jump to navigation

COVID-19: What’s Going To Happen Now March 24, 2020

Posted by mwidlake in biology, COVID-19, off-topic, Perceptions, Private Life, science.
Tags: , , ,

<< COVID-19 Basics. What it is & what it does to us

  COVID-19: What can we do to Reduce Social Distancing >>

I thought I’d record what the scientific evidence and epidemiological modelling is saying about what is going to happen in respect of COVID-19 in the UK (and, to some extent, elsewhere) over the next weeks and months. As with my intro to COVID-19 this post is mostly “for me”. I’m sharing it but please, please, treat all of this post (not the science I link to!) with some scepticism.

The figures are shocking so I want to spell out right at the start that, if our governments does what it needs to do and does it right (and over the last 2 or 3 weeks the UK government has fallen a tad short on this, but it’s improving) in the end over 99% of us will be OK. If they get it wrong, it’s more like 97% of us will come through this.

And, I feel it is important to say:

90% of even high risk people will also be OK.

I strongly feel that the message is constantly that it is the at-risk people who are dying and not that most people at risk will be OK. Yes, COVID-19 is more of a danger to those over 70 and those with underlying medical conditions, but with the media and government constantly saying “the people who died are old” etc it makes it sound like COVID-19 is a death sentence to them – and it is not.

Yes, I’m quite angry about that that poor messaging.

Source of Epidemiological information

ICU beds needed per 100,000 people

My main source is This paper by Imperial College in collaboration with the World Health Organisation and British Medical Research Council. If you can, please read this paper. It spells out how COVID-19 will spread and what happens when the NHS intensive care unit (ICU) beds are all full. It’s a hard read in two ways.  It is technically dense; and it says things people are still refusing to believe:

  • If we had done nothing and had an infinite number of critical care beds, it would burn through the population of the UK (and all other countries) in 3 months, infecting 81% of people. At that point herd immunity stops it.
  • In the UK 510,000 people would die (COVID-19 kills about 1% of people even with ICU treatment). 2.2M would die in the USA.
  • At the time of publication of the report, the “mitigation” plans by the UK government would have failed to stop even more deaths (more than 1%) as the NHS would have been overwhelmed by the 2nd week of April.
  • At the peak we would have needed 30 times the number of ICU beds we have.
  • The paper does not fully spell this out, but if you need an ICU bed and there is not one, you will almost certainly die. Thus the death rate would be more like 2.3% {Note, that is my figure, I have not spotted it in the report. It is based on 4.4% of the population needing hospitalisation and 30% of them needing critical care, figures that are in the report}. I’ll let you work that out based on the UK population of 66.5 million. OK, it’s about 1.17 million.

These figures are truly scary. They won’t happen now as it shocked our government enough to ramp up the social isolation. If anyone questions why we need the social isolation, give them the figures. If they refuse to believe them,  tell them to read the paper and various articles based on it and point out where they are significantly wrong. If they won’t, thank them for their baseless “opinion”.

The calculation of 510,000 deaths in the UK did not factor in self-isolating naturally, as we all saw people fall ill and die. That would slow down the disease.

However, if the hospital is full to absolute bursting capacity with COVID-19 patients, any person who needs ICU care for other illnesses (cancer, cardiovascular disease, stroke) or accident. How do you fit them in? Deaths for other reasons will increase.

One thing I am not sure of is that in the paper critical care is stated as “invasive mechanical ventilation or ECMO”. If you need just a ventilator and one is not available, I’m pretty sure you would also be likely to die or suffer brain and other organ damage from oxygen deprivation.

As I understand it, this report is what made the UK and other governments take COVID-19 a lot more seriously and really understand the need to implement strict social isolation.

I’d like to say why I put so much trust in this source:

  1. The three organisations behind it are all highly respected (WHO, MRC, and Imperial College)
  2. They state clearly at the top their assumptions – the R number, incubation period, types of social isolation, the percentage of people who will comply with each one.
  3. They created a model that was then verified by running the numbers and seeing if it predicted what had happened in reality to that point.
  4. The subject matter experts I follow have all endorsed this piece of work.

Mitigation or Suppression

The Imperial College report spells out the distinction between Mitigation and Suppression:

Mitigation is where you reduce the R number (the number of people each infected person in turn infects) down from the natural number of around 2.4 but it is still above 1. At this rate the disease continues to spread and the number of cases per day continues to increase, but more slowly. The idea seems to be that it would lead to herd immunity. This was the UK governments aim until Monday 16th March.

Suppression is where you reduce the R number below 1. Within a few weeks the disease is no longer spreading. But it is still there in the population. This is what Wuhan did and Italy is making progress on.

To achieve mitigation the government isolated people infected, asked those who had had contact with them to self isolate, and asked us all to wash our hands and keep a distance and think about working from home. The impact on daily life, business, the economy is minimal. Further steps would be introduced later, like closing universities and schools.

The Imperial college report demonstrated that mitigation was a terrible idea as the number of cases would still explode, but just be delayed a little, and the NHS would be absolutely overwhelmed.

The graph at the top of this article shows the mitigation steps being considered and how it only shifted the curve and did not lower to anywhere like the NHS ICU capacity. It was simply not enough.

Isolation involves the sort of steps most of us would have previously thought only an authoritarian regime like China or North Korea could manage. Schools, universities and non-critical business shut, everyone not doing a critical job made to stay at home except to buy food etc. Basically, Wuhan. And now Italy is doing very similar. As of the 23rd March the UK is following suit.

Most western countries are now implementing many of the steps needed for isolation levels that will suppress COVID-19, but not all the steps needed.

The graph to the right shows the impact of two implementations of Isolation, both implementing several measures but the orange line does not include closing schools and universities. The green line does. The green line keeps the number of cases within the NHS ICU capactiy, the orange does not. That is why schools and universities were closed.

The graph also makes the point about the main problem with Isolation. It is only stopping the virus spreading, it is NOT getting rid of it. Remember, no one is immune unless they have had COVID-19. When the steps to enforce isolation are relaxed, COVID-19 will burst back.

This is potentially the position that China is in. They have locked down Wuhan province tightly and it worked. The number of cases there rocketed even after the lock-down but have since reduced, almost as fast as they increased. China as a whole now have very few new cases. The lock-down is being relaxed as I prepare this post. Epidemiologists expect the number of cases in China to increase again.

The degree to which either mitigation or suppression is enforced obviously impacts society and commerce. The Imperial College report makes the point that they are not addressing those concerns, they are simply saying what social isolation changes will have what effect on COVID_19 spread, deaths, and the ability of the NHS to cope.

Delayed impact.

UK daily cases to March 20th, Italy deaths to March 20.

This next point is being made widely, by both non-scientific observers and the scientific community, but I want to re-iterate it as it is so far being played down by government (which could be changing at the very moment I am typing).

There is no way to avoid the huge increase in COVID-19 cases and deaths that are going to happen in the UK over the next 2-3 weeks. Expect our levels to be the same levels as Italy. In fact, expect them to be 20, 30% higher. This is because the UK government were too slow to lock down and did it in stages when, based on the epidemiology, we should have shut down totally on Monday 16th when the paper I reference was published, or within 2 days to allow for planning.

Up until now COVID-19 has been spreading exponentially (1 person has it, passes it to 2-3 people. They pass it to 4 people who pass it to 8…16…. 32… 64… 128… 256… 512… 1024). This has been seen in the way the number of case had double every 3-4 days, deaths are now following the same pattern.

The two graphs to the right show the number of cases in the UK to the 20th March above, and the number of deaths in Italy to the 20th. They look like the same graph as they sort of are. This is how something grows exponentially when the growth rate is the same – the same as both cases and deaths are caused by the same thing.

(these graphs are from Worldometers – I use this site as I think the John Hopkins site has more incorrect information on it).

Covid-19 takes on average 5.1 days to show symptoms from when you catch it (this can be up to 2 weeks – with all these averages there will be some cases which are two or three times as long). It takes less time, 4.6 days on average, from when you catch it to when you spread it. So you can spread the disease before you get ill. And some people do not get ill (or only very mildly) and spread it. Like “Typhoid Mary”. If you are going to be ill enough to need hospitalisation it takes 5 days from first symptoms for you to deteriorate to that point.. At this point you will be admitted to hospital, tested, and will join the number of confirmed cases. If you are going to die (I know, this sounds really callous) that is another few days. The report does not spell it out but going on the figures they use for time spent in intensive care in the model, about a week.

Add it all together and someone who dies of COVID-19 today caught it 15-20 days ago on average, so the spike will be delayed that much.

Yesterday, 23rd March, almost total lock-down in the UK was announced. Cases and deaths will rise for 20 more days in the UK. Exponentially. To Italy levels, maybe 20-30% higher. Then they will plateau for a few days and drop quickly, depending on how well people respect the social distancing or are forced to. I am expecting over 9,000 will die in this first spike, with a peak number of deaths between 750 and 900 in one day. Sadly my predictions so far have all been correct or a little too optimistic.

That is the reality and that is why we are seeing the actions of our government that have never been seen outside World Wars before.

Three choices – or is it four?

To summarise the above, there were 3 choices available to the UK (and all other countries):

  1. Let COVID-19 burn through the population in 3 months. It would kill 2-3% of the population as the NHS collapsed and also anyone who needed medical treatment during that time would probably not get it. During the 3 months lots of people would have “bad ‘flu”. 80%  of survivors would be resistant to COVID-19 for now.
  2. Mitigate the impact by the measures implemented in stages during mid-March, reduce the impact a little and stretch the curve a little, and have 1.5-2.5% of the population die over 4 months. 70% of survivors {my guess!} would be resistant to COVID-19 for now.
  3. Suppress COVID-19, 10,000 dead and everyone in lock-down until “something changes”, which could be 18 months or more.  A tiny percent, maybe 5% {my guess} resistant to COVID-19.

The UK government chose option 3, after considering 2 for a while (and thus increasing the death count by, hmmm, 3,000 in that first spike).

The “something changes” in option 3 is that scientist create a vaccine for SARS-COV-2, the underlying organism to COVID-19, or we have a quick and reliable immunity test for it that allows those who have survived the disease to move about unrestricted. See further down in this post. Most of us stay in lock-down until “something changes”

But this Imperial College paper has a solution 4:

Turning social isolation up and down

  1. sorry, 4. I can’t get the layout to work. solution 4 is to
    1. suppress.
    2. Let the known bubble of cases come and deal with it.
    3. Once it has passed, relax (not remove!) the Suppression rules to let business and normal life start up again.
    4. Monitor the number of COVID-19 cases coming into ICU.
    5. When it hits a threshold, back to total lockdown and deal with the next bubble.
    6. Repeat.

It is a clever idea. No one wants to stay at home until a vaccine is created in 18 months. Economically, total lock-down until we have a vaccine would be a disaster. So varying the lock-down based on NHS demand indicators would allow some relief from the restrictions. But not back to normal.

Option 4 comes at a cost. More people will die reach time you relax the lock-down, depending on what is allowed. Much of the rest of the paper details this plan and, based on the figures they state at the top of the report in respect of how many people will abide by the rules, what different isolation strategies and key triggers (how many new COVID-19 ICU cases in a week) to increase isolation levels, gives death rates varying from 8,700 to 120,000. This also takes into account a range of R values (how easy it spreads naturally) as there is still some uncertainty about this.

The paper makes one thing clear – we would need to maintain the isolation levels for suppression for 2 years – their cautious estimate of how long it will be until we have a widely available vaccine.

The best case is deaths creep up (after the initial surge we can no longer avoid) with very strong lockdown only relaxed at very low levels of ICU cases and deaths. I personally doubt very strongly that enough people will abide by the rules for long and, as people start ignoring them, others will feel “why should I play by the rules when they don’t”.

I do not have anything like the understanding of human nature needed to predict how people are going to react so I won’t. But the figures being bandied around a few days of keeping UK deaths to 8,000 or less seem utter fantasy to me.

The “The hammer and the dance” paper…

Some of you may have come across “The hammer and the dance”, which is based on a paper by Tomas Pueyo on “Medium”, a home for science papers that have not been verified by anyone. I would not normally look at things here very much but several people have mentioned the paper or even linked to it. If you recognise the term, you will probably recognise the “dance” part as choice 4 above.

Context is paramount

Lots of numbers are being thrown about, but to understand the true impact of COVID-19 those numbers need to be interpreted in light of some general background.

Let’s start with the base rate of mortality. In the UK there were 541,589 deaths in 2018. That give 9.3 deaths per 1,000 residents. See the office for national statistics article for this figure. Over the year that is 1,483 deaths a day, from all causes. People keep on insisting on comparing COVID-19 to influenza. I’ve struggled to get a definitive number of deaths due to Influenza in the UK but it seems to be between 8,000 and 17,000 a year. Let’s take 17,000 as a top estimate, that is 46 a day.

(you may wonder why it is hard to say how many people die of influenza. Well, influenza kills people who are already seriously ill and likely to die anyway, and I believe not every death attributed to influenza is tested for sure to be influenza.

Our key figures are 1,482 deaths by any means a day and 46 a day from influenza, in the UK.

On the 21st March 56 people in the UK died of COVID-19. More than Influenza, about 4% of the daily mortality rate. Bad, but nothing that significant. In Italy, 793 people died of COVID-19 on 21st March (and it looks like that might be the peak). Our figures in the UK for known diagnoses and deaths are following the Italy pattern very closely (for very good scientific reasons) just 2 weeks behind – 15 days to be more precise. In 15 days the death rate for COVID_19 is likely to be very similar to Italy so, despite my hunch the UK peak will be higher, let’s use Italy’s peak number:

  • 50% of the total death rate for everything in the UK.
  • And 17 times the death rate by ‘flu.

So COVID-19 is incredibly serious,  but it could have been worse. It looks like for a period at least, for each country, it will increase the daily death rate by 50% and maybe more. But it is not killing a large percentages of the population.

I’ve seen some scare stories about this disease sending us back to the dark ages as it kills half the population of the world. Rubbish. It might stop the world population growing for a year.

Why will social distancing last 18 months?

No one is naturally immune to COVID-19 until they have had it. Let’s assume that once you have had it you are immune for several years, as you are with many other viral diseases (Influenza A is a special case as changes so fast and in a way that reduces the effectiveness of both vaccines and immunity via exposure).

We could let COVID-19 spread naturally or at least in a contained way – but it will overwhelm our health services as discussed, and 1-3% of us would die.

The other way is to create a vaccine, which gives immunity or partial immunity without having the disease (or maybe a very mild version of it). Vaccination works, it rid us of smallpox totally and, until the loony anti-vaxxer movement got going, it was vastly reducing measles, rubella and many other diseases.

But creating a vaccine that works is hard. Lots of biomedical scientists are working on it and we might get lucky and someone comes up with a very effective vaccine that can be created in bulk, but by lucky we are still talking months. (There is at least one early trial running – but that absolutely does not mean it will be available next month!)

Any vaccine has to be tested, proven effective, and shown not to itself harm.

All of this is why specialist in the field all say “18 months”. It’s a guess based on science and experience. It could take longer, it could be only 12 months, it might be that an initial vaccine is only as effective as the yearly flu vaccine (the flu vaccine generally protects 40-60% of people – see  this oxford university paper).

We can test for if people currently have COVID-19, the test is accurate and relatively cheap. It checks for the RNA of the virus, an established diagnostic practice. Production of the test is being massively increased and improved and we need that so we can better track the disease and accurately identify who has the disease and put them in isolation. In the short term, wider testing will help a lot and those countries that have gone in for huge testing efforts (South Korea and Singapore are examples) have done well in containing COVID-19.

The other tool we really need is a test for immunity, which is usually for the antibodies to a disease. Again, these tests take time to devise. If we could identify those who have had the disease (but were not tested) and are now immune. They would not need to be isolating themselves. A small and growing part of our population could return to normal. But we have no idea when such a tool will be ready, how accurate it is, how cheap it is to do etc.

Finally, scientists need to work out if immunity to COVID-19 is long-lasting, for how long, and if the immunity is strong or weak. We just do not know yet.

Until we have a vaccine (ideally), or the immunity test (it would really help) we have to suppress COVID-19 via social distancing etc.

Basically we are sleeping with a tiger. Best not wake her.


All of what I put here is based on what is said by experts, scientists, epidemiologists. I’m just pulling some of it together. As I said in the previous blog, I am not an expert in any of this. I’ll make it clear when something is my opinion. I also want to highlight that I only look at sources that I feel are backed by good science. The only information I take from the government is official statistics on cases & deaths. I’m heartened that our government is now taking the spread and impact of COVID-19 more seriously but I remain angry that the experts told them what was coming weeks ago and they were slow to act, putting business concerns before lives.

Any mistakes in this blog post are mine. There are bound to be a couple.

I would love to hear about sources of information you feel are good. I had several excellent sources pointed out to me after my last post, including being corrected on a couple of counts – which I am very happy about.

However, I will probably ignore anything based on rumour or anecdote. Ginger & Garlic are not going to boost your immune system and protect you, quinine is almost certainly not a magic protector. If you have a peer reviewed article in a reputable journal or the support of a respected epidemiologist to back those opinions, then let me know.



No comments yet — be the first.

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 )

Google photo

You are commenting using your Google 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: