the House asks me what is the exit strategy, what is the way out – let me be as clear as I can: the way out is to get the R down now to beat this autumn surge and to use this moment to exploit the medical and technical advances we're making to keep it low.”
He also told MPs we could see twice as many deaths in winter than we did in the first wave of the virus, and that the NHS could crumble without immediate action since the virus is doubling faster than the Government had anticipated.
READ MORE SUN STORIES
Cabinet minister Michael Gove admitted on November 1 that the lockdown could continue beyond the start of December if the measures do not drive the R-number down.
Once the restrictions are lifted, the plan is to return to the three-tiered local lockdown system.