Sweden with a very mild form of reduced social interaction, got community immunity over the summer at a death rate equivalent to 40,000 deaths in the UK. This is significantly lower than the death rate in the UK, and in addition, it is clear that either the UK was still a significant distance from community immunity or some new epidemic mistaken for covid has occurred. Either way, locking up society caused greater deaths, but how.
It is tempting to blame the higher UK death rate on the appalling behaviour of our politicians in that they callously sent infected patients from hospital into care homes and thereby almost deliberately spread the virus. However, reports suggest that similar behaviour occurred in Sweden such that they also had a lot of infections in care homes.
Thus, the main difference between the UK and Sweden appears to be that the UK had a very draconian, human-rights infringing lockup, whereas Sweden did not.
The conventional wisdom is that lockups save lives, because it appears that by stopping the virus spreading, that people will not get infected. But unfortunately, due to the non-homogenous nature of society, there is no single R number such that the concept of “R number suppression” is totally worthless. Instead, R number suppression, doesn’t act in the way it is believed to had to reduce the death rate, but the evidence shows it increases it.
This is counter-intuitive, but so too is another form of “epidemic” which is wild fire. The conventional wisdom, is that putting out fires will reduce the damage from wild fires. However, the reality, is that fires keep happening, and if they occur where there has been no wild fire for some time, instead of a small naturally contained fire, there is sufficient debris and deadwood so that the fire takes off very rapidly and spread very rapidly. Moreover, if a large area has all had fire suppression policies for a long time, the entire area is likely to go up in one enormous blaze. Thus, putting out fires actually increases the damage. In contrast, if fires are allowed to regularly start, or are actively started in the less dry times of years, they can be controlled, at first by human intervention, but eventually, when there is a patchwork of forest all having last burnt at differing times but relatively recently, most fires then only burn in a small area and then die out naturally.
So, things can and do act in a counter-intuitive way. So, how is it possible that lockups increase deaths? After some thought, I have developed this possible model.
Model
For this simple model, I will imagine a society that consists of two groups. In the first are what I term “Young people”, which I imagine to be in their 20s to 30s and all out to have a good time socially. The second group I will call “old people”, and I will suggest this group consists of a variety of people ages perhaps 50 upwards.
For simplicity I will suggest that the young group on average meets with about 8 people a day, and that the old group meet with about 2 people a day. The younger people have a large number of social interactions because they are either single and still “having fun”, or are looking for marriage partners, or are with young families meeting other young families. Most of these social interactions are “unnecessary” in the terms of the lockup. A large number of the social interactions of the older group is with relatives, carers or for getting food or other provisions. Thus whilst they have far fewer social interactions, the vast majority of these are necessary.
For simplicity (because the exact number does not affect the argument) let us imagine that the groups are the same size and that of the 8 interactions per day, the young people drop to perhaps 1. In contrast, the older people drop from 2 to 1. If we then assume that social interactions are otherwise random so that an older person is as likely to meet a younger person as an older person. Let us also suggest that only old people die. For simplicity let’s assume community immunity is achieved at 20%.
Let us then assume two groups of 100 individuals, then with these numbers:
Before lockup
In the young group there are 800 social interactions per day.
In the old group there are 200 social interactions per day.
Therefore there are 1000 social interactions of which the chance of the next being with a young person is 800/1000 and that of being with an older person is 200/1000.
If then imagine a chain of people each one infecting the next until society gets community immunity, then out of the 200 individuals, we 40 to be infected. That chain will be 80% young people and 20% older people, which means 32 young people and 8 old people.
After Lockup
Because lockup predominantly hits the “unnecessary” social interactions of the young:
In the young group there are now 100 social interactions per day.
In the old group there are 100 social interactions per day.
Therefore there are 200 social interactions of which the chance of the next being with a young person is 100/200 and that of being with an older person is also 100/200.
If then imagine a chain of people each one infecting the next until society gets community immunity, then that chain will contain equal numbers of young and old each with 50% of the infected individuals.
We need 40 people infected to get community immunity so that 20 older and 20 younger people get infected.
Discussion
From this simple model, using the numbers selected, we find that because young people have far more social interactions which are predominantly “unnecessary” when viewed by old grumpy politicians in Parliament, that the result of lockup is to change the number of older people being infected from 8 in the non-lockup example to 20 in the lockup example. Because these are the people who die, the death rate will therefore increase massively (2.5x). In effect, the lockup tends to focus the epidemic in the group of people who are most likely to die and by doing so, it massively increases the death rate.
Of course, these numbers were chosen solely to illustrate the point and no real society has two such groups. However, if the society has an much higher death rate amongst those whose social interactions are largely necessary (so for example social care or help from family), then by restricting young people’s social life, although fewer social interactions occur, lockup tends to make far more of the social interactions with older people and therefore focusses the epidemic in those most likely to die.
Moreover, if we assume a fixed fraction of interactions result in infection, then in order to achieve community immunity, we need a fixed total number of interactions. So, although these occur more slowly, the total number of infections are the same and if the fraction in older people are greater, then the death rate is greater.
The only proviso, to this argument is that lockup policies are supposed to reduce the total death rate of the epidemic. The problem with that argument, is that as soon as society returns to normal, there will inevitably be a flare up again unless community immunity has been achieved. Moreover, the longer any lockup policies go on, the less anyone adheres to lockup rules and so the epidemic will always end with much the same lax behaviour to lockup as if there were none. And as the final death rate is determined by the behaviours at the very end, then the death rate is the same whether or not there is a lockup (unless by some miracle a vaccine comes very quickly and before the lockup policies themselves have caused many deaths).
