The data is suggestive

It’s been almost a month since George Floyd was murdered, and protests erupted across the country (and the world) very shortly after, so there’s been time enough for the coronavirus to piggy-back on the crowds and cause a surge in infection rates. But look at these plots, especially for Hennepin county!

That’s good news, but it’s a little confusing. Why aren’t those big crowds perfect petri dishes for the pandemic?

What’s more, a new analysis based on cell-phone tracking data suggests a surprising reason for the lack of protest-related spikes in COVID-19: In the cities with large protests, the wider population actually spent more time at home during the demonstrations — suggesting that any surge caused by virus transmission at the protests themselves would have been countered by an increase in social distancing among the rest of the cities’ populations.

While experts consulted by BuzzFeed News agreed that wearing masks and being outside may have reduced the risk of viral transmission at the protests, they pointed to other possible factors as well. Many of the protesters were young, for example, meaning that new infections that occurred while they were demonstrating would be less likely to cause severe disease and show up in official case counts. And even though hundreds of thousands participated in the protests, that’s still a relatively small number compared to the total population of the cities involved — so it might be hard to notice transmission of the coronavirus at the protests.

“The fact is that we will just never know for sure, because there’s too many moving parts,” Andrew Noymer, an epidemiologist at the University of California, Irvine, told BuzzFeed News.

Epidemiology is hard — too many variables, too many moving parts. This suggests, however, that you shouldn’t expect dramatic surges from the recent Republican rally, for the same reasons: small crowds relative to the greater population. That participants were generally older might have more effect, though, and BLM protest participants seem to be a lot more careful about using masks and distancing..

What’s worrying is that the article also shows recent rapid rises in states like Texas, Oklahoma, Florida, and South Carolina, and the country overall. That’s associated with states that have been generally opening up, and reducing mask and social distancing expectations. The lesson: general policy is far more influential than limited events. Republican governors are greater threats to public health than grassroots protests.


  1. blf says

    poopyhead, Is the “That’s good news” (start of 2nd paragraph) link borked? If not, then where does the block-quote come from? (Or, for that matter, the charts?)


    A related opinion column in the Grauniad a week ago, Why protests aren’t as dangerous for spreading coronavirus as you might think (June 18th):

    […] Protests will result in more infections, with the greatest burden falling on black communities. But the number of people infected is likely to be lower than you might expect. A growing body of evidence indicates that the vast majority of infections occur in so-called super-spreader events, almost all of which take place indoors. The risk of infection from outdoor protests is further reduced by the striking consistency with which demonstrators wear masks, and often make efforts to maintain personal distance. The evidence to date suggests that if people adhere to these practices, mass demonstrations will produce more coronavirus cases, but not an avalanche.

    Some US states with rising rates of infection, such as Arizona, Texas and Florida, began easing the lockdown before the recent protests began, without having met the federal government’s reopening criteria for reopening, including downward trends in cases and rates of positive tests. And scientific evidence suggests that the risk of returning to offices and shopping centres is probably greater than participating in large demonstrations. The crucial exception, of course, is when they are met with mass arrests. Throwing peaceful protesters into police vans and jails is unquestionably a recipe for mass infection.

    […] Protests will lead to infections, but our public health communications should clearly emphasise that the risk can be mitigated with symptom checks and tests (where available) before participation; masks, eyewear, and distancing, where possible; and testing and self-quarantine, where possible, afterwards. We should be actively sharing guidance on How to Protest Safely, as Amnesty International and others have done.

  2. Matt G says

    I started following the state of COVID-19 in Israel because of an interview from early March (?) with an Israeli-American Nobel Laureate (physiology, I believe). He was quoted as saying something like “I am not an epidemiologist, but I understand numbers.” He predicted that if Israelis took reasonable precautions, they should expect to have “no more than about” 10 deaths. They now have at least 308 deaths, and are entering their “second wave.” Epidemiology IS hard, and should be left to the epidemiologists. Another victim of Nobel Disease.

  3. KG says

    Yes, what we have seen, in a number of countries (Iran, Turkey, Brazil, India, Portugal, South Korea, parts of Germany and Australia) and at a range of infection levels, is a rise in cases following a relaxation of government measures to suppress the epidemic. Johnson is doing his level best to add England to that list (other parts of the UK are being more cautious, although not cautious enough in my view).

  4. stroppy says

    Any difference in dispersal between indoor and outdoor rallies?

    Outdoors you’ve got wind and…?

    Republicans seem to want to gather in caves and breath their toxified Freedom Fumes at each other.

    (or maybe the great god Reality just like liberals better)

  5. raven says

    Live updates: Federal government to scale back testing support even as hospitalizations reach new highs
    Washington Post June 24, 2020

    New coronavirus cases continued to rise in many parts of the United States, as seven states — Arizona, Arkansas, California, North Carolina, South Carolina, Tennessee and Texas — reported record-high numbers of hospitalizations since the pandemic began. Thirty-three states and U.S. territories reported a higher rolling average than last week. Even as case numbers climb, the federal government is poised to stop providing federal aid to testing sites in some hard-hit states, including Texas, where new cases have spiked sharply in recent weeks.

    PZ mentioned that Covid-19 cases are going up in the USA again.

    Cases are in fact, going up rapidly and hitting records in much of the USA.
    We are still in the first wave of the pandemic with no end in sight.
    The USA is on track to have 200,000 dead by the end of September.
    Then it gets cold outside, people spend more time indoors, the primary and secondary children’s schools open up, and the universities start again.
    What could go wrong here?

    The way it looks right now, the US pandemic is going to keep going as far as we can see into the future.
    As a high risk Boomer, I really wish I didn’t live here right now.

  6. raven says

    It’s no secret that as part of the US Exceptionalism, we are the worst hit country in the world for Covid-19 cases and deaths.
    One of the main causes was the all but nonexistent federal government response to the pandemic.

    The federal government under Trump/GOP did nothing much.
    They sort of wandered around haphazardly trying to figure out what to do, while Trump issued a few unenforceable orders with no followup.
    Then around June, the federal government just got bored with the whole pandemic virus-dead people thing and started ignoring it.

    The federal response right now is nothing much.
    Occasionally they will threaten a state governor or officials for putting some common sense social distancing rules in place.
    They are also claiming that they slowed down testing to make the case numbers look smaller, although they even probably failed at that.

    Testing is still a limiting problem in many locations, we don’t have a very large stockpile of Personal Protective Equipment, and the hospitals are starting to fill up again as the pandemic increases.
    And, there is no end in sight to any of this.

  7. says

    “It’s no secret that as part of the US Exceptionalism, we are the worst hit country in the world for Covid-19 cases and deaths.”

    Only in terms of absolute numbers at the moment. If you adjust for population, and ignore countries with less than a million people, The US is only 6th in number of cases and 7th in number of deaths. The UK is currently at second place in number of deaths, just a little behind Belgium.

    But the UK’s cases per day and deaths per day are still on a downward tend. The US cases per day are going up, and no doubt deaths per day will start rising shortly.

    I only hope your state governors take action very soon, as it’s clear that you’ll get no proper leadership from your president.

  8. Rich Woods says

    @Paul Durrant #10:

    The UK is currently at second place in number of deaths, just a little behind Belgium.

    Look, we’re working hard on this, OK? We as a country need to be world-beating, as our beloved Prime Minister rightly says. If it wasn’t for the intransigence of Scotland, Wales and Northern Ireland we’d have beaten Belgium into second place by now. That would serve them right, what with all their European institutions and polyglot citizens and small glasses of good beer.

  9. DonDueed says

    Here in Massachusetts, the state offered, and encouraged, free COVID testing for anyone who attended a protest or rally. They have reported the result, which is pretty encouraging: so far the rate of positive tests for that group is around 2.5%. For contrast, a positive rate of under 10% (for all testing, sustained for two weeks) was an important criterion for beginning the process of “opening up”.

  10. jaredcormier says

    Arizona will be interesting over the next 2 weeks; the possible transmission decrease from city-mandated masks will likely be offset by the rally this week coupled with the large number of people not following those mandates. As for leadership from governors re Paul @10; I wouldn’t hold out for governors in many states to do anything. See: Texas, Florida, and Arizona for examples.

  11. blf says

    jaredcormier@14, The current weekly average for positive Covid-19 tests in Arizona is 20%. (Yes, twenty percent.)
    Abbot in Texas has bent (a bit) to reality today and tightened lockdown somewhat (dunno details but very very probably nowheres sufficiently), apparently because hospitals are near capacity in Huston.

  12. says

    @#8, raven:

    If you take the somewhat more useful approach of reporting the US by state — which is what Canada has done for their numbers as a matter of course, which revealed early on that British Columbia was almost bizarrely safe — then a somewhat different picture emerges: the first wave is winding down in the states with major metropolitan areas, where it took off quickly, while it’s still just beginning in the states which didn’t have big numbers yet. Although I haven’t looked for it, I haven’t seen anybody talking about evidence of a second wave actually hitting any of the former hotspots which are dying down now.

  13. Ridana says

    Clearly tear gas and pepper spray suppress COVID-19 infection. These caustic bleach-like agents probably kill it in the air, just like that, in seconds, or flush it out the lungs somehow, like a kind of cleansing…

    Seriously though, if people staying home offset infections acquired during protests, I don’t think we’ll see a similar balancing of numbers with the Red Death Rallies. People who don’t go to them won’t be staying home because of them either. If they cause a spike, it’s going to be obvious. And unless they’re constantly blowing a curtain of purified air down around the Plague Dog like a bio-hood, it’s only a matter of time before he finally gets infected. If he doesn’t, then that settles it: he is the anti-Christ and is under supernatural protection.

  14. wzrd1 says

    PZ, you missed one confounder to your theorized lower infection rates at the rallies.
    Indoor with minimal mask usage vs outdoors with significant numbers wearing masks can cause significant numbers, due to outdoor attrition of microscopic droplets, as opposed to indoors being far friendlier to droplets, resulting in longer suspension times increasing the risk of exposure.

    I nearly chuckled over the megachurch’s braggadocio over their parishioners designed air conditioning system that’d remove 99%+ of particles. If it weren’t so dire, I would’ve actually chuckled. Unless the parishioners are willing to sit with a 10 – 20 MPH updraft, with the resultant audible rush of air, that fancy AC system is simply an exercise in intellectual masturbation. There’s a reason hood containment systems have such a narrow area to reach into and a specific airflow rate required to safely operate them.

    There are some rumblings that reports will be soon released for review and publication that suggest that the virus is mutating into a more easily transmitted form. Alas, no specifics, but I can think of a couple of ways it could do so that’s already present in the infection’s repertoire. A slightly more productive cough would be one way, with greater numbers of droplets allowing for easier transmission. I don’t expect to see a significant viral envelop change that’d allow longer survival time outside of the host, that’s way too many changes to reasonably expect.
    Although, there is very little about this unreasoning virus that is reasonable!

    I’ve long past concerned to gravely concerned, now I’m actually growing worried. We have a hugely successful virus potentially growing more successful. We have a imPresident that has the anti-Midas touch – everything he touches turns to shit. We have a slavishly obedient political party of essentially Jones

  15. wzrd1 says

    …level devotion, mixing the soft drinks.

    Sorry about the break, working on my smaller alternate laptop, the other one threw a wlan card on me and I’m awaiting its replacement.

  16. wzrd1 says

    RFon @20, the laugh is, any form of physicist uses much the same mathematics as epidemiologists use. The difference is, how it’s applied. So, an epidemiologist would be about as bad at quantum theory experimentation as a physicist is at epidemiology.

    @PZ, are you telling us that chemistry isn’t a sub-branch of physics?
    Let’s test that theory out by placing a small particle accelerator inside of a chemical reactor…
    Expensive hilarity ensues.