PreprintPDF Available

COVID-Period Mass Vaccination Campaign and Public Health Disaster in the USA From age/state-resolved all-cause mortality by time, age-resolved vaccine delivery by time, and socio-geo-economic data

Authors:
Preprints and early-stage research may not have been peer reviewed yet.

Abstract and Figures

All-cause mortality by time is the most reliable data for detecting and epidemiologically characterizing events causing death, and for gauging the population-level impact of any surge or collapse in deaths from any cause. Such data is not susceptible to reporting bias or to any bias in attributing causes of death. We compare USA all-cause mortality by time (month, week), by age group and by state to number of vaccinated individuals by time (week), by injection sequence, by age group and by state, using consolidated data up to week-5 of 2022 (week ending on February 5, 2022), in order to detect temporal associations, which would imply beneficial or deleterious effects from the vaccination campaign. We also quantify total excess all-cause mortality (relative to historic trends) for the entire covid period (WHO 11 March 2020 announcement of a pandemic through week-5 of 2022, corresponding to a total of 100 weeks), for the covid period prior to the bulk of vaccine delivery (first 50 weeks of the defined 100-week covid period), and for the covid period when the bulk of vaccine delivery is accomplished (last 50 weeks of the defined 100-week covid period); by age group and by state. We find that the COVID-19 vaccination campaign did not reduce all-cause mortality during the covid period. No deaths, within the resolution of all-cause mortality, can be said to have been averted due to vaccination in the USA. The mass vaccination campaign was not justified in terms of reducing excess all-cause mortality. The large excess mortality of the covid period, far above the historic trend, was maintained throughout the entire covid period irrespective of the unprecedented vaccination campaign, and is very strongly correlated (r = +0.86) to poverty, by state; in fact, proportional to poverty. It is also correlated to several other socioeconomic and health factors, by state, but not correlated to population fractions (65+, 75+, 85+ years) of elderly state residents.
Content may be subject to copyright.
1
COVID-Period Mass Vaccination Campaign
and Public Health Disaster in the USA
From age/state-resolved all-cause mortality by time,
age-resolved vaccine delivery by time, and socio-geo-economic data
Denis G. Rancourt1,2,*, PhD, Marine Baudin3, PhD, Jérémie Mercier3, PhD
1 Correlation Research in the Public Interest (correlation-canada.org)
2 Ontario Civil Liberties Association (ocla.ca)
3 Mercier Production (jeremie-mercier.com)
* denis.rancourt@alumni.utoronto.ca
This is a pre-print, to be submitted to a peer-reviewed journal.
The pre-print is to be made public at the following websites.
https://ocla.ca/
https://denisrancourt.ca/
https://archive.today/
https://www.researchgate.net/profile/Marine-Baudin
https://www.medrxiv.org/
2 August 2022
2
Abstract
All-cause mortality by time is the most reliable data for detecting and epidemiologically
characterizing events causing death, and for gauging the population-level impact of any
surge or collapse in deaths from any cause. Such data is not susceptible to reporting
bias or to any bias in attributing causes of death. We compare USA all-cause mortality
by time (month, week), by age group and by state to number of vaccinated individuals
by time (week), by injection sequence, by age group and by state, using consolidated
data up to week-5 of 2022 (week ending on February 5, 2022), in order to detect
temporal associations, which would imply beneficial or deleterious effects from the
vaccination campaign. We also quantify total excess all-cause mortality (relative to
historic trends) for the entire covid period (WHO 11 March 2020 announcement of a
pandemic through week-5 of 2022, corresponding to a total of 100 weeks), for the covid
period prior to the bulk of vaccine delivery (first 50 weeks of the defined 100-week covid
period), and for the covid period when the bulk of vaccine delivery is accomplished (last
50 weeks of the defined 100-week covid period); by age group and by state.
We find that the COVID-19 vaccination campaign did not reduce all-cause mortality
during the covid period. No deaths, within the resolution of all-cause mortality, can be
said to have been averted due to vaccination in the USA. The mass vaccination
campaign was not justified in terms of reducing excess all-cause mortality. The large
excess mortality of the covid period, far above the historic trend, was maintained
throughout the entire covid period irrespective of the unprecedented vaccination
campaign, and is very strongly correlated (r = +0.86) to poverty, by state; in fact,
proportional to poverty. It is also correlated to several other socio-economic and health
factors, by state, but not correlated to population fractions (65+, 75+, 85+ years) of
elderly state residents.
The excess all-cause mortality by age group (also expressed as percentage of pre-
covid-period all-cause mortality for the age group) for the whole USA for the entire covid
period through week-5 of 2022 is:
3
all ages
1.27M
23%
0-24
13K
12%
25-44
109K
41%
45-64
274K
27%
65-74
319K
30%
75-84
316K
24%
85+
240K
14%
The corresponding fatality risk ratios are relatively uniform with age (non-exponential
and non-near-exponential with age; and even skewed towards young adults), which
holds essentially for all states, and for all examined periods within the covid period. This
fundamental result implies that a dominant cause of excess mortality could not have
been assigned COVID-19, which consistently has been measured to have a strong
near-exponential infection fatality ratio with age. The implication is further corroborated
by the absence of correlation between all-age-group-integrated excess mortality and
age, by state. COVID-19 was not a dominant cause of excess mortality during the covid
period in the USA.
All of our observations can be coherently understood if we interpret that the covid-period
socio-economic, regulatory and institutional conditions induced chronic stress and social
isolation among members of large vulnerable groups (individuals afflicted and co-
afflicted by poverty, obesity, diabetes, high susceptibility to bacterial respiratory infection
[inferred from pre-covid-period antibiotic prescription rates], old age, societal exclusion,
unemployment, drug and substance abuse, and mental disability or serious mental
illness), which in turn caused many of these individuals to be more and fatally
immunocompromised, allowing them to succumb to bacterial pneumonia, at a time
when a documented national pneumonia epidemic raged and antibiotic prescriptions
were systemically reduced; in addition to possible comorbidity from COVID-19 vaccine
challenge against individuals thus made immunocompromised, under broad and hastily
implemented “vaccine equity” programs.
4
Table of contents
Abstract ........................................................................................................................... 2
Table of abbreviations and definitions ............................................................................. 6
1. Introduction ............................................................................................................... 11
2. Data ........................................................................................................................... 14
3. Results ...................................................................................................................... 16
3.1. USA all-cause mortality by month, 1999-2021 .................................................... 16
3.1.1. Historic trend, normal pre-covid period seasonal pattern .............................. 16
3.1.2. Anomalies in the covid period ....................................................................... 17
3.1.3. Quantifying excess mortality of the covid period, by age group and sex ...... 19
3.2. USA all-cause mortality by week, by age group, 2015-2022 ............................... 29
3.2.1. Historic trend, discontinuous break on 11 March 2020, entering the covid
period ...................................................................................................................... 29
3.2.2. Quantifying the excess mortality of the covid period, by age group .............. 30
3.2.3. Excess mortality of the covid period, by state ............................................... 37
3.3. Time and age-group variations of mortality during the covid period, and relation to
implementation of the vaccination campaign ............................................................. 42
3.3.1. All-cause mortality by week and vaccination delivery by week, by age group,
2019-2022 .............................................................................................................. 42
3.3.2. All-cause mortality by week and vaccination delivery by week, by state, 2019-
2022 ........................................................................................................................ 48
3.3.3. Quantifying excess mortality of the pre-vaccination and vaccination periods of
the covid period, by age group................................................................................ 54
3.3.4. Excess mortality of the pre-vaccination and vaccination periods of the covid
period, by state ....................................................................................................... 63
3.3.5. Difference of vaccination and pre-vaccination mortality in the covid period, by
age group and by state ........................................................................................... 68
3.4. Associations of excess mortality of the covid period with socio-geo-economic
variables..................................................................................................................... 73
4. Discussion ................................................................................................................. 85
4.1. All-cause mortality in the covid period in the USA: Sudden onset and
heterogeneity by state ................................................................................................ 85
4.2. Late-summer-2021 anomalous mortality of young adults.................................... 88
4.3. Vaccination campaign ......................................................................................... 94
4.4. Looking ahead .................................................................................................... 98
5. Conclusion ................................................................................................................ 98
5
References .................................................................................................................. 106
Data References ...................................................................................................... 106
Main References ...................................................................................................... 107
Appendix ..................................................................................................................... 114
Appendix A ACM/w and by 50-week period, by state, 2015-2022 ........................ 115
Appendix B Poverty and obesity maps of the USA ............................................... 166
Appendix C ACM/w in the USA from 2015 to most recent data ............................ 167
6
Table of abbreviations and definitions
Abbreviation
Units
Description
Notes
65+
People
Resident population estimate of people aged 65
years old and over as of July 1st, 2020
65+/pop
%
Proportion of the population aged 65 years old and
over
75+
People
Resident population estimate of people aged 75
years old and over as of July 1st, 2020
75+/pop
%
Proportion of the population aged 75 years old and
over
85+
People
Resident population estimate of people aged 85
years old and over as of July 1st, 2020
85+/pop
%
Proportion of the population aged 85 years old and
over
ACM
Deaths
Mortality from all causes of death (occurring in a
defined period and for a defined place)
ACM/m
Deaths/m
ACM occurring per month
ACM/w
Deaths/w
ACM occurring per week
At least 1 dose
People
Total count of people with at least one dose
1
Booster
People
Total count of people aged 12 years and older with
a booster dose
1
CDC
N/A
The Centers for Disease Control and Prevention is
the national public health agency of the United
States.
COVID-19
N/A
"Coronavirus disease 2019 is a contagious disease
caused by severe acute respiratory syndrome
coronavirus 2"
covid period
Period starting with the WHO announcement of a
pandemic on March 11, 2020, up to and including
the most reliable ACM data (through December
7
2021 for the data by month; through week-5 of 2022
for the data by week)
cvp1
Deaths
ACM peak occurring over March, April and May
2020
cvp2
Deaths
ACM peak occurring over the winter 2020-2021
Disability
%
Percent of Americans with a disability
2
Fully vaccinated
People
Total count of people who are fully vaccinated
1
m22c
Deaths
Integrated ACM from March 2020 to December
2021, included
m22c-1
Deaths
Integrated ACM from May 2018 to February 2020,
included
m22c-2
Deaths
Integrated ACM from July 2016 to April 2018,
included
MHI
$
Estimated median household income in US dollars
Obesity
%
Prevalence of self-reported obesity among U.S.
adults (BRFSS (Behavioral Risk Factor Surveillance
System), 2020)
pCVD
Deaths
corresponds to w50c-2
pop
People
Resident population estimate for the states of the
USA as of July 1st of 2020
Poverty
%
Percent of the population living in poverty
pVax
Deaths
corresponds to w50c-1
pVax-pCVD
Deaths
pVax-pCVD = w50c-1 - w50c-2
3
pVax-
pCVD/pCVD
%
pVax-pCVD/pCVD = (w50c-1 - w50c-2) / w50c-2
(Equation 9)
4
smp1
Deaths
ACM peak occurring over the summer 2020
smp2
Deaths
ACM peak occurring over the late-summer and fall
5
8
2021
SSDI
People
Number of all disabled beneficiaries aged 18-64 of
the SSDI program
SSDI/pop
%
SSDI normalized by population
SSI
People
Number of recipients of the SSI program
SSI/pop
%
SSI normalized by population
USA
N/A
USA is composed of 51 states, including the District
of Columbia, Alaska and Hawaii
VAERS
N/A
United States program for vaccine safety, co-
managed by the U.S. Centers for Disease Control
and Prevention (CDC) and the Food and Drug
Administration (FDA)
Vax
Deaths
corresponds to w50c
Vax-pCVD
Deaths
Vax-pCVD = w50c - w50c-2
6
Vax-
pCVD/pCVD
%
Vax-pCVD/pCVD = (w50c - w50c-2) / w50c-2
(Equation 10)
7
Vax-pVax
Deaths
Vax-pVax = w50c - w50c-1 (Equation 11)
Vax-
pVax/pCVD
%
Vax-pVax/pCVD = (w50c - w50c-1) / w50c-2
(Equation 12)
w100c
Deaths
Integrated ACM from week-11 of 2020 (week of
March 9, 2020) to week-5 of 2022 (week of January
31, 2022), included
w100c-1
Deaths
Integrated ACM from week-15 of 2018 (week of April
9, 2018) to week-10 of 2020 (week of March 2,
9
2020), included
w100c-2
Deaths
Integrated ACM from week-19 of 2016 (week of May
9, 2016) to week-14 of 2018 (week of April 2, 2018),
included
w50c
Deaths
Integrated ACM from week-8 of 2021 (week of
February 22, 2021) to week-5 of 2022 (week of
January 31, 2022), included
8
w50c-1
Deaths
Integrated ACM from week-11 of 2020 (week of
March 9, 2020) to week-7 of 2021 (week of February
15, 2021), included
9
w50c-2
Deaths
Integrated ACM from week-13 of 2019 (week of
March 25, 2019) to week-10 of 2020 (week of March
2, 2020), included
10
WHO
N/A
The World Health Organization is a specialized
agency of the United Nations responsible for
international public health.
xDc(22)1
Deaths
xDc(22)1 = m22c - m22c-1 (Equation 1)
xDc(22)1%
%
xDc(22)1% = xDc(22)1 / m22c-1 (Equation 3)
xDc(22)2
Deaths
xDc(22)2 = m22c - m22c-2 (Equation 2)
xDc(22)2%
%
xDc(22)2% = xDc(22)2 / m22c-2 (Equation 4)
xDc(100)1
Deaths
xDc(100)1 = w100c - w100c-1 (Equation 5)
11
xDc(100)1%
%
xDc(100)1% = xDc(100)1 / w100c-1 (Equation 7)
12
10
xDc(100)1/pop
xDc(100)1 normalized by population
13
xDc(100)2
Deaths
xDc(100)2 = w100c - w100c-2 (Equation 6)
xDc(100)2%
%
xDc(100)2% = xDc(100)2 / w100c-2 (Equation 8)
1 In Figures 10 and 11, it is presented as the cumulative number of people by week
2 Disability is defined as a long-lasting sensory, physical, mental, or emotional condition or conditions that make it difficult
for a person to do functional or participatory activities such as seeing, hearing, walking, climbing stairs, learning,
remembering, concentrating, dressing, bathing, going outside the home, or working at a job.
3 Also called "pre-vaccination-period excess mortality" in the text
4 Also called "covid-period pre-vaccination-period relative excess mortality" in the text
5 Also called "late-summer-2021 peak" in the text
6 Also called "vaccination-period excess mortality" in the text
7 Also called "covid-period vaccination-period relative excess mortality" in the text
8 Also called "integrated mortality in the vaccination period of the covid period" in the text
9 Also called "integrated mortality in the pre-vaccination period of the covid period" in the text
10 Also called "pre-covid-period integrated mortality" in the text
11 Also called "100-week covid-period excess mortality" in the text
12 Also called "covid-period fatality risk ratio" in the text
13 Also called "100-week covid-period fatality ratio" in the text
N/A stands for not applicable
11
1. Introduction
Following Rancourt’s 2 June 2020 article critically assessing circumstances of the
declared pandemic using all-cause mortality (ACM) (Rancourt, 2020), more and more
researchers are recognizing that it is essential to examine ACM by time, and excess
deaths from all causes compared with projections from historic trends, to help make
sense of the events surrounding COVID-19 (Kontis et al., 2020; Rancourt, Baudin and
Mercier, 2020; Villani et al., 2020; Rancourt, Baudin and Mercier, 2021a, 2021b;
Achilleos et al., 2021; Chan, Cheng and Martin, 2021; Faust et al., 2021; Islam, Jdanov,
et al., 2021; Islam, Shkolnikov, et al., 2021; Jacobson and Jokela, 2021; Joffe, 2021;
Karlinsky and Kobak, 2021; Kobak, 2021; Kontopantelis et al., 2021; Locatelli and
Rousson, 2021; Sanmarchi et al., 2021; Woolf et al., 2021; Woolf, Masters and Aron,
2021; Kontopantelis et al., 2022; Ackley et al., 2022; Johnson and Rancourt, 2022; Lee
et al., 2022; Wang et al., 2022).
Rancourt (2020) argued that ACM by time and by jurisdiction data for many countries
and states of the USA in the months that followed the WHO 11 March 2020 declaration
of a pandemic:
(1) was inconsistent with the dominant view of the characteristic features of a
pandemic (high contagiousness and spread by person-to-person “contact”), and
(2) gave clear evidence of synchronous local “hot spot” (jurisdictional)
response-induced mortality.
Likewise, in our further prior analyses of ACM by time (by day, week, month, year) for
many countries (and by province, state, region or county), we found that both the initial
and long-term ACM data in the covid period is inconsistent with a viral respiratory
disease pandemic, where the time-integrated mortality per capita is highly
heterogeneous between jurisdictions, with no anomalies in the first many months in
most places, and hot spots or hot regions having death rate increases that are
synchronous with aggressive local or regional responses, both medical and
12
governmental, which accompanied the 11 March 2020 WHO declaration of a pandemic
(Rancourt, Baudin and Mercier, 2020, 2021a, 2021b; Johnson and Rancourt, 2022).
The initial surges in ACM are highly localized geographically (by jurisdiction) and are
precisely synchronous (all starting immediately after the 11 March 2020 WHO
declaration of a pandemic, across continents), which is contrary to model pandemic
behaviour; but is consistent with the surges being caused by the known government
and institutional responses (Rancourt, 2020; Rancourt, Baudin and Mercier, 2020,
2021a, 2021b; Johnson and Rancourt, 2022).
The ACM by time data for the USA in the covid period has extraordinary features,
including large peaks occurring in the summer seasons, and dramatically different state
to state behaviours. State-to-state heterogeneity in integrated covid-period health-
status-adjusted mortality is well illustrated by Johnson and Rancourt (Johnson and
Rancourt, 2022; their Figure 7). Above-decadal-trend mortality in the covid period is
massive. Nothing like this occurs in neighbouring Canada (Rancourt, Baudin and
Mercier, 2021a). Nothing like this occurs in Western European countries. Similar
anomalies occur in some Eastern European countries and in Russia. The large
differences in covid-period mortality in the USA compared to other Western countries
are probably related to the known relatively poor health-status of the USA population,
suggesting large groups of particularly vulnerable residents (Roser, 2020).
We found that in the USA the state-wise integrated excess ACM of all main age groups
in the summer seasons (2020 and 2021) especially was largest (on a per capita basis)
in the southern states, and was correlated to state-specific obesity and poverty rates,
strongly correlated to the product of obesity and poverty rates, and correlated to mean
climatic temperature of the state, and to state-wise pre-covid-period antibiotic
prescription rate per capita (Rancourt, Baudin and Mercier, 2021b). We postulated that
vulnerable groups became more immune-deficient due to increased experienced
physio-psychological stress and social isolation, and mostly succumbed to bacterial
pneumonia, which is the dominant comorbidity (40-60%) reported in the CDC covid
13
mortality data, at a time when antibiotic prescription rates show an unprecedented
decrease (Rancourt, Baudin and Mercier, 2021b).
In the present article, we extend our epidemiological analysis using consolidated ACM
data (by month, by week, by state, and by age group) up to week-5 of 2022 (week
ending on February 5, 2022), which gives us 100 weeks since the WHO’s 11 March
2020 declaration of a pandemic.
Our goal is three-fold:
(1) Accurately quantify excess mortality (ACM) during the covid period in the USA
(2) Look for socio-economic factors that correlate to time-integrated excess ACM per
capita, by state
(3) Examine whether any impact of the COVID-19 vaccination campaign, which was
implemented in 2021, can be detected and quantified
Presently (as of July 14, 2022), a total of 221,924,152 people are fully vaccinated
against COVID-19 (Johns Hopkins, 2022) in a population of 332,878,208 (US Census
Bureau, 2022a), following an unprecedented vaccination campaign, which was largely
accomplished in the last 50 weeks of the covid period up to week-5 of 2022.
Has this massive campaign had any measurable impact, positive or negative, on
the all-cause mortality in the USA, for any discerned age group?
Can such an impact be detected in delayed or immediate synchronicity with the
dose delivery rates for the different age groups?
Are there important differences in ACM by time and by age group for the periods
(within the covid period) prior to and following vaccine dose delivery, and how
should such differences be interpreted if they occur?
14
2. Data
Table 1 describes the data used in this work and the sources of the data.
Data
Country
Period
Time unit
Filters
Source
ACM
USA
1999-2021*
Month
State, sex,
age group1
CDC, 2022a
ACM
USA
2015-2022**
Week
State, age
group2
CDC, 2022b
Vaccines
USA
2020-2022+
Day
Age group3
CDC, 2022c
Vaccines
USA
2020-2022++
Day
State, age
group4
CDC, 2022d
Obesity
USA
2020
Year
State
CDC, 2021
Population
USA
2010-2020§
Year
State, sex,
age group5
US Census
Bureau,
2021
Poverty
USA
2020
Year
State
US Census
Bureau,
2022b
MHI
USA
2020
Year
State
US Census
Bureau,
2022b
SSI
USA
2020
Year
State
SSA, 2022a
SSDI
USA
2020
Year
State
SSA, 2022b
Disability
USA
-
-
State
Disabled
World, 2020
Table 1. Data retrieved. In this work, USA is composed of 51 states, including the District of
Columbia, Alaska and Hawaii, unless otherwise stated in the text.
* These data are a combination of the data found in CDC 2022a: data for the years 1999 to
2020 were downloaded under the “Current Final Multiple Cause of Death Data” section of the
reference (on November 17, 2021 for the years 1999 to 2019 and on May 18, 2022 for the year
2020), and data for the year 2021 was downloaded under the “Provisional Multiple Cause of
Death Data” section of the reference on May 18, 2022. The complete series is thus from
January 1999 to December 2021.
** At the date of access, data were available from week-1 of 2015 (week ending on January 10,
2015) to week-19 of 2022 (week ending on May 14, 2022). Usable data are until week-5 of 2022
15
(week ending on February 5, 2022) due to unconsolidated data in later weeks, which gives a
large artifact (anomalous drop in mortality).
+ At the date of access, data were available from Sunday December 13th 2020 to Wednesday
May 4th 2022.
++ At the date of access, data were available from Sunday December 13th 2020 to Sunday April
24th 2022.
§ In this work, we use the population data of the year 2020.
1 11 age groups: <1, 1-4, 5-14, 15-24, 25-34, 35-44, 45-54, 55-64, 65-74, 75-84, 85+
2 6 age groups: 0-24, 25-44, 45-64, 65-74, 75-84, 85+
3 9 age groups: <5, 5-11, 12-17, 18-24, 25-39, 40-49, 50-64, 65-74, 75+
4 4 age groups: 5+, 12+, 18+, 65+
5 86 age groups: by 1 year age group, from 0 to 85+
The vaccines data are daily cumulative data; when shown together with all-cause
mortality by week data, the last day of the week is used (the Saturday) as a data point,
so that both ACM and vaccination data correspond to the same time point (end of week
for both).
The vaccines data presented in this work correspond to three data type (CDC, 2022c):
At least 1 dose, corresponds to the total count of people with at least one dose”.
Fully vaccinated, corresponds to the “total count of people who are fully
vaccinated”.
Booster, corresponds to the “total count of people aged 12 years and older with a
booster dose”.
According to the CDC, a person is considered fully vaccinated when they “have second
dose of a two-dose vaccine or one dose of a single-dose vaccine”.
A booster dose is an additional dose given to a fully vaccinated person.
For all the scatter plots presented in this article, the following color-code is applied for
the 51 states of the USA:
16
3. Results
3.1. USA all-cause mortality by month, 1999-2021
3.1.1. Historic trend, normal pre-covid period seasonal pattern
Figure 1 shows the all-cause mortality by month (ACM/m) for the USA from January
1999 to December 2021.
Figure 1. All-cause mortality by month in the USA from 1999 to 2021. Data are displayed
from January 1999 to December 2021. The vertical dark-blue line indicates the month of
February 2020, intended to point the beginning of the covid period. Data were retrieved from
CDC (CDC, 2022a), as described in Table 1.
The usual seasonal variations are evident, exhibiting a regular pattern of mortality
maximums in winter and mortality minimums in summer. The summer troughs follow a
straight-line trend on a decadal or shorter timescale. On Figure 1 we discriminate two
such periods: 2000-2008 and 2009-2019.
17
ACM/m has artifacts caused by the months having different numbers of days, unlike
weeks (which always have 7 days). The most noticeable such artifact is the dip for the
month of February, which usually has only 28 days. This allows the viewer to spot
February in each winter season.
The regular seasonal pattern of mortality by month in the USA since 1999 is broken
after February 2020 (Figure 1, vertical dark-blue line) when large anomalies occur. The
anomalies occur in what we define as the covid period, starting after the 11 March 2020
WHO declaration of a pandemic.
We showed and discussed these anomalies in detail recently, for ACM by week
(ACM/w) for the USA from week-1 (beginning of January) of 2013 to week-37 (mid-
September) of 2021 (Rancourt, Baudin and Mercier, 2021b).
3.1.2. Anomalies in the covid period
In the covid period, after February 2020, we note the same peaks or features that we
have previously described and interpreted (Rancourt, Baudin and Mercier, 2021b),
using the nomenclature from our previous article:
cvp1 (March-May 2020)
smp1 (summer 2020)
cvp2 (winter 2020-2021)
smp2 (late-summer 2021)
Figure 2 shows those features with their labels.
18
Figure 2. All-cause mortality by month in the USA from 2016 to 2021. Data are displayed
from July 2016 to December 2021. The cvp1, smp1, cvp2 and smp2 features discussed in the
text are indicated. The vertical dark-blue line represents the month of February 2020, intended
to point the beginning of the covid period. Data were retrieved from CDC (CDC, 2022a), as
described in Table 1.
The anomalies in the covid period are as follows:
A mortality peak late in the 2020 winter season, cvp1 (from February to June
2020, Figure 2).
Peaks of mortality in the summers 2020 and 2021, smp1 and smp2, respectively,
when mortality values are usually at their lowest. On Figure 2 specifically:
o smp1 from June to September 2020
o smp2 from July to November 2021 (connecting with the winter 2021-2022)
A large mortality peak in the winter 2020-2021, cvp2 (from September 2020 to
April 2021, Figure 2), which surpasses in magnitude any single winter mortality
peaks since at least 1999 (Figure 1).
In the next section, we use the monthly ACM data to quantify the total excess mortality
that occurred in the covid period, which contains these anomalies.
cvp1
smp2
smp1
cvp2
19
3.1.3. Quantifying excess mortality of the covid period, by age group and sex
We use the ACM/m data of Figure 1 to quantify the excess deaths of the covid period
to date, compared to the historic trend, as follows.
For a given age group and sex, we add all the monthly deaths together, for the months
of March-2020 (start of the pandemic period; announced by the WHO on 11 March
2020) through to the latest useable month (December 2021). This is a total for 22
months (the covid period “to date”). We call this total “m22c”. Then we perform a similar
total for the 1st-prior 22-month period, immediately preceding the covid period, for the
22 months up to and including February 2020. We call this total “m22c-1”. And we do
the same for the 2nd-prior 22-month period, and we call this total “m22c-2”. We
continue moving back in time, to the end of the useable data in 22-month periods:
m22c-3, etc.
Figure 3 shows the graph of “m22c-x” versus time, together with the ACM/m for the
USA where each 22-month period has been emphasized with a different color.
Figure 3. All-cause mortality by month (colors) and by 22-month period (black) in the
USA from 2000 to 2021. Data are displayed from January 2000 to December 2021. The
20
different colors indicate the successive 22-month periods. The last light-blue color corresponds
to the covid period. All the other previous colors are in the pre-covid period. The black line
shows the integration of these successive 22-months periods. Data were retrieved from CDC
(CDC, 2022a), as described in Table 1.
Figure 3, based on more than two decades of data, dramatically illustrates the sudden
change in regime of ACM by time, both in magnitude of time-integrated ACM and in
seasonal behaviour of ACM by time, occurring as soon as the WHO on 11 March 2020
announced a pandemic. In addition, the covid-period regime of ACM by time is
characterized by large (and unprecedented in the historic record) heterogeneity by state
of ACM, which is not shown in such a figure for the whole USA, but which can be
appreciated in the ACM/w by state graphs of Appendix A.
In Figure 3, each dot of the 22-month period deaths corresponds to the integration of
deaths by month from the month of the dot to the previous month of the next dot,
included. So the integrated deaths are shown at the beginning of each integration
period (emphasized with colors).
With the integrated mortality by 22-month periods, we can spot a plateau of deaths from
2000 to 2010, an increase from 2010 to 2019, and the break between the pre-covid
period and the covid period (2020).
Figure 4 shows the integration of the 22-month periods with ACM/m for each of the 10-
year age groups.
Figure 4. All-cause mortality by month (light-blue) and by 22-month period (dark-blue) in
the USA from 2000 to 2021, for each of the age groups. Data are displayed from January
2000 to December 2021. Panels below: (A) for the 0-14 years age group; (B) for the 15-24
years age group; (C) for the 25-34 years age group; (D) for the 35-44 years age group; (E) for
the 45-54 years age group; (F) for the 55-64 years age group; (G) for the 65-74 years age
group; (H) for the 75-84 years age group; (I) for the 85+ years age group. Data were retrieved
from CDC (CDC, 2022a), as described in Table 1.
21
A
B
22
C
D
23
E
F
24
G
H
25
Figure 4 is hard raw data that allows one to robustly evaluate a covid-period fatality risk
(covid-period excess mortality compared to the historic trend of pre-covid-period
mortality) for each age group. Here, with an eye to more than two decades of data for
the whole USA.
Except for the younger age group (the 0-14 year-olds, Figure 4A), we can see the break
in mortality from the covid period for all the age groups: mortality by month reaches a
new higher plateau and mortality by 22-month period has an increase beyond the one
expected from the historic trend (Figure 4B, C, D, E, F, G, H, I). This is especially true
for the 25-34 and 35-44 year-olds (Figure 4C and D), which experience close to a 50%
increase in the covid period compared to the period of same duration immediately
before.
Next, for a given age group and sex, we calculate the excess deaths of the covid period
“to date” using two different assumptions, as follows.
In the first assumption, we take the “excess deaths of the covid period” to mean the (all-
cause) deaths above the deaths that would have occurred if the same circumstances
I
26
would have prevailed during the covid period as prevailed in the 1st-prior 22-month
period, immediately preceding the covid period. Under this assumption, the excess
deaths of the covid period, due to everything different or extraordinary that occurred or
was imposed during the covid period, for a given age group and sex, is simply:
xDc(22)1 = m22c - m22c-1 (1)
In the second assumption, we take the “excess deaths of the covid period” to mean the
(all-cause) deaths above the deaths that would have occurred if the same
circumstances would have prevailed during the covid period as prevailed in the 2nd-
prior 22-month period, the period preceding the 1st-prior 22-month period before the
covid period. Under this assumption, the excess deaths of the covid period, due to
everything different or extraordinary that occurred or was imposed during the covid
period, for a given age group and sex, is:
xDc(22)2 = m22c - m22c-2 (2)
These formulas (Equations 1 and 2) are justified because m22c-1 and m22c-2 are
different and fair estimates of what mortality would have been in the 22-month covid
period if the events associated with the declared pandemic had not occurred. In other
words, m22c-1 and m22c-2 are fair historical projected values of what the covid-period
mortality “would have been”. Judging from Figure 3 and Figure 4, there would be little
benefit from applying a more mathematically sophisticated extrapolation method, while
using both reference values allows one to estimate the uncertainty in our determinations
of excess mortality for the covid period.
The relative magnitudes of the covid-period extra deaths above the historic trend are:
xDc(22)1% = xDc(22)1 / m22c-1, expressed as a percentage, (3)
and
xDc(22)2% = xDc(22)2 / m22c-2, expressed as a percentage, (4)
27
Table 2 contains the calculated covid-period excess mortality, for each age group and
sex for the USA, and for all ages and both sexes for the entire USA (“Total”), using each
assumption described above, and the relative changes also, as percentages of the
reference values in Equations 1 and 2 (m22c-1 and m22c-2, respectively).
28
Table 2. Estimated excess mortality of the covid period in the USA, by age group and by
sex. m22c is the total deaths during the covid period (from March 2020 to December 2021,
included). m22c-1 is the total deaths during the 1st-prior 22-month period before the covid
period (from May 2018 to February 2020, included). m22c-2 is the total deaths during the 2nd-
prior 22-month period before the covid period (from July 2016 to April 2018, included). xDc(22)1
and xDc(22)2 correspond to the excess mortality in the covid period, calculated from Equation 1
and Equation 2, respectively. xDc(22)1% and xDc(22)2% correspond to the relative changes,
29
calculated from Equation 3 and Equation 4, respectively. ACM data were retrieved from CDC
(CDC, 2022a), as described in Table 1.
One of the most surprizing results from the above calculations is that young adults were
severely negatively impacted in the covid period, more so in comparative terms (percent
mortality increase relative to pre-covid values) than elderly persons. This is explored
further, below.
In the next section, we follow the same method to estimate the excess mortality of the
covid period in the USA from a different dataset: the all-cause mortality by week
(ACM/w).
3.2. USA all-cause mortality by week, by age group, 2015-2022
3.2.1. Historic trend, discontinuous break on 11 March 2020, entering the covid period
Figure 5 shows the all-cause mortality by week (ACM/w) for the USA from January 2015
to January 2022.
30
Figure 5. All-cause mortality by week in the USA from 2015 to 2022. Data are displayed
from week-1 of 2015 to week-5 of 2022. The vertical dark-blue line indicates the week-11 of
2020 (week of 11 March 2020, when WHO declared a pandemic), intended to point the
beginning of the covid period. The cvp1, smp1, cvp2 and smp2 features discussed in the text
are indicated. Data were retrieved from CDC (CDC, 2022b), as described in Table 1.
The regular seasonal variation of mortality is seen from 2015 to early 2020, and from
week-11 of 2020 (the week the WHO declared a pandemic), a new pattern of mortality
(new regime of ACM by time) occurs (Figure 5, after the vertical dark-blue line). This
new pattern includes the previously discussed features: cvp1, smp1, cvp2, smp2.
In the next section, we use the weekly ACM data to quantify the total excess mortality
that occurred in the covid period, which includes these anomalous features.
3.2.2. Quantifying the excess mortality of the covid period, by age group
We use the ACM/w data of Figure 5 to quantify the excess deaths of the covid period “to
date”, compared to the historic trend, as follows.
For a given age group, we add all the weekly deaths together, for the weeks of 11
March 2020 (week-11 of 2020, start of the pandemic period; announced by the WHO on
cvp1
smp2
smp1
cvp2
31
11 March 2020) through to the latest useable week (week-5 of 2022, beginning of
February 2022). This is a total for 100 weeks (the covid period “to date”). We call this
total “w100c”. Then we perform a similar total for the 1st-prior 100-week period,
immediately preceding the covid period, for the 100 weeks up to and including week-10
of 2020. We call this total “w100c-1”. And we do the same for the 2nd-prior 100-week
period, and we call this total “w100c-2”. We cannot move back further in time with this
dataset, as the “w100c-3” would be incomplete (less than a 100 weeks, with the
available data).
Figure 6 shows the graph of “w100c-x” versus time, together with the ACM/w for the
USA where each 100-week period has been emphasized with a different color; thus
applying the same method as in producing Figure 3.
Figure 6. All-cause mortality by week (colors) and by 100-week period (black) in the USA
from 2016 to 2022. Data are displayed from week-19 of 2016 to week-5 of 2022. The different
colors indicate the successive 100-week periods. The light-blue color corresponds to the covid
period. The dark-blue and the orange colors are in the pre-covid period. The black dots show
the integrated ACM on these 100-week periods. Data were retrieved from CDC (CDC, 2022b),
as described in Table 1.
32
Figure 6, based on more than 7 years of data, here time-resolved by week, again (like
Figure 3) dramatically illustrates the sudden change in regime of ACM by time, both in
magnitude of time-integrated ACM and in seasonal behaviour of ACM by time, occurring
as soon as the WHO on 11 March 2020 announced a pandemic. In addition, the covid-
period regime of ACM by time is characterized by large (and unprecedented in the
historic record) heterogeneity by state of ACM, which is not shown in such a figure for
the whole USA, but which can be appreciated in the ACM/w by state graphs of
Appendix A.
For the whole USA (all states and all ages together), the increase in ACM between the
pre-covid and the covid period is close to 25% (Figure 6).
The mortality data (Figure 6) can be resolved by age group, which is shown, as follows,
in Figure 7.
Figure 7. All-cause mortality by week (light-blue) and by 100-week period (dark-blue) in
the USA from 2016 to 2022, for each of the age groups. Data are displayed from week-19 of
2016 to week-5 of 2022. Panels below: (A) for the 0-24 years age group; (B) for the 25-44 years
age group; (C) for the 45-64 years age group; (D) for the 65-74 years age group; (E) for the 75-
84 years age group; (F) for the 85+ years age group. Data were retrieved from CDC (CDC,
2022b), as described in Table 1.
A
33
B
C
34
D
E
35
Except for the younger age group (the 0-24 year-olds, Figure 7A), the integrated
mortality of the covid period is much larger than in any of the two previous 100-week
periods (Figure 7B, C, D, E, F). These results are comparable to those illustrated in
Figure 4.
It is interesting to note that the sudden rise in ACM, immediately following the WHO’s 11
March 2020 declaration of a pandemic, which we have discussed in several previous
articles (Rancourt, 2020; Rancourt, Baudin and Mercier, 2020, 2021a, 2021b), occurs in
all the age groups for the whole USA (Figure 7; and see Figure 4), not solely in the most
elderly populations as reports of severe COVID-19 morbidity might lead one to conclude
(e.g., Elo et al., 2022; Sorensen et al., 2022). This, in itself, suggests that the covid-
period deaths are not predominantly explained by the postulated SARS-CoV-2
pathogen.
Next, for a given age group, we calculate the excess deaths of the covid period “to date”
using our simplest assumption from above. We take the “excess deaths of the covid
period” to mean the (all-cause) deaths above the deaths that would have occurred if the
same circumstances would have prevailed during the covid period as prevailed in the
1st-prior 100-week period, immediately preceding the covid period. Under this
F
36
assumption, the excess deaths of the covid period, due to everything different or
extraordinary that occurred or was imposed during the covid period, for a given age
group and state, is:
xDc(100)1 = w100c - w100c-1 (5)
In the second assumption, we take the “excess deaths of the covid period” to mean the
(all-cause) deaths above the deaths that would have occurred if the same
circumstances would have prevailed during the covid period as prevailed in the 2nd-
prior 100-week period, the period preceding the 1st-prior 100-week period before the
covid period. Under this assumption, the excess deaths of the covid period, due to
everything different or extraordinary that occurred or was imposed during the covid
period, for a given age group and state, is:
xDc(100)2 = w100c - w100c-2 (6)
As with Equations 1 and 2 above, these formulas (Equations 5 and 6) are justified
because w100c-1 and w100c-2 are different and fair estimates of what mortality would
have been in the 100-week covid period if the events associated with the declared
pandemic had not occurred. In other words, w100c-1 and w100c-2 are fair historical
projected values of what the covid-period mortality “would have been”. Judging from
Figure 6 and Figure 7, there would be little benefit from applying a more mathematically
sophisticated extrapolation method, while using both reference values allows one to
estimate the uncertainty in our determinations of excess mortality for the covid period.
The relative magnitudes of the covid-period extra deaths above the historic trend are:
xDc(100)1% = xDc(100)1 / w100c-1, expressed as a percentage, (7)
and
xDc(100)2% = xDc(100)2 / w100c-2, expressed as a percentage, (8)
37
Table 3 contains the thus calculated covid-period excess mortality, for each age group
for the USA, and for the entire USA (“Total”), using each assumption described above,
and the relative changes also, as percentages of the reference values in Equations 5
and 6 (w100c-1 and w100c-2, respectively).
Table 3. Estimated excess mortality of the covid period in the USA, by age group. w100c
is the total deaths during the covid period (from week-11 of 2020 to week-5 of 2022, included).
w100c-1 is the total deaths during the 1st-prior 100-week period before the covid period (from
week-15 of 2018 to week-10 of 2020, included). w100c-2 is the total deaths during the 2nd-prior
100-week period before the covid period (from week-19 of 2016 to week-14 of 2018, included).
xDc(100)1 and xDc(100)2 correspond to the excess mortality in the covid period, calculated
from Equation 5 and Equation 6, respectively. xDc(100)1% and xDc(100)2% correspond to the
relative changes, calculated from Equation 7 and Equation 8, respectively. ACM data were
retrieved from CDC (CDC, 2022b), as described in Table 1.
Equivalents to Table 3 for each of the states of the USA can be found in Appendix A.
Not surprisingly, we find the same results as with the ACM/m data, where young adults
were relatively more impacted in the covid period than elderly persons.
In the next section, we explore the excess mortality of the covid period at the state level.
3.2.3. Excess mortality of the covid period, by state
Figure 8 shows USA maps of the state-wise values of the covid-period excess mortality
(xDc(100)1), as relative changes in percentage of the pre-covid period mortality
(xDc(100)1%) (Panel A), and xDc(100)1 per state population (Panel B), for comparison.
38
Figure 8. Maps of the excess mortality of the covid period in the USA, as percentages of
the pre-covid period mortality (panel A) and as normalized by state population (panel B).
Alaska and Hawaii are excluded. The darker the color (black), the more intense is the relative
change. ACM data were retrieved from CDC (CDC, 2022b) and population data were retrieved
A
B
39
from US Census Bureau (US Census Bureau, 2021), as described in Table 1. xDc(100)1 and
xDc(100)1% are calculated from Equation 5 and Equation 7, respectively.
These maps (Figure 8) can be compared to the maps of poverty and obesity shown in
Appendix B; and to the maps from Rancourt et al. (Rancourt, Baudin and Mercier,
2021b) of life expectancy (their Figure 38a), antibiotic prescriptions (their Figure 38b),
average climatic temperature (their Figure 22), intensity of the smp1 mortality (their
Figure 16), intensity of the cvp1 mortality (their Figure 15). Some of these comparisons
are discussed further below.
Generally, high 100-week covid-period mortality per capita or per baseline mortality
occurs in the Southern states, and in the hottest climatic state of Arizona. This is similar
to what we have reported previously for summer-season covid mortality (Rancourt,
Baudin and Mercier, 2021b). Below we show that state-wise covid-period mortality is
very strongly correlated (r = +0.86) to state-wise poverty, and also correlated to median
household income, obesity, disability, and government subsidy programs; which in turn
are known to be correlated to each other and to diabetes prevalence, life expectancy,
and antibiotic prescriptions. All of this is consistent with the geographical pattern shown
in Figure 8.
Figure 9 shows the xDc(100)1% (Equation 7) values from Table 3 by age group, for the
whole USA (Panel A), and for the ten most populous states (Panel B), ordered from the
most populous to the less populous (US Census Bureau, 2022a): California, Texas,
Florida, New York, Pennsylvania, Illinois, Ohio, Georgia, North Carolina and Michigan.
The horizontal dashed line represents the value for the whole USA (all ages and all
states).
40
Figure 9. Excess mortality of the covid period in the USA (panel A) and in the ten most
populous states of the USA (from left to right in each band: California, Texas, Florida,
New York, Pennsylvania, Illinois, Ohio, Georgia, North Carolina, Michigan) (panel B), as
percentages of the pre-covid period mortality, by age group. The constant dashed line
B
A
41
represents the value for the whole USA. ACM data were retrieved from CDC (CDC, 2022b), as
described in Table 1. xDc(100)1% is calculated from Equation 7.
Figure 9 illustrates one of the most striking features of mortality in the covid period: The
relative covid-period excess mortality (covid-period fatality risk ratio, relative to pre-covid
mortality) is broadly distributed to all age groups and is not exponential or near-
exponential with age as determined for viral respiratory diseases, including COVID-19,
when these are the verified dominant cause of death.
Indeed, we note that all age groups were significantly differentially affected in the covid
period, which is inconsistent with the reported infection fatality ratios (morbidity) that
generally increase exponentially with age, as is also the case for many chronic diseases
and for all-cause mortality risk itself (e.g., Richmond et al., 2021; Elo et al., 2022;
Sorensen et al., 2022). Again, this suggests that the covid-period deaths are not
predominantly explained by the postulated SARS-CoV-2 pathogen. Rather, risk of death
in the covid period appears to result from distributed aggression against vulnerable
populations in all the age groups, not predominantly (or exponentially) the elderly.
We see from Figure 9 that young adults (25-44 years) were particularly devastated by
the events and conditions of the covid period. It is not unreasonable to postulate that
this age group would have been most impacted by the large-scale life-changing
economic and job-loss changes that occurred in the covid period, or that this age group
would have been most devastated by social isolation and institutional abandonment for
those who are mentally disabled or otherwise dependent on a fragile social support
network.
Next, we examine whether any impact of the mass and age-distributed USA vaccination
campaign can be detected and quantified.
42
3.3. Time and age-group variations of mortality during the covid period, and
relation to implementation of the vaccination campaign
3.3.1. All-cause mortality by week and vaccination delivery by week, by age group,
2019-2022
In our previous article about ACM in the USA (Rancourt, Baudin and Mercier, 2021b),
we stated the following about the vaccination campaign:
“Readers who would be tempted to ascribe the downturn in the cvp2 peak to the
vaccination campaign should note that the downturn coincides with the
expected seasonal downturn of every seasonal winter maximum that has ever
been observed by epidemiologists in the last century or more.
More importantly, the largely completed vaccination campaign did not prevent a
second surge of summer deaths (2021, “smp2”) (Figure 31). The mortality in the
said second surge appears to be comparable to or more than the mortality for
summer-2020. Furthermore, the COVID-19-assigned deaths (CDC, 2021a) are
significantly greater in number in summer-2021 than in summer-2020 (Figure
34), and, unlike at any other time in the COVID-era, account for virtually all the
excess (above-SB) deaths, in the summer-2021 feature (smp2) (Figure 34),
following the vaccination campaign.
There is no sign in the ACM/w that the vaccination campaign has had any
positive effect. However, given that the vaccination campaign starts well after
the 2020 summer and essentially ends mid-summer-2021 prior to the start of
the smp2 feature, given that the 2021 excess (above-SB) summer deaths
(smp2) occur in significantly younger individuals than the excess summer-2020
deaths, and given that the smp2 feature is significantly larger than the smp1
feature for the said younger individuals (35-54 years, Figures 33d and 33e; and
55-64 years, Figure 33f, to a lesser degree), it is possible that vaccination made
35-54 year olds and others more vulnerable to death, especially summer death
in disadvantaged individuals in hot-climate states (Montgomery et al., 2021)
(Simone et al., 2021).”
43
Here, we examine this question again, via the time and age-group variations in structure
of the ACM/w (Figure 5 and Figure 7) in the covid period, using the most up-to-date
consolidated data.
Figure 10 shows the all-cause mortality by week (ACM/w) for the USA from January
2019 through January 2022, together with vaccination data, for all the available age
groups.
Figure 10. All-cause mortality by week (light-blue), cumulated number of people with at
least one dose of vaccine (dark-blue), cumulated number of fully vaccinated people
(orange) and cumulated number of people with a booster dose (yellow) by week in the
USA from 2019 to 2022, for all and each of the age groups. Data are displayed from week-1
of 2019 to week-5 of 2022. The vertical solid line indicates week-11 of 2020 (week of 11 March
2020, when WHO declared a pandemic), indicating the beginning of the covid period. The
vertical dashed line indicates week-8 of 2021, dividing the covid period into two periods of 50
weeks each: the pre-vaccination period (before the dashed line) and the vaccination period
(after the dashed line). Panels below: (A) for all ages; (B) for the 0-24 years age group; (C) for
the 25-44 years age group; (D) for the 45-64 years age group; (E) for the 65-74 years age
group; (F) for the 75+ years age group. Data were retrieved from CDC (CDC, 2022b, 2022c), as
described in Table 1.
A
44
The booster data for this age group only concern people aged 12 years and older.
For the vaccination data of this age group, the solid lines are for the 25-39 year olds and the
dashed lines are for the 25-49 year olds. That is because the available age groups for the
mortality data don’t exactly match the available age groups for the vaccination data.
C
B
45
For the vaccination data of this age group, the solid lines are for the 40-64 year olds and the
dashed lines are for the 50-64 year olds. That is because the available age groups for the
mortality data don’t exactly match the available age groups for the vaccination data.
D
E
46
Figure 10 is a key figure in the present article because it allows an investigation of
whether accelerations of vaccine delivery are synchronous or near-synchronous with
surges (vaccine-induced death) or subsequent drops (vaccine-induced protection
against death) in ACM, for all ages (Figure 10A) and by age group (Figure 10 B, C, D,
E, F). In this regard, we make the following observations.
First, one might be tempted to mechanistically associate the initial and most
important surge in 1st-dose vaccine delivery with the large drop in mortality that
for several age groups occurs at about the same time (in March-2021 for all
ages, Figure 10A). This is incorrect for the following reasons:
o The drop in mortality is expected from purely seasonal considerations:
high mortality in the winter always drops eventually.
o The cvp1 at the end of the winter occurring in the pre-vaccination period of
the covid period saw an eventual large decrease, months before the start
of the vaccination campaign.
o There is an increase in ACM in the 0-24 years age group, rather than a
decrease (Figure 10B), and similarly for the 25-44 years age group (Figure
10C). The vaccine would need to be harmful or beneficial regarding death,
depending on the age group.
F
47
o For the 45-64, 65-74 and 75+ years age groups, the 2020-2021 winter
peak in ACM occurs in the same way even though the vaccine-delivery
upsurge is at different times, because the most elderly were vaccinated
first (Figure 10D, E, F). The vaccine’s life-saving properties would need to
be strongly dependent on age for these ages.
Second, it is clear that the prominent late-summer-2021 peak in ACM (all ages,
and all age groups except 0-24 years) is far in excess of any proportionate
increase in vaccination-dose delivery. The said late-summer-2021 peak occurs in
a period during which the cumulative vaccine dose delivery is essentially regular,
without a large fractional step-wise increase.
Third, the latter observation notwithstanding, there is nonetheless a modest but
statistically significant stepwise increase in 1st-dose vaccine delivery, which is
synchronous with the late-summer-2021 peak in ACM, visible for all ages and for
the 25-44 and 45-64 years age groups (Figure 10A, C, D). This temporal
association is prominent in the data for many specific states (e.g., Figure 11),
and cannot easily be dismissed. It is discussed below.
While the second and third bullet points above appear to be contradictory, they
are not. On the one hand (second bullet point), neither large increases in ACM
(upsurge of the late-summer-2021 peak) nor large decreases in ACM (drop in
ACM ending the late-summer-2021 peak) can be interpreted as proportionately
driven by vaccine adverse effects, while on the other hand (third bullet point), a
modest stepwise upsurge in cumulative vaccine dose delivery may be causally
associated with a peak in ACM if the said stepwise upsurge includes increased
capture of immunocompromised residents. The two propositions (second and
third bullet points) and their implications are simultaneously possible because the
number of delivered vaccine doses is large compared to the number of excess
deaths (the per-dose fatality toxicity ratio of the vaccine is much smaller than 1),
as discussed more below.
Fourth, one might be tempted to mechanistically associate the increase in
cumulative booster-dose delivery with irregular increases in ACM in the late
stage of the covid period. This is incorrect for the following reasons:
48
o The apparent association is confounded by the 2021-2022 winter
increase. Every winter, including during the covid period, has always had
increased ACM, in the entire recorded history of mid-latitude countries and
jurisdictions.
o Booster and concomitant first-series dose increases have an apparent
insignificant effect on ACM in the 0-24 years age group, and cause a
decrease if anything in the winter 2021-2022 season (Figure 10B).
o Boosters cause no special increase in ACM in the 25-44 years age group
(Figure 10C), which is the age group with the largest vaccination-period
relative increase in integrated ACM (see below).
o The 2021-2022 winter peaks in all the >24 years age groups have their
maxima at a time when the cumulative booster-dose delivery has
plateaued, after its period of most rapid increase (Figure 10A, C, D, E, F).
Data by age group shown in Figure 10 were only available at the national level. In the
next section, we look at vaccination data at the state level, with less defined age groups.
3.3.2. All-cause mortality by week and vaccination delivery by week, by state, 2019-
2022
Vaccination delivery by week data is available at the state level for the 18+ and the 65+
age groups (CDC, 2022d). By subtracting the data for the 65+ age group from the data
for the 18+ age group, we can calculate data for the 18-64 age group.
Figure 11 shows the all-cause mortality by week (ACM/w) for some states of the USA
from January 2019 through January 2022, together with vaccination data, for the 25-64
years or the 65+ years age groups.
Figure 11. All-cause mortality by week (light-blue), cumulated number of people with at
least one dose of vaccine (dark-blue), cumulated number of fully vaccinated people
(orange) and cumulated number of people with a booster dose (yellow) by week from
2019 to 2022, and by age group for some states. Data are displayed from week-1 of 2019 to
week-5 of 2022. Panels below: (A) Alabama, 25-64 years age group; (B) Mississippi, 25-64
49
years age group; (C) Georgia, 25-64 years age group; (D) Florida, 25-64 years age group; (E)
Louisiana, 25-64 years age group; (F) Louisiana, 65+ years age group; (G) Michigan, 25-64
years age group; (H) Michigan, 65+ years age group. For the 25-64 years age group graphs,
the vaccination data is for the 18-64 years age group; because the available age groups for the
mortality data do not exactly match the available age groups for the vaccination data. The
discontinuous breaks in cumulative number of vaccinated individuals are artifacts. Data were
retrieved from CDC (CDC, 2022b, 2022d), as described in Table 1.
A
B
50
C
D
51
E
F
52
Figure 11 illustrates the late-summer-2021 peak in ACM/w for the states of Alabama,
Mississippi, Georgia, Florida and Louisiana; and the unique spring-2021 (April-centered)
peak in ACM/w occurring for Michigan.
Here, the modest but significant stepwise increase in 1st-dose vaccine delivery, which
is synchronous with the late-summer-2021 peak in ACM, visible for all ages and for the
G
H
53
25-44 and 45-64 years age groups (Figure 10A, C, D) discussed above for the whole
USA is now examined through the ACM/w and cumulative vaccine dose delivery by
week data for the states of Alabama, Mississippi, Georgia, Florida and Louisiana, where
the feature is prominent (Figure 11A, B, C, D, E, F), in the 25-64 years age group in
particular. These five states are examples of states in which the late-summer-2021 peak
is the most intense feature (largest peak) in the ACM/w data. In each case, the
synchronous stepwise increase in cumulative vaccine dose delivery is evident.
This association between late-summer-2021 peak and stepwise increase in vaccine
dose delivery is present throughout all the states: Where this is a most prominent late-
summer-2021 peak there is an evident synchronous stepwise increase in vaccine dose
delivery, and vice versa. The case of the state of Michigan shows a counter example:
There is no late-summer-2021 peak and there is no stepwise increase in vaccination
(Figure 11G, H).
However, the case of Michigan is shown for an additional reason: Michigan is the only
state that has a spring-2021 (April-centered) peak in ACM/w (Figure 11G, H). This is
arguably the most remarkable feature in all of the ACM data for the USA, since it occurs
only in one state and does not correspond to a local intense summer heatwave
phenomenon.
Michigan’s said spring-2021 peak in ACM/w occurs synchronously with Michigan’s
fastest increase in vaccine dose delivery for 18-64 year olds (Figure 11G). It occurs
when the vaccination campaign was “turned on” for this age group. This is also the time
(April-2021) when, for this age group, for the whole USA, vaccine delivery was at its
highest, and all reported vaccine adverse effects, including death, peaked (Hickey and
Rancourt, 2022; their Figure S2). The Janssen-shot deliveries (shots administered), in
particular, peaked strongly in approximately April-2021 (whole USA) (Hickey and
Rancourt, 2022; their Figure S1), and were CDC-recommended to be “paused”, and
then re-authorized at approximately that time, also (FDA, 2021, 2022).
54
For Michigan, therefore, one is tempted to directly assign the unique spring-2021 peak
in mortality as directly caused by the vaccine injections. The vaccine fatality toxicity per
dose would need to be approximately 10 times greater than the known value for non-
immunocompromised subjects (Hickey and Rancourt, 2022; their Table 1). However, if
immunocompromised young adults (stressed and mentally disabled, and such, see
below) were captured by the vaccination campaign, then the causal link is entirely
possible.
Coming back to the big picture: The massive vaccination campaign in the USA did not
reduce all-cause mortality to a pre-covid-period level, overall or in any of the age
groups; nor does it appear to have substantially increased ACM during the vaccination
campaign, compared to the pre-vaccination period of the covid period (Figure 10).
In the next section, we use the method described above (in section 3.2.2) to
quantitatively assess whether the vaccination campaign measurably affected integrated
ACM.
3.3.3. Quantifying excess mortality of the pre-vaccination and vaccination periods of the
covid period, by age group
We adapt our method described in section 3.2.2 and use the ACM/w data of Figure 5 to
quantify the excess mortality of the vaccination period “to date”, compared to the excess
mortality of the pre-vaccination period of the covid period, as follows.
The idea is to test whether there is a significant systematic increase in mortality, by
state and by age group, occurring after the large increase in vaccination injections,
compared to the (equal duration) part of the covid period prior to the surge in
vaccination delivery, and compared to a pre-covid period of same duration occurring
immediately prior to the 11 March 2020 start of the covid period.
55
For a given age group, we add all the weekly deaths together, for the weeks of 22
February 2021 (week-8 of 2021, inflection point of the vaccination period) through to the
latest useable week (week-5 of 2022, beginning of February 2022). This is a total for 50
weeks (the vaccination period “to date”). In analogy with our previously introduced
notation (above in section 3.2.2), we call this total “w50c”. Then we perform a similar
total for the 1st-prior 50-week period, immediately preceding the vaccination period, for
the 50 weeks up to and including week-7 of 2021. We call this total “w50c-1”. These two
50-week periods of the covid period, divide the covid period into equal-duration pre-
vaccination (w50c-1) and vaccination (w50c) periods, which can be visualized with the
help of Figure 10A and Figure 12 (below). And we do the same for the 2nd-prior 50-
week period, and we call this total “w50c-2”. We continue moving back in time, to the
end of the useable data in 50-week periods: w50c-3, etc.
Figure 12 shows the graph of “w50c-x” versus time, together with the ACM/w for the
USA where each 50-week period is distinguished using a different color.
Figure 12. All-cause mortality by week (colors) and by 50-week period (black) in the USA
from 2015 to 2022. Data are displayed from week-21 of 2015 to week-5 of 2022. The different
colors indicate the successive 50-week periods. The light-blue color corresponds to the
vaccination period of the covid period. The dark-blue color corresponds to the pre-vaccination
period of the covid period. All the other colors are in the pre-covid period. The black dots show
56
the integrated ACM on these 50-week periods. Data were retrieved from CDC (CDC, 2022b), as
described in Table 1.
Equivalents to Figure 12 (without the color-code) for each of the states of the USA can
be found in Appendix A.
Contrary to what would be expected if we assumed that the injections themselves
induced a large (dominant) measurable positive or negative change in ACM, over a 50-
week integration period the integrated ACM in the vaccination period of the covid period
is comparable to and lower than in the pre-vaccination period of the covid period, for the
USA as a whole (Figure 12). Indeed, there is a much greater and discontinuous change
in ACM in going between the pre-covid period and the covid period than in going
between the pre-vaccination period of the covid period and the vaccination period of the
covid period.
The mortality data (Figure 12) can be resolved by age group, which is shown, as
follows, in Figure 13.
Figure 13. All-cause mortality by week (light-blue) and by 50-week period (dark-blue) in
the USA from 2015 to 2022, for each of the age groups. Data are displayed from week-21 of
2015 to week-5 of 2022. Panels below: (A) for the 0-24 years age group; (B) for the 25-44 years
age group; (C) for the 45-64 years age group; (D) for the 65-74 years age group; (E) for the 75-
84 years age group; (F) for the 85+ years age group. Data were retrieved from CDC (CDC,
2022b), as described in Table 1.
57
A
B
58
C
D
59
The ACM by 50-week period resolved by age group shows that integrated ACM is
higher in the vaccination period of the covid period than in the pre-vaccination period of
the covid period for all the younger age groups, under 75 years old (Figure 13A, B, C,
D).
E
F
60
The integrated mortality by consecutive 50-week periods is shown for all the age groups
together in Figure 14, by normalizing all the 50-week periods by the first 50-week period
for each age group.
Figure 14. All-cause mortality by 50-week period normalized by the first 50-week period in
the USA, from 2015 to 2022, for each of the age groups. Data are displayed from week-21 of
2015 to week-8 of 2021 (beginning of the vaccination period). ACM data were retrieved from
CDC (CDC, 2022b), as described in Table 1.
The only age groups for which ACM in the vaccination period of the covid period is
lower than ACM in the pre-vaccination period of the covid period are the 75-84 and 85+
age groups. All the other age groups show otherwise (Figure 14).
In order to quantify and directly compare the pre-vaccination period and the vaccination
period within the covid period, we define the following quantities:
pVax-pCVD/pCVD = (w50c-1 - w50c-2) / w50c-2, expressed as a percentage,
(9)
and
61
Vax-pCVD/pCVD = (w50c - w50c-2) / w50c-2, expressed as a percentage,
(10)
Where w50c is the integrated ACM of the vaccination period of the covid period (50
weeks), w50c-1 the integrated ACM of the pre-vaccination period of the covid period (50
weeks) and w50c-2 the integrated ACM of the first pre-covid period of 50 weeks
(immediately preceding the covid period).
Table 4 contains the calculated vaccination-period excess mortality (Vax-pCVD) and
pre-vaccination-period excess mortality (pVax-pCVD) of the covid period, for each age
group for the USA, and for the entire USA (“Total”), and the relative changes also, using
each equation described above (Equations 9 and 10), as percentages of the pre-covid-
period reference values (w50c-2).
Table 4. Estimated excess mortality of the pre-vaccination and vaccination periods of the
covid period in the USA, by age group. w50c is the total deaths during the vaccination period
of the covid period (from week-8 of 2021 to week-5 of 2022, included). w50c-1 is the total
deaths during the pre-vaccination period of the covid period (from week-11 of 2020 to week-7 of
2021, included). w50c-2 is the total deaths during the pre-covid period (from week-13 of 2019 to
week-10 of 2020, included). pVax-pCVD and Vax-pCVD correspond to the excess mortality in
the pre-vaccination period of the covid period and to the excess mortality in the vaccination
period of the covid period, respectively. pVax-pCVD/pCVD and Vax-pCVD/pCVD correspond to
the relative changes, as percentages of the pre-covid-period mortality, calculated from Equation
9 and Equation 10, respectively. ACM data were retrieved from CDC (CDC, 2022b), as
described in Table 1.
Equivalents to Table 4 for each of the states of the USA can be found in Appendix A.
62
The numbers in Table 4 are represented graphically in bar charts, below, and are
discussed below.
Figure 15 shows those quantities together with the relative excess mortality change in
the covid period (xDc(100)1%, Equation 7) for each of the age groups for the whole
USA.
Figure 15. Excess mortality of the covid period (xDc(100)1%) (light-blue), of the pre-
vaccination period of the covid period (pVax-pCVD/pCVD) (dark blue) and of the
vaccination period of the covid period (Vax-pCVD/pCVD) (orange) in the USA, as
percentages of the pre-covid-period mortality, by age group. The constant dashed line
represents the value of xDc(100)1% for the whole USA. ACM data were retrieved from CDC
(CDC, 2022b), as described in Table 1. xDc(100)1%, pVax-pCVD/pCVD and Vax-pCVD/pCVD
are calculated from Equation 7, Equation 9 and Equation 10, respectively.
The excess mortality in the pre-vaccination period of the covid period is relatively lower
than the excess mortality in the vaccination period of the covid period and lower than
the excess mortality of the covid period for the younger age groups (0-24, 25-44, 45-64,
65-74) (Figure 15). The opposite is true for the older ages (75-84, 85+ years) (Figure
15). This qualitative difference can be interpreted as possibly associated to the
63
vaccination program, along the lines discussed above (Figure 10; Figure 11), in relation
to the late-summer-2021 peak and the synchronous modest stepwise increase in
cumulative vaccine dose delivery (administered). However, it is also possible that the
said qualitative difference results instead (or concomitantly) as being due to the impacts
of cumulative socio-economic pressures. Younger adults will have more resilience than
older adults, such that the deadly toll of life-changing circumstances will take longer to
materialize.
Next, we look at the excess mortality in the pre-vaccination period of the covid period
and in the vaccination period of the covid period at the state level.
3.3.4. Excess mortality of the pre-vaccination and vaccination periods of the covid
period, by state
Figure 16 shows USA maps of the covid-period pre-vaccination-period relative excess
mortality (pVax-pCVD/pCVD) (Panel A) and of the covid-period vaccination-period
relative excess mortality (Vax-pCVD/pCVD) (Panel B), as relative changes in
percentages of the pre-covid-period mortality by state.
64
Figure 16. Maps of the excess mortality in the pre-vaccination period of the covid period
(panel A) and in the vaccination period of the covid period (panel B) in the USA, as
percentages of the pre-covid-period mortality. Alaska and Hawaii are excluded. The darker
the color (black), the more intense is the relative change. ACM data were retrieved from CDC
A
B
65
(CDC, 2022b), as described in Table 1. pVax-pCVD/pCVD and Vax-pCVD/pCVD are calculated
from Equation 9 and Equation 10, respectively.
Figure 16 shows a striking “positive-negative” effect in which many states that have
relatively large relative mortality in the first half of the covid period (Panel A) have a
relatively small relative mortality in the second half of the covid period (Panel B), and
vice versa. This suggests a long-term (2 year) “dry tinder effect” in which vulnerable
populations are decimated early or late during the 100-week covid period, but that once
decimated cannot be re-decimated.
Figure 17 shows the covid-period pre-vaccination-period excess mortality (Equation 9)
and the covid-period vaccination-period excess mortality (Equation 10) as percentages
of the pre-covid-period mortality by age group, for the whole USA (Panel A), and for the
ten most populous states (Panels B and C), ordered from the most populous to the less
populous (US Census Bureau, 2022a): California, Texas, Florida, New York,
Pennsylvania, Illinois, Ohio, Georgia, North Carolina and Michigan.
A
66
Figure 17. Excess mortality in the pre-vaccination period of the covid period (pVax-
pCVD/pCVD) and in the vaccination period of the covid period (Vax-pCVD/pCVD) in the
USA (Panel A) and in the ten most populous states of the USA (from left to right in each
band: California, Texas, Florida, New York, Pennsylvania, Illinois, Ohio, Georgia, North
B
C
67
Carolina, Michigan) for the pre-vaccination period of the covid period (Panel B) and for
the vaccination period of the covid period (Panel C), as percentages of the pre-covid-
period mortality, by age group. ACM data were retrieved from CDC (CDC, 2022b), as
described in Table 1. pVax-pCVD/pCVD and Vax-pCVD/pCVD are calculated from Equation 9
and Equation 10, respectively.
Figure 15 and Figure 17 strikingly illustrate a large systematic change in going between
the pre-vaccination period of the covid period (first 50 weeks) and the vaccination period
of the covid period (second 50 weeks): The age structure of relative excess mortality
changes significantly, from being largely uniform with age (pre-vaccination) to being
highly weighted towards young adults (vaccination).
Regarding the evident change in age structure of the relative mortality in going from the
pre-vaccination period of the covid period into the vaccination period of the covid period
(Figure 17), the same possible interpretations apply as discussed above for Figure 15:
The said change in age structure can be interpreted as possibly associated to the
vaccination program, along the lines discussed above (Figure 10; Figure 11), in relation
to the late-summer-2021 peak and the synchronous modest stepwise increase in
cumulative vaccine dose delivery (administered). However, it is also possible that the
said change in age structure results instead (or concomitantly) as being due to the
impacts of cumulative socio-economic pressures. Younger adults will have more
resilience than older adults, such that the deadly toll of life-changing circumstances will
take longer to materialize. Both of these hypotheses (resilience in youth and vaccine
assault of vulnerable-group individuals), in turn, are consistent with the fact that the
prevalence of serious mental illness is large and highly skewed towards young adults in
the USA (NIMH, 2022).
In the next section, we explore the differential integrated mortality between the
vaccination and pre-vaccination periods of the covid period at the state level.
68
3.3.5. Difference of vaccination and pre-vaccination mortality in the covid period, by age
group and by state
For a given age group and state, we calculate the difference (Vax-pVax) between
integrated mortality in the vaccination period of the covid period (w50c) and integrated
mortality in the pre-vaccination period of the covid period (w50c-1):
Vax-pVax = w50c - w50c-1 (11)
This difference (Vax-pVax) normalized by the pre-covid-period integrated mortality
(w50c-2) is:
Vax-pVax/pCVD = (w50c - w50c-1) / w50c-2, expressed as a percentage,
(12)
Table 5 contains the calculated difference in mortality between the vaccination and pre-
vaccination periods of the covid period (Vax-pVax), for each age group for the USA, and
for the entire USA (“Total”), and the relative change also, as percentages of the pre-
covid-period reference values (w50c-2).
Table 5. Difference of vaccination and pre-vaccination mortality in the covid period in the
USA, by age group. w50c is the total deaths during the vaccination period of the covid period
(from week-8 of 2021 to week-5 of 2022, included). w50c-1 is the total deaths during the pre-
vaccination period of the covid period (from week-11 of 2020 to week-7 of 2021, included).
w50c-2 is the total deaths during the pre-covid period (from week-13 of 2019 to week-10 of
69
2020, included). Vax-pVax corresponds to the difference between the vaccination-period
mortality and the pre-vaccination-period mortality, calculated from Equation 11. Vax-pVax/pCVD
corresponds to the relative change, as percentage of the pre-covid-period mortality, calculated
from Equation 12. ACM data were retrieved from CDC (CDC, 2022b), as described in Table 1.
Equivalents to Table 5Table 4 for each of the states of the USA can be found in
Appendix A.
In the covid period, vaccination-period ACM is greater than pre-vaccination-period ACM
for younger people, and smaller for older people (Table 5). In terms of deaths
predominantly caused by the vaccines, this would be opposite to the known exponential
increase with age of vaccine-associated deaths (Hickey and Rancourt, 2022).
Figure 18 shows a USA map of the state-wise difference between vaccination and pre-
vaccination mortality (Vax-pVax), as relative changes in percentage of the pre-covid-
period mortality (Vax-pVax/pCVD).
Figure 18. Map of the difference of vaccination and pre-vaccination mortality in the covid
period in the USA, as percentages of the pre-covid-period mortality. Alaska and Hawaii are
70
excluded. The darker the color (black or yellow), the more intense is the relative change
(positive or negative, respectively). ACM data were retrieved from CDC (CDC, 2022b), as
described in Table 1. Vax-pVax/pCVD is calculated from Equation 12.
Figure 18 is a geographical representation of where (by state) the differences between
the mortality per pre-covid mortality (pCVD) of the first half of the covid period (pVax;
pre-vaccination period) and the second half of the covid period (Vax; vaccination period)
are largest, both negative (pVax > Vax; darkest yellow) and positive (Vax > pVax;
darkest grey). The known initial hot spots of New Jersey and New York are bright
yellow, whereas the states with comparatively large late-covid-period mortality show up
in dark grey: Maine, Oregon, Idaho, Washington, Florida…
In our view, it is not tenable to propose that the structure represented in Figure 18
arises from the national vaccination campaign as the dominant causal factor. There is
no logical reason to propose, as the dominant excess-mortality-determining factor, that
the vaccines saved lives in the states that have the largest initial (first 50 weeks of the
covid period) mortality per capita or per pre-covid mortality and/or caused massive
mortality per capita or per pre-covid mortality in the states that had relatively small initial
covid-period mortality per capita. However, the map (Figure 18) does suggest a “dry
tinder effect” for vulnerable populations, over the course of approximately two years
under covid-period conditions, as discussed above for Figure 16.
Figure 19 shows the Vax-pVax/pCVD (Equation 12) values from Table 5 by age group,
for the whole USA (Panel A), and for the ten most populous states (Panel B), ordered
from the most populous to the less populous (US Census Bureau, 2022a): California,
Texas, Florida, New York, Pennsylvania, Illinois, Ohio, Georgia, North Carolina and
Michigan. The horizontal dashed line represents the value for the whole USA (all ages
and all states).
71
Figure 19. Difference of vaccination and pre-vaccination mortality in the covid period in
the USA (panel A) and in the ten most populous states of the USA (from left to right in
each band: California, Texas, Florida, New York, Pennsylvania, Illinois, Ohio, Georgia,
North Carolina, Michigan) (panel B), as percentages of the pre-covid-period mortality, by
age group. The constant dashed line represents the value for the whole USA. ACM data were
B
A
72
retrieved from CDC (CDC, 2022b), as described in Table 1. Vax-pVax/pCVD is calculated from
Equation 12.
Figure 19 is another way (by difference) to illustrate the dramatic change in age
structure of relative (i.e., age-group specific) excess mortality from being largely uniform
with age (pre-vaccination) to being highly weighted towards young adults (vaccination),
which is shown in Figure 17.
Figure 19 shows that more young adults died (relative to their population or to their pre-
covid death rate) in the second half (Vax) of the covid period relative to the first half
(pVax) of the covid period, for most states and for the whole USA. This is consistent
with a long-term (2-year) dry tinder effect for elderly populations, and greater resilience
against the assault of the covid-period conditions for younger populations, such as to
take longer for mortality to be experienced in younger residents. It is also consistent
with the hypothesis that immunocompromised young adults were captured by the
vaccination campaign, including the so-called “vaccine equity” programs, which would
also explain the large late-summer-2021 ACM peak for young adults discussed above.
Both of these hypotheses, in turn, are consistent with the fact that the prevalence of
serious mental illness is large and highly skewed towards young adults in the USA
(NIMH, 2022).
Therefore, from all of the above, it does not appear that the USA vaccination campaign
has had a dominant impact, positive or negative, on integrated all-cause mortality,
although it may have participated or predominantly caused the change in age structure
of mortality risk, and may have contributed to maintaining a large covid-period ACM.
The changes in mortality per pre-covid mortality, which occur between the first (pVax)
and second (Vax) halves of the covid period may be due to temporal changes in both
quantity (dry tinder effect) and quality (age, resilience) of the vulnerable populations
during a sustained covid-period assault on living conditions, and may have been
significantly modulated by vaccine-campaign capture of immunocompromised young
adults from vulnerable groups. In order to explore these hypotheses, regarding
73
vulnerable groups, we next quantify excess mortality per capita for the entire 100-week
covid period and examine its correlations with various socio-economic factors, in the
following section.
3.4. Associations of excess mortality of the covid period with socio-geo-
economic variables
In our previous article (Rancourt, Baudin and Mercier, 2021b), we described
associations of integrated excess (with respect to an extrapolated summer baseline
mortality) all-cause mortality per capita in anomalous features (cvp1, smp1, cvp2, smp2)
of all-cause mortality by time in the covid period with socio-geo-economic and climatic
parameters:
[] we have shown is that, in the COVID-era, during summer-2020 (smp1),
fall-winter-2020-2021 (cvp2) and summer-2021 (smp2), combined factors
including poverty, obesity and hot climate became deadly associations for
excess (above-SB) deaths, beyond the deaths that would have occurred from
the pre-COVID-era background of preexisting risk factors.
Therefore, here again we examine associations with such factors.
The following factors normalized by state population are tested against the quantified
excess mortality of the covid period (xDc(100)1) normalized by the state population:
Poverty
Median Household Income (MHI)
Obesity
Population aged 65 and over (and 75+, and 85+)
Supplemental Security Income (SSI)
Social Security Disability Insurance (SSDI)
Disability
74
Figure 20 shows the scatter plot for poverty (on two different scales, A and B), defined
as the estimated percentage of the population of people of all ages living in poverty (US
Census Bureau, 2022b). The Y-axis is the fraction xDc(100)1/pop, the 100-week covid-
period excess mortality by population, which is the “100-week covid-period fatality ratio”
for the USA population.
A
B
75
Figure 20. Excess mortality of the covid period normalized by population versus poverty
in the USA. The axes are optimized for the dataset (Panel A) and for the intercept between
trend line and X-axis (Panel B). Each point is for one state of the USA. The parameters of the
least squares fitted linear trend line are given in Table 6. The color-code of the 51 states is
shown in section 2. Data were retrieved as described in Table 1. xDc(100)1 is calculated from
Equation 5.
Figure 20 is a striking result. Such a result is rarely so clear in epidemiological studies.
The Pearson correlation coefficient is r = +0.86 (Table 6). Beyond this very strong
correlation, we note that the least squares fitted straight line passes virtually through the
origin (Table 6), implying that the integrated excess mortality per capita per state for the
whole 100-week covid period (i.e., what we have termed the “100-week covid-period
fatality ratio” for the USA population) is directly proportional to poverty of the state, not
merely very strongly correlated to poverty. Such proportionality suggests a fundamental
relationship, which is causal in nature; in which poverty captures or is an accurate proxy
for the dominant factor or factors that determine mortality arising from all the conditions
occuring during the covid period.
The said proportionality (Figure 20) means that a state with zero poverty would have
experienced zero excess mortality in the 100-week covid period, and that doubling
state-wise poverty (the fraction of state residents living in poverty) doubles excess
mortality in the 100-week covid period, for example.
Furthermore, we note that it is unlikely that this strong epidemiological relationship with
poverty arises from a viral respiratory disease. The classic development of a viral
respiratory disease, leading to death, is one in which the infection fatality ratio is
approximately exponential with age, with the main co-factors being comorbidity, not
economic hardship itself, irrespective of age. There is no known viral respiratory
disease in which the pathogen targets poverty, while being insensitive to age (see
scatter plot versus age of the state population, Figure 23 below).
Figure 21 shows the scatter plot for median household income (MHI) (on two different
scales, A and B), defined as the estimated median household income in US dollars (US
76
Census Bureau, 2022b). The Y-axis is the fraction xDc(100)1/pop, the 100-week covid-
period excess mortality by population, which is the “100-week covid-period fatality ratio”
for the USA population.
Figure 21. Excess mortality of the covid period normalized by population versus median
household income (MHI) in the USA. The axes are optimized for the dataset (Panel A) and for
the intercept between trend line and X-axis (Panel B). Each point is for one state of the USA.
B
A
77
The parameters of the least squares fitted linear trend line are given in Table 6. The color-code
of the 51 states is shown in section 2. Data were retrieved as described in Table 1. xDc(100)1 is
calculated from Equation 5.
Here, the Pearson correlation coefficient is r = −0.71 (“strong”) (Table 6). The graph
(Figure 21) suggests that a USA state with a MHI of approximately $130K or more
would have zero excess mortality integrated over the 100-week covid period. Likewise,
the states with smallest MHI attain a “100-week covid-period fatality ratio” of
approximately 0.005, or 0.5%, which is very large, since this is over and above non-
covid-induced mortality for such states.
Income (Figure 21) and poverty (Figure 20) are clearly determinative factors predicting
excess 100-week covid-period mortality in a state of the USA, occuring since a
pandemic was announced on 11 March 2020 by the WHO.
Figure 22 shows the scatter plot for obesity, defined as the prevalence of self-reported
obesity among U.S. adults (CDC, 2021). The Y-axis is the fraction xDc(100)1/pop, the
100-week covid-period excess mortality by population, which is the “100-week covid-
period fatality ratio” for the USA population.
78
Figure 22. Excess mortality of the covid period normalized by population versus obesity
in the USA. Each point is for one state of the USA. The parameters of the least squares fitted
linear trend line are given in Table 6. The color-code of the 51 states is shown in section 2. Data
were retrieved as described in Table 1. xDc(100)1 is calculated from Equation 5.
Here the positive correlation is “strong”, although less than for MHI, at r = +0.62 (Table
6). The least squares fitted straight line suggests that a USA state that would have an
obesity rate of approximately 7% or less would have zero excess 100-week covid-
period mortality. This implies that certain groups of obese residents do not contribute to
100-week covid-period excess mortality, presumably wealthy obese residents, for
example.
Figure 23 shows the scatter plot for the proportion of the population aged 65 years old
and over. The Y-axis is the fraction xDc(100)1/pop, the 100-week covid-period excess
mortality by population, which is the “100-week covid-period fatality ratio” for the USA
population.
Figure 23. Excess mortality of the covid period normalized by population versus the
proportion of people aged 65 and over in the USA. Each point is for one state of the USA.
The color-code of the 51 states is shown in section 2. Data were retrieved as described in Table
1. xDc(100)1 is calculated from Equation 5.
79
There is no significant correlation (r = +0.046, “very weak”, Table 6). This is also true for
the proportion of the population aged 75 years old and over (75+/pop) and for the
proportion of the population aged 85 years old and over (85+/pop) (data not shown).
Excess all-cause mortality of the 100-week covid period in the USA has no relation to
old age, on a state-wise basis.
This lack of correlation with age again shows that the excess mortality is not consistent
with having been caused by a viral respiratory disease, including COVID-19, since the
known infection fatality ratios are exponential with age (Elo et al., 2022; Sorensen et al.,
2022).
Other factors which we did not consider in our previous article (Rancourt, Baudin and
Mercier, 2021b) are Supplemental Security Income (SSI) and Social Security
Disability Insurance (SSDI). Those factors are state-provided benefits in case of
disability or blindness (SSA, 2020). They can be interpreted as indicators or proxies for
the proportion of frail populations in the USA. Whitaker (Whitaker, 2015) has interpreted
that the majority of SSI and SSDI recipients can be classified as mentally disabled and
receiving prescription psychiatric medication. He reports that some of these drugs are
definitely associated with obesity. See also a current report about the prevalence of
mental illness in the USA (NIMH, 2022).
Figure 24 shows the proportion of people receiving SSI versus the proportion of people
receiving SSDI by state.
80
Figure 24. SSI recipients normalized by population versus SSDI recipients normalized by
population in the USA. Each point is for one state of the USA. The color-code of the 51 states
is shown in section 2. Data were retrieved as described in Table 1.
Although SSI and SSDI are independent programs, they are positively correlated to
each other, showing that states that have more of one type of recipients also have more
of the other type of recipients. Also, the two programs are not mutually exclusive, as
some people called “concurrent” are eligible for both (SSA, 2020), and there is an
approximately 10% overlap (data not shown).
Figure 25 shows the scatter plot for SSI recipients by population (SSA, 2022a). The
Y-axis is the fraction xDc(100)1/pop, the 100-week covid-period excess mortality by
population, which is the “100-week covid-period fatality ratio” for the USA population.
81
Figure 25. Excess mortality of the covid period normalized by population versus SSI
recipients normalized by population in the USA. Each point is for one state of the USA. The
parameters of the least squares fitted linear trend line are given in Table 6. The color-code of
the 51 states is shown in section 2. Data were retrieved as described in Table 1. xDc(100)1 is
calculated from Equation 5.
Here, the Pearson correlation coefficient is r = +0.51 (“moderate”) (Table 6). The graph
(Figure 25) suggests that a USA state with a SSI/pop of zero would nonetheless have a
“100-week covid-period fatality ratio” of approximately 0.2%. This implies that the SSI
population cannot account for all the excess mortality in the 100-week covid period:
Other groups must also contribute to the said excess mortality.
Figure 26 shows the scatter plot for SSDI recipients by population, defined as the
number of all disabled SSDI beneficiaries aged 18-64 (SSA, 2022b). The Y-axis is the
fraction xDc(100)1/pop, the 100-week covid-period excess mortality by population,
which is the “100-week covid-period fatality ratio” for the USA population.
82
Figure 26. Excess mortality of the covid period normalized by population versus SSDI
recipients normalized by population in the USA. Each point is for one state of the USA. The
parameters of the least squares fitted linear trend line are given in Table 6. The color-code of
the 51 states is shown in section 2. Data were retrieved as described in Table 1. xDc(100)1 is
calculated from Equation 5.
Here, the Pearson correlation coefficient is r = +0.47 (“moderate”) (Table 6). The graph
(Figure 26) suggests that a USA state with a SSDI/pop of zero would nonetheless have
a “100-week covid-period fatality ratio” of approximately 0.2%. Like with the population
of SSI recipients (Figure 25), this implies that the SSDI population cannot account for all
the excess mortality in the 100-week covid period. Other groups must also contribute to
the said excess mortality.
Figure 27 shows the scatter plot for disability, defined as the percentage of Americans
living with a disability (Disabled World, 2020). Disability is defined as a long-lasting
sensory, physical, mental, or emotional condition or conditions that make it difficult for a
person to do functional or participatory activities such as seeing, hearing, walking,
climbing stairs, learning, remembering, concentrating, dressing, bathing, going outside
the home, or working at a job. The Y-axis is the fraction xDc(100)1/pop, the 100-week
83
covid-period excess mortality by population, which is the “100-week covid-period fatality
ratio” for the USA population.
Figure 27. Excess mortality of the covid period normalized by population versus
disability in the USA. Each point is for one state of the USA. The 8 apparent bottom outliers
are: Hawaii, Massachusetts, New Hampshire, Washington, Rhode Island, Vermont, Oregon,
and Maine. The color-code of the 51 states is shown in section 2. The parameters of the least
squares fitted linear trend line are given in Table 6. Data were retrieved as described in Table 1.
xDc(100)1 is calculated from Equation 5.
Here, the Pearson correlation coefficient is r = +0.59 (“moderate”) (Table 6). The graph
(Figure 27) suggests that a USA state with no one living with a disability would have a
near-zero “100-week covid-period fatality ratio” (estimated at 0.01%). This is similar to
the situation with poverty (Figure 20), in that if there were no disabled persons in the
USA, the excess ACM of the covid period would have been essentially zero.
Table 6 gives parameters of the correlations discussed above.
84
Factor (units)
Slope (units)
Intercept
Pearson
coefficient (r)
Strength
(Evans, 1996)
Poverty (%)
+0.0331 (per %)
-0.00008
+0.855
Very strong
MHI ($)
-6E-08 (per $)
+0.008
-0.706
Strong
Obesity (%)
+0.0152 (per %)
-0.0011
+0.618
Strong
65+/pop (%)
+0.0023 (per %)
+0.0034
+0.046
Negligible to
very weak
SSI/pop (%)
+0.069 (per %)
+0.0022
+0.512
Moderate
SSDI/pop (%)
+0.0546 (per %)
+0.0022
+0.466
Moderate
Disability (%)
+0.028 (per %)
+0.0001
+0.590
Moderate
Table 6. Parameters of the least squares fitted straight lines for xDc(100)1/pop (Y-axis)
versus Factor (X-axis), where xDc(100)1/pop is dimensionless. Here: xDc(100)1/pop =
Slope x Factor + Intercept.
In this article, we did not apply a strict separation between sections, which would
exclude any discussion of results in the Results section, in order to facilitate
appreciation for the often novel features of the data being presented.
In the next section, we continue, organize and supplement our discussion of the above
results.
85
4. Discussion
4.1. All-cause mortality in the covid period in the USA: Sudden onset and
heterogeneity by state
The covid period in the USA discontinuously starts immediately after the WHO’s
11 March 2020 declaration of a pandemic, and is a period exhibiting extraordinarily
large and time-wise (by week, by month, by season) anomalous ACM, compared to the
historic record since at least 1999 (Figure 1). The sudden discontinuity is synchronous
everywhere that it occurs, and its occurrence (presence and magnitude) is highly
heterogeneous across state, provincial, regional and national jurisdictions, in North
America and Europe, where the best ACM by time data is available (Rancourt, 2020;
Rancourt, Baudin and Mercier, 2020, 2021a, 2021b; Johnson and Rancourt, 2022).
Such a large discontinuity, into a qualitatively different long-term (2-year) regime of
ACM behaviour, has previously not been observed in epidemiology, so clearly. The
break occurs between two regimes of ACM, between two distinct types of mortality
behaviours by time, by age group and in terms of heterogeneity by jurisdiction, and it
occurs at or near the date (11 March 2020) of the WHO’s declaration of a pandemic;
which is the date at which hospital, care-home and public health protocols were
discontinuously, somewhat permanently and broadly changed, while lockdowns
(jurisdiction-wide shelter-in-place or stay-at-home orders) were often and
heterogeneously (by state) applied soon after this same date (Johnson and Rancourt,
2022), accompanied by massive restructuring of local economic activity.
Rancourt (Rancourt, 2020) seems to have been the first to point out this discontinuity in
ACM by time and to have associated it to the measures installed on or near 11 March
2020, rather than to a pandemic spread of a contagious disease. We have discussed
this break in detail previously, following Rancourt, and further associated it with the
imposed structural changes in the society and the economy (Rancourt, Baudin and
Mercier, 2020, 2021a, 2021b).
86
The heterogeneity by jurisdiction of the ACM by time behaviour following the said
discontinuity is a striking phenomenon compared to remarkably uniform behaviour of
ACM by time across jurisdictions, indeed across continents (at mid-latitudes), in pre-
covid time (before 11 March 2020) (Rancourt, Baudin and Mercier, 2020). One has to
go back to 1918 to observe a possibly similar phenomenon, at a time when less data
was available (Rancourt, Baudin and Mercier, 2021b). In the USA, there are particularly
large state-to-state differences in ACM by time behaviour during the covid period,
compared to very similar state-to-state behaviour in pre-covid time (Rancourt, Baudin
and Mercier, 2021b). For example, Johnson and Rancourt (Johnson and Rancourt,
2022) find covid-period health-status-adjusted integrated ACM per capita to vary by
approximately 20% from state to state for the covid period, while a state-to-state
variation of only approximately 2% occurs for corresponding integration windows prior to
11 March 2020 (their Figure 7).
The USA state-wise heterogeneity in ACM behaviour is a further demonstration of the
abrupt change in ACM regime that occurred on or near 11 March 2020. Given the
complexities of the comparative behaviours between states, there is no substitute for
showing the all-ages data for each of the states. This is done for the ACM/w data in
Appendix A.
We previously showed that in the USA the ACM by time and by state jurisdiction in the
covid period is contrary to the expected behaviour for a viral respiratory disease
pandemic, and that the extra deaths, when and where they occur in the USA, were
likely due to the government and medical responses, including constructive denial of
treatment of an unprecedented bacterial pneumonia epidemic that predominantly
affected poor and obese individuals living in hot-climate states (Rancourt, Baudin and
Mercier, 2021b).
More specifically, we proposed the following interpretive scheme:
The covid response and measures created stressful socio-economic, regulatory
and institutional conditions. For example, studies report increased unemployment
87
and worsening mental health (Czeisler et al., 2020; Jewell et al., 2020; Giuntella
et al., 2021). This would result in chronic psychological stress in many
individuals, during the covid period.
As we have discussed and reviewed previously (Rancourt, Baudin and Mercier,
2021a), chronic stress debilitates the immune system and is arguably the
dominant determinant of individual health (Cohen, Tyrrell and Smith, 1991; Ader
and Cohen, 1993; Cohen et al., 1997; Sapolsky, 2005; Cohen, Janicki-Deverts
and Miller, 2007; Dhabhar, 2014; Prenderville et al., 2015). Furthermore, the
molecular and physiological mechanisms for suppression of the immune system
by experienced chronic stress are being elucidated more and more (Devi et al.,
2021; Udit, Blake and Chiu, 2022).
In terms of assigning actual cause of death for covid-period excess mortality in
the USA, we argued that bacterial pneumonia was a likely candidate, attacking
vulnerable groups subjected to debilitating stress, during a massive pneumonia
epidemic evident in the CDC data, combined with a dramatic drop in antibiotic
prescriptions (Rancourt, Baudin and Mercier, 2021b). The said pneumonia
epidemic is also seen, directly or indirectly, in other studies (Di Gennaro et al.,
2021; Bradley et al., 2022).
Further studies have since established a sustained drop in antibiotic prescriptions
(e.g., (Buehrle et al., 2021; King et al., 2021; Kitano et al., 2021; Van Laethem et
al., 2021, 2022; Gisselsson-Solen and Hermansson, 2022; Givon-Lavi et al.,
2022; Gottesman et al., 2022; Knight et al., 2022; Winglee et al., 2022).
Those conclusions are supported by the present study, which has the added benefits of:
month-wise time-resolved ACM by age group and by sex back to 1999;
more recent consolidated week-wise time-resolved ACM, up to and including
week-5 of 2022;
closer examination by age group; and
cumulative vaccine dose delivery data time-resolved by week, by injection series
or status, by age group and by state.
88
In particular, the correlations between xDc(100)1/pop (the 100-week covid-period
excess mortality by population, which is the “100-week covid-period fatality ratio” for the
USA population) and poverty (Figure 20), median household income (MHI, Figure 21),
obesity (Figure 22), SSI/pop (SSI recipients per population) (Figure 25), SSDI/pop
(SSDI recipients per population) (Figure 26), disability (Figure 27), and the absence of
significant correlations with population fractions of elderly residents (Figure 23, and
above discussion) (Table 6), provide compelling support for the said conclusions. For
example, the absence of significant correlations with population fractions of elderly
residents (65+, 75+, or 85+ years) is incompatible with the reported exponential age-
dependence of the COVID-19 infection fatality ratio (Elo et al., 2022; Sorensen et al.,
2022), and contrary to all the studies finding that the dominant factors are age and age-
associated comorbidities for viral respiratory diseases, including COVID-19. Whereas,
no known respiratory-disease virus specifically targets residents living in poverty (Figure
20), irrespective of age (Figure 23).
4.2. Late-summer-2021 anomalous mortality of young adults
The time-structure of the all-cause mortality by month (ACM/m) from 2000, into the
covid period, by age group is shown in Figure 4. Here, the relative magnitude of the
covid-period excess mortality above the historic trend is particularly large for the age
groups 25-34y (Figure 4C), 35-44y (Figure 4D), and 45-54y (Figure 4E).
See also Figure 7, Figure 9, Figure 10, Figure 11, Figure 13, Figure 15, Figure 17 and
Figure 19. Basically, we observe the same age-group-differential and seasonal-
differential ACM by time phenomena with higher time resolution and in more detail in
the all-cause mortality by week (ACM/w).
Similarly with the ACM/m (Figure 4) data, and as is evident from Table 3, the ACM/w
(Figure 7) data also shows that the relative magnitude of the covid-period extra deaths
above the historic trend is particularly large for the age group 25-44y (Figure 7B), and to
a lesser degree 45-64y (Figure 7C), especially the late-summer-2021 feature (smp2).
89
These covid-period young-adult age group large excesses in ACM by time, especially in
the late-summer-2021 (smp2) feature, are a central feature of mortality during the covid
period in the USA.
The age-group-dependent relative magnitude of the covid-period excess mortality is
contrary to the age dependence of mortality for viral respiratory diseases, including that
reported for COVID-19, in which mortality strongly increases exponentially or near-
exponentially with age (Elo et al., 2022; Sorensen et al., 2022).
These results are contrary to, incompatible with, and irreconcilable with an interpretation
in which excess mortality (by age-group) in the covid period in the USA is mostly or
predominantly caused by COVID-19; or any known viral respiratory disease (see
Rancourt et al.’s discussion about the 1918 declared pandemic (Rancourt, Baudin and
Mercier, 2021b), and references therein). Either one must admit that the declared
COVID-19 pandemic is not the main cause of death to explain the excess mortality
data, or ignore the well-established data showing that COVID-19-assigned mortality
increases exponentially or near-exponentially with age, and that young people
essentially (comparatively) do not die from COVID-19, as the primary assigned cause of
death in a controlled clinical and laboratory verified setting.
Furthermore, relative mortality is particularly large for the late-summer-2021 feature
(smp2) in the 35-44y age group (Figure 4D), compared to any other time in the covid
period, and more so than with any other age group. This feature (large smp2 in the 35-
44y age group), however, is highly variable from state to state, being prominent or very
prominent in states such as Texas, Florida, Georgia, North Carolina, South Carolina,
Alabama, Arkansas, Hawaii, Idaho, Kentucky, Louisiana, Mississippi, Missouri, Nevada,
Oklahoma, Oregon, Tennessee, Washington, Virginia, West Virginia and Wyoming,
while being absent in New York and New Jersey, intermediate in California, and mostly
intermediate or absent in other states, while Michigan uniquely has a spring-2021 peak
in mortality for that age group centered in April (Figure 11). The latter observations are
confirmed in the ACM/w data for 25-64y age group (not shown). Generally, the 2020
90
and 2021 summers were most deadly in the Southern states, as previously described
(Rancourt, Baudin and Mercier, 2021b). Such state-to-state heterogeneity is
inconsistent with the pandemic paradigm of rapid spread, extensive coverage and
complete immune susceptibility. It is more understandable in terms of the driving forces
described above.
Coming back to age-groups: Why would this be? Why would mortality in this young-
adult age group suddenly spike in late-summer-2021, in many states and as seen on
the basis of the whole USA, to an unprecedented large value, after 18 months of the
declared pandemic, compared to anything in the earlier covid period or the last 20
years, approximately doubling all-cause mortality for several months for 35-44 year olds
(Figure 4D), both male and female (not shown)? See also Figure 7, Figure 9, Figure 10,
Figure 11, Figure 13, Figure 15, Figure 17 and Figure 19.
In attempting to answer this question (Why are young adults dying more than ever in the
second half of the covid period, and in the late-summer-2021 ACM peak in particular?),
we submit that the answer is probably not “variants of concern”, or any such theoretical
proposal from immunology. Instead, we describe two preferred hypotheses to explain
the observation:
i. The first is that young adults are more resilient than old adults against the
cumulative impact of persistent covid-period conditions that cause chronic
psychological stress that, in vulnerable groups, causes emergent or
worsening immunodeficiency that enables death by bacterial pneumonia. In
support of this hypothesis, one of the largest vulnerable groups in the USA
those afflicted by serious mental illness (5.6% adults = 14.2 million aged 18+,
in 2020) has a heavily skewed prevalence towards young adults (see
below).
ii. The second is that the vaccination campaign, including the “vaccine equity”
campaigns, captured many thus made immunocompromised young adults
from vulnerable groups and that the vaccine challenge against many of these
individuals constituted a significant comorbidity, which was absent in the first
91
(pre-vaccination) half of the covid period, thus increasing the death toll of
young adults, overall, in the second (vaccination) half of the covid period.
Note that the second hypothesis (vaccine toxicity) relies on the conditions described in
the first hypothesis (cumulative stress-induced immunodeficiency). This is because the
vaccine toxicity for subjects who are not immunocompromised (fatality risk per dose,
inferred from VAERS data) is too small to quantitatively explain the observed ACM
increases that are synchronous with increases in vaccine-delivery (administered doses),
assuming avoidance of immunocompromised subjects (see above, and below).
As mentioned above, we do not believe that any “variant of concern” (CDC, 2022e)
emerging in 2021 could produce such a result in the mortality data, or that the
explanation is viral. Rather, we prefer to propose that the same forces that appear to
generally determine the exceptionally large excess mortality in the covid period in the
USA namely the impact on the immune systems of individuals in populations of those
most vulnerable to psychological stress and social isolation during life-changing covid-
period circumstances, combined with an essentially untreated mass bacterial-
pneumonia epidemic (Rancourt, Baudin and Mercier, 2021b) also largely determined
the jurisdictional, age and time structures of excess mortality in the covid period, on the
background of the demographics of highly vulnerable groups. Here, the hypothesis is
that, while the “conditions = stress = immune-vulnerability = death from pneumonia”
scenario existed from the very start of the covid period (Rancourt, Baudin and Mercier,
2021b), the prolonged conditions and associated chronic stress eventually has more
relative impact on young adults that are more resilient at first, thus changing the age
structure of mortality as the covid period advances.
We expect, therefore, that the change in age structure of mortality during the course of
the covid period (Figure 15, Figure 17) is driven by such factors as a dry tinder effect
among the elderly and differential youth resilience to chronic stressors (relative
endurance over long periods), on the background of the demographics of highly
vulnerable groups, rather than driven by the vaccination campaign via general-
92
population vaccine toxicity (fatality risk per dose for non-immunocompromised subjects)
acting alone and irrespective of these circumstances.
There is also a cumulative effect on young adults, which is irreversible to some extent
(Giuntella et al., 2021). Our interpretation of Figure 15 is consistent with the fact that the
hardships (expenses, housing and food insecurity) are sustained in the USA during the
covid period (CBPP, 2022). In the words of the OECD (OECD, 2022):
“The COVID-19 pandemic has triggered one of the worst jobs crises since the
Great Depression. There is a real danger that the crisis will increase poverty
and widen inequalities, with the impact felt for years to come.”
Also, socio-economic factors may have caused young adults to have higher
experienced stress in the second half of the covid period, compared to older adults. For
example, pressures inducing bankruptcies and associated losses of livelihood and
personal identity would increase as the restrictive conditions persist in many sectors,
although analysis of the macroeconomic data is complex (Martos-Vila and Shi, 2022).
It is also possible that the age-structure change phenomenon partly results from or is
significantly contributed to by vaccine-campaign (including so-called “vaccine equity”
campaigns) capture of vulnerable young adults made immunocompromised by the said
chronic psychological stress.
Both of the latter hypotheses (relative resilience to stress of young adults in vulnerable
groups and vaccine capture of young adults made immunocompromised by chronic
stress) advanced to explain the increased skewness of mortality towards young adults
in the vaccination period (second half) of the covid period are consistent with the fact
that the prevalence of serious mental illness is large and highly skewed towards young
adults in the USA (NIMH, 2022). Indeed, the age distribution in covid-period fatality risk
that we observe, which is skewed towards young adults in the vaccination period of the
covid period (Figure 17), should be put in the context of the prevalence of serious
mental illness, which was 14.2 million adults aged 18 or older in the USA in 2020,
93
representing 5.6% of all USA adults, and which is highly skewed towards young adults
(NIMH, 2022):
Figure 28. Prevalence of serious mental illness among U.S. adults in 2020. Data are shown
for the entire USA (Overall), by sex (Female, Male), by age (18-25, 26-49, 50+) and by
race/ethnicity (Hispanic, White, Black or African American, Asian, Native Hawaiian/Other Pacific
Islander, American Indian/Alaskan Native, Two or more races). Serious mental illness is defined
as a mental, behavioral, or emotional disorder resulting in serious functional impairment, which
substantially interferes with or limits one or more major life activities. This figure is from NIMH
(NIMH, 2022).
Basically, any model of excess mortality, which relies on the mentally disabled as a
source group and serious mental illness as a major cofactor, will be biased towards
mortality risk that is skewed towards young adults. It is well established that in the USA
younger people are disproportionately affected by diagnosed mental disorders
(Merikangas et al., 2010).
We advance that the tragic excess deaths of 35-44 year olds (and 25-64 year olds) in
late-summer-2021 in the USA, extraordinarily exhibited as an actual peak (smp2) in
94
ACM by time, for example, needs to be explained by specific health-status and socio-
psycho-economic circumstances in the different jurisdictions, and not solely in terms of
theoretical proposals from virology and immunology (e.g., “variants of concern”, etc.).
The needed actual community-level field work is not being sufficiently funded or
undertaken, to our knowledge.
4.3. Vaccination campaign
The time-resolved and age-group resolved vaccination campaign, together with the
similarly resolved ACM/w show that the vaccination campaign did not reduce mortality
during the covid period (Figure 10; Figure 11; Figure 12; Figure 13; Figure 14; Figure
16; Figure 17; Table 4; Table 5).
We conclude with a high degree of certainty that the COVID-19 vaccination campaign in
the USA was ineffective in reducing all-cause mortality. The mass vaccination campaign
was not justified in terms of reducing excess all-cause mortality. The large excess
mortality of the covid period, far above the historic trend, was maintained irrespective of
the unprecedented vaccination campaign.
Furthermore, the vaccination campaign may have affected the age structure of ACM by
contributing to the deaths of young adults in vulnerable groups but the same dominant
forces that caused the large excess ACM in the first (50-week, pre-vaccination) half of
the covid period appear to have continued to cause the large excess ACM in the second
(50-week, vaccination) half of the covid period.
In the ACM/w data (Figure 7), similarly to the ACM/m data (Figure 4), relative mortality
is particularly large for the late-summer-2021 feature (smp2) in the 25-44y age group
(Figure 7B), compared to any other time in the covid period, and more so than with any
other age group. It is also anomalously large, to a lesser degree, for the age group
45-64y (Figure 7C).
95
This feature in ACM by time during the covid period (an exceptionally large late-
summer-2021 peak, smp2) is highly variable from state to state (see above, Figure 11,
and see Appendix A), and occurs after the majority of the main-series vaccination
campaign has been mostly completed.
Nonetheless, is some or most of the exceptional mortality occurring in the late-summer-
2021 period (smp2) consistent with having been caused by the vaccination campaign?
Likewise, is any feature in ACM by time consistent with having been caused by the
vaccination campaign?
Figure 10 allows a direct comparison, on the same time axis, of all-cause mortality by
week and cumulative number of vaccinated individuals, by vaccine sequence (1st dose,
fully vaccinated, booster), for separate age groups. Figure 11 allows the same for
specific states.
A study of the Vaccine Adverse Event Reporting System (VAERS) data of the USA has
shown that the deaths associated with the COVID-19 vaccine in the USA typically occur
first in a large initial peak within 5 days or less following the injection; followed (~5 days
to ~60 days post injection) by a shoulder of exponential decay in deaths, with a fitted
half-life decay time typically in the range 13-30 days (Hickey and Rancourt, 2022; their
figures S3 through S5).
This means that deaths associated with the injections in the USA occur essentially
immediately following delivery of the injection (mostly within days, with a decaying
residual risk of fatality lasting weeks).
In addition, it is usually postulated that the alleged life-saving benefits of the vaccine
become operative 7-14 days from the time of injection, and should last several months,
similarly to the 90 days or so of efficacy claimed for flu vaccines (Rambhia and
Rambhia, 2019).
96
In this way, any measurable positive or negative impact of the vaccination campaign on
death rate (all-cause mortality by week) should be temporally associated with times of
large or maximum slope in cumulative vaccine dose delivery (or vaccinated status
acquisition), if vaccine fatality toxicity is large enough (deleterious impact) or vaccine
protection against death is large enough (positive impact).
An increase in mortality from the vaccination campaign would be seen within 5 days or
so of a large slope in cumulative vaccine dose delivery, whereas a smaller mortality
would be seen to follow a large slope in cumulative vaccine dose delivery (or vaccinated
status acquisition) by a few weeks or more and should be persistent after having
attained significant vaccine dose coverage.
As discussed above in presenting Figure 10 and Figure 11, there is a modest but
significant stepwise increase in 1st-dose vaccine delivery (administration), which is
synchronous with the late-summer-2021 peak in ACM, visible for all ages and for the
25-44 and 45-64 years age groups (Figure 10A, C, D). This temporal association is
prominent in the data for many specific states (e.g., Figure 11), and cannot be
dismissed as noise.
We estimate that, in order to achieve quantitative agreement between the outcome
(late-summer-2021 peak integrated excess mortality) and factor (additional vaccine
doses over the period of occurrence of the peak), the vaccine adverse-effect fatality
toxicity per dose would need to be approximately 100 times the estimated non-
immunocompromised fatality toxicity per dose (Hickey and Rancourt, 2022; their
Table 1), assuming that only non-immunocompromised resident were injected. There
are many more doses administered than deaths. This means that if
immunocompromised residents from vulnerable groups were captured in the
vaccination doses delivered in the relevant period, then it is possible that the late-
summer-2021 mortality peak is entirely or partly due to vaccine challenge of vulnerable
young adults.
97
In this regard, it is relevant that the so-called “vaccine equity” campaigns in the USA
were operating in the relevant period:
A JAMA Editorial of 29 January 2021, entitled “Vaccine Distribution—Equity Left
Behind?” recommended, among other things “1. Prioritize vaccine distribution to
zip codes that have been most severely affected by COVID-19 and that have
high indexes of economic hardship”, and so on (Jean-Jacques and Bauchner,
2021).
The New York Times provided extended reporting on county-wise vaccine
coverage (The New York Times, 2022).
Large foundations such as the Rand Corporation were significantly involved
supporting “vaccine equity” programs (Faherty et al., 2022).
Louisiana, for example, had fully launched its comprehensive “vaccine equity”
program, as did virtually all states to varying degrees (Louisiana Launches
Grassroots COVID Vaccine Campaign to Ensure No Community Gets Left
Behind | Office of Governor John Bel Edwards, 2021).
Similarly, as discussed above in introducing Figure 11, Michigan has a unique feature in
its ACM by time data, not seen for any other state. Michigan has a unique April-2021-
centered spring-2021 peak in ACM for young adults, which coincides with the large
main onset of the vaccination campaign for these ages (Figure 11G, H). In this case, in
order to achieve quantitative agreement between the outcome (spring-2021 peak
integrated excess mortality) and factor (additional vaccine doses over the period of
occurrence of the peak), the vaccine adverse-effect fatality toxicity per dose would need
to be approximately 10 times the estimated non-immunocompromised fatality toxicity
per dose (Hickey and Rancourt, 2022; their Table 1), assuming that only non-
immunocompromised resident were injected. There are many more doses administered
than deaths. This means that if immunocompromised residents of Michigan from
vulnerable groups were captured in the vaccination doses delivered in the relevant
period, then it is possible that the unique spring-2021 mortality peak for Michigan is
entirely due to vaccine challenge of vulnerable young adults. We consider this to be the
most likely hypothesis we can make, with the available information, to explain the
98
unique spring-2021 mortality peak for Michigan. If the hypothesis is correct, then this
demonstrates the principle that vaccine challenge of residents made
immunocompromised by chronic stress can explain large features in ACM by time, in
the covid-period and vaccine-campaign circumstances.
4.4. Looking ahead
Unavoidable questions are: “When will the covid period end?” and “Will ACM by time
and by jurisdiction return to the pre-covid-period normal?”
We quickly looked at the latest ACM/w data for the USA, which appears to be reliable
through to April-2022, in order to give tentative answers, looking forward.
The data (shown in Appendix C) has the 2021-2022 winter peak in ACM dropping
precipitously in February-2022, down to a level, in March and April 2022, which is
typical of pre-covid-period summer baseline values. Such a low value did not occur at
any time in the USA in the covid period that we studied in the present article.
It would seem that, in terms of all-cause mortality, the covid period ended, at least
momentarily, in March and April 2022. It will be interesting to see whether there will be a
summer-2022 peak in ACM when more data becomes available.
Late reporting of mortality to the CDC could alter the above tentative observation.
5. Conclusion
Our results show the following overall large-scale features:
All-cause mortality by time in the USA is heterogeneous by state and persistently
far in excess of the recent historic decadal trend, starting immediately when a
pandemic was declared by the WHO on 11 March 2020, and continuing
99
throughout the entire covid period that we examined, up to the week ending on
February 5, 2022 with a total of 1.27M excess deaths (Figure 1, Figure 3,
Figure 5, Figure 6; Table 2, Table 3).
Throughout the covid period, all-cause mortality is heterogeneous by state and
anomalous in its time (by week, by month) and seasonal variations, compared to
historic behaviour. The anomalies include winter and summer peaks, which are
highly variable in magnitude from year to year in the covid period, and from state
to state (Figure 8, Figure 9; Appendix A); as we observed previously (Rancourt,
Baudin and Mercier, 2021b). The broad summer peaks” of ACM by time in 2020
and 2021 are of a nature that has not previously been observed in mortality data
for the USA or any country, historically, since quality data has been available for
more than 100 years. The anomalous heterogeneity by state in integrated
mortality over the covid period was recently demonstrated by Johnson and
Rancourt (Johnson and Rancourt, 2022; their Figure 7).
Unlike for viral respiratory diseases, including the presumed SARS-CoV-2 virus
itself (Elo et al., 2022; Sorensen et al., 2022), the covid-period excess mortality
risk by age group is not predominantly confined to the elderly population; and the
inferred age-group-specific infection fatality ratios are not exponential or near-
exponential with age, as they would be (Table 2, Table 3, Figure 4, Figure 7,
Figure 9). On the contrary, overall for the covid period, mortality risk is broadly
distributed to all age groups and is significantly larger for younger adults
compared to the eldest adults (Figure 9). The non-exponential-with-age (more
age-uniform) distribution of mortality risk to all age groups holds for both the first
half (pre-vaccination) and second half (vaccination) of the 100-week covid period
(Figure 10, Figure 13, Figure 15, Figure 17, Table 4). The observed age-group
distribution of all-cause mortality risk constitutes proof that the covid-period
excess mortality cannot predominantly be due to the presumed SARS-CoV-2
virus or to any viral respiratory disease. The alternative would be to abandon the
accepted body of research on mortality risk by age.
100
Instead of the covid-period excess all-cause mortality risk being predominantly
(or even moderately) determined by age of the population, the state-wise
integrated excess all-cause mortality for the entire 100-week covid period
normalized by state population (outcome) is correlated to socio-economic factors
that are macro-indicators of state-wise resident vulnerability (Table 6):
o The covid-period excess all-cause mortality risk is very strongly correlated
to poverty (r = +0.86) (Figure 20).
o The said mortality risk is strongly correlated to MHI (Median Household
Income) (r = -0.71) (Figure 21).
o The said mortality risk is strongly correlated to obesity (r = +0.62) (Figure
22).
o The said mortality risk is not correlated simply to age of the population.
This is shown for 65+ ages in Figure 23, and is maintained for 75+ and
85+ ages (not shown).
o The said mortality risk is moderately correlated to the number of SSI
(Supplemental Security Income) recipients by population (r = +0.51)
(Figure 25).
o The said mortality risk is moderately correlated to the number of SSDI
(Social Security Disability Insurance) recipients by population (r = +0.47)
(Figure 26).
o Whitaker (Whitaker, 2015) has interpreted that the majority of SSI and
SSDI recipients can be classified as mentally disabled and receiving
prescription psychiatric medication.
o The said mortality risk is moderately correlated to disability (r = +0.59)
(Figure 27).
Despite the fact that there are significant changes in age structure of ACM by
time during the course of the covid period, the overall qualitative behaviour of
ACM by time (anomalously large excess mortality and presence of anomalous
summer and winter seasonal variations in ACM by time) and the 50-week-
integrated excess ACM are not substantially different in the first half of the
101
100-week covid period (first 50 weeks of the covid period), in which there was
essentially no vaccination campaign, and in the second half of the 100-week
covid period (second 50 weeks of the covid period), in which most of the
vaccination campaign was accomplished (Figure 4, Figure 7, Figure 13 , Figure
15, Figure 17, Figure 19). Therefore, the suddenly applied and massive
vaccination campaign (Figure 10) did not induce a large change of regime from
one type of ACM by time to another, on the scale of the dramatic change in
regime from pre-covid to covid period.
Regarding mortality averted by vaccination, the COVID-19 vaccination campaign
in the USA did not cause any seasonally unambiguous temporally associated
decrease in all-cause mortality, for all ages or in any age group (Figure 10; see
also Figure 11). The vaccination campaign did not measurably cause any deaths
to be averted. This is contrary to the notion that the vaccines are “effective” in
reducing “serious illness” (and presumably death), becoming operative 7-14 days
following the time of injection, with the protection presumably lasting at least
several months.
Therefore, although much messaging attention is directed towards life-saving
consequences arising from the mass vaccination campaign in the USA, clearly
such effects are both undetectable in all-cause mortality and necessarily small
compared to the overwhelming harm from the extraordinary covid-period
conditions themselves.
Conversely, regarding vaccine-induced mortality, the COVID-19 vaccination
campaign in the USA did not cause the 50-week-integrated excess ACM in the
second half of the 100-week covid period (second 50 weeks of the covid period),
in which most of the vaccination campaign was accomplished, to be
systematically larger (systematically across all age groups, or all states) than in
the first half of the 100-week covid period (first 50 weeks of the covid period), in
which there was essentially no vaccination campaign. The persistent socio-
economic, regulatory, institutional… changes associated with the covid period
102
(relative to pre-covid behaviour) had a large effect compared to changes
associated specifically with the period of the vaccination campaign, positive or
negative (Figure 4, Figure 7, Figure 13 , Figure 15, Figure 17, Figure 19).
Despite the fact that there is no large systematic effect of the vaccination
campaign on either 50-week-integrated mortality or main qualitative features of
ACM by time, positive or negative, we nonetheless detect significant seasonally
unambiguous local temporal associations between increases in number of
vaccinated residents and synchronous increases in all-cause mortality, for certain
age groups, and most prominently in certain states:
o The largest of these local temporal associations is seen in the data for the
whole USA and all age groups, as an accelerated increase in cumulative
number of residents having received at least one dose (or being fully
vaccinated), which is synchronous with the late-summer-2021 surge in
ACM by time (Figure 10A).
o The said local temporal association is most evident for the 25-44 years
age group (Figure 10C), also prominent for the 45-64 years age group
(Figure 10D), and discernible for the 65-74 years age group (Figure 10E).
o The said local temporal association is most prominent for the 25-64 years
age group in Southern states which typically have the smallest
vaccination rates including: Florida, Georgia, Louisiana, Mississippi and
Alabama (Figure 11).
o The special case of Michigan is also noteworthy (Figure 11G, H), as
discussed above.
The latter observations lead us to conclude that the large changes in age structure of
ACM by time (first half versus second half of the covid period) (esp. Figure 17) may be
partly (see Discussion section) or largely due to aggressive “vaccine equity” campaigns
that captured immunocompromised young adults in Southern states, thus causing
disproportionate mortality among vulnerable young adults in late-summer-2021.
103
The entire picture of mortality during the covid period in the USA, which included
implementation of the vaccination campaign after the first 50 weeks or so, can be
modelled as:
covid-period socio-economic, regulatory, institutional… conditions
psychological stress / social isolation
severely suppressed immune system in most vulnerable residents
(+ vaccine assault of thus immunocompromised vulnerable residents)
mortality from untreated bacterial pneumonia (+ vaccine-
assault comorbidity) in most vulnerable residents
The model arises as follows.
We infer from the temporal and jurisdictional characteristics of age-group-
resolved excess ACM that large structural changes in the living and care
conditions of residents of the USA directly enacted by state and institutional
players (including employers) during the covid period and including secondary
consequences of the said directly enacted changes are causally associated
with the large and sustained excess mortality in the covid period.
We infer from correlations with socio-economic factors that severe harm and
death were induced by the said covid-period changes in particular classes of
residents, such as isolated, sick, disabled, dependent, obese, poor, seriously
mentally ill or elderly individuals; and of course residents who are co-afflicted by
such conditions.
104
We postulate that the mechanistic connection between the said covid-period
changes and high risk of all-cause death in vulnerable residents is the well-
established link between experienced psychological stress and social isolation
(factor) and suppressed immunity, ill-health and death (outcome).
We postulate that the end-point mechanistic cause of death in the thus
immunocompromised vulnerable groups is bacterial pneumonia, in the midst of a
recorded mass epidemic of bacterial pneumonia, at a time when antibiotic
prescription rates showed an unprecedented decrease, in addition to aggressive
vaccine challenge (“vaccine equity” programs) in late-summer-2021.
The model is developed and contextualized in more detail in the Results and Discussion
sections. It provides a plausible and consistent explanation for all the aspects of the
ACM data for the USA, including the large change in age structure of the ACM on
entering the vaccination-campaign part of the covid period.
The model is predictive in that any type of comparable sudden socio-economic
upheaval, such as war or a Great Depression, in societies with large pools of vulnerable
residents, would give rise to this kind of large and rapid increase of mortality, targeting
the most vulnerable, with bacterial pneumonia playing a major role. We have previously
advanced that 1918 was such an episode in mid-latitude nations (Rancourt, Baudin and
Mercier, 2021b).
In conclusion, in terms of all-cause mortality, the covid-period socio-economic,
regulatory, institutional… conditions in the USA (from 11 March 2020 to week-5 of 2022)
were in-effect a large-scale deadly assault against vulnerable groups, which killed
approximately 1.27M members of the said groups. The temporal, jurisdictional and age-
group characteristics of the mortality are incompatible with the excess mortality having
been primarily caused by the presumed SARS-CoV-2 viral respiratory disease virus. In
the absence of poverty or if the covid-period socio-economic, regulatory, institutional…
conditions had not been imposed, there most probably would not have been excess
mortality in the USA, which was essentially the case in neighbouring Canada (Rancourt,
105
Baudin and Mercier, 2021b; their Section 4). The COVID-19 vaccination campaign,
accomplished in the second half of the covid period, did not avert any deaths, and may
have been a significant contributing factor causing excess mortality in vulnerable-group
young adults during late-summer-2021.
In regard to the fundamental results of this study, we would recommend a transparent
and accountable large-scale state, county and community level independent forensic
investigation of the deaths, excluding the involvement of interested government
agencies and private corporations. The mandate should include broad systemic
considerations, in addition to specific circumstances, and the investigators should have
the necessary powers and resources consistent with the magnitude and extent of the
catastrophe, in the hope of preventing any similar public health disaster in the future.
106
References
Data References
CDC (2021) “Adult Obesity Prevalence Maps | Overall Obesity: Prevalence of Self-
Reported Obesity Among U.S. Adults by State and Territory, BRFFS, 2020”. Page last
reviewed: May 16, 2022 (accessed September 24, 2021).
https://www.cdc.gov/obesity/data/prevalence-maps.html#states
CDC (2022a) “National Center for Health Statistics | Mortality Data on CDC WONDER”.
Page last reviewed: January 6, 2022. (accessed on May 18, 2022)
https://wonder.cdc.gov/mcd.html
CDC (2022b) “Weekly Counts of Deaths by Jurisdiction and Age”. Page updated: May
25, 2022. (accessed on May 26, 2022)
https://data.cdc.gov/NCHS/Weekly-Counts-of-Deaths-by-Jurisdiction-and-Age/y5bj-
9g5w
CDC (2022c) “COVID-19 Vaccination Demographics in the United States, National”.
Page updated: June 12, 2022. (accessed on May 5, 2022)
https://data.cdc.gov/Vaccinations/COVID-19-Vaccination-Demographics-in-the-United-
St/km4m-vcsb
CDC (2022d) “COVID-19 Vaccinations in the United States, Jurisdiction”. Page
updated: July 29, 2022. (accessed on April 25, 2022)
https://data.cdc.gov/Vaccinations/COVID-19-Vaccinations-in-the-United-States-
Jurisdi/unsk-b7fc
Disabled World (2020) “U.S. Disability Statistics by State, County, City and Age”.
Published: May 30, 2017. Updated: February 20, 2020. (accessed on July 27, 2022)
www.disabled-world.com/disability/statistics/scc.php
SSA (2022a) SSI Recipients by State and County, 2020”. Released: December 2021.
(accessed on June 24, 2022) https://www.ssa.gov/policy/docs/statcomps/ssi_sc/
SSA (2022b) Annual Statistical Report on the Social Security Disability Insurance
Program, 2020”. Released: November 2021. (accessed on July 6, 2022)
https://www.ssa.gov/policy/docs/statcomps/di_asr/index.html
US Census Bureau (2021) “State Population by Characteristics: 2010-2020”. Page
updated: October 8, 2021. (accessed on September 24, 2021)
https://www.census.gov/programs-surveys/popest/technical-
documentation/research/evaluation-estimates/2020-evaluation-estimates/2010s-state-
detail.html
107
US Census Bureau (2022b) “Small Area Income and Poverty Estimates (SAIPE) State
and County Estimates for 2020”. Page last revised: December 16, 2021. (accessed on
July 6, 2022). https://www.census.gov/data/datasets/2020/demo/saipe/2020-state-and-
county.html
Main References
Achilleos, S. et al. (2021) ‘Excess all-cause mortality and COVID-19-related mortality: a
temporal analysis in 22 countries, from January until August 2020’, International Journal
of Epidemiology, p. dyab123. Available at: https://doi.org/10.1093/ije/dyab123.
Ackley, C.A. et al. (2022) ‘County-level estimates of excess mortality associated with
COVID-19 in the United States’, SSM - Population Health, 17, p. 101021. Available at:
https://doi.org/10.1016/j.ssmph.2021.101021.
Ader, R. and Cohen, N. (1993) ‘Psychoneuroimmunology: conditioning and stress’,
Annual Review of Psychology, 44, pp. 5385. Available at:
https://doi.org/10.1146/annurev.ps.44.020193.000413.
Bradley, J. et al. (2022) ‘Pneumonia Severity Index and CURB-65 Score Are Good
Predictors of Mortality in Hospitalized Patients With SARS-CoV-2 Community-Acquired
Pneumonia’, CHEST, 161(4), pp. 927936. Available at:
https://doi.org/10.1016/j.chest.2021.10.031.
Buehrle, D.J. et al. (2021) ‘Trends in Outpatient Antibiotic Prescriptions in the United
States During the COVID-19 Pandemic in 2020’, JAMA Network Open, 4(9), p.
e2126114. Available at: https://doi.org/10.1001/jamanetworkopen.2021.26114.
CBPP (2022) Tracking the COVID-19 Economy’s Effects on Food, Housing, and
Employment Hardships, Center on Budget and Policy Priorities. Available at:
https://www.cbpp.org/research/poverty-and-inequality/tracking-the-covid-19-economys-
effects-on-food-housing-and (Accessed: 17 July 2022).
CDC (2022e) SARS-CoV-2 Variant Classifications and Definitions, Centers for Disease
Control and Prevention. Available at: https://www.cdc.gov/coronavirus/2019-
ncov/variants/variant-classifications.html (Accessed: 2 August 2022).
Chan, E.Y.S., Cheng, D. and Martin, J. (2021) ‘Impact of COVID-19 on excess mortality,
life expectancy, and years of life lost in the United States’, PLOS ONE, 16(9), p.
e0256835. Available at: https://doi.org/10.1371/journal.pone.0256835.
108
Cohen, S. et al. (1997) ‘Chronic social stress, social status, and susceptibility to upper
respiratory infections in nonhuman primates’, Psychosomatic Medicine, 59(3), pp. 213
221. Available at: https://doi.org/10.1097/00006842-199705000-00001.
Cohen, S., Janicki-Deverts, D. and Miller, G.E. (2007) ‘Psychological Stress and
Disease’, JAMA, 298(14), pp. 16851687. Available at:
https://doi.org/10.1001/jama.298.14.1685.
Cohen, S., Tyrrell, D.A.J. and Smith, A.P. (1991) ‘Psychological Stress and
Susceptibility to the Common Cold’, New England Journal of Medicine, 325(9), pp. 606
612. Available at: https://doi.org/10.1056/NEJM199108293250903.
Czeisler, M.É. et al. (2020) ‘Mental Health, Substance Use, and Suicidal Ideation During
the COVID-19 Pandemic - United States, June 24-30, 2020’, MMWR. Morbidity and
mortality weekly report, 69(32), pp. 10491057. Available at:
https://doi.org/10.15585/mmwr.mm6932a1.
Devi, S. et al. (2021) ‘Adrenergic regulation of the vasculature impairs leukocyte
interstitial migration and suppresses immune responses’, Immunity, 54(6), pp. 1219-
1230.e7. Available at: https://doi.org/10.1016/j.immuni.2021.03.025.
Dhabhar, F.S. (2014) ‘Effects of stress on immune function: the good, the bad, and the
beautiful’, Immunologic Research, 58(23), pp. 193210. Available at:
https://doi.org/10.1007/s12026-014-8517-0.
Di Gennaro, F. et al. (2021) ‘Increase in Tuberculosis Diagnostic Delay during First
Wave of the COVID-19 Pandemic: Data from an Italian Infectious Disease Referral
Hospital’, Antibiotics, 10(3), p. 272. Available at:
https://doi.org/10.3390/antibiotics10030272.
Elo, I.T. et al. (2022) ‘Evaluation of Age Patterns of COVID-19 Mortality by Race and
Ethnicity From March 2020 to October 2021 in the US’, JAMA Network Open, 5(5), p.
e2212686. Available at: https://doi.org/10.1001/jamanetworkopen.2022.12686.
Evans, J.D. (1996) Straightforward Statistics for the Behavioral Sciences. Belmont, CA,
US: Thomson Brooks/Cole Publishing Co, pp. xxii, 600. ISBN: 0-534-23100-4.
Faherty, L.J. et al. (2022) The U.S. Equity-First Vaccination Initiative: Early Insights.
RAND Corporation. Available at:
https://www.rand.org/pubs/research_reports/RRA1627-1.html (Accessed: 2 August
2022).
Faust, J.S. et al. (2021) ‘All-Cause Excess Mortality and COVID-19Related Mortality
Among US Adults Aged 25-44 Years, March-July 2020’, JAMA, 325(8), pp. 785787.
Available at: https://doi.org/10.1001/jama.2020.24243.
FDA (2021) FDA and CDC Lift Recommended Pause on Johnson & Johnson (Janssen)
COVID-19 Vaccine Use Following Thorough Safety Review, FDA. FDA. Available at:
109
https://www.fda.gov/news-events/press-announcements/fda-and-cdc-lift-recommended-
pause-johnson-johnson-janssen-covid-19-vaccine-use-following-thorough (Accessed: 2
August 2022).
FDA (2022) Coronavirus (COVID-19) Update: FDA Limits Use of Janssen COVID-19
Vaccine to Certain Individuals, FDA. FDA. Available at: https://www.fda.gov/news-
events/press-announcements/coronavirus-covid-19-update-fda-limits-use-janssen-
covid-19-vaccine-certain-individuals (Accessed: 2 August 2022).
Gisselsson-Solen, M. and Hermansson, A. (2022) ‘Trends in upper respiratory tract
infections and antibiotic prescriptions during the COVID-19 pandemic -- a national
observational study’, Authorea [Preprint]. Available at:
https://doi.org/10.22541/au.165545645.58948908/v1.
Giuntella, O. et al. (2021) ‘Lifestyle and mental health disruptions during COVID-19’,
Proceedings of the National Academy of Sciences of the United States of America,
118(9), p. e2016632118. Available at: https://doi.org/10.1073/pnas.2016632118.
Givon-Lavi, N. et al. (2022) ‘Disproportionate reduction in respiratory vs. non-respiratory
outpatient clinic visits and antibiotic use in children during the COVID-19 pandemic’,
BMC Pediatrics, 22(1), p. 254. Available at: https://doi.org/10.1186/s12887-022-03315-
0.
Gottesman, B.-S. et al. (2022) ‘Community antibiotic prescriptions during COVID-19 era:
a population-based cohort study among adults’, Clinical Microbiology and Infection
[Preprint]. Available at: https://doi.org/10.1016/j.cmi.2022.02.035.
Hickey, J. and Rancourt, D.G. (2022) ‘Nature of the toxicity of the COVID-19 vaccines in
the USA’, ResearchGate [Preprint]. Available at:
https://www.researchgate.net/publication/358489777_Nature_of_the_toxicity_of_the_C
OVID-19_vaccines_in_the_USA.
Islam, N., Jdanov, D.A., et al. (2021) ‘Effects of covid-19 pandemic on life expectancy
and premature mortality in 2020: time series analysis in 37 countries’, BMJ, 375, p.
e066768. Available at: https://doi.org/10.1136/bmj-2021-066768.
Islam, N., Shkolnikov, V.M., et al. (2021) ‘Excess deaths associated with covid-19
pandemic in 2020: age and sex disaggregated time series analysis in 29 high income
countries’, BMJ, 373, p. n1137. Available at: https://doi.org/10.1136/bmj.n1137.
Jacobson, S.H. and Jokela, J.A. (2021) ‘Beyond COVID-19 deaths during the COVID-
19 pandemic in the United States’, Health Care Management Science, 24(4), pp. 661
665. Available at: https://doi.org/10.1007/s10729-021-09570-4.
Jean-Jacques, M. and Bauchner, H. (2021) ‘Vaccine Distribution—Equity Left Behind?’,
JAMA, 325(9), pp. 829830. Available at: https://doi.org/10.1001/jama.2021.1205.
110
Jewell, J.S. et al. (2020) ‘Mental Health During the COVID-19 Pandemic in the United
States: Online Survey’, JMIR formative research, 4(10), p. e22043. Available at:
https://doi.org/10.2196/22043.
Joffe, A.R. (2021) ‘COVID-19: Rethinking the Lockdown Groupthink’, Frontiers in Public
Health, 9. Available at: https://www.frontiersin.org/articles/10.3389/fpubh.2021.625778
(Accessed: 31 July 2022).
Johns Hopkins (2022) Coronavirus resource center. (accessed on July 14, 2022)
https://coronavirus.jhu.edu/region/united-states
Johnson, J. and Rancourt, D.G. (2022) ‘Evaluating the Effect of Lockdowns On All-
Cause Mortality During the COVID Era: Lockdowns Did Not Save Lives’, ResearchGate
[Preprint]. Available at: https://doi.org/10.13140/RG.2.2.34191.46242.
Karlinsky, A. and Kobak, D. (2021) ‘Tracking excess mortality across countries during
the COVID-19 pandemic with the World Mortality Dataset’, eLife. Edited by M.P.
Davenport et al., 10, p. e69336. Available at: https://doi.org/10.7554/eLife.69336.
King, L.M. et al. (2021) ‘Trends in US Outpatient Antibiotic Prescriptions During the
Coronavirus Disease 2019 Pandemic’, Clinical Infectious Diseases, 73(3), pp. e652
e660. Available at: https://doi.org/10.1093/cid/ciaa1896.
Kitano, T. et al. (2021) ‘The Impact of COVID-19 on Outpatient Antibiotic Prescriptions
in Ontario, Canada; An Interrupted Time Series Analysis’, Open Forum Infectious
Diseases, 8(11), p. ofab533. Available at: https://doi.org/10.1093/ofid/ofab533.
Knight, B.D. et al. (2022) ‘The impact of COVID-19 on community antibiotic use in
Canada: an ecological study’, Clinical Microbiology and Infection, 28(3), pp. 426432.
Available at: https://doi.org/10.1016/j.cmi.2021.10.013.
Kobak, D. (2021) ‘Excess mortality reveals Covid’s true toll in Russia’, Significance
(Oxford, England), 18(1), pp. 1619. Available at: https://doi.org/10.1111/1740-
9713.01486.
Kontis, V. et al. (2020) ‘Magnitude, demographics and dynamics of the effect of the first
wave of the COVID-19 pandemic on all-cause mortality in 21 industrialized countries’,
Nature Medicine, 26(12), pp. 19191928. Available at: https://doi.org/10.1038/s41591-
020-1112-0.
Kontopantelis, E. et al. (2021) ‘Excess deaths from COVID-19 and other causes by
region, neighbourhood deprivation level and place of death during the first 30 weeks of
the pandemic in England and Wales: A retrospective registry study’, The Lancet
Regional Health - Europe, 7, p. 100144. Available at:
https://doi.org/10.1016/j.lanepe.2021.100144.
Kontopantelis, E. et al. (2022) ‘Excess years of life lost to COVID-19 and other causes
of death by sex, neighbourhood deprivation, and region in England and Wales during
111
2020: A registry-based study’, PLOS Medicine, 19(2), p. e1003904. Available at:
https://doi.org/10.1371/journal.pmed.1003904.
Lee, W.-E. et al. (2022) ‘Direct and indirect mortality impacts of the COVID-19 pandemic
in the US, March 2020-April 2021’, medRxiv: The Preprint Server for Health Sciences,
p. 2022.02.10.22270721. Available at: https://doi.org/10.1101/2022.02.10.22270721.
Locatelli, I. and Rousson, V. (2021) ‘A first analysis of excess mortality in Switzerland in
2020’, PLOS ONE, 16(6), p. e0253505. Available at:
https://doi.org/10.1371/journal.pone.0253505.
Louisiana Launches Grassroots COVID Vaccine Campaign to Ensure No Community
Gets Left Behind | Office of Governor John Bel Edwards (2021). (18 March 2021)
Available at: https://gov.louisiana.gov/index.cfm/newsroom/detail/3040 (Accessed: 2
August 2022).
Martos-Vila, M. and Shi, Z. (2022) Bankruptcy Filings During and After the COVID-19
Recession. Available at:
https://www.americanbar.org/groups/business_law/publications/blt/2022/03/bankruptcy-
filings-during-and-after-the-covid-19-recession/ (Accessed: 17 July 2022).
Merikangas, K.R. et al. (2010) ‘Lifetime prevalence of mental disorders in U.S.
adolescents: results from the National Comorbidity Survey Replication--Adolescent
Supplement (NCS-A)’, Journal of the American Academy of Child and Adolescent
Psychiatry, 49(10), pp. 980989. Available at:
https://doi.org/10.1016/j.jaac.2010.05.017.
NIMH (2022) Mental Illness, National Institute of Mental Health (NIMH). Available at:
https://www.nimh.nih.gov/health/statistics/mental-illness (Accessed: 2 August 2022).
OECD (2022) The impact of COVID-19 on employment and jobs, OECD. Available at:
https://www.oecd.org/employment/covid-19.htm (Accessed: 17 July 2022).
Prenderville, J.A. et al. (2015) ‘Adding fuel to the fire: the impact of stress on the ageing
brain’, Trends in Neurosciences, 38(1), pp. 1325. Available at:
https://doi.org/10.1016/j.tins.2014.11.001.
Rambhia, K.J. and Rambhia, M.T. (2019) ‘Early Bird Gets the Flu: What Should Be
Done About Waning Intraseasonal Immunity Against Seasonal Influenza?’, Clinical
Infectious Diseases: An Official Publication of the Infectious Diseases Society of
America, 68(7), pp. 12351240. Available at: https://doi.org/10.1093/cid/ciy748.
Rancourt, D.G. (2020) ‘All-cause mortality during COVID-19: No plague and a likely
signature of mass homicide by government response’, ResearchGate [Preprint].
Available at: https://doi.org/10.13140/RG.2.2.24350.77125. Archived at:
https://archive.ph/PXhsg
112
Rancourt, D.G., Baudin, M. and Mercier, J. (2020) ‘Evaluation of the virulence of SARS-
CoV-2 in France, from all-cause mortality 1946-2020’, ResearchGate [Preprint].
Available at: https://doi.org/10.13140/RG.2.2.16836.65920/1.
Rancourt, D.G., Baudin, M. and Mercier, J. (2021a) ‘Analysis of all-cause mortality by
week in Canada 2010-2021, by province, age and sex: There was no COVID-19
pandemic and there is strong evidence of response-caused deaths in the most elderly
and in young males’, ResearchGate [Preprint]. Available at:
https://doi.org/10.13140/RG.2.2.14929.45921.
Rancourt, D.G., Baudin, M. and Mercier, J. (2021b) ‘Nature of the COVID-era public
health disaster in the USA, from all-cause mortality and socio-geo-economic and
climatic data’, ResearchGate [Preprint]. Available at:
https://doi.org/10.13140/RG.2.2.11570.32962.
Richmond, P. et al. (2021) ‘Mortality: A physics perspective’, Physica A: Statistical
Mechanics and its Applications, 566, p. 125660. Available at:
https://doi.org/10.1016/j.physa.2020.125660.
Roser, M. (2020) Why is life expectancy in the US lower than in other rich countries?,
Our World in Data. Available at: https://ourworldindata.org/us-life-expectancy-low
(Accessed: 31 July 2022).
Sanmarchi, F. et al. (2021) ‘Exploring the Gap Between Excess Mortality and COVID-19
Deaths in 67 Countries’, JAMA Network Open, 4(7), p. e2117359. Available at:
https://doi.org/10.1001/jamanetworkopen.2021.17359.
Sapolsky, R.M. (2005) ‘The influence of social hierarchy on primate health’, Science
(New York, N.Y.), 308(5722), pp. 648652. Available at:
https://doi.org/10.1126/science.1106477.
Sorensen, R.J.D. et al. (2022) ‘Variation in the COVID-19 infectionfatality ratio by age,
time, and geography during the pre-vaccine era: a systematic analysis’, The Lancet,
399(10334), pp. 14691488. Available at: https://doi.org/10.1016/S0140-
6736(21)02867-1.
SSA (2020) Social Security - The Red Book - Overview of our Disability Programs, SSA.
Available at: https://www.ssa.gov/redbook/eng/overview-
disability.htm?tl=0%2C1%2C2%2C3#! (Accessed: 27 July 2022).
The New York Times (2022) ‘See How Vaccinations Are Going in Your County and
State’, 28 July. Available at: https://www.nytimes.com/interactive/2020/us/covid-19-
vaccine-doses.html (Last updated: 21 July 2022) (Accessed: 2 August 2022).
Udit, S., Blake, K. and Chiu, I.M. (2022) ‘Somatosensory and autonomic neuronal
regulation of the immune response’, Nature Reviews. Neuroscience, 23(3), pp. 157
171. Available at: https://doi.org/10.1038/s41583-021-00555-4.
113
US Census Bureau (2022a) “U.S. and World Population Clock”. (accessed on July 14,
2022) https://www.census.gov/popclock/
Van Laethem, J. et al. (2021) ‘Antibiotic Prescriptions Targeting Bacterial Respiratory
Infections in Admitted Patients with COVID-19: A Prospective Observational Study’,
Infectious Diseases and Therapy, 10(4), pp. 25752591. Available at:
https://doi.org/10.1007/s40121-021-00535-2.
Van Laethem, J. et al. (2022) ‘Antibiotic prescriptions in the context of suspected
bacterial respiratory tract superinfections in the COVID-19 era: a retrospective
quantitative analysis of antibiotic consumption and identification of antibiotic prescription
drivers’, Internal and Emergency Medicine, 17(1), pp. 141151. Available at:
https://doi.org/10.1007/s11739-021-02790-0.
Villani, L. et al. (2020) ‘Comparison of Deaths Rates for COVID-19 across Europe
During the First Wave of the COVID-19 Pandemic’, Frontiers in Public Health, 8.
Available at: https://www.frontiersin.org/articles/10.3389/fpubh.2020.620416 (Accessed:
31 July 2022).
Wang, H. et al. (2022) ‘Estimating excess mortality due to the COVID-19 pandemic: a
systematic analysis of COVID-19-related mortality, 2020–21’, The Lancet, 399(10334),
pp. 15131536. Available at: https://doi.org/10.1016/S0140-6736(21)02796-3.
Whitaker, R. (2015) Anatomy of an Epidemic: Magic Bullets, Psychiatric Drugs, and the
Astonishing Rise of Mental Illness in America. Broadway Books, New York, pp. xii, 420.
ISBN: 978-0-307-45242-9.
Winglee, K. et al. (2022) ‘Decrease in Tuberculosis Cases during COVID-19 Pandemic
as Reflected by Outpatient Pharmacy Data, United States, 2020’, Emerging Infectious
Diseases, 28(4), pp. 820827. Available at: https://doi.org/10.3201/eid2804.212014.
Woolf, S.H. et al. (2021) ‘Excess Deaths From COVID-19 and Other Causes in the US,
March 1, 2020, to January 2, 2021’, JAMA [Preprint]. Available at:
https://doi.org/10.1001/jama.2021.5199.
Woolf, S.H., Masters, R.K. and Aron, L.Y. (2021) ‘Effect of the covid-19 pandemic in
2020 on life expectancy across populations in the USA and other high income countries:
simulations of provisional mortality data’, BMJ, 373, p. n1343. Available at:
https://doi.org/10.1136/bmj.n1343.
114
Appendix
The Appendix is in three parts:
Appendix A shows for each state of the USA:
o ACM/w versus time together with ACM by 50-week period, from 2015 to
2022 (equivalent to Figure 12 without the color-coded periods)
o Excess mortality of the pre-vaccination and vaccination periods of the
covid period, by age group (equivalent to Table 4Table 5)
o Excess mortality of the covid period, by age group (equivalent to Table 3)
Appendix B shows state-wise maps of poverty and obesity in the USA
Appendix C shows ACM/w in the USA with most recent data, from 2015 to 2022
The states in Appendix A are ordered alphabetically.
115
Appendix A ACM/w and by 50-week period, by state, 2015-2022
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142