Martin Widlake's Yet Another Oracle Blog

Covid-19: The New Variant and the NHS December 29, 2020

<<- Long term hopeful, short term worried

As I said in my blog post a couple of days ago, I’m very concerned about the new variant of SARS-CoV-2 that has been spreading through the UK and is now being found in countries all over the world. My main concern is that this could be what pushes our health services beyond the limit of what they can stretch to and, as a result deaths will jump up – and not just from Covid-19.

New Variant Impact

In my last post I highlighted the new variant of SARS-CoV-2 that is more infections (spreads more easily), but said that there was little evidence that it was any more fatal. Understanding what was going on was hampered at that point as we had hit the festive period and, with the best will in the world, everyone needs a break at some point. New data on hospital admissions, virus sequencing, case numbers were all missing or affected. Scientists studying aspects of Covid-19 were reminding themselves what their partners, kids, and pets looked like after what must have been a heavy year. Now the new information is coming out, as is the analysis by relevant experts.

There is a paper detailing this new variant by Public Health England which was published on 28/12/20. Much of the below is derived from that, but is backed up from many tweets and bits of evidence from the scientific community.

This new variant is know by a few names:

• VOC 202012/01 – Variant of Concern identified in 2020 month 12, number 1
• B.1.1.7 – the phylogenetic name of the variant (I think!)
• 20B/501Y.V1 or simply 501Y.V1 – the identifier given by Nextstrain

B.1.1.7 has many mutations from the original SAR-CoV-2 virus (this STAT article states 17 mutations, the tracking page I mention below lists 17 SNP mutations, this overview by the CDC on VOC 201212/01 lists 20 SNPs and 3 deletions and seems to be the best source of information on this. I’ll explain all the mutations better in a later post) . Mutation is not unusual, viruses change all the time. Each time a virus is copied (and that is how viruses like coronaviruses reproduce, there is no sex, they are identical clones of their only parent) the RNA is copied and occasional mistakes are made and thus changes, mutations, happen. The most common change is a Single Nucleotide Polymorphism, or SNP. One letter of the 30,000 in the viral genome changes.

A single SNP change to the SARS-CoV-2 RNA does not seem to be enough to change the virus into a significantly more infective version (or more lethal, or more likely to infect children, or change it’s behaviour in a way to make it more dangerous). If it did, we would have seen this already – the virus has been so successful in spreading in humans and thus reproducing and so those SNPS occurring, that most individual SNP mutations that are possible will have happened by now (there is evidence for minor change by them though, but that’s for another time). It is going to be a combination of two or more changes I think that has altered the transmissibility.

B.1.1.7 has several changes to the gene that creates the spike protein.

The paper from Public Health England I reference reviews the data that was initially presented to the UK government (on around the 19th December I presume) and resulted in their initial analysis of the 21st, which this paper links to.  This review considers the degree to which the new strain transmits more easily and possible reasons why. It can  be summed up as saying:

• This new variant is indeed spreading faster.
• it is becoming the most dominant strain in all the areas it is in.
• It’s ability to spread to others (secondary attack rate) is increased by about 55%.
• It is not spreading faster as it is more successful in re-infecting people who have already had Covid-19.
• There is no evidence it results in more hospital stays or is more fatal.

I’m not sure the evidence is yet firm that this new variant does not also increase the severity of the illness a little as there are too few cases to go on, but it does not like there can be a huge increase. Usual caveat, I’m no epidemiologist.

I’ve also looked at a paper by Nick Davies’ team at the London School of Hygiene and Tropical Medicine.

They considered 4 possible methods by which the new variant (they use the VOC202012/01 name) could be causing the rapid spread of the new variant

• A) Increased Transmissibility
• B) avoids current immunity
• C) Children being more susceptible
• D) shorter viral generation time

As you can see from the graphs, the model based on (A) Increased Transmissibility fitted the date better than anything else.

You may be aware of the new variant in South Africa that is also more transmissible. This is not the same as B.1.1.7, for example, it does not have the 69/70 deletion mentioned in Public Health England paper that is used as a proxy to identify B.1.1.7 in the UK population (again, more information later on the details of the new mutation). So this deletion either is not key to the increased transmission or else there are two methods by which the transmissibility can be increased (now, that’s a worrying thought).

There has been a lot of other analysis and commentary from the scientific community to back up the hypothesis that B.1.1.7 spreads 50%-55% faster.

Why Is 50% Faster Spread So Significant?

Why is this significant? Wouldn’t 50% more lethal be more of a worry?

No. The reason an epidemic is so scary and has such an impact is down to exponential growth. To use an extreme example such as exists before a new disease is recognised and steps taken to control it, If 1 person infects 2 people who infect 4, 8,16,32… Ten duplications later and you are at 1,024 infected people. If each person infects 3 then it goes 1 person, 3, 9,27, 81…ten tripling’s is 59,049. If you know how many people each infected person will infect (the R number) and how long it takes for an exposed person to themselves become infectious, then you can calculate how quickly the disease will spread and grow. So the transmissibility is key.

Adam Kucharski put it better than I can (if you are on twitter and you are interested in Covid-19 science, if you are not already following Adam then I highly recommend you do, and then follow some of the people he follows). This is how he explained it:

Here in the UK the number of cases and, more importantly, hospital admissions have been shooting up. You cannot compare case from the spring to now as testing now is orders of magnitude improved compared to the shambles back in April. But hospital beds occupied is a very powerful metric and can be compared. Up to a point.

I showed a graph in my last post about how many people are ill in hospital with Covid-19, going up to 24th December. The below is the graph up until the 28th December. We still don’t have data for Wales, Northern Ireland, and Scotland beyond the 22nd December – but England on it’s own ( 20,426) is not far off matching the UK peak of 21,683 back on 12th April. If we optimistically only add on 1,727 for Wales, 1,045 for Scotland, and 451 for Northern Ireland (their figures for the 22nd) we are at 23,649. I’m seeing a lot of stuff on social media and the BBC news about hospitals running out of capacity, cancelling routine work, calling staff in from holiday (and remember, this is staff who have nearly all been pulling extra and double shifts for 9 months already). We suspect are approaching 100% hospital capacity for the NHS.

Patients in hospital with Covid-19 across the UK, 28/12/20

Update, 30/12/20 – we now have the Welsh & Northern Irish data to 28/12, Scotland & England to the 29th . The UK total for the 28/12 is 23,771 (slightly above my optimistic lower threshold of 23,649, as is to be expected. Northern Ireland shows a modest increase that could just be random variation, all three other nations show an increase.

Patients in hospital with Covid-19 across the UK, 29/12/20

100% Hospital Capacity is a Really, Really Bad Thing

I said that hospital beds occupied is a powerful metric up to a point. Why up to a point? At some point that metric stops increasing so fast or even at all – but not because of a lack of patients to treat, but because you are running out of capacity in your hospitals.

I’m sorry, I’m going to go on a bit of a rant here. The below is why I get so vexed at people saying “I need to go on holiday” or “I must have my nails painted” or decide it’s OK if they have a party or that we don’t need a lockdown.

If you get Covid-19 and are badly affected, you may well need supplemental oxygen. You may also need treating for various blood clotting conditions, or to stop your immune system over-reacting, and several other things. That can only be done in hospital. If it is done, most people treated survive (though some of course still sadly die). If you are not treated, you will die. As some of you know, I had personal experience of this late last year when influenza type A and pneumonia landed me in intensive care for a week, on very powerful CPAP ventilators. If I had not had that treatment, I would not be typing this (or anything). So we can (and do) treat and save many people with compromised breathing and the other things that come with Covid-19. Until we run out of trained hospital staff. It’s not beds per se that are the issue, or ventilators, or really any equipment. It is people who have the skills to run that equipment, monitor you, keep you ticking over and otherwise not-dead whilst supporting the broken parts of your body until they heal. Once the capacity of the health service is exceeded, they have to pick who dies. And of course, we do not just have Covid-19, hospitals are dealing with all the other sick patients we always have – car accidents, cancer, influenza, septicaemia, heart attacks…

I’ve seen the stuff by some people about how “only” 377 or so healthy, young people have died of Covid-19. Part of me can’t be bothered explaining to them why they are selfish, clueless idiots right now but what I can say is if we run out of hospital staff capacity, you can be as young and fit and bloody callous as you like but you will die if you need oxygen treatment and do not get it.

I’ve seen some tweets by people who say things like “well, just get more nurses and doctors”. I checked, they are not made in a factory. Training to be a nurse is not like going on a week-long course to learn to use a chainsaw. Doctors and nurses and radiologists and lab staff (and all the others people forget about who are vital to the NHS) are trained for many years. Being an ICU doctor or nurse is particularly technical and needs months or years of training ON TOP of being a standard doctor or nurse.

The UK was desperately short of all NHS clinical staff before Covid-19. One of my closest friends organises the lab rotas for a very large hospital and she never has enough people to fill the rotas. She has to beg and hassle people to do more than their fair share of weekend and night shifts. They constantly have not just one or two but a dozen or more open positions for staff. I’m not getting political here but there was a crisis in care long before the pandemic.

If you see figures saying ICU capacity is at 90% you would probably naturally think “well, they still have 10% spare, it’s fine”. It’s not. One of my first jobs was writing bed management software for hospital systems and teaching hospital staff how to use the software. The software was a god-send for them. A hospital bed is not just a bed. It’s a type of bed, and there are several types in hospitals. Some are for children, most are for adults, some are powered to help move the patient about, some are specialist for ICU (such as being able to pass air around incapacitated patients to reduce bed sores)… And beds move. For my spell in ICU I was initially admitted to A&E and held in a storage room as there was no spare capacity. They brought a suitable bed to me and squeezed it into the storage room. About 12 hours later, 6 or 7 nurses took the bed with me and a shit load of equipment through the hospital to the ward.

You have to know who is in which bed, the consultant & specialty treating them. For very, very good reasons, the specialist or someone in their group needs to approve a lot of what is done to you in a hospital. To administer a drug to a patient you have to find the bed they are in and you have no time to go wandering around the ward as you have 101 other things to do. The same is true of feeding the patient. You have to track when a patient moves (either with their bed or moving from one bed to another) and you need to know where you can move them to, so you need to know what beds are spare or, more likely, probably going to come spare. I worked on another part of the hospital system, “notify patient as dead”. It was horribly complex, lots of stuff has to happen when a patient dies, for example some lab tests get cancelled, others get created. The bed is noted as empty pending a deep clean. Sometimes, heartless though it sounds, the staff need to know when a bed is likely to become available via that route.

The people in charge of beds need to know ASAP when a bed is free so they can try and do all the juggling above that I mentioned. The fewer spare beds they have the harder it gets to make use of the few spare ones you still have and move people around efficiently. Or even inefficiently.

When I moved out of ICU it was a rush job. Someone needed one of the very most critical ICU beds (yes, there are tiers to what we non-medics think of as ICU), they felt able to move another person into my intermediate dependency bed as they were improving – IF they could get me out of it and into the Respiratory Medicine ward. Which they did, at about midnight. The sticking point was I needed to be isolated to I could not give someone with COPD influenza and finish them off. Another complication. It being night there were fewer staff so only 2 people could be spared to move me. Admittedly, less equipment came with me but half of it (including a heavy oxygen cylinder) was on the bed with me, I had hold of something on wheels, the 2 nurses somehow corralled the bed and other equipment.

The point I am making is that the closer a hospital gets to 100% capacity, the harder all that juggling becomes, and you actually end up having to move patients to other hospitals – and moving a sick patient to a different hospital is generally not in the best interest of the moved patient – or discharge patients who could really benefit from being there longer (but don’t need it as much as the person who is dying that they can’t find a bed for).

I’ve only ranted about beds. I have no idea how they keep track of other equipment, plan who is allocated to do what, how to cover for say a member of staff going ill, a major road traffic accident when all ICU is full…

If we do not see some sort of miraculous downturn in hospital admissions (and all indicators are against this happening) I’m expecting the UK to be in full national lockdown in a week, kids returning to schools cancelled. If we hit 100% hospital capacity and are not in a strict lockdown, then our government will have failed us in this crisis once more.

Even more distressingly, we may see avoidable deaths.