Covid-19 Fatalities and Excess Fatalities

The most recent “excess fatality” count remains solidly in the positive region, despite the severe under-reporting bias in the most recent observations. To see this, consider the most recent estimates for each of the previous vintages of “excess fatalities” calculated as actual-expected.

Figure 1: Excess fatalities, 10/21 vintage (chartreuse), 10/7 vintage (purple red), 9/30  vintage (violet), 9/23 vintage (chartreuse), 9/16 vintage (red), 9/9 vintage (green), 9/2 vintage (orange), 8/25 vintage (blue). Note excess fatalities differ from CDC series which are bounded below at zero. Source: CDC , various vintages, and author’s calculations.

This pattern suggests to me we should take with circumspection (1) the most recent counts of excess fatalities as they are likely to be revised substantially upward; and (2) administrative counts, either from CDC or from alternative compilations, as they are possibly missing many actual Covid-19 related deaths. Extending point (1), it is likely that excess fatalities were indeed falling for some time in August, but are possibly rising again given the size of the revisions in recent weeks’ data (The Economist has a discussion of excess fatalities around the world; the Excess Death Tracker is here.)

Here are the various series of interest, latest available.

Figure 2:  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 Our World in Data (dark red). Note excess fatalities differ from CDC series which are bounded below at zero. Light green shading denotes CDC data that are likely to be revised. Source: CDC  10/21/2020 vintage, OurWorldinData version of 10/22 accessed 10/22/2020 and author’s calculations.

One conclusion that seems obvious: Cumulative excess fatalities through week ending 8/25 are substantially higher than administratively defined Covid-19 cumulative fatalities.

Woolf et al. (JAMA, October 12, 2020) reports that March through July (ex-CT, NC):

Of the 225 530 excess deaths, 150 541 (67%) were attributed to COVID-19. Joinpoint analyses revealed an increase in deaths attributed to causes other than COVID-19, with 2 reaching statistical significance. US mortality rates for heart disease increased between weeks ending March 21 and April 11 (APC, 5.1 [95% CI, 0.2-10.2]), driven by the spring surge in COVID-19 cases. Mortality rates for Alzheimer disease/ dementia increased twice, between weeks endingMarch 21 and April 11 (APC, 7.3 [95% CI, 2.9-11.8]) and between weeks ending June 6 and July 25 (APC, 1.5 [95% CI, 0.8-2.3]), the latter coinciding with the summer surge in sunbelt states.

Discussion | Although total US death counts are remarkably consistent from year to year, US deaths increased by 20% during March-July2020.COVID-19was a documented cause of only 67% of these excess deaths. Some states had greater difficulty than others in containing community spread, causing protracted elevations in excess deaths that extended into the summer. US deaths attributed to some noninfectious causes increased during COVID-19 surges. Excess deaths attributed to causes other than COVID-19 could reflect deaths from unrecognized or undocumented infection with severe acute respiratory syndrome  coronavirus2ordeaths among uninfected patients resulting from disruptions produced by the pandemic. Study limitations include the reliance on provisional data, inaccuracies in death certificates, and assumptions applied to the model.

225,530 excess fatalities reported by Woolf et al. for this period exceeds the 221,848 in the CDC excess fatalities estimates I’m reporting above, and obviously far above the 160,092 administratively defined Covid-19 deaths.

Update, 5/12 3:13PM Pacific:

CDC reports excess deaths by age and ethnicity, in this report dated October 23 (for data running through October 3).

8 thoughts on “Covid-19 Fatalities and Excess Fatalities

  1. ltr

    October 20, 2020

    The Pandemic’s Real Toll? 300,000 Deaths, and It’s Not Just From the Coronavirus
    A C.D.C. analysis finds that overall death rates have risen, particularly among young adults and people of color.
    By Roni Caryn Rabin

    The coronavirus pandemic caused nearly 300,000 deaths in the United States through early October, federal researchers said on Tuesday.

    The new tally includes not only deaths known to have been directly caused by the coronavirus, but also roughly 100,000 fatalities that are indirectly related and would not have occurred if not for the virus.

    The study, published by the Centers for Disease Control and Prevention, is an attempt to measure “excess deaths” — deaths from all causes that statistically exceed those normally occurring in a certain time period. The total included deaths from Covid-19, the illness caused by the coronavirus, that were misclassified or missed altogether.

    Many experts believe this measure tracks the pandemic’s impact more accurately than the case fatality rate does, and they warn that the death toll may continue an inexorable climb if policies are not put in effect to contain the spread.

    “This is one of several studies, and the bottom line is there are far more Americans dying from the pandemic than the news reports would suggest,” said Dr. Steve Woolf, director emeritus of the Center on Society and Health at Virginia Commonwealth University, whose own research recently reached similar conclusions about excess deaths.

    “We’re likely to reach well over 400,000 excess deaths by the end of the year” if current trends continue, Dr. Woolf said….

  2. Moses Herzog

    This is a prototypical example of why I LOVE this blog and the efforts of Sirs Chinn and Hamilton. Menzie keeps “beating the drum” on excess deaths. As well he should. Why do I have to go to an economics blog to get what should be one of the core messages from journalists right now, who have completely dropped the ball on a figure (although yes, a rough and not exact benchmark) which does the best job at gauging how much we can trust this year’s (or “pro rata”, if that is the right term, to the current date) Covid-19 death numbers?? The reality is, which I can’t seem to get local journalists such as t*** [edited MDC] Kassie McClung to understand or overexert the finger muscles she uses for a keyboard to even broach the topic, the Covid-19 death numbers in America right now are being significantly undercounted. And if journalists weren’t so afraid of math, or calling experts who reside outside of their own state of residence, they would realize there’s a huge investigative journalism type story to be addressed there. The type of story ProPublica often does. But “why do that??” when you’re a dumb Okie journalist who can just transcribe numbers off of OSDH??

  3. Willie

    I don’t expect this pandemic to be over or under control for another year or so. If there’s an effective vaccine, which is a big if, it will start being available in the new year at the soonest. It will go to a variety of groups of people, including the well connected, that the majority of people on the planet. Including most of us who read this blog. What will keep the death toll under some semblance of control is improved treatments. Thank goodness for science and people who actually learn from science.

    Which all boils down to a million excess deaths or more by this time next year. Not the same percentage dead as from the Spanish flu, but then there will be the long haul sufferers and the people who have undiagnosed damage that will either disable or kill them later. Too bad the administration sat on its thumbs early. We might be in better shape by now. There’s a huge mess to clean up, and it won’t get cleaned up without a new administration.

    1. baffling

      “I don’t expect this pandemic to be over or under control for another year or so.”
      as biden says, the winter will be bad. and why? we are now sending in a generation of young kids to school with absolutely no plan in place to protect them, their teachers or the community when the virus ultimately spreads. this will carry on through the spring. the virus will be endemic again in our population before biden ever gets a chance to implement a control plan. donald will leave us one more turd before he leaves.
      and this did not have to play out this way. china showed the virus can be controlled. did they take drastic measures that could not be repeated here in the usa? sure. but the virus can be controlled and the economy can return to growth. we have empirical evidence in china. instead we get the trump rot of a floundering economy with sickness and death all around. i hope rick stryker, bruce hall, corev and all the other assorted nut jobs are happy with the current state of affairs created by donald.
      as my father once told me, democrats get elected to clean up the mess left behind by republicans.

  4. Moses Herzog

    How many policeman do we have reading this blog?? I’m guessing a percentage under 2% ?? Be that as it may, I wanted to show/remind any police out there what their job actually is~~~”protect and serve the public”. That means your main focus in life shouldn’t be searching out unarmed minorities and low income groups to hassle and shoot. It means you actually attempt to make the world a better place (odd experimental concept, I know, maybe just run some “field tests”, I know it’s counter-intuitive for most redneck hick knucklehead police).

    Just to give you a conceptual idea and let you know it’s not all the “fun” of shooting at Black people like you’re playing the early ’80s Atari game of Galaga, here is officer Pheng Ly of Davis California Police Dept. (I think the footage is Solano County) showing you how real police work is supposed to be done.

    THIS video above is how America is SUPPOSED to be~~~Pheng Ly makes me proud to be an American. Not 20 white a$$-holes with the maturity of a 12 year old playing “supercop”, surrounding one unarmed black man.

Comments are closed.