Covid-19 Fatalities, Excess Fatalities and Likely Revisions

From CDC and the Atlantic/Covid Tracking Project.

Figure 1: Weekly fatalities due to Covid-19 as reported to CDC for weeks ending on indicated dates (black), excess fatalities calculated as actual minus expected (teal), fatalities as tabulated by The Covid Tracking Project/Atlantic (dark red), all on log scale. Source: CDC accessed 8/14/2020 vintage, Covid Tracking Project/Atlantic accessed 8/15/2020 and author’s calculations.

Three observations: (1) the official CDC count for recent weeks is likely to rise toward the unofficial count as certificates arrive, (2) the unofficial count remains elevated at a high plateau of about 7470/week; (3) recent weeks’ (about a month’s worth) CDC data are subject to severe undercounting, so inferring recent trends on the basis of CDC data is not advisable; (4) the IHME forecast from 7 August were on the mark for the week just passed (according to the unofficial count). IHME is right now about in the middle of the pack of models surveyed by Fivethirtyeight. Current forecast is for over 200,000 fatalities by September 19th.

From week ending February 29th through the week ending August 1st (the latest CDC data, which incorporates severe undercounting in the last three weeks), the cumulative CDC fatality tally is 154.7 thousand. Cumulative excess fatalities is 204.7 thousand, implying an additional 50.0 thousand Covid-19 fatalities above the official tally over this period.

See also NY Times article, for a regional breakdown, and additional discussion.

42 thoughts on “Covid-19 Fatalities, Excess Fatalities and Likely Revisions

  1. Moses Herzog

    Wow, like if it was just a regular stat, maybe not terribly surprising. But the difference in the CDC cumulative count and the cumulative excess, to MY eyes that’s nearly in the shocking category. I haven’t clicked on that NYT link (probably the same story), but I saw one that was dated maybe a week ago that broke down the excess deaths by state, and I did think that was one of the more interesting stories I have read about it up to now, and something that even the better journalists in my state have been absentee in discussing. Those excess death numbers are really pretty huge (even on the state level).

    I’m hoping this will get more journalistic attention as time goes on—because I think excess deaths is our best “rule of thumb” way to keep track of COVID-19 deaths that aren’t making it into the official counts.

    1. Dr. Dysmalist

      Leave out the “nearly,” and I agree with you fully.

      The more I see that graph, I’m further convinced that more than one data series is being cooked. Considering Trump’s wild, bogus claims of falsified economic data during his first campaign, I can only say again, “Every accusation is a confession.”

      1. Willie

        I am not sure this administration is competent enough to cook the books like you suggest. Washington State has been struggling with data reporting. This state has been reasonably competent in its response to the landemic, especially in light of the fact that we got hit early, before anybody knew how to respond to an infection. Even so, it has been an ongoing struggle to decide how to report Covid infections and deaths. I would not suspect the Inslee administration in Washington of carrying any water for bunker boy.

        1. Dr. Dysmalist

          Oh, I think that any book-cooking is occurring in red states and with the new ‘custodians’ of the data that was formerly reported to the CDC.

          Blue states in general, especially certain ones like Washington, have so far shown themselves to be straight shooters and therefore above suspicion.

          1. Baffling

            In houston the data being reported suddenly changed about a month ago, after gov abbott had a discussion with the hospitals and their release of information. Overnight the data changed from dire to not so dire. There is definitely hancky pancky going on with red state reporting.

    1. Willie

      And the Mariners can lose small as well as lose big. I hope they are not the Cubs of this century. I want to live to see the World Champion Mariners. Better start taking vitamins.

      1. Moses Herzog

        Usually you can judge from the owner and the general manager. The owner has the wallet and the GM figures out how to get player value out of it. If those two aren’t looking good then you’ll be waiting for that “lucky” year to come around. Pay attention to your minor league team also, sometimes if that looks really good then it’s a good sign for the future.

      2. 2slugbaits

        Ah, but as a lifelong Cubs fan I can tell you that 2016 was worth the wait. Still, the pain of 1969 never goes away. To this day I can no more stand the NY Mets for winning the ’69 Series anymore than a good Cub fan can stand the St. Louis Cardinals.

        1. Willie

          I am not a lifelong Mariners fan. But 30 years is close enough for scars. The playoff collapse of 2001 left a mark. But there is nobody to hate. The Mariners are just hapless. They are building a decent farm system now, for the first time ever. Maybe there is hope.

          Back to reporting. I would not be at all surprised to find that GOP governor’s are attempting to fiddle the numbers. But i am not as convinced that they will be successful. Of course, I live in a state that has so far had decent leadership and that owns up to mistakes and has the mental flexibility to change course when it recognizes mistakes and setbacks.

      3. Dr. Dysmalist

        I don’t know … I’ve read and heard from several (overall dependable) sources that dietary supplements serve only two purposes: 1) make profits for the manufacturers and purveyors, and 2) create expensive urine from the consumers.

  2. Kate

    I don’t understand the fascination with IHME models, especially the fatality charts. IHME appears to be doing nothing but taking the current daily rate of death multiplying it over the number of days in a time period and producing a number. They make it look scientific and give themselves lots of cover by throwing a huge margin of error around their center line number. If the fatality rate stays constant they are fairly close to the mark. If the fatality rate changes, they are way off (like they were with all their July 1 projections). I stopped paying any attention to these projections back in May. They are simply silly and have proven to be damn near worthless.

    What’s the point other than to get TV talking heads who can’t do 5th grade math riled up so they can get their audience, who can’t do kindergarten level math, riled up?

    170k people have died…1k are dying per day…in 30 days we project 200k will have died…SHOCKING!!!

    1. 2slugbaits

      What the fivethirtyeight site is showing is the “current projection” model, which assumes no change in the state’s intervention policy. If you go to the actual IMHE website you will find a dropdown box under the “Compare” tab that allows you to invoke different intervention regimes.

      If you want a more robust simulation I would recommend the COVID-19 simulator, which is another one of the fivethirtyeight projections.
      Again, the fivethirtyeight forecasts use the current intervention regimes projection, but there are other simulation options.

      Also, the fivethirtyeight site offers the US Army model that makes use of Bayesian probabilities within a SEIR compartmental model, which they explain here:
      As a retired Army Research & Development Command analyst I’m required to put in a plug for my former colleagues. 🙂

      1. Moses Herzog

        You’re wasting your time. Kate (“Karen”) sounds like the type that avidly watches Eric Bolling every weekend, in some demented fantasyland where she has the same sensation people used to get listening to FDR fireside chats.

        Kate (“Karen”) is like those people whose close relative dies of disease X, and then the very next day insists all Americans need to be contributing dollars to disease X and that it is suddenly the world’s most pressing need. The only way she will feel COVID-19 is a “real” urgent issue is when someone SHE loves, like an immediate family member or lifetime friend gets COVID-19, THEN she will start bawling her eyes out to everyone, telling everyone she was “on board” with COVID-19 precautions back in late January of 2020. Otherwise all anyone who dies of COVID-10 is, is a political cartoon for Kate.

        All world events and deaths are biewed through ONE SINGLE lens “How doe this affect ME, Kate (‘Karen’)”.

  3. Alan Goldhammer

    What is worrisome to me is the current mortality rate still hovers at 2% based on documented cases and deaths. I use the Washington Post published data each day. Of course, if we really knew what the correct number of cases was, the mortality rate would likely drop significantly. At 2%, seniors citizens such as myself will continue to not dine out, go to movies, etc.

    1. Baffling

      I think it will remain elevated. Treatment like remdisivere will help some. But the old and compromised, like in nursing homes, will have no treatment. They will continue to die and keep the rate elevated. Unless we can figure out how to keep the virus out of nursing homes. The are the dominant figure in death counts.

      1. pgl

        “Unless we can figure out how to keep the virus out of nursing homes.”

        This should be an achievable result if we try. The NHL and NBA are keeping the virus out of their playoff bubbles.

        1. baffling

          this can only be achieved if we make a concerted effort to minimize the spread in the general public. if we continue to fight against masks, work from home orders and community lockdown periods, we will NOT control the spread of the virus in the community. it will be impossible to keep the virus out of the nursing homes under those conditions. personally, i do not think it is possible until trump is removed from office and the red state governors are able to act responsibly with their stay at home orders. therefore, the death rate will remain elevated by events which will continue to unfold in the nursing homes. it will be impossible to create a bubble around nursing homes, when most of the care workers are making slightly above minimum wage.

    1. pgl

      Australians are well aware of the crack pot views of Sky News host Rowan Dean. He is not a doctor but rather a former ad agency type pushing Hamlet cigars. Dean is a full bore Trump supporters so it is no wonder that his latest rightwing rant began with herd immunity and hydroxychloroquine.
      I guess Bruce Hall needs company in the Cuckoo’s Nest!

    2. pgl

      CoRev’s new expert is a real hoot. Let’s see – he claims New Zealand’s approach has been a failure. Seriously? Never mind his endorsement of Bruce Hall’s snake oil – his evidence that the Swedish approach works is that the death count in Sweden is less than the death count in the UK. What? Does this moron think there are only two countries in Europe? But the big lie from this dork is the claim that herd immunity worked. Let’s see:
      Herd Immunity to COVID-19 Fails to Materialize in Sweden
      Sweden’s policy of allowing the controlled spread of Covid-19 viral infection among the population has so far failed to deliver the country’s previously stated goal of herd immunity. Commenting on recent antibody testing clinical and research findings, authors of a paper published by the Journal of the Royal Society of Medicine, write that Sweden’s higher rates of viral infection, hospitalization and mortality compared with neighboring countries may have serious implications for Scandinavia and beyond.
      Rather than imposing a hard lockdown in March as most European and Scandinavian countries did, Sweden’s strategy in dealing with the pandemic has been to rely on people’s individual responsibility to curtail the spread of the disease. This follows the Swedish sociocultural concept of ‘folkvett’; the common sense of the people as a collective.
      The health authorities predicted that 40% of the Stockholm population would have had the disease and acquired antibodies by May 2020. However, the actual prevalence figure was around 15%. While clinical and research findings suggest that severely infected Covid-19 patients do acquire antibodies in the immediate and early recovery phase of their illness, antibodies are much less commonly found in only mildly ill or asymptomatic patients. This means they are very likely not to be immune, and so cannot act as a bulwark against further spread of infection amongst the community.
      Lead author Professor David Goldsmith said: “It is clear that not only are the rates of viral infection, hospitalization and mortality (per million population) much higher than those seen in neighboring Scandinavian countries, but also that the time-course of the epidemic in Sweden is different, with continued persistence of higher infection and mortality well beyond the few critical weeks period seen in Denmark, Finland and Norway.” He added that in these countries, rapid lock-down measures brought in from early March seem to have been initially more successful in curtailing the infection surge and thus the malign consequences of Covid-19 on the country as a whole.
      Prof Goldsmith said: “We in the UK would do well to remember we nearly trod the same path as Sweden, as herd immunity was often discussed here in early March. Right now, despite strict (but tardy) lock-down in the UK, and the more measured Swedish response, both countries have seen high seven-day averaged Covid-19 death rates compared to other Scandinavian and European countries.”
      The authors do say, however, that only once the pandemic and impact of measures taken are fully understood, after one or two years at least, can we begin fairly then to judge what was done correctly.
      Reference: “Four months into the COVID-19 pandemic, Sweden’s prized herd immunity is nowhere in sight” by Eric J W Orlowski, David J A Goldsmith and David J A Golds, 11 August 2020, Journal of the Royal Society of Medicine.

      1. 2slugbaits

        Comparing Sweden’s death rate with the UK’s is a real howler. Initially the UK chose a herd immunity strategy as well, but a few weeks into it the number of deaths became so horrific that Boris Johnson had to abandon it. If you want to look at success stories, try looking at Denmark or Germany of South Korea or New Zealand. Those countries locked down hard, took their lumps early on and are now in a decent position. Germany has seen some isolated outbreaks, but they appear to be reasonably well contained and are nothing like what we see in the US.

        1. pgl

          “If you want to look at success stories, try looking at Denmark or Germany of South Korea or New Zealand. ”

          If you have the stomach to listen to Rowen Dean, you might note he mentions New Zealand as a country that will end up doing very poorly. I guess this howler impressed CoRev as well!

    3. 2slugbaits

      CoRev First of all, it’s a Murd(er)och story, so you should be suspcious. Second, nominally Sweden says it’s following a herd immunity strategy, but even Sweden’s chief official and proponent of herd immunity admitted that it was a failure. Swedes effectively abandoned herd immunity on their own and took to wearing masks, sheltering in place, social distancing, etc. Try to keep up.

      1. pgl

        Check out:

        “Four months into the COVID-19 pandemic, Sweden’s prized herd immunity is nowhere in sight” by Eric J W Orlowski, David J A Goldsmith and David J A Golds, 11 August 2020, Journal of the Royal Society of Medicine.

        It seems CoRev failed to do before he ran this rant from Australia’s number 1 MAGA hat wearer.

      2. pgl

        CoRev thinks Rowan Dean is a news reporter? Isn’t that like calling Glenn Beck the new Walter Cronkite?

        1. CoRev

          PGL, check worldometer and Sweden to see if they have reached herd immunity. I know you are required to dispute any analysis that points to a successful approach to contgrolling Covid-19, but Sweden’s daily death rate has ben in the low single digits for quite some time.

          “Sweden’s seven-day rolling average of daily new deaths has been mostly declining since as far back as April 16, when the average was at 99. The average dropped to two on August 2, according to Worldometer.”

          2slugs: “Swedes effectively abandoned herd immunity on their own and took to wearing masks, sheltering in place, social distancing, etc. Try to keep up.” These with the exception of masks are almost the normal Swedish social state. What you forgot was these were not mandated now were lock downs, except to protect the susceptible nursing homes. Schools also stayed open.

          There are a growing number of reports citing Sweden’s success:
          “As the pandemic struck, the Swedes normalized as much as they could, while maintaining a measure of social distancing. What Sweden didn’t do, unlike most of the globe, is reflexively dive into a full-lockdown scenario. Instead, they allowed life to go continue as ordinarily as possible. Stores remained open. Restaurants remained open. Most importantly, schools remained open as well. The state banned public gatherings of more than 50 people, but any further social distancing measures were largely unenforced and voluntary.”

          It must be frustrating to watch your two strongest political issues, Covid-19 deaths and the rebounding economy, start to disappear.

          1. pgl

            None of this has anything to do with herd immunity:

            “As of Sunday, the latest death rate in Sweden (deaths per 100,000 people) was reported to be 56.40. The figure is lower than that reported in the U.K. (69.60), Spain (60.88) and Italy (58.16), according to the latest report Sunday by Johns Hopkins University.

            The U.K. currently has the world’s fourth highest death toll, while Spain and Italy (which have the sixth highest and eighth highest death tolls, respectively) were formerly Europe’s two countries worst hit by the outbreak.

            Sweden’s latest case-fatality ratio (portion of deaths compared to total cases) was reported to be 7.1 percent. The figure is more than half the percentage reported in the U.K. (15.1 percent), half that of Italy and Belgium (each reporting 14.2 percent) and nearly half that of France (13.4 percent), according to Johns Hopkins University.”

            Sweden is not the highest is your evidence? What about its Northern European neighbors? Oh wait – Sweden is much worse. BTW – daily death counts may be declining because the average Sweden is not as stupid as you are. I hear a lot of them did decide to wear masks.

          2. pgl

            CoRev’s second link was written by this dude:

            Sumantra Maitra is a doctoral researcher at the University of Nottingham, UK, and a senior contributor to The Federalist. His research is in great power-politics and neorealism. He also writes for Quillette, Providence Magazine, Spectator US, The Telegraph, Claremont Review of Books, International Affairs, Washington Examiner, and other publications.

            Come on CoRev – you are even worse than Bruce Hall in terms for checking your sources. Writing for the Federalist after all. Right!

          3. 2slugbaits

            More idiocy from CoRev. Italy, Spain and France (and NYC) got clobbered early on. That’s why their initial death rate was so high. They got control of the virus because they went into a hard lockdown and reached near zero deaths long before Sweden. Despite having the advantage of advance warning Sweden’s total deaths per million is almost identical to Italy’s (590 vs 578) and is significantly higher than France’s total deaths per million (454). And Sweden isn’t even in the same ballpark with other Nordic countries. Germany’s death rate is less than one-fifth of Sweden’s. It is true that Sweden is doing better than Britain, but remember that Britain initially tried the Swedish approach and abandoned it after deaths were unacceptably high. The British experience is probably the best case study against herd immunity.

            Swede’s effectively went into a total lockdown. They kept businesses open, but over 50% of Sweden’s workforce was able to work from home. Sweden also has a fine healthcare system. Sweden also has the most single member households in Europe. Let’s repeat…Sweden’s own health minister who was the biggest proponent of herd immunity admitted it was a mistake and that Swede’s had effectively abandoned it. We’ve been over this before. I’ve provided Bruce Hall with the Swedish minister’s quote. Sweden also had one big advantage…they don’t have stupid redneck MAGA hatters parading in front of state capitol buildings with AR-15s.

            BTW, have you ever looked at Sweden’s data. It’s very lumpy and very slow to come in. And I don’t mean a few days. Swedish data typically gets continually revised for 3-4 weeks, so you shouldn’t put a lot of weight on recent data. Wait a few weeks and you’ll see huge revisions. That’s because Sweden’s data is decentralized and not reported on a regular basis.

            The only place where you’ll find stories about Sweden’s success with herd immunity is internet sites for knuckle draggers.

  4. Bruce Hall

    I’ll continue to provide weekly updates from the CDC for pgl’s benefit; however, another way of looking at the data comes by way of this source:

    Interestingly, the U.S. ranks 49th out of the top 50 countries.

    While the number of deaths in the U.S. is increasing on a cumulative basis and the death rate will increase, it falls far short of many European countries praised for their response to the epidemic. That is partly due to the smaller geography of the European countries which meant the epidemic moved quickly through the population. In the U.S., the epidemic has been more of a “slow roll” starting off, for the most part, in New York City.

    Meanwhile, it appears that the weekly death rate in the U.S. is somewhere around 40% of the April peak when Cuomo and de Blasio were leading the New York effort by pushing the sick elderly back into nursing homes and refusing help from the hospital ship Mercy.

      1. Bruce Hall

        pgl, currently that assertion [the US has the highest daily death count per capita] is correct. The U.S. “slow roll” is continuing. Cumulatively, however, … I guess your argument is with

        BTW, good news for you: New York is now only #2 in deaths per 100K; NJ is first. Of course, a lot of people from NJ worked in NYC. Other good news, Cuomo did reverse his policy of shoving infected elderly back into nursing homes; unfortunately, Gretchen Whitmer didn’t.

        Got to keep those “expendable” old people dying so that Trump gets blamed.

    1. pgl

      Brucie – could you share this with your good buddy CoRev:

      The 7-average of US daily death counts was rising up to August 10, 2020 when it almost reached 3.5 per million. Yea it has come down a wee bit since but it is still much higher than it was in early July. And it still is far higher than any other in that graph.

      Which basically means Brucie Boy – you have been lying for weeks and CoRev’s feeble excuses to defend you are an utter joke.

    2. Ulenspiegel

      “While the number of deaths in the U.S. is increasing on a cumulative basis and the death rate will increase, it falls far short of many European countries praised for their response to the epidemic. ”

      That is nonsense. The European countries which are actually praised have a much lower deaths rate than the USA, per capita even the whole of the EU-27 is performing better than the USA.

      Why is it so hard for you to get even simple problems right?

  5. Moses Herzog

    I don’t know how cerebral (not being facetious) commenter noneconomist feels about this, but I always had this dark fear we might be reading about one of JBH’s family members in the newspaper one day.

    I haven’t seen “crazy eyes” like that since Michelle Bachmann was a media darling.

    OK, one F-bomb in the video link Menzie, you can truncate this sentence and the link if it’s better:
    [link edited by MDC]

    1. noneconomist

      Possible she’s one of JBH’s million dollar clients who seek his vital expertise on many matters. Or, like many Qers, she’s simply, uh, nuts

Comments are closed.