Imperial College modelling that led to nationwide lockdowns based on ‘inaccurate’ data

Scientists did not have accurate Covid data when they predicted that 500,000 people could die if the UK took no action during the first wave of the pandemic.

Modelling from Professor Neil Ferguson and colleagues at Imperial College London published on March 16, 2020, predicted the NHS would be overwhelmed within weeks and a terrible death toll would arise if nothing was done to stop the spread of the disease.

But minutes from a SPI-M (Scientific Pandemic Influenza Group on Modelling) meeting released to The Telegraph following a Freedom of Information request have shown that, a week earlier, the modellers remained ‘uncertain’ of case numbers ‘due to data limitations’.

Following the release of its model, Imperial College held a press conference, followed by Prime Minister Boris Johnson ordering the public to avoid pubs, restaurants and non-essential social gatherings later the same day.

At the briefing, Prof Ferguson said new conclusions had been drawn as ‘the last few days’ had provided ‘refinements’ in estimates of intensive care and hospital demand.

Minutes now show, though, that SPI-M did not believe the data was complete.

Imperial College London published a paper in March 2020 on the potential impact of coronavirus. It weighed up options on how a lockdown could reduce demand on hospitals

The scientific paper published by Professor Ferguson and his colleagues on the Imperial College COVID-19 Response Team was credited for persuading Boris Johnson’s Government to ramp up their response to the coronavirus.

The paper, released on March 17, and titled Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand, predicted that the Government’s original plan to ‘mitigate’ the outbreak instead of trying to stop it could have led to a quarter of a million people dying.

Using data from Italy and China, the scientists predicted how different Government measures would have different impacts on the outbreaks.

If no action at all had been taken against the coronavirus it would have claimed 510,000 lives, the team’s report said. Had the Government stuck with their strategy of trying to ‘mitigate’ the spread – allowing it to continue but attempting to slow it down – with limited measures such as home isolation for those with symptoms this number would be roughly halved to 260,000.

If the strictest possible measures are introduced, the number of deaths over a two-year period will fall below 20,000, the scientists said.

Other points in the Imperial College report, titled Impact of non-pharmaceutical interventions (NPIs) to reduce COVID19 mortality and healthcare demand, included:

  • Lockdown measures could be brought back if the virus resurfaces after this epidemic is over
  • The coronavirus outbreak is worse than anything the world has seen since the 1918 Spanish Flu pandemic 
  • Dramatic measures to suppress an outbreak carry ‘enormous social and economic costs which may themselves have significant impact on health and well-being’
  • Virus transmission happens evenly – one third of cases are caught in the home, one third at work or school, and one third elsewhere in the community
  • People are thought to be infectious from 12 hours before symptoms start, or from four days after catching the infection if someone doesn’t get symptoms
  • Patients who do get symptoms are thought to be 50 per cent more infectious than those who don’t 
  • People are thought to develop at least short-term immunity after catching the virus, meaning they can’t catch it again
  • Approximately 4.4 per cent of patients need hospital care. 30 per cent of those need intensive care, and 50 per cent of intensive care patients can be expected to die, according to data from China
  • The average length of a hospital stay for a coronavirus patient is 10 days – eight days for those who recover quickly; 16 days for those who need intensive care

Helped refine the code before the paper was published online two years ago, said some parts of the code looked like they were machine translated from Fortran’, an old coding language.

After growing pressure, the Imperial team released their code, which simulates homes, offices, schools and people movement, and sceptics were quick to point out it was 13 years old.

Bob Seely, MP for the Isle of Wight, today described the the modelling as ‘a national scandal’

-->