Health Protection (Coronavirus) (Restrictions) (England) (No. 4) Regulations 2020

Baroness Noakes Excerpts

Wednesday 4th November 2020 Lords Chamber

[Baroness Noakes (Con)]

My Lords, my amendment is about the evidence for the lockdown in the order before us. The Government have no easy task in finding the optimal policy responses to the virus. They deserve the best advice they can get, but I am not sure they are getting it.

I hope that at least some noble Lords have read the book published earlier this year by the noble Lord, Lord King of Lothbury, and John Kay, entitled Radical Uncertainty. It warns of excessive reliance on probabilistic reasoning and modelling. Its core insight is that we need to face not knowing the answers when confronted with massive uncertainty. Instead people should stand back and ask themselves, “What is going on here?”. If that had been the focus of policy discussions, I do not believe this destructive lockdown would have been the solution.

On Monday, my right honourable friend the Prime Minister said in the other place that the data now suggest that our health system will be overwhelmed. The so-called data on which he was relying were not facts but modelled numbers. Modelled outcomes are only as good as their base data and assumptions.

As we have heard, last weekend the Chief Scientific Adviser showed a slide that suggested a scenario of 4,000 deaths per day in December. According to that chart, there should have been 1,000 daily deaths last weekend. There were fewer than 300. Four thousand deaths per day is virtually impossible, even using SAGE’s inflated infection fatality rate of 0.7%, and implies that around 600,000 people will be infected every day next week. None of the data points that we have on infection levels come anywhere close to that.

The adviser then showed something called the SPIM—medium-term projections of hospital admissions, deaths and NHS bed usage. These indicated bed demand later this month apparently shooting way ahead of the March-April peak and ahead of the assumption that only about 20% of NHS beds can be used for coronavirus patients. However, the detailed modelling assumptions have not been made public. Instead, the small print says that this is a consensus forecast, based on several models, none of which assumptions has been made public. Does that sound like a good basis for a momentous decision to close the country down?

We learned yesterday from evidence given to the Science and Technology Committee in the other place that these models were based on earlier, not up-to-date, data. The assumptions took no account of the recently introduced tier system, despite those areas already showing reduced infection rates and hospital admissions. Leaked NHS data show that, despite a few local hotspots, intensive care bed capacity is around normal for this time of year.

Some have suggested that those scenarios and forecasts were deliberately calibrated to produce the maximum fear in the general public and thereby generate support for another national lockdown. The Daily Mail has also called out the way in which last Saturday’s presentation cherry-picked data and presented it in a way that would make even Liberal Democrats blush. Is this all a deliberate plot to provide cover for the curtailment of our liberties? I could not possibly comment. However, I know that the Government should be alert to dangers of groupthink and the self-reinforcing nature of scientific cliques. The history of science is littered with views, such as whether the earth is flat, that remained widely held beliefs long after clear evidence to the contrary emerged.

There is no independent challenge to the SAGE analysis. There ought to be a place for techniques such as red teaming that robustly challenge house views. There certainly are scientists out there, for example in the Oxford Centre for Evidence-Based Medicine, who could provide that challenge. Was anyone asking, “What is going on here?” I do not think so. Otherwise, they would have compared infection rates and R numbers in the models with the latest data points and would have noticed that the models use R numbers that are ahead of current numbers, even though the R numbers from the ONS data have been falling. As we heard, the Liverpool R number is already below 1. The Prime Minister said several times in the other place on Monday that an R number of 1 or less was the aim of the lockdown. As we have heard, the latest ZOE survey data show that infections are past their peak and the R rate is already at the magic number, 1.

If the Government had sought independent challenge, they might well have concluded that this heartless order was unnecessary and, as a minimum, dialled back their scary charts. The Prime Minister cares sincerely about civil liberties but I suspect that he has more of a way with words than numbers and is in thrall to a tightly knit group of scientists with a single world view. It is time for him to ask, “What is going on here?”, and take back control of the coronavirus agenda.

--- Later in debate ---

Tabled by

[Baroness Noakes]

At end insert “but that this House regrets that the modelling used to support the claims that (1) the National Health Service would be overwhelmed, and (2) daily deaths from COVID-19 would be 4,000 or more, has not been subjected to independent review and challenge.”

Baroness Noakes’s amendment to the Motion not moved.