SAGE Modelling From May Last Year Said Approach Recommended in Great Barrington Declaration Was Least Bad Alternative to Lockdown

18 April 2021

by Glen Bishop

In May last year, after the UK had been plunged into lockdown for the first time, the Warwick SAGE SPI-M team released a paper analysing the potential exit strategies and their projections for deaths under each scenario. Whether their projections were correct is debatable because of the vagueness of their assumptions, but more importantly the paper inadvertently highlights why lockdowns are farcical.   

Below is the graph projecting the pandemic course under different scenarios of lifting lockdown last year. The graph is poorly labelled and hard to follow, so bear with this explanation. It models five scenarios of releasing lockdown, where restrictions are eased on the May 7th by varying degrees, from no release to full release at intervals of 25% i.e., 0%, 25%, 50%, 75% and 100%. With these levels of restrictions maintained until January 1st 2021, at which point, they are all released fully. The effect of this eventual release is shown in the ‘no controls’ side of the graph which is a continuation of the five same lines into 2021.

As shown on the left-hand side representing 2020, the lines of 50%, 75% and 100% release of restrictions, projected an immediate resurgence in the virus with predictions of daily summer death tolls between 1000 and 4000 at their peaks. Clearly this did not happen – deaths last summer in the UK flatlined. Whether this is a failure of the modelling depends on the percentage by which we released lockdown last summer, which is hard to quantify. Google mobility trends for the UK in 2020 show that visits to retail and recreational venues had increased from 77% below normal in April to only 14% below normal by August. This would imply a lifting of restrictions far greater than 50% occurred. So, there is certainly a good argument that this modelling was over-pessimistic, though it was a gradual return last summer unlike the modelling assumptions of immediate relaxation from the May 7th

Perhaps more interestingly though is how these projections admit lockdowns to be no more than delaying the inevitable. The two lines (0% and 25% lift) which kept the virus suppressed while in place through 2020, are the two lines showing a large resurgence in the months following when they are eventually lifted on January 1st 2021 on the ‘no controls’ side. Whereas the three scenarios which lead a pandemic last summer (50%, 75% and 100%) experience no resurgence after January. This is essentially SAGE modelling showing that, at best, restrictions were a delaying tactic and a gamble on a vaccine arriving. Were the Government and SAGE lying when they said, “Three weeks to flatten the curve!” or did their approach change?  

This paragraph from the paper is even more important:  

A completely uncontrolled outbreak is predicted to lead to around 200,000 deaths, approximately 2 million QALY losses but no lockdown impacts. If the current controls are maintained until the end of 2020, then we predict 39,000 deaths this year [2020], but a further 159,000 if controls were then completely removed. Regional switching and age-dependent strategies provide alternative exit strategies in the absence of pharmaceutical interventions. Of these, the age-dependent shielding of those age 60 or over generates the lowest mortality and also the lowest lockdown scale, thereby minimising socio-economic disruption. However, it is unclear if a protracted lockdown of this age-group would be practical, ethical or politically acceptable.

This is a clear acknowledgement that the Great Barrington Declaration approach of shielding the elderly would have led to the “lowest mortality” and “minimising [of] socio-economic disruption”, while querying whether it would be “practical, ethical or politically acceptable”. Ironically, most of the over-60s have been in strict lockdown for longer than this approach would have required, alongside the whole population. Apparently, what wasn’t ethical to do to the over-60s, became ethical to do to everybody and for much longer. Why did the SPI-M teams ignore their own models showing that was the best option? My guess is that this approach would have made the Government’s scientific advisers largely redundant and required the politicians to stand up, make a decision and lead, rather than kicking the can down the road and virtue signalling about saving the NHS.  

Additionally, their estimates of 2 million QALY (quality adjusted life year) losses from an “uncontrolled outbreak” show how disproportionate the response has been. Sixty seven million people in the UK will have lost 1.25 years of their life to harsh restrictions between March 2020 and June 2021. That is 84 million quality years of life that we will not get back and a massive degradation of quality of life for many years to come, not just for the 67 million brits alive today but those yet to be born who will be paying the Covid debts.  

The current death toll stands at 127,000 in the UK, so despite all the damaging restrictions, we are only 73,000 deaths away from the 200,000 deaths estimate for the ‘let it rip’ with no shielding approach, used to approximate two million QALYs losses. So, only 730,000 life years (assuming 10 QALYs per death but ignoring lives lost due to lockdown effects) have been saved, assuming their estimates are correct, and the cost has been 84 million life years spent under harsh restrictions on life or complete house arrest. In other words, 115 people have been placed under house arrest for a year, so one person can live one year longer – and the person who’s life has been extended is overwhelmingly likely to have been in their 80s with more than one underlying health condition. It is even more damning to look at the estimates for ‘letting it rip’ in the young last year and shielding the over-60s until January 2021. Here they predict 138,000 deaths. Only 11,000 more than we have actually had. Were all healthy young people in the UK placed under house arrest for a year to save fewer lives than are lost in a bad flu season?  When you bring the numbers down to a level the human mind can comprehend, it is clear that hysteria is ruling the day.  

There are about 100 professors on SAGE. Have none of them broken down the numbers to an understandable level and thought about what they’ve been advocating rationally, or are they all just carrying this on for the research funding and the chance to appear on television? Unfortunately, when healthy people lose a year of life to lockdown measures, it is ignored in the cost-benefit analyses. Quality of life is considered an irrelevance; only longevity of life is assumed to have value in this brave new world.

Glen Bishop is a second year maths students at Nottingham University.