How much of the country-to-country difference in COVID-19 is simply random variation?

Some countries are suffering more from the coronaplague than others. Italy, for example, is a hotspot while Greece is scarcely touched. Parishioners in the worldwide Church of Shutdown would say that this shows the excellence of the Greek government’s early and eager shutdown (they are not afraid to relax at home and borrow some more money that the Germans can work extra hours to pay off?).

What if we are celebrating the lucky rather than those with special insight?

A map of influenza in Europe for 2015-2016 shows apparently similar countries with radically different levels of flu. Greece was slammed while Italy was barely scratched. Portugal was flu-free compared to adjacent Spain. Ireland suffered much more than adjacent Wales/England/Scotland. Finland had more flu than adjacent Sweden.

The 2016-2017 map, on the other hand, shows no difference between Portugal and Spain. France was hit hard. Greece was hit hard again.

If we step back one year, to 2014-2015, we find that Sweden and Finland have swapped places. Germany and Italy are hit hard while Greece is comparatively better off.

How about within the U.S. states? The CDC offers a map of “Influenza/Pneumonia Mortality by State”, adjustable by year. North Dakota and South Dakota may have dramatically different rates, despite being similarly situated. Vermont is always lower than New Hampshire, despite the geographic and demographic similarities (maybe southern NH gets infected by commuting into Boston?). Nevada is bad in most years, but not all. Florida seems never to be touched by flu and Colorado hardly ever. (It can’t be Florida’s tropical climate that saves it, however, because Hawaii usually has a high prevalence.) It seems that there is a significant amount of random variation and also a consistent pattern for some states. We could certainly look at this map and say that Florida, Vermont, and Colorado are examples of superb governance. Washington and Oregon are always much better off than California. What are they doing right?

If COVID-19 behaves like the flu, are a lot of the policy attributions that we’re making the result of accidents of fate?

Related:

18 thoughts on “How much of the country-to-country difference in COVID-19 is simply random variation?

  1. Interesting maps. Is there a correlation here?

    Bad flu then mild coronaplague: Greece, Finland
    Mild flu then bad coronaplague: Italy, Spain

    Suggests that some of the corona death rate is due to the accumulation of a larger proportion of fragile elderly people who wouldn’t have survived a bad flu season and have instead succumbed to coronaplague.

    • Aahhh Event 201. People have been talking about these things for a long time. From the way we’ve been responding, it doesn’t seem like many people were listening. Perhaps a few people now responding to COVID benefited from attending. The PDF implies all of this was planned, to wipe us all out financially and transfer all our future wealth away. I don’t know whether the former is true, but it’s doing a damn good job of the latter so far.

      I know the Bill Gates has been concerned about pandemics for many years now. They told the BBC they were personally stockpiling food in their basement. He’s talked about it many times: 2010, 2015, 2018. He did a TED talk on it. For a while there I thought he was going to ask Microsoft to release Windows Pandemic, with a browser pre-loaded with pandemic bookmarks to keep everyone riveted. Maybe he should have!

      https://www.cnbc.com/2020/04/16/bill-gates-stored-food-in-basement-for-years-to-prep-for-a-pandemic.html

      Here’s the link to the Event 201 pandemic exercise he references at the beginning of that PDF. The Johns Hopkins Center for Health Security has run several of these, with names like “Clade X”, “Dark Winter”, “Atlantic Storm”. It was covered by Forbes, Bloomberg, Axios, the Guardian, Gizmodo, the BBC, et. al.

      https://www.centerforhealthsecurity.org/event201/about

      “Event 201 was a 3.5-hour pandemic tabletop exercise that simulated a series of dramatic, scenario-based facilitated discussions, confronting difficult, true-to-life dilemmas associated with response to a hypothetical, but scientifically plausible, pandemic. 15 global business, government, and public health leaders were players in the simulation exercise that highlighted unresolved real-world policy and economic issues that could be solved with sufficient political will, financial investment, and attention now and in the future.

      The exercise consisted of pre-recorded news broadcasts, live “staff” briefings, and moderated discussions on specific topics. These issues were carefully designed in a compelling narrative that educated the participants and the audience.”

      Here’s the Forbes coverage from December 12, 2019:

      https://www.forbes.com/sites/judystone/2019/12/12/how-prepared-are-we-for-the-next-pandemic-not-very-experts-show/#308ba1703524

      “I have been particularly concerned about:
      — our reliance on other countries (like China) to provide the active ingredient for essential medicines.”

      A little late now!

    • By the way, it’s appropriate to reference the funding for Event 201:

      “Event 201 was supported by funding from the Open Philanthropy Project.”

      “Facebook co-founder Dustin Moskovitz and his wife, Cari Tuna, announced late Thursday that they plan on donating $20 million to political action committees that support Hillary Clinton in order to make sure Donald Trump never becomes president. It is the first time that the Silicon Valley power couple have publicly endorsed a candidate, although Moskovitz wrote that they have always voted for Democrats.”

      https://heavy.com/news/2016/09/cari-tuna-dustin-moskovitz-wife-bio-20-million-hillary-clinton-net-worth-facebook-founder-who-is-married-to-good-ventures-megadonor/

  2. My contacts in Greece say they need to register with the police through sms even for a trip to the supermarket with only essential outings allowed. Non compliance could result in a 5000 euro fine if your are found outside without the necessary permission and the police have been vigorously enforcing the edict. Maybe that’s why there’s more compliance since the culture there is not normally big on following rules that closely.

    • Demetri: The Germans will have to work extra hard in 2021 to pay off all of those 5000 euro fines!

  3. The country to country variation visible right now is entirely due to where they happen to be on the growth curve and differences in the shape of their growth curves. These differences are related to variations in environmental/demographic conditions, different interventions/where they happened to be on the growth curve when the interventions occurred, and luck. Still early days so probably mostly measurement error and luck, but to me these data suggest that the timing of an intervention may define its effectiveness (earlier=more effective).

    • Sigh. A better question might be: “What is Taiwan doing right?”

      It may be that no man is an island, but sometimes it’s good to be an island, man. They are either very good, or very lucky, or both. They really knocked it out of the park.

      1) Early inspection of travelers from Wuhan (Dec. 31, 2019) – before any cases were reported, they started inspecting all the passengers.
      2) Early activation of their Central Epidemic Command Center (CECC).
      3) First case: January 21, 2020. They *banned* the export of face masks.
      4) By the end of March they were making 13 million face masks *PER DAY* and they instituted a rationing system.
      5) It also appears they designated at least one hospital to specifically handle coronavirus cases.
      “In early February 2020 Taiwan’s Central Epidemic Command Center requested the mobilization of the Taiwanese Armed Forces to contain the spread of the virus and to build up the defenses against it. Soldiers were dispatched to the factory floors of major mask manufacturers to help staff the 62 additional mask production lines being set up at the time.”

      How many mask production lines does America have, even at this moment?

      Read the rest of what they did:
      https://en.wikipedia.org/wiki/2020_coronavirus_pandemic_in_Taiwan

      As of 04/23/2020 they’ve logged 427 cases, most of which were imported, and a whopping 6 deaths out of a population of 23.78 million people. They’ve only tested approximately 0.00217 of the population (2,170 tests per million). If the U.S. was as good/lucky as Taiwan, we’d have about 85 deaths right now, a handful of people in various ICUs across the country, no surge, and I’ll guarantee we wouldn’t be passing another $428 billion dollars of stimulus by unanimous consent.

      And all of that is “…despite being unable to gather WHO information due to being barred by China.”

      Also interesting: what happens when a Taiwanese Navy vessel experiences a coronavirus outbreak? Their Defense Minister offers to resign for letting his sailors get sick.

      https://www.reuters.com/article/us-health-coronavirus-taiwan-defence-idUSKCN22321R

      “I have asked President Tsai for punishment, and the president has asked the Defence Ministry to face the incident honestly and reflect thoroughly,” Yen said, adding that he would accept any punishment including being asked to resign.

    • Sorry Neal, I meant that reply to go to the main thread, not specifically to you. But Taiwan is a good example of early intervention also. They were very good and very lucky.

  4. “…the Germans can work extra hours…”

    I realize this is beside point, but it is very important to note that the Greeks work the MOST hours in the EU and Germans work the *FEWEST* hours.

    Long working hours are strongly negatively correlated with an economy’s productivity and individual health (both physical and cognitive). Perhaps there is a reason for the early and frequent references to the sabbath in both the Torah and New Testament.

    https://www.weforum.org/agenda/2018/02/greeks-work-longest-hours-in-europe/
    https://papers.ssrn.com/sol3/papers.cfm?abstract_id=2737742
    https://en.wikipedia.org/wiki/Biblical_Sabbath

  5. I don’t see the relevance for the U.S. of comparisons to Taiwan. There are roughly 22 million undocumented immigrants here (https://insights.som.yale.edu/insights/yale-study-finds-twice-as-many-undocumented-immigrants-as-previous-estimates ) plus a lot more who’ve come here and returned back to their native land. If Taiwan’s approach requires inspecting people who arrive in their island nation then by definition we could never do that (millions of people arrive in the U.S. by traveling across the border without ever stopping at a checkpoint).

    Many of my friends on Facebook who are Democrats say that they would choose a world in which more people die from coronavirus as long as it means that a Democrat (or fossil Democrat!) will be elected as president. In what alternate universe would Democrats have ever agreed to border controls (or a wall!) early enough in this process to effectively seal off the U.S. the way that Taiwan is sealed off by its island geography?

    On February 1, a day after Trump announced travel restrictions for people who had been to China within the preceding 14 days, the #1 Democrat attacked him for “xenophobia”: https://twitter.com/JoeBiden/status/1223727977361338370

    On January 31, the NYT ran a story quoting an epidemiologist saying that it was too late to stop coronavirus in the U.S. with an island mentality. (He was probably right, since now we know the first death in California from a locally-acquired disease was February 6.) https://www.nytimes.com/2020/01/31/business/china-travel-coronavirus.html

    So I think that there was never a situation in which the U.S. could have “pulled a Taiwan”.

    • I am not sure if Phil is trying to be funny, but Taiwan has done an excellent job in responding to this epidemic. There are lessons to be learned, including screening all arriving air passengers, wearing masks, and contact tracing. All of these could be implemented in the US.

      Anecdotal reports from San Francisco International in the last month said that we are not even checking temperature of arriving passengers. If true, that’s just pitifully weak.

      I don’t believe the number of people currently crossing the boarder illegally from Mexico or Canada is all that high, relative to the US population. Sure, there are some, but not millions per month.

      This isn’t the time for “Not Invented Here” syndrome. Taiwan has done very well, and we should learn from them.

    • G C: If a country wants to essentially eliminate coronavirus, as Taiwan seems to have done, even a handful of infected, but not tracked, people can re-seed the nation with infection, no? https://abcnews.go.com/Politics/tens-thousands-migrants-evaded-border-patrol-year-record/story?id=66616450 says that “In total, nearly 1 million unauthorized migrants were stopped at the border in the last budget year” (2019). If we assume 50 percent efficiency, that means that roughly 1 million migrants came in without being stopped and maybe 800,000 of the migrants who were stopped ending up staying by claiming refugee or asylum status? Perhaps the 800,000 asylum-seeking migrants can be screened, but what about the ones who never interact with American officialdom?

      On the other hand, if a country is on the “flatten the curve” plan, as the U.S. supposedly is, what difference does another few million infections make? Why exert ourselves to pick through individual humans as Taiwan does if we have literally millions of Americans infected in NYC alone?

  6. “If COVID-19 behaves like the flu, are a lot of the policy attributions that we’re making the result of accidents of fate?”

    I would suggest that early action can make a big difference. In particular, you want hospitals and airports to be on the alert. A column from January 26 compares the response of two different Canadian provinces to SARS in 2003:

    On March 7, 2003, two men with SARS were admitted to hospital in Canada.

    At 4:55 p.m., Mr. C., 55, was transported by ambulance to Vancouver General Hospital. Three hours later, Mr. T., 43, walked into the emergency room of Scarborough Grace Hospital.

    The aftermath of the SARS outbreak in Ontario is well known: The province would record 247 SARS cases (half of them health workers) and 44 deaths (including one doctor and two nurses), and the province’s economy virtually ground to a halt as SARS panic spread.

    British Columbia didn’t have an outbreak; it recorded four cases and no deaths.

    The difference was that B.C. was prepared.

    The BC Centre for Disease Control was proactive, monitoring the spread of SARS in China closely, and sending bulletins to emergency-room staff. As a result, Mr. C. was isolated immediately. Health workers took appropriate precautions, from paramedics to janitorial staff. The tracing of contacts was swift and rigorous.

    By contrast, Ontario had no central agency monitoring infectious diseases. Mr. T. sat in a crowded emergency room, then was placed in a ward. A nurse who was an avid reader of Chinese media was the one who alerted the hospital that the patient probably had SARS and more caution was required.

    By then, Mr. T. had already become a “super-spreader,” infecting many health workers and patients who crossed paths with him. It took a while to trace his contacts and, in a domino effect, the infection spread into the community and to other hospitals, with tragic consequences.

    It’s true that the US is in a very different situation from Taiwan.

    Canada invoked the Quarantine Act on March 25, requiring everyone arriving in Canada (whether Canadian or not) to quarantine themselves for 14 days, with legal enforcement. Prior to that, travelers were asked to quarantine themselves, but it was voluntary. Does the US have a similar requirement?

    The February 1 ban on travel from China exempted US citizens, permanent residents, and their families. That’s most of the people traveling between the US and China. So the travel ban didn’t do much.

    The Economist:

    Governments need to choose from the menu of options by comparing the costs of each measure with the benefits it brings — and the calculus will differ in different countries. Should masks be mandatory? (Yes, if supplies are adequate.) Should schools take pupils back? (More research is needed.) Which industries can safely reopen? (Factories can; hospitality is harder.) Do you shut borders or quarantine travellers? (Quarantine is better.) Policies will evolve as the science improves or the disease flares up. Sometimes, that will mean tightening the rules again.

Comments are closed.