5000 deaths a day or just incorrect assumptions in modelling

[Garysmum]

Our favourite team at Imperial college have predicted this wave will lead to 5000 deaths a day if we carry on the status quo without further restrictions. I would imagine that means 5000 deaths a day without lockdown.
Best case scenario shows 3000 deaths a day if we continue as is.

Neil Ferguson also believes that Omicron is as serious as Delta - ie not a milder disease.

Are these numbers way off base - is the modelling based on incorrect or too wide parameters?

Personally when deaths hit over 1000 a day from Covid (apparently the peak was 1800 but I had given up with news by that point) I found it deeply unsettling.

Maybe I'm an idiot, but I would appreciate someone who understands statistics to explain some of this.

As we were all told at primary school, "After you've done your maths problem, look at your answer, look at the problem again, and ask yourself 'Does my answer make sense?' If it doesn't, you've gone wrong somewhere."

Ferguson has a track record for coming up with massive figures that don't pan out. I don't think I'll be taking this one too seriously; the SA data so far looks really different to their previous waves.

The irony is that yes, even assuming a good outcome, there will be a rise in hospitalizations and a fair amount of pressure in hospitals, but nobody will really care or bother because "Yes well, but we were told ten squillion deaths a day--hasn't happened, has it? The numbers like very good compared with the dire warnings."

We know that two doses of AZ means ZERO vaccine efficiency (vaccine efficiency being presenting with symptoms)

Severe disease/hospitalization is the metric we need, not "any symptoms at all"--people getting a cold is not a national crisis. This kind of post is irresponsible.

People have stopped listening. They need to end this ridiculous scaremongering.

I wouldn't trust SAGE to accurately predict what day it will be tomorrow. They howled their protest about various lockdown listings with dire warnings of thousands of cases/hospitalisations/deaths, only for the numbers to drop off a cliff and for them to admit they had no idea why.

They are the Boy Who Cried Wolf and they can collectively fuck right off.

@Viviennemary

People have stopped listening. They need to end this ridiculous scaremongering.

People might stop listening has to be the worst possible reason for deciding to discard scientific advice.

Dealing with a virus doesn't run to timelines like that. It doesn't matter how short some people's attention spans are, or how many times we have to do the same thing.

And blaming the scientists (which does seem to be what a section of the Tories are doing a lot of at the moment) is going for the wrong target. Especially when it's pretty clear that the whole point of a reasonable worst case prediction has been throughly misunderstood (tabloid headlines) and the full range of what the modelling really shows is barely reported on.

Its bad media reporting. Does show that a lot of people just read the headlines.

@IWannaWishYouANutNutsChristmas

Also,

Correct me if I'm wrong, but Neil Ferguson is correct isn't he?

There is no data that omicron is inherently any milder than delta?

Just data about the early stages of an omicron wave in one country, markedly different from ours in several ways, whee other variables are almost certainly in play.

As well as some unknown unknowns.

The news coming from South Africa has been positive over the past few days and I can't be dismissed as simply "SA is younger, it's summer over there too etc., and really does appear to suggest that Omicron is significantly milder.

The key metric is comparing to how SA is faring against Omicron compared to Delta. That automatically adjusts for all variables (apart from perhaps the season but we know that only impacts on transmissibility and Omicron is so transmissible that growth is extremely high irrespective of season) and hospitalisations are 91% below where they were at the same point in their Delta wave. And that's reported in Bloomberg, not in a tweet of some Covid denier.

Apart from in inimitable catastrophist Dr Ding, report after report from SA all
report something similar.

Sorry, I really should have proof read my last post - lots of typos!

Those of you who think Ferguson's models are spectacularly wrong & outdated etc...

Whose models do you like? There are lots of rival models to choose from. LSHTM?

Please link to the models that you think are better.
This must be easy to find, right?

SA and UK populations aren’t very comparable imo (I know some people in this thread think they are).
Similar population sizes but SA only has 5% pop over 65, it’s 17% in UK.
This could be a huge, huge difference when the main concern is not swamping your health service (3mil over 65 vs 10mil+).
Big differences between vaccine rate between countries as well, in UK’s favour, so that may offset our older population problem.
Too early to know anything at this point. See what the data from London looks like in 1-2 weeks.

Haven't his models always been wildly out, why is this one going to all of a sudden be accurate?

@herecomesthsun

On the one hand, all the models are potentially wrong, as the future is changeable and no one can know exactly what it holds.

On the other hand, people change their behaviour according to what the models say, so it has been suggested that the suggestions of much higher deaths in the first wave may have been quite accurate, if we hadn't had the lockdown.

If they helped people change their behaviour so that 1000s of lives were saved, they had a value. Even if the predictions therefore didn't come true.

This.

The worse case scenarios predict what will happen if nothing changes. Everything carries on like "normal". Then people change their behaviour and restrictions are imposed - so the worse case scenario doesn't happen.

It doesn't mean the prediction was inaccurate.

@Toastmost

Haven't his models always been wildly out, why is this one going to all of a sudden be accurate?

Haven't his models always been wildly out, why is this one going to all of a sudden be accurate?

This is the most fundamental and important type of critical question that could be asked.

I bet Johnson and his team have not even thought to ask it.

It could have a really thorough and credible answer. Having said that, if Fergusson had radically changed his model in some way, you’d have thought that itself would have filtered out of academia into the media.

If only the numbskulls of SAGE would defer to MNet's collective wisdom then the pandemic would come to a magical end. That said fuck this fucking bastard Omicron fucker, only a month ago I was truly believing learned minds that Delta was the end of the line mutation wise and we are on the way out. Maybe fingers crossed Omicron would turn out to be relatively mild and the last play of the fucker before it turns into a cold. That's gotta be the hope.

Better to be prepared with contingency plans than assume the best.

@Hazelnutbean

There is no credible information that omicron is inherently milder than delta

And there won't be for at least another 5 or 6 doublings.

SA has had a very different pandemic from us.

There's no reason to assume our omicron wave will look like theirs since our original covid, alpha, and delta waves didn't.

I don't believe that figure. I think it's scaremongering. It's mild for most.