Looking at covid stats again

Our World in Data has an interesting section on covid infection modelling, that looks at estimates of the ‘true’ number of infections based on the available data. https://ourworldindata.org/covid-models

Meanwhile, here is an update of the graph I showed in my previous post.

The same caveats apply. The exact number of confirmed cases isn’t comparable between countries for multiple reasons. However, the trajectory of cases tells us a lot and to that end I’ve left Singapore and Sweden on the graph for comparison. As I said previously, both countries have different circumstances and different approaches to the pandemic but they’ve ended up with similar curves that show slow growth in cases but not rapidly surging growth.

Here is a similar graph but with a modified choice of countries that shows a variety of different patterns each of which tell their own national story.

Israel, Spain, the UK and Australia each have had situations of initial growth in infections that was then met with various lockdown measures, leading to a flattening of the curve, leading to a degree of relaxation of measures…then being hit by a second wave. I hadn’t been aware of how badly the situation had changed in Israel. Abigail Nussbaum has an excellent account of what went wrong here https://www.lawyersgunsmoneyblog.com/2020/07/from-new-zealand-to-america-covid-19-in-israel It looks like the UK may be on a similar path but potentially it may flatten out again.

Poor leadership, confused strategy and confused public health messaging do appear to be a common theme in countries struggling to control infection rates. I wonder if that will be confirmed when we finally get past this pandemic and researchers have had time to sift through how policy responses impacted both infection rates and death rates.

Our World in Data also has a page on the economic impact of the virus: https://ourworldindata.org/covid-health-economy I’ve been sceptical of the data on death rates for multiple reasons but that page has an interesting graph comparing confirmed deaths per million people and GDP growth compared with 2019 (the ‘growth’ is all negative of course). The article contends:

“But among countries with available GDP data, we do not see any evidence of a trade-off between protecting people’s health and protecting the economy. Rather the relationship we see between the health and economic impacts of the pandemic goes in the opposite direction. As well as saving lives, countries controlling the outbreak effectively may have adopted the best economic strategy too.”

I’d still be sceptical about making a stronger conclusion there i.e. that poor controls of the virus lead to worse economic impact. However, I think we can safely conclude that framing health measures as a trade-off with economic growth is misleading. Avoiding taking measures to control the rate of infection brings no observable national economic benefits.

9 thoughts on “Looking at covid stats again

  1. My friend in Tel Aviv was pretty despondent when we chatted on Monday. He thinks they’re heading into sustained hardcore lockdown for some weeks. In contrast, friends in Chile aren’t despondent, they’re broke and angry at a government that has done nothing for them.

    Liked by 2 people

      1. The “3% club” consists of Qatar, Bahrain, Aruba, French Guiana, Louisiana, Mississippi, Alabama, Florida, Brasilia, and 9 Brazilian states. Last time I dug around I couldn’t find any other substantial regions in that category per officially reported cases – though I wouldn’t be surprised if Lombardy, Madrid/Castile, New York/Boston area, and hotspots in Ecuador, Peru and Chile have been comparatively badly hit.

        I suspect that the Mexico is the worst hit country. They’ve got CFR’s comparable to the UK and Italy (Spain’s has plummeted to 5%, which fits with the hypothesis is that their “new” cases are mixed with old ones) with a much younger population and a test positivity rate of over 40%, both of which suggest that there is a disproportionate number of unrecognised cases. But several other Latin American countries are in a bad way.

        Liked by 1 person

  2. A comment on the Wikipedia talk page alleges that “new” case numbers in Spain include both current (PCR) and past (antibody) tests, which makes the official figures rather unhelpful for monitoring the progress of the epidemic there. (I was trying to work out why the case rate fell off a cliff two weeks ago, and then suddenly had two consecutive record daily case numbers.)

    According to coronavirus.data.gov.uk, the UK is doing the same, which makes interpreting numbers difficult, but as far as I know the UK isn’t doing as many antibody tests.

    Liked by 1 person

  3. Canada is in kind of the same state as you mentioned for Israel, Spain, the UK and Australia: early lockdown, flattening of cases, and then a rising of the curve again starting in late august or so. Now, that said, the positivity rate for tests is nowhere near as high as it was back when the daily number of cases was at its peak in April/May, so the current rise in confirmed cases is probably due at least in part to increased testing.

    Now, the U.S., on the other hand, depending on how you read the data, either still hasn’t left the first wave, or is starting a third. Their first peak in mid-April was mostly in the big cities, and got interfered with because that was mostly in Democratic states. Their second peak in July was when it hit a lot of the smaller towns which also had even less adequate health care or infrastructure to handle it. And now it’s rising again with the push to get ‘back to normal/school’.

    Liked by 1 person

    1. Germany also has rising infections, but tests have been massively increased and the number of hospitalisations and deaths remains thankfully low. Most of the infected are also young people where the risk of complications and death is extremely low (except in the US).

      Still a lot of wailing and gnashing of teeth and pointing fingers, particularly at weddings and private parties and protests (but only anti-covid measure protests, apparently the virus knows not to spread at Black Lives Matter, Fridays for Future and pro-refugee protests). Even though people returning from ill-advised holidays and football games are a much bigger problem.


      1. Yeah, outdoor protests of any kind don’t appear to be responsible for many cases anywhere.
        A lot of the concern about people heading to the beach was probably not warranted either.


      2. In Canada the recent increases have also been mostly in the (relatively) young, 20s or so. Which is another piece of evidence suggesting that the increase has been primarily coming from asymptomatic people getting tested.

        As for protests, well, the BLM protests have been wearing masks and keeping appropriate distancing, which the anti-covid measure protests obviously weren’t doing (because that was kind of their point)… being outdoors certainly helps, but that doesn’t mean you should otherwise act stupidly about it.

        Liked by 1 person

      3. Well Friday for Future protests have in their favour that if you see them, they wear masks and are keeping distance from each other. It is dificult to be the bad guy if you play by the rules.
        Football games are insane, but probably not that much responsible for the now, they are mostly starting to let the people back in.
        Holidays are imho the main reason for the rise in numbers.


Comments are closed.