Dear Ed,   

Thank you for the letter of 19 February on the presentation of modelling within the UK Covid-19 Inquiry’s Module 2ABC report. Your Office is clear that public bodies must use statistics, data and wider analysis transparently and with integrity, clarity and accuracy. You are concerned that the Inquiry conveyed too great a level of certainty when referring to the results of modelling.  

I note that you are not questioning the Inquiry’s conclusions, including those on the timing of lockdowns or the contents of the report itself.  

The focus of your letter is this extract on page 5 of the Executive Summary of the report: “Had a mandatory lockdown been imposed on or immediately after 16 March 2020, modelling has established that the number of deaths in England in the first wave up until 1 July 2020 would have been reduced by 48% – equating to approximately 23,000 fewer deaths”.  

I accept that in trying to condense and simplify the 800-page Module 2ABC report into a short, accessible Executive Summary, framing of the statistic through use of the word “established” was too definitive and did not reflect the uncertainty inherent in all modelling. The main body of the report uses more appropriate language in saying it is “estimated” that locking down one week earlier “could have” led to a reduction of 23,000 deaths.  

The Inquiry’s findings are grounded in the whole of the evidence received. Baroness Hallett concluded that the outputs of this model were useful in illustrating a broader finding about the timing of the first lockdown based on the totality of the evidence before her – both written and oral. She observed:  

“Indeed, the Inquiry accepts the consensus of the evidence before it that a mandatory lockdown should have been imposed one week earlier”1 and that doing so would have saved many thousands of lives during the first wave. Omitting, in its Executive Summary, the inherent uncertainty within one model does not change that overall finding of the Module 2ABC report.  

The academic paper (“Knock et al”) on which Professor Ferguson’s evidence was based – and from which the 23,000 figure was derived – did not make clear that it accounted for the impact of voluntary restrictions in the week before a mandatory lockdown. You say that you confirmed this with Prof Ferguson but this does not change the position. In the body of the Module 2ABC report, the Inquiry sets out the inherent uncertainty in modelling of this kind and the fact it is based on assumptions. The Inquiry accepts, however, that it did not sufficiently express this uncertainty in the Executive Summary.  

Having identified the issue about the framing of the modelling data in the Executive Summary to the Module 2ABC report, the Inquiry has introduced stronger quality assurance of the presentation of statistical data in the documents that accompany main reports such as summaries. The Inquiry will publish eight more reports in the coming months and in each we will seek to ensure the highest standards of statistical analysis and communication.  

The Inquiry will be publishing this letter on our website alongside the Module 2ABC report page so that future readers are aware.   

Yours sincerely,  

Ben Connah  

Secretary  

UK Covid-19 Inquiry