His point is a good one. And it prompts me to spend much of the day digging into data that, unexpectedly, highlights exactly this point.

Per capita, the worst spots have remained the places with some of the highest percentages of African Americans—namely the Gulf South and the Mississippi River Valley—and Latinx populations—namely Texas through southern California. As we’ve already shown throughout our pandemic year, these are largely poor areas ruled over by a Republican elite without much access to healthcare or employment that stresses workplace safety. Over the last half-year, there has been little improvement in these areas at all; high caseloads per capita in early May, high caseloads per capita in late September. America may have moved on from their interest in these areas, but coronavirus has not. 

There’s another unwelcome indicator to add to the story this week. The New York Times recently tabulated “excess deaths” through the first five months of the pandemic. It’s especially important to reveal this statistic now, since right-wing conspiracy mongers seized on the statistic released by the CDC last month that “only” 6% of COVID-related deaths listed coronavirus as the exclusive cause of death.[1] Ninety-four percent of the 200,000 deaths by coronavirus, in other words, listed more than one condition or co-morbidity contributing to death—something we’ve known since the first public reports on coronavirus patients out of China in January. This string of co-morbidities is not suspicious at all; it’s actually the standard death certificate coding procedure mandated by the WHO and the CDC throughout this pandemic.[2] But those searching for reasons to disbelieve the severity of COVID in order to dispense with mask mandates, business closures, or stay-at-home orders are hyping this 6% as a secret admission by Fauci and his supposed left-winged minions.

As we already saw in April, the measurement of excess deaths bypasses one of the weak links in any public health system: the improper coding of death certificates. Even if a physician, coroner, or medical scribe is reluctant to place the “U07.1” indicating SARS-CoV-2 on the death certificate, every death counts toward the total of excess deaths no matter the cause.[3] “Excess death” could potentially stand for three different things, however. Mid-nineteenth century statisticians and public health officials compared deaths from one year to the year before it or, alternatively, used a multi-year “average expected mortality” and compared actually counted deaths against that number to measure the impact of diseases, accidents, wars, and other tragedies.[4] In the last few decades, British public health statisticians attempted to account for under- and over-counting by formulating an “upper bound” threshold.[5] The CDC uses this upper bound algorithm to warn of impending outbreaks instead of merely reporting a difference between tabulated deaths and the expected average mortality because, they say, “The upper bound more readily identifies areas experiencing statistically significantly higher than normal mortality.”[6] The New York Times chose to use the expected average mortality instead of the typical upper bound threshold, however. That turned out to be a significant decision. Unlike the more conservative estimate usually adopted by the CDC, yesterday’s New York Times article identifies 257,900 deaths above the expected average mortality from the second week of March through the final week of August.[7] (I did my own count using National Center for Health data and beginning the count two weeks earlier—when the first deaths started. My count comes out only slightly higher.[8]) That excess death count reveals that over one quarter of a million people more than expected have died in America this year. That’s the equivalent of a Madison, Wisconsin, a Buffalo, New York, or a St. Petersburg, Florida exterminated in under a half-year. Whatever is killing Americans this year, it’s bad in an unprecedented, once-in-a-century sort of way.

Those on the political right might proclaim the “cure” of shutdowns and masks is worse than the coronavirus disease. Fortunately, there’s a long-practiced method by which to assess the stringency of our nation’s response to coronavirus. National “p-scores” rank excess mortality by the difference between the weekly deaths as actually counted and the average number of deaths in that same week averaged over the previous five years. Compared to Italy, which suffered a p-score more than double their yearly average in February and March, the USA appeared to weather the first wave proportionally better. By April, while the USA was more or less in lockdown, we averaged only 30%-40% over the ordinary number of deaths. But, unlike Italy, whose p-scores dropped precipitously given their own quarantines, the level of death in the United States persisted in that 30% range for weeks, then months. In fact, America’ p-score over these last few months appears similar to Sweden’s. But Sweden elected not to shut down its schools, restaurants, and other drivers of the economy at first and, buffered by its quasi-socialistic system, suffered comparatively less economic damage than Germany, Italy, or the UK. In other words, the USA suffered the negative economic consequences of closing most businesses, but because we didn’t follow through with truly restrictive lockdown measures, we suffered the worst long-term epidemiological consequences in tandem with the economic problems. The United States suffered the worst effects of both Italy and Sweden.

Notes

[1] BMartin1776, “Updated CDC Covid Numbers Show Only 6% of Total US Deaths Actually Died Solely From Virus,” SavingtheRepublic.Com: Video News & Opinion (blog), August 30, 2020, http://savingtherepublic.com/blog/2020/08/updated-cdc-covid-numbers/.

[2] National Vital Statistics System, “Guidance for Certifying Deaths Due to Coronavirus Disease 2019 (COVID-19),” Vital Statistics Reporting Guide (U.S. Department of Health and Human Services, April 2020), https://www.cdc.gov/nchs/data/nvss/vsrg/vsrg03-508.pdf; WHO, “International Guidelines for Certification and Classification (Coding) of COVID-19 as Cause of Death” (Geneva, Switzerland: World Health Organization, April 16, 2020), https://www.who.int/classifications/icd/Guidelines_Cause_of_Death_COVID-19.pdf?ua=1.

[3] Steven Schwartz, “New ICD Code Introduced for COVID-19 Deaths,” COVID-19 Alert (Washington, DC: Division of Vital Statistics, National Center for Health Statistics, March 24, 2020).

[4] “Mortality From London Fogs,” The British Medical Journal 1, no. 1103 (February 18, 1882): 235–36, http://www.jstor.org/stable/25258965.

[5] Angela Noufaily et al., “An Improved Algorithm for Outbreak Detection in Multiple Surveillance Systems,” Online Journal of Public Health Informatics 5, no. 1 (April 4, 2013), https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3692796/.

[6] National Center for Health Statistics, “Excess Deaths Associated with COVID-19,” Centers for Disease Control and Prevention, September 9, 2020, https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm. Emphasis in original.

[7] Josh Katz, Denise Lu, and Margot Sanger-Katz, “The Toll Since Coronavirus Struck: 250,000 More Deaths Than Normal,” The New York Times, September 22, 2020, sec. U.S., https://www.nytimes.com/interactive/2020/05/05/us/coronavirus-death-toll-us.html.

[8] https://www.cdc.gov/nchs/nvss/vsrr/covid19/excess_deaths.htm