Coronavirus: Downing Street admits error in graph used to justify lockdown

Mistake does not invalidate underlying analysis that action needed, says No 10

Downing Street has admitted that one of the graphs used to justify the move to an England-wide lockdown was incorrect.

The chart, displayed at Boris Johnson’s televised press conference last Saturday, indicated that England could see up to 1,500 deaths a day by early December if additional action was not taken, but this figure has now been revised down to 1,000 deaths a day by 8 December.

A No 10 spokesperson insisted that the error did not affect the underlying analysis of the prime minister’s Scientific Advisory Group for Emergencies (Sage) that tighter restrictions were needed to avoid hospital admissions and deaths topping the first-wave peak in the spring.

The spokesperson also rejected claims that a test which is central to the prime minister’s Operation Moonshot plan for mass Covid testing was unfit for use because it missed more than 50 per cent of positive cases in a trial in Salford.

The Lamp test had been trialled in three other labs and been shown to have nearly 80 per cent sensitivity, rising to 96 per cent among those with a high viral load, making it an “important tool for testing for coronavirus”, said the spokesperson.

The Lamp test is not being used in the current mass testing trial in Liverpool, where individuals are being given “lateral flow” rapid-turnaround tests alongside conventional swabs.

The No 10 spokesperson rejected suggestions that the error in the chart produced by Sage modelling group SPI-M meant that the decision to put England into lockdown was premature.

The medium-term projections remained the same, but the line showing the upper end of the range for potential hospitalisations has been revised downwards from 9,000 to 6,000 a day.

“In the specific graph, SPI-M’s central medium-term projections remain the same for admissions of the next six weeks,” said the spokesperson.

“But we accept there was a mistake in plotting the confidence intervals … which we corrected as soon as it was identified.

“There was no error in the underlying analysis. The data still shows and the consensus remains that without action we would breach the first-wave hospital admissions and deaths within a matter of weeks.”

The error emerged a day after chief scientific adviser Sir Patrick Vallance was issued with a public warning from the Office for Statistics Regulation over the use of data to justify the current lockdown.

Ed Humpherson, ODR director general for regulation, warned there was a danger the public could become “confused” by a long-term projection of a possible 4,000 daily deaths, undermining confidence in official statistics.

The No 10 spokesperson said Mr Johnson continues to have “full confidence” in Sir Patrick.