Day 287: The Long Haul

MedPage Today calls for more clarity and less credulousness over presumed COVID "long haul" cases:

First, we need clear definitions. What counts as impaired memory? Can we measure it with available scales? And, what is needed to diagnose someone with prior COVID? Some people report long COVID symptoms with negative PCR and Ab tests to SARS-CoV-2. Of course, there can be false negatives, but there can also be true negatives. How can we separate the two to better understand this problem?

A paper in Nature analyzes the sample bias in "numerous observational studies" of COVID risk factors:

Numerous studies have reported risk factors associated with COVID-19 infection and subsequent disease severity, such as age, sex, occupation, smoking and ACE-inhibitor use1,2,3,4,5,6,7,8,9,10. But to make reliable inference about the causes of infection and disease severity, it is important that the biases which induce spurious associations in observational data are understood and assessed. Bias due to confounding remains well-understood and attempts to address it are typically made (bar rare exceptions e.g. ref. 11). But the problem of collider bias (sometimes referred to as selection bias, sampling bias, ascertainment bias, Berkson’s paradox) has major implications for many published studies of COVID-19 and is seldom addressed.

A collider is most simply defined as a variable that is influenced by two other variables, for example when a risk factor and an outcome both affect the likelihood of being sampled (they “collide” in a Directed Acyclic Graph, Fig. 1a). Colliders become an issue when they are conditioned upon in analysis, as this can distort the association between the two variables influencing the collider. [...]

As illustration, consider the hypothesis that being a health worker is a risk factor for severe COVID-19 disease. Under the assumption of a higher viral load due to their occupational exposure, healthcare workers will on average experience more severe COVID-19 symptoms compared to the general population. The target population within which we wish to test this hypothesis is adults in any occupation (or unemployed); the exposure is being a health worker the outcome is COVID-19 symptom severity. The only way we can reliably estimate COVID-19 status and severity is by considering individuals who have a confirmed positive polymerase chain reaction (PCR) test for COVID-19. However, restrictions on the availability of testing especially in the early stages of the pandemic mean that the available study sample is necessarily restricted to those individuals who have been tested for active COVID-19 infection. If we take the UK as an example (until late April 2020), let us assume a simplified scenario where all tests were performed either on frontline health workers (as critical vectors for disease among high-risk individuals), or members of the general public who had symptoms severe enough to require hospitalisation (as high-risk individuals). In this testing framework, our sample of participants will have been selected for both the hypothesised risk factor (being a healthcare worker) and the outcome of interest (severe symptoms). Our sample will therefore contain all health workers who are tested regardless of their symptom severity, while only non-health workers with severe symptoms will be included. In this section of the population, health workers will therefore generally appear to have relatively low severity compared to others tested, inducing a negative association in our sample, which does not reflect the true relationship in the target population (Fig. 2b). It is clear that naive analysis using this selected sample will generate unreliable causal inference, and unreliable predictors to be applied to the general population.

The cities and towns data came out yesterday. The map colors are toned down a bit (things were getting fairly dark), and bear even less of a relationship to the grey/green/yellow/red color scheme the state has been using to declare some of us "red" since that, too, has changed. Instead of continuing to use simple case rates for the colors and just increasing the case rate cutoffs so the entire state is not red, the MDPH now does a complicated calculation involving population size, absolute case numbers, case rates, and positivity (the most nonsensical, dependent variable of all) that takes an entire page of the report to explain.

There's nothing particularly notable this week, unless it's the relatively low case rates in a couple of the Stop the Spread cities: Taunton, Randolph, and Marlborough are currently below the state average in daily case counts.

(Pop out.)

P.S. Massachusetts cases were up 1.6% today.

Day 282: Man Walks Dog

We are not alone in our nonsensical nocturnal COVID curfew. In Czechia, a man was caught walking a stuffed dog as a way around their 9pm curfew. Upon being caught, he claimed to be joking and was let off with a warning.

Massachusetts cases are up one percent today.

Some locals are annoyed that, for this week's cities and towns report, the Commonwealth has defined "red" down to the point that not every town in the state is red, so the map is still marginally useful for detecting hot spots. Out of respect for these incensed innumerates, PlagueBlog has used the same color scale this week as last week for easier comparison. As our color scale was finer than red-yellow-green, our map is still marginally useful in its darkened state.

(Pop out.)

Day 277: Welcome to Efrafra

PSA: The CDC says the infected can vote in person today.

Greg Cochran blogs briefly at West Hunter; the meat is in the comments. On his COVID-19 post yesterday, Free Parameter, he made a passing reference to the rabbit disease myxomatosis, a.k.a. white blindness, which inspired a commenter to compare our SARS reactions to the rabbit military dictatorship of Efrafra in Watership Down:

Woundwort thinks that the white blindness is a disease used by humans to harm rabbits (shades of how some think COVID-19 is a Chinese-made weapon), so he tries to stop it by keeping his rabbits underground, out of human sight.

Although his theory’s off, his measures would limit Efrafra’s exposure to natural reservoirs of myxomatosis. [...]

[But] he’s too strict, too cruel. No Efrafran rabbit will report feeling sick, because they’ll be punished. And if they did, their Owsla officer would likely try to keep it a secret, for the same reason. White blindness will become deeply entrenched at Efrara before Woundwort even knows it’s there.

That’s the irony of being a totalitarian dictator: you want obedient underlings, but instead you get skilled liars. To fight disease you need clear information about where it began and how far it’s spread, and Woundwort has set up an incentive structure that ensures he’ll never get it.

A real-life parallel happened in 2002, when SARS broke out in Guangdong in China. The province had a 12% economic growth target to hit, come hell or high water, and “negative news” was suppressed by the state. As a result, the disease snowballed and became far more devastating than it had to be.

It is left to the reader to draw the obvious parallels to SARS-CoV-2.

Like the weekly data, Massachusetts' daily numbers are now coming out later, in perpetuity. While we wait, here's last week's long-awaited map, with a notable increase in the orangeness of the percent change panel (showing the area of increasing case counts):

(Pop out.)

P.S. Massachusetts cases were up two thirds of a percentage point today.

Day 268: Doubling Down

Thursday's cities and towns data (for Wednesday) reveals a slowdown of tourism COVID cases on Nantucket and the Cape. The red city rate (8 cases per 100k per day) remains below the state average (still 9 cases per 100k); this week's big outlier was Lawrence at 46 cases per 100k per day. (Middleton at 62 cases is just another small town outlier.) The data also includes county-level summaries for most real counties, plus the fake COVID county of Dukes and Nantucket.

(Pop out.)

Day 264: More COVID Fatigue

On the COVID fatigue front, the Commonwealth seems to have tired of Wednesday as cities and towns day, and now will be releasing the data on Thursdays, allegedly by 4pm.

Day 257: In the Red

Since last we met, the US has exceeded 8 million cases, and several small-to-middling countries have passed the China point, including Portugal, a nation of ten million (only 1/144th its size). Despite the best efforts of Maryland and Indiana, Massachusetts remains at #20 among the states in case count.

Speaking of Massachusetts, there has been much hand-wringing today over the increasing "redness" of the state. It's not about politics but about cities and towns with average daily case rates of 8 or more (per 100,000) over the past two weeks. In fact at 9 cases per 100k the whole state is currently in the red, with outliers having case rates of up to 100, so the current rating system doesn't say much about a particular town's performance. Bigger, denser cities have more cases for the obvious reasons.

(Pop out.)

Day 251: More Great Barrington

Brazil (at #3) has reached 5 million cases; #2 India is approaching 7 million, and at a pace that puts it in position to overtake the US eventually.

Governor Whitmer has been digging up Spanish Flu laws in an attempted end-run around last week's state supreme court decision striking down a 1945 law and her coronavirus restrictions along with it. PlagueBlog senses that this is not the right legal or temporal direction to be going in, and that the "burn your masks" lawyer will be back before their supreme court sooner rather than later.

AIER, after letting slip their involvement in the Great Barrington Declaration, has been reporting on the reaction to it, from scientists, the media, and the social media peanut gallery. Off-Guardian was not involved in the declaration, but assembles a lot of prior art of scientists and health care workers recommending against lockdowns.

Yesterday's weekly cities and towns data was the twenty-sixth such missive from the state. That's half a year, and the weekly reports only started in April. The coronavirus drama in Boston has actually been going on for 251 days since our first case, which is more than two-thirds of the year. Although the news just keeps on coming so does the COVID fatigue, so PlagueBlog will be reducing our publication schedule for the foreseeable future.

On to the data: while in recent weeks the Stop the Spread cities (with dotted borders) haven't been consistently ahead of other cities and large towns in case counts, this week you can see they're hotspots, along with Nantucket and the usual random smattering of small town outliers:

(Pop out.)

Day 244: Even the Neanderthals

An article in Nature claims that The major genetic risk factor for severe COVID-19 is inherited from Neanderthals. But not all post-Neanderthals have it; the particularly problematic genes seem to be more common in South Asians than in any other out-of-Africa group. (Africans rarely carry Neanderthal genes.)

MedPage Today reports that ECMO (extracorporeal membrane oxygenation) wasn't as universal a failure for COVID patients as previously suspected. (It made ventilators look helpful.) It turns out that only about 40% of patients died in hospital afterwards. However, the follow-up on the allegedly surviving 60% wasn't really solid or long-term enough to be entirely convincing.

Forbes reports that Mississippi's mask mandate expired peacefully in its sleep last night.

Massachusetts' cases are up four-sevenths of a percentage point today. While the general grouping of our "red" cities (Lawrence, Framingham, Revere, New Bedford, and the Islands) hasn't changed much, things are getting a little more infected. (Also, maps from last week have been corrected due to PlagueBlogger error.)

(Pop out.)

Day 237: Variolation

In world news, the recent Israeli lockdown is getting tighter after the first week of lockdown produced 7,000 cases a day and stressed hospital capacity. Unemployment there is currently at 11%.

The Guardian reports a "nearly 100% accuracy" rate for COVID sniffer dogs in Helsinki, and it's also cheaper and easier than laboratory testing. The exact mechanism is unknown, but a prior study suggests the sweat of the infected smells different.

Here at PlagueBlog Headquarters, we've been intentionally ignoring the variolation story for a couple of weeks now as too silly to report, but it made ProMED earlier this week so here goes nothing: A "perspective" in the New England Journal of Medicine compares "universal facial masking" to variolation. While it's interesting as an admission that dose matters, as well as an implied admission that SARS-CoV-2 is getting through your face masks, it's less interesting in its examples of masking successes in crowded, germy settings like hospitals and ships. In reality, any of the host of COVID precautions can reduce viral loads, and the degree of reduction is poorly understood for novel approaches like universal mask usage or lockdowns. PlagueBlog is duly impressed that 2020, the year that keeps on giving, has managed to produce a vaccination technique that's even less reliable than the 15th-century institution of variolation.

Massachusetts' cases were up three sevenths of a percentage point today. Yesterday's cities and towns data shows the usual pattern of small town outliers, tourism COVID, and above-average case rates in some of the larger cities, though not in all of the Stop The Spread cities (the ones with dotted outlines):

(Pop out.)

Day 230: Should Fido Wear a Mask?

The Netherlands have jumped a couple of places past China (now #41) over the course of the week. Next up to surpass China are Guatemala and the United Arab Emirates. Michigan also surpassed Massachusetts this week, pushing us down to #18 in the US. Next up is Maryland, though Missouri and Wisconsin are moving faster at the moment.

Today's title question is an apparently serious one from Reader's Digest: Should Your Dog Be Wearing a Pet Mask? The answer, of course, is no. Dogs need to pant as well as breathe, and some of them have been bred to be dangerously bad at the latter even without a mask. Of course no one asks whether your cat should wear a mask because clearly your cat will refuse.

On the science front, not only does smoking protect you from COVID-19, but wearing glasses does as well, at least according to this report in JAMA Ophthalmology.

Massachusetts' cases are up a third of a percentage point today. Here's the graph of the cities and towns data from Wednesday; things are still happening in Williamstown and on the islands, as well as in Worcester, Lawrence, Revere, and New Bedford:

(Pop out.)

Day 224: 9/11

Aside from the obvious analogy of a misdirected overreaction (in the wars on terror and coronavirus), 9/11 and COVID also share flight from Manhattan and an economic threat to the World Trade Center.

On the science front, the Lancet questions improbable fomite research, while others question the Lancet paper about Sputnik V:

“There are very strange patterns in the data,” Enrico Bucci, a biology professor at Temple University in the U.S. who has published an open letter highlighting the concerns told The Moscow Times.

“By strange patterns I mean there are duplicate values for different [groups of] patients … which cannot be,” Bucci said, in reference to results concerning the production of antibodies by groups of patients who had been tested with different formulations of the vaccine.

The cities and towns data for this week has a few tiny town outliers (most notably Montgomery), but also some activity in the bigger cities, in Williamstown (where Williams College is mostly in session, albeit with a reduced course load and activities), and on the Cape and islands:

(Pop out.)

P.S. Massachusetts' cases were up a third of a percentage point again today.

Day 216: More Zombies

Yesterday's data change at the MDPH not only raised 4 COVID-19 victims from the dead, but also set Massachusetts back about 7,750 cases. A mysterious lack of such CDC-based corrections in other states means that Massachusetts is suddenly #16 in case counts, newly behind #13 Alabama, #14 Ohio, and #15 Virginia, with #17 South Carolina close behind. [They surpassed us as well later in the day.]

The cities and towns data considers only confirmed cases, so should not be affected by this change. Watch this space.

P.S. Massachusetts' cases are up a third of a percentage point today. Some cities and towns are up a bit as well this week, but the increase doesn't particularly correspond to the Stop the Spread cities any more. The map:

(Pop out.)

Day 209: This Sort of Vague, Uneasy Feeling

Back in the height of the pandemic, CTV News reported on the phenomenon of losing track of time during lockdown. (PlagueBlog reminds readers that it's not a quarantine if you're not actually ill or the contact of a known ill person.)

Steve Joordens, a psychology professor in the department of Cognitive Neuroscience at the University of Toronto, described [daily routines] as a sort of “anchor” that gives people a sense of where they are in time.

“They are kind of like a rhythm to our lives,” he explained to CTVNews.ca during a telephone interview from Toronto on Tuesday. “They tell us where we are within a given day, you know if it's lunchtime, but also within the given week, you know if it's Thursday or Friday, they feel very different than a Monday or Tuesday do.”

When suddenly all of those rituals are gone, as in the case of a pandemic, Joordens said it’s easy for people to lose that sense of where they are in the workday or the workweek.

“They also lose sense of who they are,” he said. “This is what a lot of people kind of feel too, is that they feel a little adrift. They’re not really sure what they're supposed to be doing and they have this sort of vague, uneasy feeling.”

Although by now we may all have daily routines of going nowhere and doing nothing, but the article addressed that eventuality with the help of a Manitoba psychologist:

On the other hand, Abdulrehman said even those people who do develop a routine in quarantine may be susceptible to losing track of time if there isn’t enough variety within it.

“They don't know what day it is because it's the same thing every day,” he said.

This is all to say that the staff at PlagueBlog headquarters somehow forgot it was cities and towns day yesterday, so we're a bit behind with the cities and towns data. We have, however, updated last week's map to show Stop the Spread cities (outlined in maroon) and to include the old whole-pandemic case rates in the popup info (though not on the map). This week coming soon...

P.S. Massachusetts' cases are up three-tenths of a percentage point today. Our death toll has reached 9,008 (including presumptive cases). Here is the new daily/biweekly map:

(Pop out.)

Day 201: Even the Pepperoni

The Brazilian news reports that Brazil has exceeded our death rate; however Worldometers still has them nine deaths per million behind us, with 111,000 deaths total. Neither death rate tops the charts; the big performers there are Belgium and Peru (plus nominal country San Marino at #1).

USA Today reports that our latest shortage is pepperoni. So far it's only driven up the price for pizzerias, who, thus far, have not passed the costs on to their customers.

Massachusetts' cases are up about a quarter of a percentage point today. It's cities and towns day, for which event we've corrected some text on last week's maps and are still working on rearranging this week's.

P.S. After discovering and correcting an error in the raw data, PlagueBlog presents the state of the Commonwealth (modulo any remaining errors):

(Pop out.)

Day 199: The Rotten Apple

Long-time PlagueBlog readers may recall Day 2 of the coronavirus pandemic (February 2nd), on which New York City had a COVID-19 false alarm. They didn't really get going until Day 32 (March 3rd), and now, despite being the most recovered and immune place in the nation, they have a new problem: the death of New York City. A few days ago, James Altucher wrote up a resigned obituary of the city that never sleeps, recounting its death spiral in scenes from a rats-deserting-the-sinking-ship scrapbook he's been keeping. It's depressing. And I say that as a Bostonian who wants New York City to die just on principle.

As a Bostonian I also have to worry about whether it will happen here. Somerville has already made The Atlantic as a poster city for bad COVID decisions about sending children to school (one of the most basic functions of a modern society), not to mention our governor being the very first example in their article against the Fun Police.

The combination of criminalization and unscientific moralism is ineffective and counterproductive, and often leads authorities to take actions that may yield more infection.

We even need a map to track how many local colleges and universities have gone online this fall, despite our state being just as recovered as New York. I think the only thing we really lack here is a chattering class to navel-gaze about how many of their friends have moved to Phoenix or the woods and how many of their favorite restaurants have closed.

Massachusetts' cases are up a fifth of a percentage point today. Here is a map incorporating the new columns of the cities and towns data; the daily rate seems to be just a fourteenth of the bi-weekly rate (that we were computing for you out of the bi-weekly case counts), with certain towns omitted. As usual, we have estimated the rates for the omitted towns. The MDPH is also tracking the change in percent positivity explicitly. Though PlagueBlog considers this a relatively silly value, it is included in the map below.

(Pop out.)

Day 196: The Mask on the Soapbox

Opinions seem mixed on whether New Zealand has traced its outbreak to a new strain (to them) or has yet to do so because many of their previous samples have yet to be sequenced. It's still up in the air whether COVID came in on cold freight, on people (through the not-entirely-sealed borders or a failure of quarantine), or had been percolating undetected for their 102 case-free days.

At a presser in Wilmington, Virginia, Democrat Presidential candidate Joe Biden called for a nationwide mask mandate:

“Every single American should be wearing a mask when they’re outside for the next three months at a minimum — every governor should mandate mandatory mask wearing,” Biden declared.

PlagueBlog doubts this is going to win him many swing voters, and reminds readers that in the vast majority of situations there is no reason or evidence for wearing a mask outdoors.

Though there's no new germ boat today, the Washington Post reports that the plague ships of Italy are getting ready to sail again. (Tourism is 13% of the Italian economy.) In passing they mention the Hurtigruten cruise line apologizing for "fail[ing] to follow its own procedures" to prevent the cases aboard Germ Boat #30, the MS Roald Amundsen.

Locally, the Commonwealth's deadline for school districts to figure out what they're doing about coronavirus is today. NBC reports that 31 districts have already decided to go all-online. (There are 292 regular school districts in Massachusetts, plus some vocational and charter districts that may or may not be treated separately for this purpose.)

Aside from some strangeness involving fewer than five people in Hancock (at the far west of the state), the old case map is looking pretty plateaued as usual. The newer, two-week positivity map has some hot-looking spots, including the cities involved in the Stop the Spread testing effort. A daily rate map is in the works but not ready yet.

(Pop out.)

(Pop out.)

P.S. Massachusetts' cases were up a quarter of a percentage point today.

Day 188: Hard Labor

The world jumped well past 19 million cases today, with the US hitting 5 million and #3 India reaching 2 million cases. Bolivia (now #29) has surpassed China in cases, and the Daily Mail reports that failed state Venezuela (#64) is forcing non-mask-wearers into hard labor.

In the US, Governor DeWine of Ohio has tested positive. ABC News reports that the student who took some viral video of students not wearing masks in a crowded Georgia high school hallway has been suspended. (PlagueBlog recommends saving your education budget for PPE, not blowing it on the inevitable lawsuit.)

On the high seas, a third plague ship has joined the new fleet: passengers and "most" crew of Germ Boat #32, the SeaDream I out of Bodø, Norway, have been quarantined after a passenger from their previous cruise tested positive.

Massachusetts' cases are up a fifth of a percentage point today. Yesterday was cities and towns day; the new maps are below. In addition to being late, the report from the MDPH came with no comment about the reporting issues from late last week or whether they affect these numbers at all. Caveat lector.

(Pop out.)

(Pop out.)

Day 182: The Half-Year Mark

Ecuador surpassed China in cases yesterday to become #28. Even though Sweden is #30 now, Bolivia (#32) and the Dominican Republic (#35) are much bigger movers and more likely to be the next to pass China.

Despite the excitement of having reached half a year of coronavirus in Boston, it's been a slow news day. Massachusetts cases are up 0.44% today, due to the ongoing reporting delay issues. Here are the usual maps for the cities and towns data from Wednesday, which came with no warnings about the reporting delay:

(Pop out.)

(Pop out.)

Day 174: Just Say No To Lockdowns

South Africa has exceeded 400,000 cases; it is now more than halfway to #4 Russia. Peru (#6) and #7 Mexico are also climbing steadily towards 400,000, and #15 Colombia and #20 Argentina are racking up more than 5,000 cases a day.

In the US, California is not quite #1 yet, but they're looking like a shoe-in for tomorrow. Louisiana at #12 has just broken the 100,000 case mark, and has joined the queue of states waiting to dethrone #9 Massachusetts. For unknown reasons, the MDPH has not published our case numbers for today yet (an unprecedented 5 hours after the usual time).

A paper in EClinicalMedicine evaluating policies and socio-economics in 50 countries concludes that lockdowns don't help, smoking does help, GDP is bad for you, and obesity definitely hurts. Despite not helping with mortality, lockdowns did improve recovery rates, which the authors speculate had to do with flattening the curve (as the kids say). They write off the GDP correlation as possibly testing-related, or too much disposable income for vacationing in COVID hotspots like ski resorts.

The authors are also mystified by the smoking data, wave their hands a bit about average age in smokier countries, and think more study is necessary, when plenty has been done already. That they can't read the literature doesn't necessarily mean they can't do the math (multivariable negative binomial regression), though the 50 countries with the most cases as of May 1st is not the most promising data set, all their statistical precautions aside. Still, it's encouraging that they reproduced a well-known pro-smoking result despite not knowing about it.

On the mapping front, now that there are two weeks of the new data we can restore the case rate map. These case rates will be a bit different from the previous ones provided by the MDPH, based on their super-secret population numbers; the new map is based entirely on US Census numbers for our cities and towns (rather than only partially in those cases where the state was not previously reporting its own rate).

(Pop out.)

On the down side, the testing rate map is lost and gone forever, because the MDPH no longer reports test rates on a per-person basis, but on a per-test basis. While perhaps still informative about testing policy, the new data is no longer particularly relevant to coronavirus, so has not been mapped.

P.S. Massachusetts cases are up 0.3% percent today.

Day 173: NO Means Nitric Oxide

The most notable development on the international front is Bolsonaro's third positive coronavirus test, taken yesterday and reported today.

California is still a day or two away from New York's #1 spot, though nothing can ever take away Governor Cuomo's title of Most Likely To Kill Your Grandma. In fact, California remains remarkably inefficient at killing people, with only 200 deaths per million, and will probably never achieve New York's high of 1,676 deaths per million. (The US average is about 440.) Of course the New York Times pats itself and the rest of the Northeast on the back for having already killed our low-hanging fruit our amazing pandemic control. Yelp, of all places, is unimpressed by the grandma slaughter's lockdown's effects on the economy, though; they report 132,500 business closures at the moment, 55% of them permanent.

On the science front, a future article in the official journal of the International Nitric Oxide Society already available online posits (you guessed it) nitric oxide (NO) as a prophylactic against COVID-19. While the paper references some previous in vitro and inhalation experiments with NO against the previous SARS epidemic, they go into far more reviewing detail about the efficacy of smoking as a COVID prophylactic, and why NO might be the reason. PlagueBlog readers may recall that nicotine has also been proposed as the smokers' secret weapon, and the authors also mention carbon dioxide as a potential mechanism, though their hearts are clearly in the NO theory.

We hypothesize, in view of our knowledge of NO and positive experience with NO inhalation in the SARS epidemic, that the intermittent bursts of high NO concentration in cigarette smoke may be a likely mechanism in protecting against the virus. To copy the intermittent high NO concentration by breathing in NO from a gas tank is problematic, since NO₂ will build up during dilution with air to potentially toxic concentrations. Pulsed short bursts of high NO concentration will require a delivery system for inhaled NO that is independent on supply from a gas tank. Such a tankless system does exist, and has been approved by the US FDA.

In a rare moment of honesty for the scientific literature (which may have something to do with the majority of the paper's authors hailing from Sweden), the authors address the ongoing misinterpretation of the clear evidence in favor of smoking:

The reluctance may be due to a general opposition to smoking and fear of using tobacco to prevent the ongoing pandemic.

As the panicked masses would put it, smoking is better than dying.

Also on the science front is an interesting preprint out of Ireland, in which the region reporting experienced a significant drop (73%) in premature births (with no corresponding increase in stillbirths). The authors enumerate quite a few potential factors brought on by the lockdown, most of which were already known to be associated with premature birth, but have no specific data on those points. If PlagueBlog had to guess, we would go with the reduction in maternal work stress and physical workload as the most pervasively changed by the pandemic, rather than with more external, seasonal, and variable factors like air pollution, exposure to infection, and partner support.

Massachusetts cases are up a quarter of a percentage point again today. The cities and towns report is out; here's the data, set up the same as last week. (The PlagueBlog punch-card girls computers may grind out a few more details tomorrow.)

(Pop out.)