Content uploaded by Scott McLachlan
Author content
All content in this area was uploaded by Scott McLachlan on Jun 30, 2021
Content may be subject to copyright.
Analysis of COVID-19 vaccine death reports from the Vaccine
Adverse Events Reporting System (VAERS) Database
Interim Results and Analysis
Scott McLachlan, Magda Osman, Kudakwashe Dube, Patience Chiketero, Yvonne Choi,
Norman Fenton
Risk and Information Management, Queen Mary University of London, UK
Birmingham Law School, University of Birmingham, UK
School of Biological and Chemical Sciences, Queen Mary University of London, UK
School of Fundamental Sciences, Massey University, NZ
Occupational Health and Wellbeing, Network Rail, UK
Health Informatics and Knowledge Engineering Research (HiKER) Group
Abstract
Clinically trained reviewers have undertaken a detailed analysis of a sample of the early deaths reported in
VAERS (250 out of the 1644 deaths recorded up to April 2021). The focus is on the extent to which the
reports enable us to understand whether the vaccine genuinely caused or contributed to the deaths.
Contrary to claims that most of these reports are made by lay-people and are hence clinically unreliable,
we identified health service employees as the reporter in at least 67%. The sample contains only people
vaccinated early in the programme, and hence is made up primarily of those who are elderly or with
significant health conditions. Despite this, there were only 14% of the cases for which a vaccine reaction
could be ruled out as a contributing factor in their death.
2
Executive Summary
There have been multiple conflicting claims made about the safety of the COVID-19 vaccines that were
rolled out world-wide from Dec 2020. However, there is no universally agreed system for reporting either
deaths or serious side-effects for which these vaccines may have been the cause or a contributory factor,
and hence, as a result, there are concerns about variability in the quality of reports and the credibility of
the sources submitting them. Reports can be submitted by physicians involved in administering the vaccine
or helping treat side effects that may have consequentially arisen, clinical and non-clinical health service
employees, or pharmaceutical professionals involved in the investigation. Likewise, lay people, such the
patient or their family and friends, may have submitted a report independently of medical carers. It has
been suggested that a third category of submission may have been made by members of anti-vaccine, or
other groups, motivated by ill-intent, who may exaggerate case numbers reported. Critics of safety
reporting cite the fact that lay people, or those with malign intent, may form the bulk of reports and hence
statistics on side effects must therefore be exaggerated because they come from non-credible sources. Set
against this, research suggests that as few as 1% of the true adverse reactions ever get formally recorded.
In early April 2021 we downloaded the 2021 Vaccine Adverse Events Reporting System (VAERS) dataset
with the aim to analyse these reports to determine the range and frequency of health problems potentially
caused by the vaccines but also the quality of the reports, and by inference the credibility of the reporters
lodging them. For each patient cited in a report, a clinically trained reviewer manually examines the report
to determine its source and clinical credibility and to identify and record medical history, current illness,
and symptoms. Each is then checked by a second reviewer. This process is ongoing, as there are 1644
deaths in the April VAERS deaths dataset that have been reported in patients who had recently received
their first or second COVID-19 vaccination, and over 28,000 serious adverse events that did not result in
death. This interim report presents the results of our analysis of the first 250 reported deaths that have
been reviewed and coded by our team. We identified health service employees as the reporter in at least
67% of the reports, while pharmaceutical employees were identified as the reporter in a further 5%. Lay
people were identifiable as the reporter in only 28% of the reports. This suggests an intention for clinical
applicability and usefulness and goes some way towards addressing the common disclaimer that many
VAERS reports are made by aggrieved family members and anti-vaxxers, both with an axe to grind. The
sample is heavily biased because these were all people vaccinated very early in the programme when only
the elderly, those with significant or chronic health conditions and frontline health service staff were being
vaccinated. Yet, our analysis shows that the patients can be grouped into three main types: (i) those where
the vaccine was most likely not a factor; (ii) those where the vaccine may have been a factor; and (iii) those
where the vaccine was the most likely factor in their deaths. We found that in 34 of the 250 deaths (14%)
a vaccine reaction could be ruled out as a contributing factor in their death; these were all patients either
already bedridden and expected to die from a serious medical condition like lung cancer, or were described
as at end of life or receiving palliative hospice care. For 203 of the 250 (81%) the vaccine may have been a
factor in their death; however, many of these patients had one or more chronic or age-related comorbid
conditions. Finally, for at least 13 of the 250 deaths (5%) the vaccine was the most likely cause of death;
these patients had strong reactions soon after vaccination and died either on the same day, or during the
next couple of days.
3
1. Introduction
Amateur critics often like to dismiss anecdotes as 'unscientific', but this is wrong:
anecdotes are weaker evidence than trials , but they are not without value and
are often the first sign of a problem (or an unexpected benefit).
Bad Pharma by Ben Goldacre, p189
In January 2020 the World Health Organisation (WHO) upgraded an outbreak of pneumonia cases which
they said were caused by Sars-CoV-2 in China to global pandemic status (WHO, 2020). The development
time for a vaccine candidate normally ranges from 3-6 months for an updated version of an already
established jab such as last year’s influenza vaccine, to three years for a completely new product (Plotkin,
Robinson, Cunningham et al, 2017). Once testing in animal models is complete, taking perhaps another 1-
2 years, testing the vaccine candidate in humans usually requires four phases of testing. Phase 1 involves
a very small number of human subjects (20-40) with a focus on identifying any potentially life-threatening
reactions and assessing immunogenicity, while Phase 4 involves collection of efficacy data from hundreds
of thousands of people who have received the vaccine in its initial 2-3 years of community use (Dermody,
DiMalo & Enquist, 2020). In total it takes anywhere from 5-8 years, or even as much as 15-20 years (Lanese,
2020), before a vaccine is licensed for community use, and several years beyond that before we have a
complete picture of its safety and effectiveness.
While the WHO said in late February it did not believe a credible vaccine would be available in less than 18
months (Grenfell & Drew, 2020), pharmaceutical companies began the first human clinical testing of
COVID-19 vaccines just three weeks later, even before most countries had felt any real impact from the
disease (Le, Andreadakis, Roman et al, 2020). Much has been made about the accelerated pace with which
COVID-19 vaccines were brought to market; including suggestions that animal testing and Phase 3
challenge trials were conducted following previously untested novel methods (Eyal, Lipsich & Smith, 2020;
Lanese, 2020), or skipped altogether (Boodman, 2020; Lanese, 2020). (Magee, 2021) refuted claims of
skipped animal testing but described that testing as using defrosted embryos and claimed that this form of
animal testing was pivotal in the vaccine’s rapid release. The implication is that animal model testing took
place during the few weeks between publishing the SARS-CoV-2 DNA in late January 2020, which was a
requirement before development of the mRNA vaccines could commence, and the first human trials that
commenced only seven weeks later.
The expedited approval of these vaccines, especially those using novel mRNA technology, has been a
source of contention for clinical and lay people alike (Doshi, 2021; Mahase, 2020). Many argue that their
novel technology should have warranted greater, not reduced, scrutiny (Doshi, 2021). Some even
suggested Vaccine-Enhanced Respiratory Disease (VAERD) and vaccine excipient-related clotting would
result from the new vaccines even before they were given emergency authorisation (Hotez, Corry, Strych
et al, 2020). Indeed, some countries suspended use of the vaccines for exactly these reasons shortly after
their vaccination programmes started (Dyer, 2021; Ostergaard, Schmidt, Horvath-Puho et al, 2021; Wise,
2021).
4
It is therefore unsurprising, that there has been great interest in the increasing number of reports of
deaths, reactions and serious side-effects for which it is claimed the vaccine was the cause or a contributory
factor. Unfortunately, there is no universally agreed system for reporting and so there can be great
variability in the quality and number of reports. On the one hand there are those who claim that vaccine
adverse event reports are primarily submitted by lay people such as the family and friends of the deceased
or members of anti-vaccine groups. On the other hand, there are those who argue that as few as 1% of the
true adverse reactions ever get formally recorded. Given that for a variety of reasons there is no other
centralised recording of COVID-19 vaccine related events, these reports are the only source from which
knowledge about patient outcomes may be drawn.
In early April 2021 we accessed and downloaded the dataset and accompanying documents from the US
VAERS reporting system. Our objective is to perform an analysis of the reports to determine not just the
range and frequency of health problems caused by the vaccines, but also the quality of the reports. Quality
was assessed through analysis of the clinical information provided for each patient: whether it incorporates
historic and current medical conditions, current medications, and details of the vaccination, onset of
symptoms and the death events. This process is ongoing, as there are 1644 deaths in the dataset that have
been reported in patients who had recently received their first or second COVID-19 vaccination, and over
28,000 serious adverse events, including severe tinnitus, inflammation of the heart muscle (myocarditis
and pericarditis) and clots, that did not result in death. This interim report presents the results of our
analysis of the first 250 reported deaths that have been reviewed and coded by our team.
The paper is structured as follows: In Section 2 we provide an overview of VAERS and compare it to the
systems in other countries. In Section 3 we describe our detailed approach to the analysis and the interim
results on the range and frequency of health problems caused by the COVID-19 vaccines. A discussion of
the clinical results is provided in Section 4. In Section 5 we present the results of our analysis into the quality
of reporting. In Section 6 we discuss what is generally known about the scale of COVID-19 vaccine deaths
and adverse reactions compared to other vaccines. A discussion of the media narrative and ‘fact checking’
of VAERS is presented in Section 7, with conclusions and recommendation in Section 8.
2. Vaccine Adverse Events Reporting and Dissemination
Most countries operate some form of vaccine adverse event reporting system for post-licensure safety
surveillance.
In the United States of America (USA) for several decades this function has been provided by the Vaccine
Adverse Events Reporting System (VAERS) which is administered by the Food and Drug Administration
(FDA) (Varricchio, Iskander, Ball et al, 2004). As shown in Figure 1a, significant and sometimes quite
descriptive data is publicly available from the VAERS website that includes clinician’s narrative notes along
with medical history, current illnesses, medications and symptoms experienced by the patient. Figure 1b
provides an example of a low-detail family reported incidence. Figure 1c is an example, discussed later,
where the VAERS call centre employee (the Recorder) may have become the Reporter by imputing their
own inferences and interpretations for whether the vaccine may or may not have been involved.
5
(a)
(b)
(c)
Figure 1: VAERS clinical narratives for patients who died within hours of receiving a COVID-19 vaccine where the reporter
was: (a) a health service employee; (b) a family member of the deceased; and (c) where the VAERS call centre staff have
included their own inferences.
In contrast, the publicly available data from the United Kingdom (UK), provided by the Medicines and
Healthcare products Regulatory Authority (MHRA), and Australia (AUS), provided by the Therapeutic
Goods Administration (TGA), is sparse. It consists in both cases of a simple list of symptoms and the
frequency with which they have been reported. As shown in the examples in Figure 2, this data is
provided with no information about the patients, clinical outcomes or vaccine batch numbers, and no
ability to analyse and identify clusters that may present with serious consequences for individual
patients, including death (Wise, 2021).
6
Figure 2: Vaccine Adverse Event data. Top: AUS (TGA, 2021) and Bottom: UK (MHRA, 2021)
3. Analysis of the VAERS data for COVID-19 Vaccines
In early April 2021 this team accessed and downloaded the raw 2021 VAERS dataset along with the
corresponding Symptoms and Vaccine data from the VAERS website (https://vaers.hhs.gov/). Based on
their consistent use of an anonymised patient identifier across all three datasets, we were able to
aggregate them into a single dataset for analysis. For each patient a clinically trained reviewer manually
examines the medical history, current illness, symptoms and clinician’s narrative to identify and record
individual medical conditions, symptoms of interest, and seeks to identify whether the Reporter is, for
7
example, a health service employee or family member. Each is then checked by a second reviewer who
ensures the details have been correctly coded. This process is ongoing, as there are 1644 deaths in the
dataset that have been reported in patients who had recently received their first or second COVID-19
vaccination, and over 28,000 serious adverse events that did not result in death.
This interim results paper presents information on the first 250 reported deaths that have been reviewed
and coded by our team. Obviously, these results cannot be generalised as the sample is heavily biased -
these were all people vaccinated very early in the programme when only the elderly, those with significant
or chronic health conditions and frontline health service staff were being vaccinated.
Death by Age: As shown in Figure 3, unsurprisingly the 85+ age group accounts for the largest proportion
of deaths (31.9%), almost double the next highest group, the 80-84 year olds (17.7%). In these two groups
advanced age and collected co-morbid conditions tended to be described by clinical staff as most likely to
have caused their death shortly after receiving the vaccine. However, the narrative notes contained
numerous examples of deaths in vaccine recipients in these age groups who were still active and living in
their own homes in the wider community, for example: 959568 who was found collapsed and unconscious
by her husband less than 72 hours after receiving her COVID-19 vaccination, having suffered and almost
immediately dying from a type of stroke known as a cerebrovascular accident (CVA), and 934373 who was
found dead in an armchair in their lounge room by her husband in the early morning around 12 hours after
receiving her COVID-19 vaccination.
Figure 3: Death by Age Group
Much has been made in the media and academic literature about the need for protection and early
vaccination of those aged 65 years and over. We believe this focus is the primary reason that 80% of the
post-vaccination decedents reported are in this age group. Almost one-tenth (9%) expired within only 6
hours of their vaccination and 18% died in less than 12 hours. Over one third (36%) did not survive through
to the following day.
8
Figure 4 highlights that 50% died in less than 48 hours after receiving their COVID-19 vaccination. This
increases to 80% when we extend to the first week post-vaccination. A further 10% of deaths occurred in
the second week, with the remaining 10% passing away during weeks 3 and 4. After receiving their COVID-
19 vaccination, those people who were: (a) diagnosed or specifically described as having an allergic
reaction, or whose symptoms strongly supported this diagnosis, died between 30 minutes and 4 days; (b)
described as suffering from respiratory distress, with or without symptoms of pneumonia, died between
days 2 and 9; and (c) diagnosed with or described as having a cardiac event, e.g.: myocardial infarction or
heart attack, died between days 5 and 14.
Figure 4: Duration from Vaccination to Death for the over-65s
Six decedents (921667, 921768, 943397, 944595, 956458 and 1092595) were identifiable from their
narrative notes as staff in the current employ of healthcare provider organisations.
Pre-existing Conditions: Pre-existing or co-morbid conditions should always be critically appraised in any
drug or vaccine study, but especially for these vaccines where issues of VAERD and clotting have been
publicly raised, and because many of the VAERS reported deaths are blamed on the recipients underlying
health status.
Figure 5: Pre-existing Conditions
Figure 5 shows that the most common single pre-existing condition were hypertensive disorders, or
disorders of blood pressure (36%). The most common collective group were those we would normally
9
aggregate as cardiac or heart conditions, which were identified in the deaths of 35% of the reported vaccine
recipients. These conditions include: congestive heart failure (CHF), cardiovascular disease (CVD), and
myocardial infarction (MI) including ST-Elevation Myocardial Infarction (STEMI) and Non-ST-Elevation
Myocardial Infarction (NSTEMI).
Almost one-third of all COVID-19 vaccine recipients had recorded food, drug or environmental allergies
(30%). While 23% of the patients had a diabetic disorder, it was noted that for all but two whose diabetes
was recent or sufficiently well controlled so as not to have resulted in co-morbidities at the time of
vaccination, the rest were compounded by a hypertensive disorder along with either a chronic renal or
thyroid condition; or an accumulation of all three.
Death by Vaccine Brand: Two vaccine brands had been administered to the 250 recipients in our interim
results: Moderna and Pfizer\Biontech. Figure 3 shows that both vaccine brands were administered to an
almost equal number of recipients. Eight reports failed to identify the manufacturer of the vaccine
administered to the decedent.
We noted in the overall VAERS dataset that vaccine recipients were diversely located around the entire
USA and they received doses from almost 200 different vaccine batches. However, there were eight vaccine
batches in the overall VAERS dataset that were identified in relation to 30 or more recipients
1
.
Figure 6: Deaths by Vaccine Brand
4. Discussion of clinical findings
The most significant incidental finding from our review of the symptoms and narrative provided in these
reports was that while 8% directly report temporary (transient ischemic) to complete (ischemic) occlusive
strokes and cerebral haemorrhage (ruptured aneurysm), many more report one or more clinical signs and
symptoms consistent with clotting disorders without specifically identifying them as a potential cause of
1
Pfizer\Biontech EL0140, EK9231, EL3249, EN5318 and Moderna 12L20A, 25J20A, 37K20A, 39K20A
10
death. Acute clotting disorders present from cerebral, pulmonary or abdominal embolism and the
symptoms described post vaccination included: sudden severe headache (8), dizziness and light-
headedness (16), sudden loss of balance or physical coordination with or without falls (38), difficulty
walking (5), post-vaccination confusion with or without difficulty speaking or comprehending speech (11),
sudden or unilateral weakness (33), sudden breathlessness or shortness of breath (39), coughing up blood
(6), and sudden severe abdominal pain (4). Early non-acute signs and symptoms that may indicate the
presence of deep vein thrombosis (DVT)
2
prior to presentation with acute embolism included pain
(throbbing and cramping) in a leg which was reported by 2% of those who went on to describe some of the
other more acute symptoms described above. Of the 22 patients where a stroke was explicitly identified,
20 (91%) had no mention of a pre-existing coagulative disorder, and 9 (41%) also had no mention of pre-
existing hypertensive, arrhythmic cardiac disorders or an injury that might have otherwise explained the
clotting event. There was no mention of a pre-existing clotting disorder for any of the 83 patients with one
or more signs and symptoms of a possible clot event. While it is true that the patient histories and current
illness descriptions present with variable quality and detail, this would not sufficiently explain away every
instance.
The reviewers and authors include people trained in clinical nursing, a medical practitioner, health
informaticians, a psychologist, a chartered mathematician, and a statistician. The reviewers generally
commented on several common themes observed in the vaccine recipient patient data. First, the
extremely high number of recipients who were reported as complaining of general weakness, tiredness,
malaise and lethargy in the days after receiving COVID-19 vaccines. This is apparently a common theme in
COVID-19 vaccine recipients and was even commented on in the notes on the website with the Australian
summary statistics. Second, the similarly high number of recipients whose post-vaccination and pre-death
symptoms included description of syncope (fainting, passing out or collapsing). We observed some
description of collapse, syncope or fainting in the symptom or narrative notes of 6% of VAERS death reports
reviewed. This observation is also generally consistent with a prior Institute of Medicine (IOM) report that
said the VAERS evidence convincingly supports causal relationships between vaccine injection and syncope
(Miller, Moro, Cano et al, 2015). Third, the majority (91%) of deaths are reported after administration of
the first COVID-19 vaccination.
5. Results of our analysis into the quality of reporting
The reporting person (the reporter) and quality of clinical detail in the report varied considerably. The
reporter ranged from a friend or family member of the deceased, through to clinicians, nurses,
administrative employees of healthcare provider organisations, staff at vaccine deployment centres and in
a small number of cases, staff from the pharmaceutical manufacturers who developed the vaccines. It was
possible to identify that health service employees accounted for at least 67% (n=168) of the VAERS reports
we evaluated, while pharmaceutical employees were identified as the reporter in a further 5% (n=12).
While it is often suggested in the media that VAERS and Yellow Card reports are primarily the product of
2
The USA CDC describe that half of all people with DVT have symptoms of swelling, pain, tenderness and
localised redness of the skin at the affected part of the body. It also describes that such clots can become
an acute pulmonary embolus (PE) without ever causing diagnosable signs and symptoms consistent with a
DVT - https://www.cdc.gov/ncbddd/dvt/facts.html
11
lay people such as the family and friends of the deceased or members of anti-vaccine groups, this interim
review found that lay people were involved in only 28% of the reports we reviewed. Finally, in 3% of cases
the VAERS call centre employee (the recorder) appears to provide their own narrative or interpretation
such that the recorder becomes, in effect, a reporter
3
.
The majority of death reports in this interim dataset include some narrative description for the death event;
whether a short clinical comment
4
or a more comprehensive narrative including elements of the patient’s
medical status prior to and/or post vaccination, similar in detail to that shown in Figure 1. On reading these
it became apparent that some nursing or clinical staff felt it important to expressly distance vaccine
administration from the resulting death. It was difficult to tell whether this was to avoid the negative
mainstream media perceptions of being vaccine hesitant or anti-vax, to distract potential blame being
levelled at the care organisation or healthcare provider who authorised and administered the vaccine, or
was simply because they had an honest
5
belief in absolute vaccine infallibility. For 8 recipients
6
, even
absent evidence to support the assertion, it was reported categorically that the vaccine could not have had
any causal relationship to the recipient’s death. For 4 recipients
7
it was claimed, even in spite of evidence
to the contrary, that COVID had caused their deaths because the vaccine had insufficient time to save their
lives. For many of the elderly and infirm, the description of death often asserts a pre-existing or age-related
condition as the more likely cause of death, including cancers and long-standing cardiac pathologies
8
. Also,
for 12 recipients (5%) we are expressly told that one or more negative COVID-19 PCR test results were
returned in the hours or days prior to the vaccine recipient’s death. In spite of the fact that only 11 (4%)
present with a test-confirmed and current COVID-19 infection, all 250 people in this interim collection were
reported as COVID-19 deaths. This means that all, even those who received one or more negative test
results, are erroneously counted in the officially reported national COVID-19 death tally.
6. Vaccine-related death reports 2017-2021
We downloaded and reviewed VAERS datasets for the years 2017-2021. Reports of possible vaccine-related
deaths averaged 180 annually between 2017 and 2020 (Figure 7), but have risen more than 900% in just
the first 3 months of 2021. Of the 1694 vaccine-related death reports recorded so far in 2021, 1644 (97%)
relate to COVID-19 vaccines. Only 50 reports (3%) refer to a non-COVID-19 vaccine; with influenza (26%)
3
E.g.: 940822 where the VAERS call centre staff member concludes on the basis of what they were told
that it is medically not possible to make meaningful causality assessment and it is unlikely the vaccine
could have contributed to the death of the patient based on the known safety profile.
4
E.g.: 958322 whose record describes only that shortly after receiving his first Pfizer\Biontech
vaccination the 62yo gentleman began shaking and became unresponsive and died later that same day;
and 930466 whose narrative describes fever, shortness of breath and chest pain that resulted in a heart
attack a few hours after vaccination as the progression of post-vaccination symptoms leading to an 82yo
woman’s death.
5
and unsound
6
VAERS-IDs: 917117, 917790, 917793, 924464, 926797, 935343, 964629 and 962714.
7
VAERS-IDs: 917790, 937127, 937152 and 937186.
8
E.g.: 924464 - lung cancer; 926797 - significant cardiac issues; 930487 - acute heart attack; 951518 -
significant deterioration of physical and mental state due to psychogeriatric disease; 964629 - metastatic
cancer; 932898 - cardiac arrest due to ongoing cardiac disease.
12
and zoster
9
(12%) vaccines most frequently identified, but almost half (48%) unspecified by the reporter.
Figure 8 shows that the vaccines receiving the most attention overall during the 2017-2021 period were
the influenza vaccines with a total of 203 reports during the five consecutive years - an average of 40 per
year, but still far from the volume of COVID-19 reports recorded in just the first three months of 2021.
Figure 7: Annual vaccine-related death reports in VAERS (2017-2021)
Figure 8: Non-COVID vaccine-related death reports (2017-2021)
9
Herpes Zoster is more commonly known as shingles and is a reactivation, in adults, of the childhood
varicella chicken pox virus. Several studies have confirmed that infection with chicken pox during
childhood produces a significant degree of life-long immunity that is preventative for shingles in later life,
and that universal childhood vaccination for chicken pox (which actively prevents infection with the
natural chicken pox virus) has resulted in a dramatic increase in moderate to severe shingles diagnoses
for middle aged and older adults.
13
In order to identify the scale of increase in vaccine adverse event reporting for the COVID-19 vaccines we
must consider the number of doses administered per each VAERS death report, and the number of reports
per million vaccinations administered in the same population for other vaccines. We collected data
10
for
the total number of influenza vaccinations administered during the 2017-2020 period from the USA’s
Centres for Disease Control (CDC), and COVID-19 vaccination doses administered in the USA during the first
three months of 2021 from the USA Department of Health’s COVID Dashboard. These data are shown in
Figure 9. We found that as the number of doses of influenza vaccine being administered annually was
increasing (from 146mil in 2017 to 174.5mil in 2020), the number of VAERS reports suggesting a link
between influenza vaccination and a vaccination-temporal death also increased. Overall and represented
by the orange line in Figure 9, the number of doses administered per VAERS death report dropped from
7.3mil/report in 2017, to 2.3mil/report in 2020. This equates to an increase from 0.1 to 0.4 reports per
million doses administered between the years 2017 and 2020 which is indicated by the blue line in Figure
10.
Figure 9: Doses Administered per VAERS Death Report (2017-2020)
10
https://www.cdc.gov/flu/prevent/vaccine-supply-historical.htm
14
Figure 10: VAERS Death reports per million doses administered
In stark contrast, for COVID-19 vaccines there is one VAERS death report for each 135,000 administered
doses (orange line in Figure 9). This represents a staggering 1677% increase in VAERS death reporting for
COVID-19 vaccines, or as shown by the blue line in Figure 10, an increase of 7 reports per million doses
administered in just the first three months of 2021. If this reporting trend continues, there could be at
least 6500 individual VAERS death reports by the end of 2021, which would represent a 3400% increase
in reports compared to the last five years.
7. The media narrative and “Fact checking”
During early May television reports, such as one made by Tucker Carlson on FOX News
11
, suggested we
should consider the number of adverse event and death reports being recorded in the VAERS database for
COVID vaccines when deciding whether young and healthy people who are not otherwise at risk from
COVID-19 should even receive these medications. Tucker asks a number of simple but relevant questions
in regards to the social, political and potentially legal coercion being used to promote the taking of COVID
vaccines: “How many people have died after taking COVID vaccines?”, “what are the potential risks from
taking these vaccinations?” and “what do we really know about the potential risks from taking these
vaccinations?”. These are all questions that should be asked when considering whether to take any
medication; whether that medication is an antidepressant, statin, or even a vaccine. Carlson’s thesis was
that when we consider the dramatic rise in adverse event and death reports for COVID vaccines, as
supported by our analysis in this work, the situation for COVID vaccines is demonstrably worse than for
any other commonly administered large-volume vaccine like those administered for influenza in our
example in Figure 9. Carlson goes to great pains to carefully point out that it is the VAERS data itself that
shows that more deaths have been reported for COVID-19 vaccines than any other vaccine during the last
15 years, and that what he is calling for is simply some level of government and independent scrutiny to
assess whether the increased VAERS reports are indicative of a problem. He also provides discussion on
the alternate viewpoints proposed by others that seek to explain away some of the excessively high
number of deaths in the over-65 age group: for example, that it could be coincidence, or may be wholly
expected, that some elderly recipients died shortly after receiving COVID injections. In any event, the
questions being asked here are those which are normally asked of any new medication. They are important
and necessary and should not be waved away simply by virtue of an application of the term ‘vaccine’.
11
https://video.foxnews.com/v/6252794642001#sp=show-clips
15
Health professionals are expected to warn patients regarding likely and possible side effects for any drug
or clinical intervention. They cannot perform this important and critical task and enable our ability to
provide informed consent in circumstances where that data is not being collected or analysed, or where
discussion of side effects and outcomes is being stifled or branded as misinformation in the public arena.
Self-titled fact checkers and journalists in the mainstream media immediately discredited Carlson’s (and
other’s) narratives where the VAERS data was concerned. They imputed that VAERS was a breeding-ground
for anti-vaccine misinformation and pointed to the twitter posts of generalist doctors as vaccine experts
that rejected Carlson’s VAERS death claims
12
by claiming all the deaths were coincidence and asserting
unproven facts about ongoing death rates, and without also pointing out that the same expert in a
subsequent post acknowledges that VAERS data was used to identify the clotting issue with the Johnson &
Johnson COVID-19 vaccine
13
(McCarthy, 2021). It is incredible to decry VAERS as rubbish self-reported
nonsense when that data suggests something that goes against your particular views, while also suggesting
it provides data that was relevant or helpful in other circumstances. Other journalists directed readers to
VAERS’ own disclaimers and the fact that anyone can report an adverse reaction to VAERS to say that any
number of its reports have nothing to do with the vaccine and assert that this also means the vaccines
haven’t been linked to or did not cause the deaths (Dunlop, 2021; Dupuy, 2021; Walsh, 2021). It would also
be factual to say that they haven’t been disconnected or disproved as potentially causal in the deaths
either. However, not one journalist paused to acknowledge this opposing truth. Several fact checkers
acknowledge that some people experience allergic reactions to the vaccines, and say the VAERS data shows
this most often occurs within 30 minutes of administration (Jaramillo, 2021). They also report that these
serious adverse reactions occur in only 2 to 5 people per million vaccinated, which would be only 0.0005%
(Jaramillo, 2021). The language and intonation used by these fact checkers is strongly opposed to any idea
that an allergic reaction, or anaphylaxis, to a vaccine could lead to death. They describe use of VAERS data
in any way that might suggest these vaccines might result in even a single allergic reaction that causes
death as misuse, and misinformation - even though our analysis shows that a vaccine allergic reaction is
the most likely cause for the symptoms and patient outcomes described for at least 13 of the 250 deaths
reviewed in this work (5%), who in each case strongly reacted either shortly after receiving the injection or
within the first 4 hours, and died between 30 minutes and 4 days later.
It should be noted that many of the self-titled fact checkers are simply mainstream media journalists and
other online content creators and opinion-makers
14
. They often name and shame those who speak out
against a particular policy or public narrative for not having qualifications appropriate to the scientific or
medical content in discussion (Jaramillo, 2021), while also providing what they claim are scientific or
medical facts from a soapbox similarly absent of any relevant inculcation (Noorchashm, 2021). Either way,
12
https://twitter.com/Craig_A_Spencer/status/1390113877128531970?s=20
13
https://twitter.com/Craig_A_Spencer/status/1390115599204896768?s=20
14
This is certainly the case for most who post at https://www.factcheck.org, including the prolific fact
check poster Catalina Jaramillo, a pre-COVID NPR reporter who berates others who do not present with
scientific or medical qualifications while hiding behind her journalism major with aspirations in the
domains of environmental issues and public policy but entirely lacking qualifications in immunology or
medical science to support her own opposing viewpoints. The same is also true at
https://factcheck.afp.com where career journalists and digital verification editors like W.G. Dunlop
dispense personal opinion under the banner of COVID fact checking on a range of highly technical
medical topics for which they possess no training or relevant qualifications
(https://www.linkedin.com/in/w-g-dunlop-aa867920)
16
we believe the fact checkers should be treated with the same degree of scepticism and distrust they
recommend we employ when reading any of the sources which they so strongly disprove of.
8. Conclusions
This report presented information derived from analysis of the first 250 COVID-19 vaccine related deaths
reported in the VAERS database. While interim in nature, there are sufficient reports of reasonable quality,
with 72% authored by health service employees, to support further analysis to potentially answer any
number of hypothetical questions. Because the sample is heavily biased - being made up only of those
vaccinated early, namely the elderly, those with significant or chronic health conditions and frontline health
service staff – the clinical results are not generalisable. But there are some important findings. that the
only patients where a vaccine allergic reaction be ruled out as contributing to death were 34 (14%) who
were all either already bedridden, at end of life, and expected to die anyway from a serious comorbid like
lung cancer or were on palliative hospice care. We also found that for at least 13 of the 250 deaths (5%), a
vaccine allergic reaction was indisputably the most likely direct cause for the symptoms and patient
outcomes described. Our ongoing work will continue to process all 1644 reported deaths before going on
to review the almost 28,000 significant adverse event reports. We are seeking to identify and investigate
relationships in the data between different singular and grouped pre-existing conditions and the COVID-19
vaccines, and to further code the narrative clinician’s notes to determine whether answers can be found
to the public hypotheses surrounding their continued use and role, if any, in the events that have been
reported. Our work has also highlighted the vast difference in transparency and content between the data
provided by the USA and that of the UK and Australia. The quantity and quality of data provided by the
USA VAERS dataset is capable of supporting meaningful research, while the sparsity of data provided by
the UK and Australia doesn’t even allow the most basic of conclusions to be drawn. Public pressure should
be directed towards the governments of the UK and Australia to provide greater granularity of data similar
to that of the USA.
References
Blume-Kohout, M. E., & Sood, N. (2013). Market size and innovation: Effects of Medicare Part D on pharmaceutical
research and development. Journal of public economics, 97, 327-336.
Boodman, E. (2020). Researchers rush to test coronavirus vaccine in people without knowing how well it works in animals.
Stat News. Sourced from: https://www.statnews.com/2020/03/11/researchers-rush-to-start-moderna-
coronavirus-vaccine-trial-without-usual-animal-testing/
Dermody, T. S., DiMaio, D., & Enquist, L. W. (2020). Vaccine Safety, Efficacy, and Trust Take Time. Annual Review of
Virology, 8.
Doshi, P. (2021). Pfizer and Moderna’s “95% effective” vaccines—we need more details and the raw data. The BMJ
Opinion.
Dunlop, W.G. (2021). Tucker Carlson misrepresents government data on COVID-19 vaccines. Last accessed: May 9th.
Sourced from: https://abcnews.go.com/Health/wireStory/explainer-activists-target-cdc-vaccine-tracking-system-
77563630
Dupuy, B. (2021). Explainer: How activists target CDC vaccine tracking system. Last accessed: May 7th. Sourced from:
https://factcheck.afp.com/tucker-carlson-misrepresents-government-data-COVID-19-vaccines
Dyer, O. (2021). COVID-19: EMA defends AstraZeneca vaccine as Germany and Canada halt rollouts. The BMJ Opinion.
https://doi.org/10.1136/bmj.n883
Eyal, N., Lipsitch, M., & Smith, P. G. (2020). Human challenge studies to accelerate coronavirus vaccine licensure. The
Journal of infectious diseases, 221(11), 1752-1756.
17
Grenfell, R. & Drew, T. (2020). Here’s why the WHO says a coronavirus vaccine is 18 months away. Sourced from:
https://theconversation.com/heres-why-the-who-says-a-coronavirus-vaccine-is-18-months-away-131213
Hotez, P. J., Corry, D. B., Strych, U., & Bottazzi, M. E. (2020). COVID-19 vaccines: neutralizing antibodies and the alum
advantage. Nature Reviews Immunology, 20(7), 399-400.
Jaramillo, C. (2021) Viral posts misuse VAERS data to make false claims about COVID-19 vaccines. Sourced
from: https://www.factcheck.org/2021/03/scicheck-viral-posts-misuse-vaers-data-to-make-
false-claims-about-COVID-19-vaccines/
Kennedy, R. (2019). CDC Recommended vaccine Schedule 1986 vs 2019. Sourced from:
https://childrenshealthdefense.org/child-health-topics/known-culprit/vaccines-culprit/cdc-recommended-
vaccine-schedule-1986-vs-2019/
Kremer, M., & Snyder, C. M. (2003). Why are drugs more profitable than vaccines? (No. w9833). National Bureau of
Economic Research. Sourced from: https://www.nber.org/system/files/working_papers/w9833/w9833.pdf
Lanese, N. (2020) Researchers fast-track coronavirus vaccine by skipping key animal testing first. LiveScience. Sourced from:
https://www.livescience.com/coronavirus-vaccine-trial-no-animal-testing.html
Le, T., Andreadakis, Z., Kumar, A., Roman, R., Tollefsen, S., Saville, M., & Mayhew, S. (2020). The COVID-19 vaccine
development landscape. Nature Reviews. Sourced from: https://www.nature.com/articles/d41573-020-00073-5
Magee, C. (2021). Top 10 COVID-19 Vaccine Myths. Understanding Animal Research. Sourced from:
https://www.understandinganimalresearch.org.uk/news/communications-media/top-10-COVID-19-vaccine-
myths/
Mahase, E. (2020). COVID-19: Vaccine trials need more transparency to enable scrutiny and earn public trust, say experts.
bmj, 371.
McCarthy, B. (2021). Tucker Carlson’s misleading claim about deaths after COVID-19 vaccine. Last accessed: May 6th.
Sourced from: https://www.politifact.com/factchecks/2021/may/06/tucker-carlson/tucker-carlsons-misleading-
claim-about-deaths-afte/
MHRA, (2021). Coronavirus vaccine - weekly summary Yellow Card reporting - 15 April 2021. Sourced from:
https://www.gov.uk/government/publications/coronavirus-COVID-19-vaccine-adverse-reactions/coronavirus-
vaccine-summary-of-yellow-card-reporting
Miller, E. R., Moro, P. L., Cano, M., & Shimabukuro, T. T. (2015). Deaths following vaccination: what does the evidence
show?. Vaccine, 33(29), 3288-3292.
Noorchashm, H. (2021) Fact-checking factcheck.org’s Ms. Catalina Jaramillo: Part 1. Last accessed: May 5th. Sourced from:
https://noorchashm.medium.com/fact-checking-factcheck-orgs-ms-catalina-jaramillo-b69ebebb9996
Østergaard, S. D., Schmidt, M., Horváth-Puhó, E., Thomsen, R. W., & Sørensen, H. T. (2021). Thromboembolism and the
Oxford–AstraZeneca COVID-19 vaccine: side-effect or coincidence?. The Lancet. https://doi.org/10.1016/S0140-
6736(21)00762-5
Plotkin, S., Robinson, J. M., Cunningham, G., Iqbal, R., & Larsen, S. (2017). The complexity and cost of vaccine
manufacturing–an overview. Vaccine, 35(33), 4064-4071.
TGA, (2021) COVID-19 Vaccine weekly safety report - 07-04-2021. Sourced from: https://www.tga.gov.au/periodic/COVID-
19-vaccine-weekly-safety-report-07-04-2021
Varricchio, F., Iskander, J., Destefano, F., Ball, R., Pless, R., Braun, M. M., & Chen, R. T. (2004). Understanding vaccine safety
information from the vaccine adverse event reporting system. The Pediatric infectious disease journal, 23(4), 287-
294.
Walsh, J. (2021). Tucker Carlson says vaccine-related deaths may be abnormally high - but vaccines haven’t been linked to
deaths. Last accessed: May 5th. Sourced from: https://www.forbes.com/sites/joewalsh/2021/05/05/tucker-
carlson-says-vaccine-related-deaths-may-be-abnormally-high---but-vaccines-havent-been-linked-to-deaths/
WHO, (2020). WHO COVID Timeline. Sourced from: https://www.who.int/news/item/27-04-2020-who-timeline---COVID-19
Wise, J. (2021). COVID-19: European countries suspend use of Oxford-AstraZeneca vaccine after reports of blood clots. The
BMJ Opinion. https://doi.org/10.1136/bmj.n699
Wouters, O. J. (2020). Lobbying expenditures and campaign contributions by the pharmaceutical and health product
industry in the United States, 1999-2018. JAMA internal medicine, 180(5), 688-697.
Yin, W. (2008). Market incentives and pharmaceutical innovation. Journal of Health Economics, 27(4), 1060-1077.
