Home > Covid 19, Covid Lockdown, Government, Health, Surveillance > I tackled a Sage Covid modeller on Twitter and it was quite the revelation.

I tackled a Sage Covid modeller on Twitter and it was quite the revelation.

By FRASER NELSON 19th December 2021. Find Article Here:-

Why does Sage not tell us the probability of its gloomy Covid ‘scenarios’?

Over the weekend, the latest Sage document arrived with some blood-curdling figures on what could await us if we fail to lock down. The omicron wave could be the deadliest yet, we’re told, killing up to 6,000 of us in a single day. This would be at least five times more than the peak of previous wave – and this from an omicron variant that South Africans say is far milder! We are a highly vaccinated country whose people are teeming with antibodies. Yet somehow, after all of these jabs we are, once again, sitting ducks.

But dig deeper, and the Sage story changes. The 6,000 is the top of a rather long range of “scenarios”, not predictions. The bottom is 600 deaths a day, which certainly would not pose an existential threat to the National Health Service. Why won’t they tell us how likely (or otherwise) these scenarios are? I was mulling all this when, on my Twitter feed, up pops the chairman of the Sage modelling committee Prof. Graham Medley. I thought I’d try my luck and ask him.

It’s a strange place, Twitter. People turn up who should not be there, but sometimes, if you’re lucky, you can get a response. Prof Medley is with the London School of Hygiene and Tropical Medicine (LSHTM), which last weekend published some grim omicron scenarios. A bank, JP Morgan, spotted a flaw: each LSHTM scenario assumed that omicron was just as deadly as Delta. “But evidence from South Africa suggests that omicron infections are milder,” it said. Adjust for this and Covid hospital numbers end up at just one-third of the January peak. So this omicron wave “would be manageable without further restrictions”. Quite a spot.

I asked Prof Medley: why not say so? When giving his scenarios, why couldn’t he say what JP Morgan had said: that if it’s as mild as the South Africans seem to think there could be no real problem and no need for lockdown? “What would be the point of that?” he replied. “Not a snarky question. Genuine to know what decision-makers would learn from that scenario.” To me, this seems odd: wasn’t the JP Morgan point rather relevant? That we might be able to live with Covid and carry on as usual without lockdown? Why not include that as a possible scenario?

“You know the answer,” he replied. “Decision-makers are generally only interested in situations where decisions have to be made.” But isn’t it just as vital to be told if action is not needed? I asked him straight. “So you exclusively model bad outcomes that require restrictions and omit just-as-likely outcomes that would not require restrictions?”

“We generally model what we are asked to model,” came the reply. “There is a dialogue in which policy teams discuss with the modellers what they need to inform them with their policy.” Again, quite the revelation. Until now, we’ve been told that policy is informed by the data: the impartial independent Sage scientists come down with their figures, and ministers act on this advice. Lockdown is always said to follow the science. But Prof Medley suggests the scientists are doing what they are “asked” during a “dialogue” with a pre-existing “policy”.

At this point our conversation was interrupted by “Reg”. “This entire exchange has left me open-mouthed,” he told us both. “To think of the livelihoods at stake here, mainly because they don’t see the need to model accurate outcomes as it will not make the government take any action. Scandalous.” Prof Medley gave “Reg” the same cryptic reply. “We model the scenarios that are useful to decisions.” He then left, and started to responding to others who asked him if he’d model in his Speedos if asked.

What to make of all this? Is this how Sage really works? Are we all going to be locked down again based not on evidence-based policymaking, but policy-based evidence-making? Since our Twitter exchange went viral I’ve been contacted by a few government ministers saying they were alarmed to think Sage modellers are not giving the probability of various outcomes and cooking up gloomy scenarios to order.

Perhaps we’re all reading too much into what Prof Medley said. But given what’s at stake, these issues may be worth clearing up before Sage “scenarios” are used to lock us down again.

  1. Sam
    January 3, 2022 at 12:45 pm

    Not the first time I’ve heard of this. Happens when organisations sponsor scientific studies too.

  1. February 23, 2022 at 1:25 pm

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: