The Covid testing dilemma

England’s pushing mass testing as a way to contain Covid. It’s free, it’s government approved, it’s somewhere between uncomfortable and painful, and it may or may not be a good idea. Let’s tear the numbers apart and see what we can figure out.

Since the schools reopened, secondary students–those are the older kids–have had to do quick Covid tests twice a week, and that’s been a bulwark of the program to keep the schools open while not letting the virus get out of control. 

The tests, unfortunately, have a reputation for being unreliable, especially when done by non-experts. Since the kids are doing their own tests, or asking their parents or three-year-old sisters to stick the swabs up their noses and down their throats, these are in the hands of the distilled essence of non-expert. One fear about relying on the quick tests has been that false positives will send a lot of people into isolation unnecessarily. So half of the positive tests were sent to a lab to be confirmed by the slower, more reliable tests, and only 18% of them were false positives. 

Irrelevant photo: Rhododendrons. Photo by Ida Swearingen

But wait, because we’re not done yet. Those numbers are from March, and Covid rates have fallen, at least in parts of the country. (Some hot spots remain, and I don’t know if numbers are falling there as well. Just put that possibility off to one side. The recipe may call for it later. If it doesn’t, we’ll stick it in the freezer.) The point is that where the number of cases is lower, everything changes

Why? Because the tests will crank out the same number of false positives, no matter how many people are infected. Find yourself a population of people who’ve never been exposed to Covid and the test will swear on any religious book you like that some of them are infected. 

I’m about to throw some numbers at you, so if your allergies are bad today just skip a few paragraphs.

Ready? In London, the southwest, the northeast, and the southeast of England, the prevalence of Covid ranged from 0. 08 to 0.02. In England as a whole, it was 0.12%. Using those figures (I’d assume that means the England-wide ones), it would take 16,000 tests to find one infected person. If the tests cost £10 each, that means spending £160,000 to find that one person.

Is that worth it? If we were trying to stamp the disease out and keep it stamped, as New Zealand is, it would be. Given that we treat stamping it out as the silly thought of irresponsible day dreamers, probably not. 

Meanwhile, in leaked emails (I do l love a good leak) “senior government officials” are talking about scaling back mass testing, although the Department of Health and Social Care says it has no plans to end the program. One in three infected people, they remind us, show no symptoms but is still contagious. 

That brings us neatly to the question of whether the rapid tests will spot that one person. In other words, it’s time to talk about false negatives. Administered by an expert, the tests pick up 79% of infections. Or to put that the other way around, they miss 21%, and those are mostly people with a low viral load. Or to put that another way, they’re most likely to miss people who don’t have symptoms, who are just the people the testing program is looking for.

Administered by secondary school students or their three-year-old sisters, they’re more likely to pick up 58% of infections, or to miss–umm– I think that’s 42%. Although estimates of the number of cases the test misses vary. It might be as high as 50%. 

The government denies that it has any plans to scale back anything ever and Boris Johnson is urging everyone to get tested twice a week. Even though his advisors say that in areas with low infection rates, only 2% to 10% of the positive results may be accurate. 

But what the hell, guys, we’ve got these tests. Someone’s cousin has the contract for them. Use them, will you, please? For the good of the nation.

 

News of an accurate rapid test that’s in development

A new test is being developed that’s both fast and accurate. It also tracks variants and tests for other viruses that might be mistaken for Covid. It can screen 96 samples at a time and within 15 minutes it starts to report the samples as negative or positive. In 3 hours, it will have sequenced all its samples. 

It’s also small and portable. It doesn’t make coffee, but it just might be able to make you a cup of tea.

Juan Carlos Izpisua Belmonte, a professor in Salk’s Gene Expression Laboratory where it’s being developed, said, “We can accomplish with one portable test the same thing that others are using two or three different tests, with different machines, to do.”

That’s the good news. But will it go from development to being manufactured and used?

Market analysis would be required to determine whether the initial cost of commercialization—and the constant tweaks to the test needed to make sure it detected new variants or new viruses of interest—are worth it.”

I believe that translates to “maybe.”

It’s called NIRVANA, which doesn’t seem to stand for anything, so I don’t know why it’s in all caps. 

 

High- and low-tech approaches to Covid

In New Zealand, they’re trying out an app that connects to smart watches and fitness trackers, monitoring people’s heart rate and temperature. It’s called an Elarm and the developer claims it can spot 90% of Covid cases up to three days before symptoms appear.

Does that include people who don’t go on to develop symptoms? I’m have to give you a definite maybe on that, because the article I found doesn’t address it. The company’s own website doesn’t answer the question either but says it will also let you know about stress and anxiety, although you might notice those without needing an app. Basically, it figures out your normal levels and lets you know when you’ve wandered off them, so you could end up going into isolation over the flu as easily as over Covid. That would scare the pants off you but would, at least, take a lot of the punch out of flu season.

So how do you use this? New Zealand wants its border force to try it out, since almost the only cases of Covid there are in incoming travelers, who have to go into quarantine, meaning the people who work for the border force are in the front lines.

When New Zealand says quarantine, by the way, they actually mean quarantine. It’s one reason they’ve been able to contain the virus.

*

On the other end of the scale comes the recommendation that we open windows in public places to minimize Covid transmission. It’s cheap, it’s simple, and–

Oh, hell, how many public places these days have windows that open? Okay, ventilation. The air in public indoor spaces needs to be replaced or cleaned. 

We’ve heard a lot about keeping two meters (or yards) away from people to avoid contagion, but in addition to the heavier droplets people breathe out, which can carry Covid, the tiniest particles that we breathe out can also carry it, and they can stay suspended in the air for hours. The goal is to run them outside and get some fresh air in. 

*

If you’re looking for a low-tech way to decide how far from people you should be standing, you can think of it this way: If you can smell that they’ve had garlic or peanut butter for lunch, you’re too close. 

 

Drug news

An asthma drug, budesonide, has been shown to shorten people’s Covid recovery time –and it can be used at home without anyone involved needing welding gloves, a deep-sea diver’s helmet, or a set of allen wrenches. It’s relatively inexpensive and comes in an inhaler. It shortened people’s recovery time by three days and at the end of two weeks the people who used it were in better shape than the control group.

It’s not clear yet whether it made hospitalization less likely. In the budesonide group, 8.5% were hospitalized. In the control group, that was 10.3%. That sounds like a result, but the problem with interpreting the numbers is that hospitalization rates are dropping in Britain. If you want to understand why that makes the numbers hard to interpret, you need to talk to someone who actually knows something.

Everyone in the test was over 50 and had underlying health problems. The drug can be used in the early stages of infection. 

How Covid mutates and why that might be a good thing

There’s good and bad news about the way Covid mutates, and it’s all wrapped around the same bit of information. 

Like most non-experts, I use the word mutate loosely. If something genomeish leads to change, I think it’s a mutation. Which goes to show you what I know.

Covid, it turns out, doesn’t just mutate, it also recombines, meaning it mixes large chunks of its genome, not just single genes. If a mutation’s a typo, recombination is a cut-and-paste error, dumping a largish chunk of text in the wrong place. And while the virus proofreads typos fairly well, it doesn’t catch cut-and-paste problems as effectively.

I’ve had that problem myself. I still wince at something quite horrible that I let go into print because the spelling was right and my eye didn’t pick up the change in meaning. And I’m larger and (I like to think) more complicated than a virus.

Most of those recombination errors, like most mutations, make a mess and that particular virus doesn’t get to leave little virette progeny behind. But some of them work and the virus changes.

Irrelevant photo: One of Janey’s crocuses.

Is recombination what’s happening with Covid? Possibly. The Kent variant has more than a dozen mutations and they seemed to appear all at once. Emphasis on seemed. A lot of what goes on happens in the kitchen while we’re out front cleaning the dining room. Feng Gao, a virologist from China, says we don’t yet have proof of recombination. “Diversity, no matter how much, does not mean recombination. It can well be caused by huge diversification during viral evolution.”

So let’s not get carried away with this. We’re dancing at the edges of what’s known. But (damn, that tune’s catchy, so I’ll do a few more steps) recombination may be how viruses that infect one species jump to another species: by swapping a bit of genetic code .

It’s possible that recombination means a more dangerous virus will appear–either a new one or a more dangerous form of Covid. So there’s our bad news. 

But the good news is that experiments with a mouse coronavirus show that blocking a single enzyme keeps the virus from correcting its typing errors and recombination events happen much less often. If this holds for Covid, the right drug might be able to block recombination and (or maybe that should be or) push the virus to mutate so badly that it ends up in something called error catastrophe–basically, the evolutionary equivalent of falling off a cliff. While dancing to that catchy little tune.

As a way to treat Covid, blocking the enzyme could make antiviral drugs more effective.

The enzyme goes by the name of nsp14-ExoN, which isn’t particularly catchy. If we’re going to be spending time with it, it needs a nickname. But whatever we call it, it’s common in coronaviruses, so if this works it opens up the possibility of curing other coronavirus diseases as well. 

 

Covid variants

If Britain didn’t end up with the world-beating test and trace system Boris Johnson promised us–and believe me, it didn’t–it may have come up with a world-beating strain of Covid instead: the Kent variant; the variant I mentioned that has all those mutations. Sharon Peacock, the director of the Covid-19 Genomics UK consortium tells us it looks likely to sweep the world.

And unlike the test and trace system, we didn’t pay a penny for it.

Go Britain!

The consortium is testing the genomes on a randomly selected 5% to 10% of all positive Covid samples in the country but aims to test them all in order to keep track of how the virus is mutating.

And speaking of variants, the World Health Organization says the small trial that found the AstraZeneca vaccine to be largely ineffective against the South African variant was inconclusive. They’re not saying the vaccine’s definitely effective against it, only that it isn’t definitely ineffective. 

Which is better than nothing. 

 

Covid and Coca Cola

How much space would all the Covid viruses in the world take up if they could be packed neatly for shipping? They’d fill a Coke can

They’d also fill a can of supermarket brand fizzy orange-flavored sugar water, but Covid’s a brand-name kind of virus. Coke it is. So in the scene where someone yells, “Don’t open the can!” for pete’s sake, don’t open the can. You know what happened when Pandora didn’t listen to the warnings?

I’m not telling. But I did give you a link.

 

Spreading the virus

More than half of all Covid cases are spread by people who have no symptoms. They may be less infectious than people who are sick, but they could well make up 80% of the total number of people carrying the disease. And they’re wandering through the world shedding viruses, not lying in bed at home.

*

A study by the U.S. Centers for Disease Control found that wearing two masks can reduce the chance of getting Covid by 90% or more. Yes, not just transmitting but catching the damned thing. 

The study had its limits. It tested a tight-fitting cloth mask over a surgical mask, not two surgical masks and not two cloth ones, and it only looked at one type of cloth masks, although the world’s awash with different types just now. And as the article where I first read this put it, it also didn’t consider “men with beards or children.”

Does having children interfere with the fit of men’s masks more than women’s? Hard to say. The study didn’t test that. 

To keep everything in perspective, an engineering professor says that the only reason to wear two masks is to get a better fit. But the masks most of us wear do fit loosely, so double masking might be worthwhile, no matter who’s right.  

*

Two weeks after U.S. states introduced mask mandates (they haven’t all), the weekly growth in hospitalization rates dropped by 2.9% among people who are 40 to 64. After three weeks it dropped 5.5% among people 18 to 64.

*

Every so often, you’ll find someone saying that Covid’s no more dangerous than a bad outbreak of the flu. So do we have any figures on how much more dangerous it is? 

Yup, some. The risk of death is 3.5 times higher. That number comes from comparing people who are hospitalized with the Covid against those hospitalized with flu. It ignores whatever long- term effects Covid has on the unhospitalized, so I’d say it’s undercounting. Still, it’s a number, and numbers help. 

At least they help most people. 

Covid patients also had one and a half times greater use of the intensive care unit and one and a half times longer hospital stays. And they were more likely to need a ventilator.

In case you think Covid’s only a problem for the old and the ill, not many of the hospitalized Covid patients had other illnesses and 21% were younger than 50. People under 50 made up 24% of the intensive care admissions.

As far as I can see, that doesn’t address the problem of how easily Covid spreads compared to flu. It only compares hospitalized patients.

*

Worldwide, the number of reported Covid cases is down for the fourth week in a row. Take a deep breath. The drop is uneven, it doesn’t count unreported cases, and we forgot to get a guarantee that it won’t go back up, but we have to take our good news where we can get it. This is good news.

 

Is there any news on curing the thing?

In a small study, a common asthma treatment, budesonide, cut the need for hospitalization and urgent care by 90%, and people who took it within seven days of showing symptoms recovered more quickly than the control group. Better yet, it cut the number of people with symptoms that lingered after twenty-eight days. 

As usual, it was a small study–146 people–so it’s preliminary, but budesonide is a well-known and well-studied drug, which would speed the process if it’s adopted.

*

The U.S. Food and Drug Administration has approved a combination of two monoclonal antibodies that can keep high-risk patients from developing Covid that’s severe enough to hospitalize them. A similar drug had already been approved. Both take Covid antibodies and synthesize them so they can be given to patients as a drip. 

And it’s that drip business that’s causing trouble. Initially, getting them from vial (or whatever they come in) into human took an hour. It can now be done in sixteen minutes. But some hospitals have been so overwhelmed they haven’t had time to deal with it. 

*

A team of researchers in China has identified six drugs that the FDA has already approved for other uses that could be repurposed to treat Covid. They whittled that down from 3,769. They still need to be tested in the real world, but already having FDA approval for other purposes means that if they work they could be put to use quickly.

After that, the article went over my head, but it has to do with proteases and substrates, not to mention clades. Have fun.

 

Your feelgood story

New York software developer Huge Ma tried to make his mother a Covid vaccination appointment and discovered that not only did the city and state have different systems that weren’t talking to each other but that there were dozens of separate websites, each one demanding that you sign up a different way.

So he took a couple of weeks and made a free website, TurboVax, that compiles information from the three main city and state sites and sends information on available appointments to Twitter.

It cost him $50 to make.

The difficulty of booking an appointment is one reason–although far from the only one–that vaccines are going disproportionately to white New Yorkers. 

“It’s sort of become a challenge to myself, to prove what one person with time and a little motivation can do,” he said. “This wasn’t a priority for governments, which was unfortunate. But everyone has a role to play in the pandemic, and I’m just doing the very little that I can to make it a little bit easier.”