COVID19 update, May 11, 2020: hydroxychloroquine bummer; breakthrough in understanding of the severe disease

Two major updates today, one a bummer, one confirmation of an insight at the cellular level.

(1) The first large-scale clinical trial with hydroxychloroquine, at NY Presbyterian, was just published in the New England Journal of Medicine. Dr. John Campbell comments at length on YouTube, and as is his wont, strenuously avoids politicking.

Watch the whole video. But in a nutshell: there is no statistically significant difference in outcomes between the hydroxychloroquine and control arms of the study. This is a major bummer, as many medical professionals (and not just President Trump) had high hopes based on initial positive reports and several plausible mechanisms.

As Dr. Campbell says, it sounded plausible enough at the time they were desperate for something, anything they could repurpose. Especially given the known immunomodulatory effects (cf. use in arthritis, lupus) and as it became increasingly clear people were getting killed by their own immune systems going amok rather than directly by the virus. Besides, it worked in the test tube against the old SARS.

More’s the pity, since it was something they could use off the shelf and didn’t cost an arm and a leg. So far, Remdesivir is the only thing that’s passed the double-blind test [it got FDA approval right after]— and that’s (a) only an incremental therapeutic benefit, no magic bullet; (b) a proprietary drug that Gilead themselves will have to license to other companies because they simply can’t manufacture enough. (Hoffmann-LaRoche probably can.)

(2) Now for the major insight (hat tip: Mrs. Arbel). Haaretz English Edition [*] has a write-up in popular language (archived copy here ) of a paper from the Weizmann Institute that just came out in the prestigious journal CELL.

This paper helps rationalize at the level of single cells what has become increasingly clear on an empirical, “macro” level: that COVID19 is really two diseases in one. The first stage is a unpleasant but not life-threatening viral disease — and about 80% of patients on average (fewer for older patients, but over 95% of young patients) just get better on their own, and that’s the end of it. The remainder, who proceed to the second stage, get a massive immune overreaction (“cytokine storm”, CS) that becomes life-threatening (and kills a nontrivial percentage of patients). I have linked the videos by Drs. Hansen and Seheult (both pulmonologists) about the clinical picture in previous updates; postmortem, several German and Swiss pathologists have shared the results of many autopsies. , where severe blood clotting secondary to CS was seen over and over, causing organ failures and strokes as well as ultimately death by heart attack or pulmonary embolism.
Now a paper from the Weizmann Institute, by the team of Prof. Ido Amit at the Department of Immunology, offers a glimpse at what goes on at the cellular level.

In the study, which [also involved] research assistants Amir Giladi and Pierre Bost, researchers used state-of-the-art genomic technologies which included a method known as single-cell genomics, an area developed and led by Prof. Amit. […] By obtaining a picture of the cell at a given moment, one can compare the differences between the activity of cells invaded by the coronavirus in severely and lightly affected individuals. Researchers can see which cells and genes are activated and which cells are silenced, thus learning about changes in inter-cellular communication and about cells that are activated by the virus in areas where it is active.

The key question of what differentiates biological processes and the actions of the immune system in severely ill COVID-19 patients as opposed to those who are slightly ill has been occupying researchers and physicians since the virus was first detected.
In the lungs of seriously ill patients, [Amit and coworkers] found that macrophages – cells that normally assist in ridding the lungs of infection, viruses and microbes – are replaced by cells that exacerbate the illness. The researchers also found that in seriously ill patients, the coronavirus neutralizes the immune system’s T-cells, which also fight infections, thereby allowing other viruses that are present in the body to inflict their damage. […] The researchers behind the study hope that a deeper understanding of the factors leading to a patient’s deterioration will help find weak spots in the chain of reactions initiated by the virus in severe cases, paving the way for effective treatments that would prevent or significantly curtail the impact of the disease.

The pattern of the disease among people who are hit hard is quite clear: After a week of mild symptoms, there is a rapid and sharp deterioration in their condition, characterized by hyperactivity of the immune system called a cytokine storm. This hyperactivity leads to serious damage to a patient’s health, often leading to a collapse of multiple systems, including the heart, liver and kidneys. In the lungs, the disease is characterized by damage to macrophage cells, whose role is to clear the lungs of infections.

The study analyzed hundreds of thousands of cells that were taken from the lung fluid of seriously ill patients, slightly ill patients and healthy people. The researchers discovered which types of cells are invaded by the virus and learned about its pathway. They found that the virus usually attacks epithelial cells, which in the lungs are responsible for respiration by enabling transport of oxygen from the air to the blood. “Due to the infection, the whole immune environment of the lungs undergoes a total transformation” explains Amit.

The study showed that in patients who are severely hit by the virus, there is a dramatic effect on the immune system as compared to patients who are only slightly affected. In the former, macrophages in the lung tissue are replaced by other immune system cells. “We found that they are replaced by monocytes, blood cells which accelerate a cytokine storm. They are recruited from the circulation as part of the overreaction of the immune system,” explains Amit.

The researchers found an enhanced presence of polypeptide cytokines called IL-6 and IL-8 in seriously ill patients. These cytokines are usually released by the monocytes, serving to either augment or suppress inflammation according to need. In this case, they facilitate inflammation. “The cytokine storm produced by the virus prevents the immune system in these patients from launching adaptive processes which are required for mounting an appropriate immune response,” says Amit. “In other studies we’re involved in, together with researchers from China and Italy, we see enhanced cytokine levels in the blood of severely ill patients before any pathological signs are evident.”

Another change that accompanies the cytokine storm involves the activity of T-cells. “In contrast to patients with light symptoms, seriously ill patients have T-cells that are neutralized and inactive,” says Amit. The researchers found that this dramatic change causes indirect damage, such as infection by other viruses which the immune system had previously managed to repulse.

I’d been wondering for a while for how many people who died of COVID19, secondary opportunistic infections (by viruses or drug-resistant “hospital bacteria”) were the proximate cause of death, or a contributory one, even if the root cause was still COVID19.

The researchers are now developing clinical studies that will use treatments to protect macrophages, with the hope that they will be able to prevent a deterioration in patients who are mildly impacted by the virus.

More than that: this may give another impetus to treatments that combine immunomodulators with anticoagulants (to combat the severe thromboses that appear to be a common by-product of the severe disease).

COVID19 update, May 10, 2020: more on COVID19 outbreaks at German meat processing plants; BND drops bombshell about China and WHO; miscellaneous updates

(1) COVID19 outbreaks at meat processing plants are not just a US phenomenon anymore. Apropos the report yesterday of large outbreaks at two such plants at opposite ends of Germany (here and here, both articles in German): it was pointed out that many at these plants are foreign workers living in very tight quarters. But in addition, a friend who is a Ph.D. biologist as well as a volunteer EMT responded: “Meat packing is one of those physical jobs (so high respiration rate) which happens in close quarters, in a cool and air[-conditioned] environment. Most other airconditioned environments are probably not so close together and/or do not involve the level of physical labor. The other possible idea is that meat surfaces and the aerosols generated cutting with band-saws might be a good place for the virus to survive and thrive.”

(2) RedState, quoting German weekly Der Spiegel, has a bombshell: The BND (Bundesnachrichtendienst or Federal Intelligence Serivce, Germany’s equivalent of the CIA — in a report that is otherwise critical of Trump— says the following (my translation from the original German):

“Nevertheless, to the BND’s knowledge, China urged the World Health Organization (WHO) at the highest level to delay a global warning after the outbreak of the virus. On 21st January China’s Head of State Xi Jinping, during a telephone conversation with WHO leader Tedros Adhanom Ghebreyesus, asked the WHO to withhold information on human-to-human transmission and to delay a pandemic warning. According to the BND, China’s information policy has resulted in the loss of four to six weeks worldwide to fight the virus.” [*]

Confirmation of what was obvious to many of us.

(3) Miscellaneous updates:

{*] original wording: “Nach Erkenntnissen des BND drängte China die Weltgesundheitsorganisation WHO allerdings nach dem Ausbruch des Virus auf höchster Ebene dazu, eine weltweite Warnung zu verzögern. Am 21. Januar habe Chinas Staatschef Xi Jinping bei einem Telefonat mit WHO-Chef Tedros Adhanom Ghebreyesus gebeten, Informationen über eine Mensch-zu-Mensch-Übertragung zurückzuhalten und eine Pandemiewarnung zu verschleppen. [new paragraph] Nach Einschätzung des BND sind durch die Informationspolitik Chinas weltweit vier bis sechs Wochen für die Bekämpfung des Virus verloren gegangen.”

UPDATE: via masgramondou, a second analysis of Neil Ferguson’s COVID19 model code that is even “better” (ahem) than the first. I’ve encountered enough modeler hubris in my day job that I believe I recognize it when I see it.

COVID19 update, May 9, 2020: Spreading reconsidered; German pathologist and Swiss doctors identify thrombosis as #1 killer in severe COVID19, Swiss clinical trial with aggressive anticoagulation treatment; meat supply chain disruptions in Germany

(1) The Daily Telegraph has a long article about an epidemiologist at St Andrews U., Scotland, who has been analyzing a number of epidemiological “contact tracing” studies on how COVID19 spreads.


An analysis of such studies was recently posted on Twitter by Dr Muge Cevik, an infectious diseases clinician and researcher from St Andrews University. It was promoted by Sir Jeremy Farrar, the head of the Wellcome Trust and a member of the Scientific Advisory Group for Emergencies (Sage). “If you read one thread,make it this one,” Sir Jeremy said.

(Twitter thread unrolled here)

Dr Cevik starts with a big Chinese study that traced 2,147 close contacts of 157 confirmed Covid-19 cases. The overall infection rate was six per cent, but it was much higher among friends (22 per cent) and family members (18 per cent). In terms of location, the main risk factors were homes (13 per cent) transport (12 per cent) and dinner and entertainment (seven per cent).

Broadly similar findings emerge in several other papers. Risk of infection is much higher within households or other enclosed environments in which contact is close and sustained. In the outdoors, it falls to something in the 0-5 per cent range.

[…]A Chinese study of 392 contacts of 105 confirmed cases found that the attack rate in children was just four per cent within the home, compared with 17.1 per cent in adults, for example. The infection of spouses was super high, at 28 per cent. 

Emphasising the link with age, another study found that household members over 60 were much more likely to become infected (18 per cent) than those under 20 (five per cent).

Children, it seems, are not only better able to resist the infection within the home but also less likely to bring it back with them. A study, funded by the Australian Research Council, of 31 household clusters including children found that only 10 per cent had been sparked by children.

Using these and other studies, Dr Cevik concludes that they suggest (not prove) the following:

[emphases mine in the quote below]
  • Close and prolonged contact is required for transmission of the virus. 
  • Risk is highest in enclosed environments such as houses, care facilities, public transport, bars and other indoor spaces where people congregate.
  • Casual, short interactions are not the main driver of the epidemic. 
  • Susceptibility to infection increases with age.

(2) Die Welt has a video interview (in German) with Prof. Nils Kucher at the Zurich university hospital. 

Summarizing his remarks:

  • We got severe COVID19 completely wrong in the past. This is changing as overwhelming evidence accumulates, e.g. from autopsies, and is published
  • Severe COVID19 is not a conventional viral disease at all, but a coagulation disease
  • Endothelial cells have ACE2 receptors. The virus docks there, cause inflammation of the endothelium, this leads to release of cytokines. (Cf. “cytokine release syndrome”, a.k.a. “cytokine storm”.) Thrombosis ensues, and eventually the patient dies of embolisms
  • It has taken us way too long to figure that out, which has cost needless lives
  • The way to prevent thrombosis is treatment with [?]molecular weight heparin. Guidelines traditionally limit this to hospital settings because of the risk of hemorrhage 
  • Often patients seem to be in decent shape, are sent to home isolation with just antifebriles, and then some die of thrombotic events
  • Now we [=Zurich U. Hospital] are running a clinical trial on 1,000 ambulatory COVID19 patients. Half get standard care, the other half also get heparin injections. All are closely monitored by telephone in case something goes awry. For the heparin treatment with that patient group, the risk of hemorrhage is rated as less than 1%
  • Sadly, doing a control group creates an ethical problem if it’s a deadly disease and you are pretty sure you have the answer

Related: a Hamburg coroner, Dr. Klaus Püschel , talks about what he has learned from 192 autopsies on COVID19 patients (Under a law in the Free City of Hamburg, autopsies are mandatory for deaths from an infectious disease.)

“Of the first dozen I autopsied, seven died of thrombose, four directly of lung embolisms.” 

Those results were just published [following peer review by four experts, which takes a while] in the Annals of Internal Medicine.

Independently of his colleague in Zurich, Hamburg University Hospital Prof. Stefan Kluge wonders if treatment of severe COVID19 patients should not primarily feature anticoagulant therapy.

[My “gut” tells me: a combination of anticoagulants and immunomodulators.]

Coroner Klaus Püschel : “This isn’t a killer virus, we’re not at war. We must not fear and should not give up.”

(3) Now even De Standaard (in Dutch) has woken up to the problem of food insecurity in the US. The article is a mixed bag: the journalist is clearly  floored by the generosity of Americans (in “nanny states” like Belgium, there is a tendency on the part of people to say, “oh, helping people, that’s the government’s job”). At the same time gets taken in by socialist snake oil peddled by the likes of “Beta” O’Rourke.

(4) On a related note, supply chain disturbances for meat are not just a US phenomenon: Here are two articles telling of outbreaks at two meat processing plants, one in Schleswig-Holstein in the North, the other in Baden-Württemberg in the South.

Consequently, meat prices are rising in Germany.

COVID19 update, VE-Day 2020 edition: software engineer on simulation code; what exactly does “peer-reviewed” mean?

Today, 75 years ago on May 8, 1945, World War Two ended in the European theatre with the unconditional surrender of Nazi Germany. “The instrument of surrender signed 7 May 1945 stipulated that all hostilities must cease at 23:01 (CET), 8 May 1945, just an hour before midnight.” However, since that was already past midnight, EET and Moscow time, the USSR and its satellite states marked VE Day on May 9, and Russia does so to this day. In Israel the day is unofficially marked on May 9, owing to the large number of elderly Russian immigrants who had actually fought in what Russians call “The Great Patriotic War”.

Meanwhile, some happenings on the COVID front.

(1) (Hat tip: masgramondou.) An experienced software engineer, formerly at Google, reviews Neil Ferguson’s simulation code in detail.Yes, the one that predicted two million dead in the US, which later had to be revised downward by a factor of twenty.. Read the whole thing, and weep. A few teasers:

My background. I wrote software for 30 years. I worked at Google between 2006 and 2014, where I was a senior software engineer working on Maps, Gmail and account security. I spent the last five years at a US/UK firm where I designed the company’s database product, amongst other jobs and projects. I was also an independent consultant for a couple of years. Obviously I’m giving only my own professional opinion and not speaking for my current employer.

The documentation says: “The model is stochastic. Multiple runs with different seeds should be undertaken to see average behaviour.” “Stochastic” is just a scientific-sounding word for “random”. That’s not a problem if the randomness is intentional pseudo-randomness, i.e. the randomness is derived from a starting “seed” which is iterated to produce the random numbers. Such randomness is often used in Monte Carlo techniques. It’s safe because the seed can be recorded and the same (pseudo-)random numbers produced from it in future. Any kid who’s played Minecraft is familiar with pseudo-randomness because Minecraft gives you the seeds it uses to generate the random worlds, so by sharing seeds you can share worlds.

Clearly, the documentation wants us to think that, given a starting seed, the model will always produce the same results.

Investigation reveals the truth: the code produces critically different results, even for identical starting seeds and parameters.

I’ll illustrate with a few bugs. In issue 116 a UK “red team” at Edinburgh University reports that they tried to use a mode that stores data tables in a more efficient format for faster loading, and discovered – to their surprise – that the resulting predictions varied by around 80,000 deaths after 80 days[…]

(2) “It’s not peer-reviewed!” You hear a lot in debates about COVID19 nowadays. But what does this really mean?

For a scientific paper to get published in a reputable scientific journal, it needs to undergo peer review: the editor (or an associate/section editor) sends the submitted paper out to (usually between two and four) experts in the field for their frank evaluation of the science. They write verbal reports, passed back anonymously to the author, and may also answer a questionnaire grading the paper on various criteria (novelty, technical correctness, quality of presentation, appropriate length,…). They also make a summary recommendation which is one of the following

  • Publish as is (rarely do all reviewers recommend this on 1st pass)
  • Publish subject to minor revisions detailed in the report. (Further review is typically not expected.)
  • May be publishable subject to major revision (and usually re-reviewing of the revised manuscript).
  • Not suitable for the journal, but may be publishable in _____
  • Not suitable for publication in any form

Where does one draw the line between “minor” and “major” revision? In practice, if (nontrivial amounts of) additional experiments/computer simulations/… are required, or if the interpretation needs to be radically overhauled, it’s considered “major”, otherwise minor. One round of the process easily takes a month or more, doubled if one or more reviewers insist on major revision, or if the paper is initially rejected and resubmitted to another journal. In fast-moving research areas (not just the present global pandemic), this causes frustrating delays. So sometime in the 1990s, when the web was still in its infancy, a group of particle physicists developed an online preprint server that, after a period under the rather confusing URL (which suggested a sideline of Los Alamos National Laboratory into adult entertainment), became known as Here scientists could share their freshly submitted manuscripts with colleagues ahead of publication, or even circulate drafts. Anybody wishing to comment on such a “preprint” could just email the author.Over time, similar sites came online for the life sciences (, medicine ( and finally chemistry ( Sure, there are spam and crank submissions to these sites (site managers try to keep out the obvious ones), but for the most part, submissions are legitimate papers in their original, pre-peer reviewed, form. Many of them, if the journal (publisher) allows this, update their submission with a “postprint”, i.e., the revised manuscript after peer review. ( and similar sites are set up such that the original and revised uploads are always preserved and accessible, to forestall a “Oceania is not at war with Eurasia” scenario.) Many journals nowadays, once a paper is accepted for publication, immediately put the accepted manuscript “postprint” online, and in priority disputes this date counts as the date of first publication.

Copy editing by the production staff, typesetting in journal format, proofreading by the author (often with some last-minute changes) may take several weeks more, after which the final “version of record” comes online, often at first with placeholder page numbers ahead of inclusion in a journal issue. (No further change is made after this other than updating the placeholder page numbers to the final ones upon inclusion in an issue. If the authors find a mistake in their own paper at this point, their only option is to publish an erratum.)

Peer review is definitely valuable, and there may be substantial changes between an online preprint and the version of record — but that does not necessarily mean the preprint is worthless, especially if it comes from an established research group, in which case it’s best regarded as a “beta release” — some changes may be expected, but the paper may already be quite useful. The anonymous peer review system has its own issues with bias and (both benign and malignant) “gatekeeping”, but for the most part has served the scientific community well. Its primary weakness at this point is that qualified reviewers become over-burdened with manuscripts to review — keep in mind this is unpaid service to the scientific community, and reviewers quickly learn not to respond too fast, or they get “rewarded” with more refereeing requests. And after all, you need to perform, manage, and publish your own research, aside from teaching and any administrative duties you might have.

Alternatives have been sought. Public open peer review is one of them, where the reviewers’ reports and critiques are visible online. This could potentially become a hybrid alternative to both the preprint system and anonymous peer review, with radical transparency to the reader. In the discussion on the community testing effort in Santa Clara County, we saw an interesting example.

COVID19 update, May 7, 2020: risk of severe case presentation increases with age too; meat processing plants; fraying lockdowns; Georgia (the country)

Busy day at work, so just some quick updates:

(1) There is a commonly quoted rule of thumb that 80-85% of COVID19 cases are mild, and the rest severe and life-threatening. But how constant is that ratio really?

I was emailed a copy of a report (in Hebrew) by a group that was consulted for our national COVID19 planning. In the section on expected hospital load was a table with a breakdown of hospital and ICU admissions by age bracket, apparently taken from a CDC Morbidity and Mortality Weekly Report, Screenshotting the table here:

* Lower bound of range = number of persons hospitalized, admitted to ICU, or who died among total in age group; upper bound of range = number of persons hospitalized, admitted to ICU, or who died among total in age group with known hospitalization status, ICU admission status, or death.

Needless to say, these are data early in the epidemic (when the group had to make its recommendations). But if we use ICU admissions as a proxy for the number of severe cases, then we see a clear increase with age, the way it is seen for mortality.

(2) Elsewhere on the CDC site, one finds a report about the conditions and challenges at meat processing plants
Some quotes:

During April 9–27, aggregate data on COVID-19 cases among 115 meat or poultry processing facilities in 19 states were reported to CDC. Among these facilities, COVID-19 was diagnosed in 4,913 (approximately 3%) workers, and 20 COVID-19–related deaths were reported. Facility barriers to effective prevention and control of COVID-19 included difficulty distancing workers at least 6 feet (2 meters) from one another (2) and in implementing COVID-19-specific disinfection guidelines.* Among workers, socioeconomic challenges might contribute to working while feeling ill, particularly if there are management practices such as bonuses that incentivize attendance. 

 Facility challenges included structural and operational practices that made it difficult to maintain a 6-foot (2-meter) distance while working, especially on production lines, and in nonproduction settings during breaks and while entering and exiting facilities. The pace and physical demands of processing work made adherence to face covering recommendations difficult, with some workers observed covering only their mouths and frequently readjusting their face coverings while working. Some sites were also observed to have difficulty adhering to the heightened cleaning and disinfection guidance recommended for all worksites to reduce SARS-CoV-2 transmission.

Solutions to structural and operational challenges that some facilities adopted included adjusting start and stop times of shifts and breaks to increase physical distance between workers. Outdoor break areas were added at some facilities to decrease contact between workers. Some facilities installed physical (e.g., plexiglass) barriers between workers; however, this was not practical for all worker functions. Symptom and temperature screening of workers was newly instituted in some facilities and improved in others.

Sociocultural and economic challenges to COVID-19 prevention in meat and poultry processing facilities (Table 2) include accommodating the needs of workers from diverse backgrounds who speak different primary languages; one facility reported a workforce with 40 primary languages. This necessitates innovative approaches to educating and training employees and supervisors on safety and health information. In addition, some employees were incentivized to work while ill as a result of medical leave and disability policies and attendance bonuses that could encourage working while experiencing symptoms. Finally, many workers live in crowded, multigenerational settings and sometimes share transportation to and from work, contributing to increased risk for transmission of COVID-19 outside the facility itself. Changing transportation to and from the facilities to increase the number of vehicles and reduce the number of passengers per vehicle helped maintain physical distancing in some facilities.

(3) Lockdowns — useful as they undeniably are in densely populated urban regions — are not something that can be maintained forever. In the “hammer and dance” strategy of Tomas Pueyo, the “hammer” — the lockdown to break the epidemic’s back — is supposed to be hard and short, followed by a maintenance phase — the “dance” — that favors such social distance measures as yield the maximum reduction for minimal economic cost. (Face masks are one example.)
There are increasing signs that lockdowns in the US are fraying. Bethany Mandel, who lives in New York, speaks for many who express a sense that politicians of a certain stripe now keep “moving the goalposts” way beyond the original justification for lockdowns, and that they are completely oblivious to the staggering and still mounting economic costs for those who do not have guaranteed government paychecks. “We are tired of being treated like children,” one reads numerous times in the comments.[*]

Days ago, a hairdresser in Texas who had reopened her business made a tearful plea that she not be punished for wanting to feed her children. She was convicted to seven days in prison and a $7K fine. Now in a dramatic turn of events, not only have both the state Attorney-General and the Governor criticized the “excessive” punishment (the lockdown over hardressing salons ends Friday anyhow), but the state’s Lt.-Gov. donated the money from his own pocket and offered to serve the 7-day sentence himself as a proxy for the woman.

(4) DIE WELT looks at what it calls the “Coronavirus Model Pupil,” Georgia (the country, not the US state). The country, knowing it could ill afford such a calamity, locked down proactively rather than reactively, and is now exiting. (Possibly the most prescient thing it did was cut air links to China before they even saw their first case.) Now, despite a social culture much like Italy, it got a sum total of 610 cases, with just 9 (nine) dead.

(5) Finally, hard-hit Belgium is reopening after a few false starts. Summarizing the report from De Standaard (in Dutch):

  • starting May 10, every household can receive and host four designated people (a fixed list of four). Recommended to sit outdoors. No travel distance limitation
  • May 11, shops will open. One customer per 10m^2 (110 sq.ft.) floor area. Wearing a mask is recommended but not mandatory; generally recommended in situations where 1.5m (5ft) distance cannot be maintained (e.g. on public transit). If lines develop, elderly, handicapped, and care workers get priority
  • public transit in principle reserved for people who have no private means of transportation (cars, motorcycles). In practice, this will not be enforced
  • restaurants, cafés, cultural centers remain closed for now
  • public sports events are put off until July 31

[*] Without engaging in partisan political rhetoric: one reason the lockdown in Israel was largely successful, and saw a compliance well above what one might expect of our garrulous nation, was that we were treated like adults. Economic trade-offs were honestly discussed, including the limits to how long we could lock down before irreparable damage to our economy would ensue — and we were given a realistic time horizon from the start. At no point was there a sense of “bait and switch”.

UPDATE: via the Jerusalem Post, this interesting paper in the Journal of Medical Virology has an interesting theory about why the SARS-nCoV-2 coronavirus may elicit potentially fatal “cytokine storm” so much more often than seasonal influenzaviruses:

We have applied mathematical modeling to investigate the infections of the ongoing COVID‐19 pandemic caused by SARS‐CoV‐2 virus. We first validated our model using the well‐studied influenza viruses and then compared the pathogenesis processes between the two viruses. The interaction between host innate and adaptive immune responses was found to be a potential cause for the higher severity and mortality in COVID‐19 patients. Specifically the timing mismatch between the two immune responses has a major impact on the disease progression. The adaptive immune response of the COVID‐19 patients are more likely to come before the peak of viral load, while the opposite is true for influenza patients. This difference in timing causes delayed depletion of vulnerable epithelial cells in the lungs in COVID‐19 patients while enhancing the viral clearance in influenza patients. Stronger adaptive immunity in COVID‐19 patients can potentially lead to longer recovery time and more severe secondary complications. Based on our analysis, delaying the onset of adaptive immune responses during early phase of infections may be a potential treatment option for high risk COVID‐19 patients. Suppressing the adaptive immune response temporarily and avoiding its interference with the innate immune response may allow the innate immunity to more efficiently clear the virus.

COVID19 update, May 6, 2020: Dr. Hansen on what we learn from autopsies of COVID19 patients; infection fatality rate; Neil Ferguson’s downfall

(1) In a previous episode, I partly translated an interview with a German pathoogy professor and his assistant about what they learned from autopsies on COVID19 patients pathology professor discusses what can be learned about COVID19 from autopsies

Now Mike Hansen MD discusses the same in English

Blood clots, endothelial inflammation, and in general the immune system trying to kill you in order to save you. The blood clots account for much if not all of the damage to all sorts of organs. If only we could reliably stop this from happening, COVID19 would be a nuisance, not an emergency.

And Roger Seheult MD just posted this video on the exact same subject.

The rule of thumb is that 80-85% of symptomatic cases get the mild version that affects the upper respiratory tract but not the lower lungs. These patients, seemingly without exception, get better on their own — it’s the remaining 15-20% that get lower pneumonia who may be fighting for their lives against bad odds. The sooner such cases are recognized, the better to start administering oxygen and anticoagulants, and perhaps remdesivir if available.

(2) But is this “80% rule” constant across all age brackets? A breakdown of case fatality rates by age cohort makes this seem quite implausible. The figures below are taken from the daily report of the Korean CDC for South Korea, from the RKI’s daily report for Germany, and from an informally shared report for Israel.

Age groupIsraelSouth KoreaGermany
80+13.023.918.4 average 70+
60-691.82.61.65 average 30-69
case fatality rates (%) per age bracket

It is pretty obvious with the low mortalities in the lower age brackets that severe cases cannot account for 15-20% of the age bracket. (I would love to lay my grubby hands on statistics of ICU admissions in Germany as a proxy for severe cases.)

Note also, by the way, that Israel’s mortality seems to be lower across the board than South Korea’s and especially Germany’s. While Israel’s medical system punches above its weight, it would be silly to claim it is that much better than Germany’s or South Korea’s. I also can’t think of any genetic reason, and in terms of vaccination schedules, Israel is similar to the other two countries. Now it is quite possible that despite pretty aggressive testing, Germany’s denominator is too low; South Korea had meticulous test & tracking from the beginning (one reason why they were able to weather the storm without hard lockdowns). Adjusting for population pyramid (Israel’s is by far the youngest of the three) accounts for much of the difference between Israel and South Korea. What remains with Germany?

It couldn’t possibly have anything to do with this fireball in the sky, 150 million kilometer away, could it? Israel of course enjoys a much warmer and sunnier climate — its largest population centers are near the 32nd parallel, compared to the 52nd parallel for Berlin. Lots of sun means more vitamin D and a stronger immune system; warmer and especially sunnier weather means virus survives less time on surfaces, clothes,…

(3) And as long as we’re crunching numbers:

German virologist Prof Hendrik Streeck, from community-wide testing in Gangelt in North Rhine-Westphalia, derives 0.36% as the infection fatality rate, but allows it might be as low as 0.24%. He suggests the ratio between his IFR estimate and the official CFR as an estimate for the Dunkelziffer or stealth infection rate.

The small country that tested the highest percentage of it s population is, of course, Iceland, which according to now has tested over 14% of their entire population (of about 360K). With ten (10) dead out of 1,799 confirmed infections, this suggests an upper limit of 0.56% to the infection fatality rate (IFR). German virologist Prof. Hendrik Streeck derived 0.36% from nearly whole-population testing of a single hard-hit village (Gangelt), but in an interview on Unherd I’m now watching, he allows it might be overestimated and be closer to 0.24%.

(Streeck also talks about the initial outbreak in Gangelt through mass spreading at a carnival celebration: packed indoors, loud singing, loud shouting over the “music”, traditional kissing on the mouth [UGH],… None of this is representative of going about regular business of life in a country where people keep each other at arm’s length, like the US.)

The infamous “two million will die” prediction of Neil Ferguson was based, among other assumption we now know to be incorrect, on an IFR of 0.9% Swedish epidemiologist Johan Giesecke [sp?] speaks of 0.1%, a number similar to what some of the community testing studies in California imply. I am leaning toward 0.3±0.2% as a rule of thumb.

(4) Speaking of Neil Ferguson: despite his vocal advocacy of social distancing and lockdown, he was forced to resign after being caught meeting with his lover , a younger woman in an allegedly “open” marriage to someone else. Toby Young in the Spectator cannot suppress a degree of Schadenfreude.

COVID19 update, May 5, 2020 edition: protective effect of MMR vaccination? strokes, vitamin D; ACE inhibitors and ARBs

Some quick updates, as today is very busy at work:

(1) Intriguing preprint from Cambridge U., via Yves [not Cohen]:

Franklin, R.; Young, A.; Neumann, B.; Fernandez, R.; Joannides, A.; Reyahi, A.; Modis, Y. Homologous Protein Domains in SARS-CoV-2 and Measles, Mumps and Rubella Viruses: Preliminary Evidence That MMR Vaccine Might Provide Protection against COVID-19. medRxiv 2020, 2020.04.10.20053207.

The COVID-19 disease is one of worst pandemics to sweep the globe in recent times. It is noteworthy that the disease has its greatest impact on the elderly. Herein, we investigated the potential of childhood vaccination, specifically against measles, mumps and rubella (MMR), to identify if this could potentially confer acquired protection over SARS-CoV-2. We identified sequence homology between the fusion proteins of SARS-CoV-2 and measles and mumps viruses. Moreover, we also identified a 29% amino acid sequence homology between the Macro (ADP-ribose-1’’-phosphatase) domains of SARS-CoV-2 and rubella virus. The rubella Macro domain has surface-exposed conserved residues and is present in the attenuated rubella virus in MMR. Hence, we hypothesize that MMR could protect against poor outcome in COVID-19 infection. As an initial test of this hypothesis, we identified that 1) age groups that most likely lack of MMR vaccine-induced immunity had the poorest outcome in COVID-19, and 2) COVID-19 disease burden correlates with rubella antibody titres, potentially induced by SARS-CoV2 homologous sequences. We therefore propose that vaccination of ‘at risk’ age groups with an MMR vaccination merits further consideration as a time appropriate and safe intervention.

Excerpt from Figure 4 of the paper:

(2) John Campbell has more on strokes as a complication of excessive blood clotting in severe COVID19,

as well as on a preprint from Indonesia that appears to show a very strong link between vitamin D deficiency and COVID19 mortality:

(3) Good news for people who are on ACE inhibitors and ARBs (angiotensin release blockers):

COVID19 update, May 4, 2020: French evidence that epidemic got to Europe in December; intriguing hydroxychloroquine find; Michael Levitt interview

(1) Via reader Yves [not Cohen], a French media report that COVID19 was found in a blood sample taken from a lung patient at a suburban Paris hospital on December 27, 2019. [Hand-corrected DeepL translation]

Invited on the set of BFMTV on Sunday, May 3, Professor Yves Cohen, head of the intensive care unit at the Avicenne hospital in Bobigny (Seine-Saint-Denis), and at the Jean-Verdier hospital in Bondy (Seine-Saint-Denis), claims to have had a patient infected with Covid-19 at the end of December 2019. […] “We had a positive case at Covid-19 on 27 December 2019, when he was hospitalised with us at Jean Verdier,” he explained on the air. An analysis of the serological PCR tests carried out on the 24 pneumonia patients in December and January in these two hospitals led to this conclusion.[…] Dr Olivier Bouchaud, an infectiologist at the same Avicenne hospital, confirms this information. “PCR samples taken from a patient clearly show that he [tested positive for] Covid. We did have a first case in France on 27 December,” he adds. […] Professor Cohen mentioned that the infected patient had been ill for 15 days and that he had infected his two children, but not his wife. This person had not made any recent trips. For Doctor Bouchaud, “this does not necessarily mean that he is patient zero in France. But it does suggest that more research is needed to find out”. […] So far, the first officially recorded cases in France are those of three people, on January 24: a Frenchman of Chinese origin and two Chinese tourists who stayed in Wuhan, the original focus of the epidemic which appeared in December.

This, together with the earlier report that COVID19 was found during autopsy of a Santa Clara County patient who had died early February (which places the infection roughly at mid January), pushes the COVID19 timeline further back. [The French story is of course being picked up in Chinese propaganda media as “proof” that the virus did not come from China. Mais bien sûr, et je m’appelle Napoléon Bonaparte.]

(2) In the middle of an article about an Italian pharmacist who claims she has uncovered the main mechanism behind severe COVID19 this interesting nugget can be found:
“Chiusolo told the Post, the Italian Society of Rheumatology interviewed 1,200 rheumatologists throughout Italy to collect statistics on contagions. Out of an audience of 65,000 chronic lupus and rheumatoid arthritis patients who systematically take hydroxychloroquine, only 20 patients tested positive for the virus [and none of those ended up in the ICU or died].”
Time for a little back-of-envelope calculation. According to Worldometers, Italy has 210,717 documented cases out of a population of 60.2 million, or 0.35% of the population. This is almost certainly a gross underestimate, but 0.35% of 65,000 lupus and rheumatoid arthritis patients would be 228 — more than ten times the observed number.

An interesting control would be to check patients on some other long-term mild immunosuppressor drug (steroids? Multiple Sclerosis patients on Copaxone? Reader Laura R. suggests )


Mike Levitt, 2013 Nobel Prize winner in Chemistry, getting interviewed about COVID19 on UnHerd. I disagree about 30% of the time (and he should hire a fact-checker — his claim that Germany did not go on lockdown is peculiar to say the least), but a lot to chew on. 

He has a rather interesting way of expressing mortality: weeks-equivalent of annual all-causes mortality. He estimates COVID19 will end up being about 4 weeks worth, which may be an easier number to grasp and keep in their head for people who don’t juggle data all day in their day jobs.

(4) Via Instapundit: a long WIRED article “inside the early days of China’s coronavirus cover-up”.

Seasoned journalists in China often say “Cover China as if you were covering Snapchat”—in other words, screenshot everything, under the assumption that any given story could be deleted soon. For the past two and half months, I’ve been trying to screenshot every news article, social media post, and blog post that seems relevant to the coronavirus. In total, I’ve collected nearly 100 censored online posts: 40 published by major news organizations, and close to 60 by ordinary social media users like Yue. In total, the number of Weibo posts censored and WeChat accounts suspended would be virtually uncountable. (Despite numerous attempts, Weibo and WeChat could not be reached for comment.)

Some people say this is China’s Chernobyl. On the contrary. While both disasters happened under totalitarian Communist regimes, COVID19 makes Chernobyl look like a kitchen ketchup spill. 

(5) Bloomberg looks at how the Hawaii tourist industry (about 20% of the island’s economy[*]) has been devastated by the pandemic. At least they have the naval and other military presence to keep the rest of their economy going, plus some agriculture.

[*] indirectly it’s more, of course. In another illustration of Bastiat’s timeless essay “That which is seen and that which is not seen” (original title: Ce qu’on voit et ce qu’on ne voit pas), the sudden drop in purchasing power of those working in the tourist industry has a knock-on effect in other sectors.

UPDATE: what does the milder version of the disease feel like? In this article from March 12, a woman age 37 from the Seattle area shares her experiences:

Schneider revealed how she first began experiencing flu-like symptoms on Feb. 25. The symptoms occurred three days after she attended a party that was later identified as the place where at least five others caught the bug.

“I woke up and I was feeling tired, but it was nothing more than what you normally feel when you have to get up and go to work, and I had been very busy the previous weekend,” she said.

She felt a headache coming on around noon, along with fever and body aches. This was enough to cause her to leave her office at her biotechnology firm and head home.

The marketing manager napped but woke with a temperature that peaked at 103 degrees Fahrenheit that night.

“And, at that point, I started to shiver uncontrollably, and I was getting the chills and getting tingling in my extremities, so that was a little concerning,” she said.

She took over-the-counter flu medication, and called a friend to be on standby in case she needed to be taken to the hospital, but the fever receded over the following days.

Schneider wrongly assumed she didn’t have COVID-19 because she didn’t experience the usual symptoms such as coughing or shortness of breath.

She was up to date with her flu shot, but thought her illness was due to a different strain. When she visited her doctor, she was instructed to go home, rest up and drink large amounts of fluids.

The way she began to suspect she had something more serious was via social media. A friend on Facebook posted that several folks from the party had developed similar symptoms. These people went to their doctors, where they tested negative for the flu, but were not offered coronavirus tests because they were not showing the common signs of coughing and difficulty breathing.

Smartly, Schneider enrolled in a research program called the Seattle Flu Study in hopes of getting to the bottom of her sickness. She was mailed a nasal swab kit by the researchers, which she mailed back. Then began a wait of several days.

On March 7, she got a call with the bad news: She had tested positive for COVID-19. Surprisingly, Schneider felt relieved. “I was a little bit pleasantly surprised, because I thought it was a little bit cool,” she told the AFP, adding that she found it interesting from a “scientific perspective.”

Her symptoms had already subsided by the time she was diagnosed.

COVID19 update, May 3, 2020: Medical bureaucrats vs. frontline medicine; Israel vs. Belgium; agriculture turns away from globalized agribusiness and toward local market

(1) John Hinderaker of Powerline contrasts the hidebound “academic medicine types” who dominate the medical agencies and healthcare bureaucracies with the more daring doctors on the frontlines of the epidemic. 

He specifically refers to a story I have been covering here at some length: the changing understanding of the severe COVID19 disease picture as primarily “immune systems killing the patients in order to have them”. (The milder disease picture without lower lung involvement — which accounts for the overwhelming majority fo COVID19 cases — is typically a self-limiting ailment ranging from a mild cold to a nasty bout of flu.)

A group of critical care physicians representing the University of Tennessee, the University of Wisconsin, Eastern Virginia Medical School, the University of Texas and a number of other institutions have formed the Front Line COVID-19 Critical Care Consortium and released a bulletin setting out a recommended treatment protocol. The protocol is based largely on the fact that it is not the virus, but the body’s reaction to the virus, that kills patients:

[I]t is the severe inflammation sparked by the Coronavirus, not the virus itself, that kills patients. Inflammation causes a new variety of Acute Respiratory Distress Syndrome (ARDS), which damages the lungs.

Practicing doctors are highly familiar with inflammatory conditions and a number of known treatments have been adapted to COVID-19. The linked bulletin advocates early intervention–the key–using Vitamin C, heparin, Methylprednisolone and Hydroxychloroquine.

Dr. Paul Marik, Chief of Pulmonary and Critical Care Medicine at the Eastern Virginia Medical School, published a Critical Care COVID Management Protocol along similar lines. As a preventive measure, Dr. Marik recommends a combination of Vitamin C, Vitamin D, zinc and melatonin. Dr. Malik notes that “[w]hile there is no high level evidence that this cocktail is effective; it is cheap, safe and widely available.” For what it is worth, I have been following this regimen for some time.

Interestingly, in my own field of science (as in many others), it is precisely the academics who push for innovation, and the industrial users of the science and tech who are hidebound. However, frontline medicine is a different matter than, say, analytical chemistry: if you have patients dying on you — and what “the book” tells you isn’t working — you start trying to think outside the book. The much-maligned steroids actually do a lot of good in acute autoimmune or allergy attacks (I’ve gotten them following a suspected allergic reaction to an antibiotic). Sure, you don’t necessarily want to rely on steroids for long-term case management if you can help it, — but here you’re trying to stop a patient from getting killed by a massive immune over-response.

(2) The director-general of Israel’s health ministry positively evaluates the outcome of Israel’s containment/mitigation strategy: we actually have one of the lowest case fatality rates in the developed world. “If we hadn’t been tough, we’d have been in the same boat as Belgium”, he said. Belgium has a slightly larger population than Israel, and so far has 7,844 dead, compared to just 230 in Israel.

To be fair, of course, Israel has several factors on its side that Belgium didn’t:

  • a much sunnier and warmer climate — this both reduces the ability of the virus to spread (as the DHS study showed, virus deposits on surfaces are neutralized by bright sunlight)   *and* the additional vitamin D boosts the immune system of humans
  • Israel has a much younger population pyramid than Belgium (or Germany, or…) — which in and of itself will reduce mortality (in Germany, just one percent of dead are below age 50).
  • Israeli can seal its borders with comparative ease
  • War and terrorism being a permanent risk in Israel, both its medical systems and its population may be more primed to respond to calamities

This is aside from Belgium counting many deaths as “of COVID19” that were really from other causes, “with” actual or suspected COVID19. (Otherwise, it is impossible to explain why its absolute mortality exceeds that of next-door Germany, which has 7x the population.)

(3) Miscellaneous updates:

  • De Standaard (in Dutch) looks at the changing agricultural landscape in Belgium and The Netherlands. Farmers are getting second thoughts about big agribusiness on low margins, becoming dependent on export markets halfway around the planet , which demand is now at zero while people in the next town or county want to buy vegetables. “Locavores” (eating local) is an expensive snobbish trend in some parts of the USA, but quite feasible in Northwest Europe. 

COVID19 update, May 2, 2020 edition: Remdesivir gets FDA approval; detailed German statistics

(1) The top news item of the day is probably that Gilead Scientific’s antiviral drug remdesivir was given FDA Emergency Approval for use in COVID19 patients. Remdesivir is not a “magic bullet”, but it’s a start.

(2) Roger Seheult MD, pulmonologist and medical school instructor, gives a 1.5 hour recap video on what we know about COVID19.

(3) Miscellaneous updates:

  • the Ma`ayanei haYeshua [Wellsprings of Salvation] hospital in Bnei Brak, Israel (a COVID19 hotspot) has deployed an Israeli-developed UV-C room sterilization system. This is of broader relevance than COVID19, and if successful, will prove very helpful in the protracted and increasingly worrying struggle against hospital “superbugs” — bacteria resistant to every known antibiotic. (Such bacteria tend to develop in hospitals and long-term care settings through “Darwinian selection”, as both infections and treatment with aggressive antibiotics are frequent.)
  • Die Welt has a detailed video (in English, with German subtitles) on significant progress with a vaccine in the USA
  • worrisome reports about some peculiar COVID19-like pediatric syndrome noted in earlier updates: these now appear to have been identified as Kawasaki’s disease, which is of uncertain origin but some sort of autoimmune etiology is suspected. Coincidence or new cases triggered by COVID19 infection?
  • disturbing reports of COVID19 “reinfections” in South Korea appear to have been false positives in the test
  • Abbott’s new rapid COVID19 test, which claims 99% accuracy, has been approved for use in Europe.
  • if you give people perverse incentives to cook the books, and don’t balance that out with a deterrent for the act of cooking — well, don’t be surprised if books get cooked. NYC funeral director on candid recording about people who obviously died from otehr causes being coded as COVID19. Mind you, I am sure the un-inflated COVID19 mortality in NYC would be quite bad enough (“thanks” to very high pollution density and the subway as “the mother of all superspreaders”) — but those numbers struck me as anomalously high from the start. (As discussed in previous updates, numbers from Italy and Belgium are inflated for different reasons.)

(4) In contrast, countries like South Korea and Germany have rather more scrupulous reporting standards. I’ve linked previously to the daily Korean CDC reports: here is the detailed daily update (in English) from the Robert Koch-Institute (Germany’s infectious diseases authority, named after the discoverer of the tuberculosis pathogen). A few highlights from the daily report:

  • Only 19% of all cases occurred in persons aged 70 years or older — but these account for 87% of deaths.
  • cases per 100,000 people in age cohorts are fairly homogenous across age cohorts 20-29 through 70-79, climb sharply in the highest age cohorts, and drop steeply for ages 10-19 and especially 0-9.
  • mortality in age cohorts 0-9 and 10-19 are ONE (1) patient each, while age cohorts 20-29 and 30-39 account for just 6 and 14 deceased out of a total of 6,472. Yes, Virginia, ages below forty account for just 0.3% of all dead, and all ages below fifty for just 1%. Fifty-somethings add another 3.2%, sixty-somethings another 9.0%.
  • Their technique of estimating the effective reproductive number R consists of dividing the 4-day moving average of new cases by the one 4 days earlier. At present it is R=0.79, with a 95% confidence interval of 0.66–0.90. Any R value below 1 implies that the epidemic will wither away, while any value over 1 implies slower or faster exponential growth.
  • the report points to a European Union website with all-causes excess mortality graphs. These serve as a useful “sanity check” on COVID19 death reporting criteria for different countries.

COVID19 update, May Day edition: reduced contagion of and from children; reopening schools in Israel; miscellaneous updates

(1) De Standaard (in Dutch) reports on a study by the Dutch institute for public health about the infection risks from and to small children. I had intuitively expected that they would be mostly asymptomatic spreaders with a minority of mild overt cases (the few exceptions of young children who went into cytokine storm appear to be ‘man-bites-dog news’: newsworthy precisely because they are rare). Now the study[s main findings, mostly based on contact tracing, can be summarized as follows:

  • children are much less likely to be infected (symptomatically or asymptomatically) than adults if both are exposed to the same infection source
  • if they are infected, viral swabs of their throats show similar virus levels as in adults, but they cough less, and owing to smaller lung volumes, coughs can’t project as far

Needless to say, this has implications for whether it is safe to open schools (apparently, more so than previously assumed).

(2) Speaking of which, Israel is reopening Grades 1-3 of primary schools on Sunday, as well as preparation classes for the bagrut (high school leaving [and college placement] exam, like the French baccalauréat or the German Abitur) in the last two years of high school. (These exams play an important role in college admission and placement, together with the psichometri which is somewhat similar to the SAT.) Kindergartens will be opened on or before May 10 — originally they were to be opened on Sunday as well, but personnel requested more preparation time. Remaining elementary and high school grades will open June 1.

(3) Miscellaneous updates:

  • According to De Standaard’s breaking news ticker, Portugal intends to start reopening on Monday. It has been much less hard-hit than neighboring Spain despite similar demographics and climate: Sarah Hoyt (herself born and raised near Porto, and an alumna of Porto University) attributes much of the difference to the continued existence of a parallel private medical system in Portugal, while Spain’s was nationalized. Choice is always good for quality.
  • Shaarei Tzedek in Jerusalem, one of the largest hospitals in Israel (and the largest Orthodox hospital) is closing down its purpose-built corona ward as the last patient in it was discharged some days ago. The hospital had already gradually resumed normal operations as its COVID19 patient load declined.
  • Le Figaro (in French) reports that France will reopen by Départements [France’s administrative regions/provinces], not one-size-fits-all. Based on various criteria such as new infections, mortality, and hospital case load to capacity ratio, the ninety-odd Départements are classified as green (most permissive unlock), orange (more limited unlock), or red (most limited unlock).
  • And in what is probably the most exasperating thing I read on Victims of Communism Day, Die Welt’s reporter in Italy describes Italians looking on Germany as “the enemy” and China as “their friend”. Italians, remember timeo Danaos et dona ferentes. [I am wary of Greeks, even if bearing gifts.][*] As much as I can understand your resentment of the EU, becoming a client state of China will make even you nostalgic for the EU one day.

[*] There is a Hebrew idiom taken from 1 Kings 21:19 for one who unleashes a calamity, then sets himself up to benefit from the resulting devastation: ratzachta ve-gam yarashta? [Literally: hast thou murdered, and also inherited?]

UPDATE: a writer at City Journal calls for people to volunteer for “Hunan challenge trials” with the new Oxford University candidate vaccine, and put his money (or life) where his mouth is. In HCTs, willing subjects get the vaccine (or a placebo) then deliberately expose themselves to infection. This way it can be definitely ascertained whether a vaccine candidate has protective value. So far, over 8,000 others have signed on to a grassroots call for volunteers.