I have again run into the issue of the news running at a faster speed than I am writing. I crunched numbers related to the schools’ reopening in the UK, or at least England, yesterday, and woke up to the news that the government has delayed until September the complete reopening of schools, amid concerns about the feasibility of social distancing. I’d also tackled the BLM protests and criticism that they would cause a significant increase in the R number (number of people each infected person is likely to infect). It occurred to me that an understanding of the role masks play in reducing transmission of the virus is relevant to both of these subjects.
This article in the Atlantic is one of the best explanations I’ve seen for why people should be wearing masks. The problem is that you need everyone to wear them, and I’m afraid that in the US and the UK, that requires a law and penalties. The UK has finally come around to the thinking of the mayor of London and required people to wear masks on public transport. People have been dithering about the feasibility of enforcement, but we’ve all been on trains and buses when fare checkers have come around with their little machines and checked our Oyster Cards or receipt. They’re not on every trip (except, for some reason, on the London Overground, where I have encountered them almost every time I’ve been on the train), but they make their presence known enough that people get used to the idea that they may be penalized for not having paid for the journey. Surely it’s much faster to check whether someone has a mask on than it is to ask people to get out their fare cards and check them on a portable machine. Mask wearing in shops, etc., has still not been made mandatory. My husband goes into our Little Waitrose, wearing a mask, and says that about 15-20% of the people inside are also wearing masks, and none of the people working in the store are wearing them. Certainly, when I go out walking, I am generally alone in my wearing a mask, even on crowded streets (which I try to avoid) with narrow pavement. I realize it’s not as easy to catch the virus outside, but when a lot of people bunch up in a small area, a mask would still lessen the possibility. Part of the problem was initially that the UK government was afraid that people would hoard masks (as they did toilet roll) and there would not be enough for NHS workers. The government spent a lot of time telling people that they just had to wash their hands and masks wouldn’t make a difference. Now that masks, including easily washable cloth masks, are readily available, it’s hard to change people’s minds. Why, if masks weren’t helpful when R was 2, are they helpful when R is below 1? Well, of course, the answer is, they would always have helped, but the focus when R was 2 was getting people to stay home completely, but now, as we start to open up, it’s more important than ever to wear a mask.
It seems that, in the UK, people are resistant because wearing a mask is too much trouble, and maybe because, if they haven’t known anyone to contract the virus, they don’t really believe in the threat. It is in this group that Dominic Cummings’s actions and the government’s subsequent justifying of them did the most harm. They might reasonably expect that if the threat were all that serious, government representatives wouldn’t be breaking government rules. The irony is that this group is also anxious to get things back to normal, and if they choose not to follow the advice of scientists and now the government to wear masks, that “normal” will be a lot farther away. Andrew Lloyd Webber is one of the highest profile names to urge the government to adapt protocols protocols similar to those in Seoul that have allowed the theatres to remain open. The measures include temperature checks, track and trace, and masks that allow the audience to be closer than 2 metres apart, a distance that would mean theatres would have to run at a loss and therefore are unlikely to open.
In order for any of this to work, the government has to move from giving advice to enforcing rules, or giving local authorities the power to enforce rules. Government ministers and representatives have to be seen as obeying those rules—no exceptions. In the US, this seems to be the hard part. It is impossible to convince the public that wearing a mask is a good and prudent thing to do when the President of the United States, who says he is a friend to small businesses, refused to wear a mask while touring Puritan Medical Products in Guildford Maine, reportedly causing all the swabs made while he was there to be destroyed. I say “reportedly” because depending on which news source you read, the account is slightly different. In more pro-Trump publications, the visit and the assertion are reported, but the point is made that they’d already limited production while the president was there because they were planning to discard the swabs. That doesn’t seem much better to me, because I assume their actions were based on the fact that they already knew the president was unlikely to follow factory guidelines. Either way, this picture makes it clear that he doesn’t think he needs to follow the rules that everyone else (except Mike Pence) follows:
As long as people refuse to wear masks and governments botch track and trace, a lot of businesses will not be able to open. Just as he forced Puritan Medical Products into throwing away their products, the president’s presenting the wearing of a mask as an infringement of rights will force companies into a position of not reopening, reopening at far lower capacity, or risking the health of their staffs and their customers. In the litigious United States, the third choice doesn’t seem like a risk most people would make.
I said at the beginning that this all relates to schools and protests. First, schools. These are less of an issue in the US right now because the long summer vacation is looming anyway. In the UK, however, with shorter summer breaks and more schools already being open for children of essential workers, the opening of schools, either now or in the autumn, remains a huge issue. On the BBC News this morning, political reporter Norman Smith brought up the “Japanese model” of opening schools, which would include requiring staff and children to wear masks. But this, like so much else in the British government’s “plan” seems to be focused, on the one hand, on not being blamed for anything (else) going wrong, and on the other, on being seen as “getting back to business.”
So how risky would it be, really, to open up the schools? How do the numbers play out? For the first 21 weeks for which COVID death numbers are available, out of the 44,401 COVID-related deaths in England and Wales, 3 were of children 0-4 years old, 2 were of children 5-14 years old, and 9 were of young people 15-19 years old. Most school-age children fall in the 5-14-year-old age group, so I thought it was useful to compare it to other causes of death in that age group. (Disclaimer: These are all what the statistician David Spiegelhalter refers to “back-of-the-envelope” calculations. They are not precise, but they provide an idea of the magnitude.)
If we annualized these numbers, this means that in that group, maybe 5 children would die of COVID, and would probably have had pre-existng identifiable conditions that made it more likely that they would be badly affected. How does this compare to the annual deaths due to non-COVID causes? The latest year I could find for a fine breakdown was 2013, but going back, the numbers are fairly consistent. In that year, a total of 515 children between the ages of 5 and 14 died in the UK. Of those, 25 died of unspecified infections, 137 died of cancer, 51 died of congenital defects, and 105 died of accidents. It would appear, then, that the COVID deaths would make about 1% of the total, while cancer deaths would account for about 27%, and accidents accounted for 20%. To break it down another way, the number of children in that age group in the UK in 2020 is approximately 8.2 million. That means that with today’s numbers, the probability of a child in that age group dying of COVID is 5/8,200,000, or 1 in 1.6 million. The likelihood of a child in that age group dying in an accident of some sort is 1 in 78,000. Every day that we send our children out of the house in a non-COVID world, we are putting those children at far greater risk of injury and death from other causes than we would be sending our children back even to a completely normal school environment.
Why, if children are at such low risk, do we need to bother with face masks, temperature checks, social distancing and hygiene? Once again, the answer is for the good of the community as a whole. Teachers may be older or may live with people who are older or vulnerable. The children themselves may live with grandparents or parents with conditions that make them more vulnerable. And even if none of these situations holds true for a particular child, if the child can transmit the virus, the more people he or she comes in contact with, the faster it will spread, just like with adults. While there is some evidence that children not only don’t tend to get severe cases of COVID, but don’t contract it at all (and if they don’t contract it, they can’t spread it), the jury is still out and the studies provide mixed evidence. But there is also plenty of harm being done to children by keeping them out of school, and that harm disproportionately affects those children on the lower end of the socioeconomic scale. All children are more susceptible to anxiety and depression in this uncertain time. Children who depend on school for nutritious meals are also susceptible to a host of secondary illnesses caused by a lack of nutrition. Children from less privileged backgrounds are less likely to have access to the kinds of technology needed to keep up with studies, and with libraries also closed, they have less access to books that the parents of their more privileged peers can order from Amazon. Once again, there seem to be two extremes being argued—open or stay closed—when the sensible thing to do would be to open with appropriate protections in place. All but the very youngest of children can rise to the challenge of wearing masks, but every new suggestion is met with arguments that it is unworkable. I would argue that it is the attitude more than the actuality in most of these cases. Wearing masks and being screened would enable the schools, just as it would enable theatres, to relax some of the extreme social distancing requirements that teachers say are impossible to enforce in young children and in old schools with narrow hallways.
Which brings me, finally, to the issue I’d originally planned to start with: the protests. How much did they mess up our R number? Obviously, pictures like the one below strike fear into all of us who have become somewhat agorophobic during the lockdown:
But if you look more closely, you’ll see that most people whose faces are visible have masks on. And even though the square is very crowded and people are closer than 2 metres apart, if you’ve been to the area around Parliament Square for other protests, or if you’ve seen them on television, you’ll note that people are keeping a more respectful distance than usual between each other.
Now, what about the numbers? (I am dealing just with the UK here, since numbers are easier to get, but the logic is the same.) The only city that put out a specific figure for the protests was Manchester, which had about 15,000 people at its protest. Most people reported that the total number of protesters throughout the UK was in the “tens of thousands.” Let’s say, at an extreme, that a total of 100,000 people, or about 0.2% of the population, participated in one of the protests. Estimates for how many people in the UK have already had COVID, including non-symptomatic COVID, go as high as 50%, but most scientists say, with lockdown measures, probably not higher than 25%.
So even though it wildly overstates the likelihood that the person next to you is infectious, let’s use that number. If 25% of the population is infectious, if the population attending protests is of similar makeup to the general population (again, probably an overestimate, given how COVID has ripped through care homes), and if the R rate is approximately 1 in most places , meaning that every person infected infects one other person (although, obviously, it would be higher in a crowd), what is the worst the protests may have done in terms of increased cases? If 25% of the protesting population was infectious and 100,000 people attended the protest, we’d be talking about an additional 25,000 cases as a direct result, which is pretty major, about 9% of the total documented cases so far. But remember, that’s assuming that the entire population who has contracted the disease is currently infectious. If that were true, the 14-day quarantine period would be useless and the disease would already have ripped through the entire population. Since we’re about 3 months into the course of the disease, and since it has an average 2-week infectious cycle, it’s more likely that at any one time, out of 25,000 people who had contracted the disease at some point in that three months, about 25,000/6 (for 6 2-week periods) would be active carriers of the disease. That brings it down to about 4,000 infected, so 4,000 additional cases as a result of people attending a protest. So in the general population, that’s about an additional 0.006 percent, probably not enough to increase the R, especially since I overstated all the “bad outcome” numbers and didn’t account for things that mitigated the risk, like social distancing in some of the protests and the fact that even in some of the more crowded demonstrations, many of the protesters wore masks. As a matter of fact, the greater the proportion of people who wore masks, the lower the likelihood that the crowd would make a signifcant difference in the number of transmissions.
Hold on a second, some of you might be saying, if the risk of the protest increasing the numbers is infinitesimal, why have we upended our lives for 3 months? Surely scenes such as the one above would pose a greater risk of transmission than what most of us go through on a daily basis.
This is the interesting thing about statistics and the difficult thing about coming out of lockdown. What one action, even by many people, does to the rate of transmission as a whole is less significant than what habitual action will do to the rate of transmission. So this demonstration, especially since so many people are wearing masks, is far less significant to overall transmission than people riding in an overcrowded tube carriage twice a day without a mask. Even though the pictures of beachgoers at Southend made for good optics, that one day was also probably less significant to transmission overall than, say, opening the pubs without any social distancing measures (or masks) in place. The problem with big group activities is that they generally accompany a cavalier attitude about lockdown measures in general. In the case of mass protests, I’d argue that that’s probably less true (hence the masks) than it is with people going to the beach. People who were at those protests were, by and large, people who knew the risk but decided that the issue was important enough to risk getting the virus. But obviously, going to a large gathering increases the risk of getting the disease for an individual who was previously observing strict social distancing measures. And coupled with the relaxing of rules to allow people to meet in groups, it’s likely that some older or more vulnerable people will be infected by people who contracted the virus at the protests.
That’s the difficult part about where we are now. First, we have no confidence in the track and trace system in the UK, where’s it’s been a series of missteps. (I won’t even bother talking about it in the US.) Second, because, when the rate of transmission was higher, the government chose to scare everyone into following lockdown rules (instead of actually giving police and councils the authority to enforce those rules consistently), people are now more scared than they should be in certain situations. We can’t stay locked up forever, we can’t ask our kids to forego their friends and their educations forever. A vaccine is still a long way off. We need to get back to normal as much as possible, but we need to do it in a sensible, scientifically defensible way. And that includes wearing masks.