Exhibit A (Dkt. 58-1, E.D.N.Y.)
Summary
This filing, Dkt. 58-1, collects Exhibits A through K, with pages Bates-stamped DOJ_23-CR-158. Exhibit G is a 2022 article from Communications Medicine that models how vaccine hesitancy could affect the COVID-19 pandemic in European countries, using survey results from 8th to 15th February 2021 and simulating a high-efficacy and a moderate-efficacy vaccine. Its authors suggest that vaccine hesitancy may have a substantial impact on the pandemic’s trajectory, deaths and hospitalisation. Exhibit H is a 2023 Cureus review article, “Unveiling the Impact of COVID-19 Vaccines: A Meta-Analysis of Survival Rates Among Patients in the United States Based on Vaccination Status,” by Ikeokwu et al. It concludes that COVID-19 vaccination confers a certain level of protection against poor outcomes in individuals infected with the virus.
Summary drafted by a model from the document's text below and checked by script against that text before publication. It is a navigation aid, not a reading of what the document proves. Where AI is used
Full text
Exhibit A
Exhibit B
Exhibit C
Exhibit D
Exhibit E
Exhibit F
DOJ_23-CR-158_0006039
Exhibit G
ARTICLE COMMUNICATIONS MEDICINE | https://doi.org/10.1038/s43856-022-00075-x
T
he COVID-19 pandemic has simultaneously resulted in Methods
high global mortality and major economic disruptions. As Vaccine hesitancy data. Attitudes towards COVID-19 vaccina-
a control measure, non-pharmaceutical interventions tion were obtained from the Imperial College London YouGov
(NPIs) such as social distancing and mobility restrictions have Covid 19 Behaviour Tracker Data10. This data set includes
been put in place worldwide and have successfully reduced weekly surveys about people’s behaviours in response to COVID-
transmission of the virus. However, these interventions are 19 (including vaccines) as well as standard demographic ques-
unsustainable in the long-term1 and current hopes to control the tions on age, gender and household structure. Ethics approval
pandemic rely heavily on vaccination. and informed consent were not required given that all data was
In December 2020, the first vaccine against SARS-CoV-2 was publicly available and de-identified.
approved; by May 2021, 14 vaccines had been licensed (https:// We extracted the survey results from 8th to 15th February 2021 for
vac-lshtm.shinyapps.io/ncov_vaccine_landscape/) and more than 10 European countries. To assess vaccine hesitancy, we used data
1.3 billion vaccination doses administered worldwide (https:// from one question pertaining to COVID-19 vaccine acceptance in
ourworldindata.org/covid-vaccinations#). Their reported efficacy which participants were asked to what extent they would definitely
against symptomatic disease ranges from 50% to over 95%2–6. get a COVID-19 vaccine, if it became available to them next week.
Given the high basic reproduction number for SARS-CoV-2 Answers were obtained on a numeric scale ranging from “Strongly
(estimates range between 3–4)1 high levels of vaccine uptake will agree – 1” to “Strongly disagree – 5”. To capture survey uncertainty,
be required to achieve herd immunity7, particularly if children are answers per age group were used to parameterise a multinomial
not vaccinated during the first phase of roll-out. distribution, from which we drew 100 replicates. To capture further
One major concern that threatens to limit the impact of vac- uncertainty associated with the translation of survey response to
cination is vaccine hesitancy8. Population surveys have found that vaccine uptake, for each replicate, coverage per age group was
between 14%9 and 27%10 of adults say that they will not accept a estimated assuming the probability of vaccination as a beta
vaccine if available, whilst between 14%9 and 19%10 say that they distribution with means: 0.98, 0.75, 0.50, 0.25 and 0.02 for survey
are uncertain. There is a large variation in vaccine hesitancy responses 1, 2, 3, 4 and 5, respectively. Coverage distributions per age
between countries, with the proportion saying that they would get group, median as well as the 10% and 90% quantiles are shown in
a SARS-Cov-2 vaccine if it became available, ranging from 40% Table S5 and Fig. S2.
for France10 to 89% for China9. In many countries, vaccine
hesitancy is heterogenous across sub-populations depending on
Mathematical model. We used a previously developed mathe-
gender, age, ethnicity, religion, or socioeconomic status9–11.
matical model for SARS-CoV-2 transmission and vaccination7
Surveys have highlighted the key drivers of SARS-CoV-2 vaccine
(Fig. S1). The age-structured deterministic SEIR-type compart-
hesitancy are related to concerns about the accelerated pace of
mental model incorporates an age-specific probability of infection
vaccine development11, side-effects10 and the spread of mis-
determined by age-based contact matrices. Susceptible individuals
information about the pandemic8. Underlying reasons of vaccine
become infected at a rate that depends on the level of infection in
hesitancy are a complex interaction between trust in government
the community. Following infection, cases proceed to mild infection
and health authorities9 coupled with new information —and
or a clinical disease pathway, which includes hospitalisation, oxygen
misinformation— on the vaccine safety and disease risk arising
support and intensive care. Waning immunity is captured by
everyday12.
recovered individuals returning to the susceptible compartment
In the present study, we aim to understand the likely impact of
following an erlang distribution.
vaccine hesitancy on future control of the pandemic, using a
Vaccination is modelled as an additional dimension disaggregating
mathematical model of SARS-CoV-2 transmission7 to explore
the population into those who have not received the vaccine (v0),
vaccine hesitancy through its impact on population coverage. We
those who have received the vaccine but are not yet protected (this
capture the effect of reduced coverage using measured levels of
stage represents the two-dose vaccine schedule and the need to wait
vaccine hesitancy from behavioural survey data10 on self-
~28 days from dose 1 for protection to develop) (v1 and v2), those
reported intention to be vaccinated. Survey results are dis-
who have received the vaccine and are protected (v3 and v4) and
aggregated by age and translated to vaccination coverage ranges
those who have received the vaccine but are no-longer protected (v5)
per age group. Pandemic trajectories with low vaccination cov-
(if vaccine-derived immunity is not life-long). In this model, only
erage due to vaccine hesitancy are compared to an ideal coun-
those who are currently infected do not receive the vaccine.
terfactual assuming no vaccine hesitancy, in which we assume
Protection due to vaccination is modelled at two stages in the
that a small proportion (5%) of the population cannot be reached
model; (1) reducing the probability of infection upon exposure
for vaccination. This value is based on maximum vaccination
(efficacy against infection) and (2) reducing the probability of
uptake reported for England’s current COVID-19 vaccine rollout
hospitalisation being indicated after developing disease (efficacy
(https://www.england.nhs.uk/statistics/statistical-work-areas/covid-
against hospitalisation and death).
19-vaccinations/). We model each scenario with both a high and a
moderate vaccine efficacy profile that represents the range of effi-
cacies of currently approved vaccines. Informed by current vaccine Parameters. Parameters for SARS-CoV-2 infection, health care
roll-out in high-income countries, we assume that vaccination capacity, age-distribution and contact patterns are based on previous
started in January 2021 and is implemented at a rate that results in work7,13 (Tables S1 and S4). Given these parameters, transmission
a total campaign of 10 months to fully vaccinate the population probability is estimated based on reproductive number (Rt), which is
above 15 years old. given as an input for each simulation as a function of time. Vaccine-
Our simulations suggest that mortality could be higher in induced immunity was assumed lifelong, while natural immunity was
countries with high vaccine hesitancy compared to an ideal assumed to last for an average of one year14. To produce simulations
vaccination and this could prolong the need for NPIs to remain representing the different vaccines approved to date, each scenario
in place. We show that to reduce this impact, vaccination was run for two vaccines: one with high efficacy (94% efficacy against
campaigns could include less vulnerable groups, like children. infection2) and one with moderate efficacy (63% efficacy against
Vaccine hesitancy is an important public health priority infection3). For both vaccines we assume an additional 60% efficacy
that needs to be addressed in order to control the current against hospitalisation for breakthrough infections, resulting in an
pandemic. overall vaccine efficacy against hospitalisation and death of 98% for
2 COMMUNICATIONS MEDICINE | (2022)2:14 | https://doi.org/10.1038/s43856-022-00075-x | www.nature.com/commsmed
COMMUNICATIONS MEDICINE | https://doi.org/10.1038/s43856-022-00075-x ARTICLE
the high efficacy vaccine and 85% for the moderate efficacy vaccine. effect size of changes in mobility on transmission (using mobility
A summary of key parameters is given in Tables S1–S6. The model data from Google (https://www.google.com/covid19/mobility)).
code is freely available at https://github.com/mrc-ide/nimue15. Model fitting was performed using a Metropolis Hastings MCMC
To mimic current vaccine rollout plans, vaccination is based sampling scheme as previously described16. The resulting
introduced in the population at the beginning of January 2021. fit generates a fitted R0 as baseline, an Rt trajectory up to the
We assumed a constant vaccination rate (κ), at which all introduction vaccination in January 2021, after which, Rt was set
individuals aged 15 years and above (~78% of the population) will to increase by 10 fixed steps, up to the theoretical herd immunity
be vaccinated over a 10-month period. This rate is implemented threshold based on an ideal vaccination schedule (as described
for all scenarios modelled, since we assume vaccination rate is above). The pandemic trajectory was evaluated using country-
constrained not by vaccine uptake but by the supply and delivery specific data on vaccine hesitancy and demography for the two
of vaccines. Therefore, lower levels of coverage, result in shorter coverage scenarios described above and assuming vaccination for
vaccination campaigns; given that in the model, once coverage individuals aged 15 years and above only.
targets are met, vaccination is ceased. To illustrate the effect
including children vaccination, vaccination rate was maintained Reporting summary. Further information on research design is
constant and vaccination period was extended such that all available in the Nature Research Reporting Summary linked to
individuals age 5–15 years could be vaccinated. this article.
Vaccines are targeted by age groups at the constant rate κ,
prioritising older age groups: with 80+ years vaccinated first and
then sequentially including additional age groups in 5-year age- Results
bands down to 15–19 years for adults only vaccination Vaccine hesitancy public health impact. We first sought to
simulations and down to 5–10 years for simulations including determine the public health impact of vaccination and vaccine hes-
children vaccination. itancy as NPIs are lifted. To do so, we allowed the time-varying
reproductive number in the absence of immunity Rt, to be increased
in steps such that the herd immunity threshold accounting for
Reproductive number profiles. To simulate a representative pre-
vaccine-induced immunity was maintained, under the assumption of
vaccination scenario, we generated a reproductive number profile in
ideal vaccination uptake (Fig. 1a, c). In this ideal scenario, NPIs can
which Rt was the same as R0 (R0 = 313) up to April 2020, subse-
be fully lifted at the end of the vaccination period with a high efficacy
quently decreased to 1 to represent the impact of NPIs against the
vaccine (94% efficacy, Fig. 1a). However, with a moderate efficacy
first wave, and then rose to 1.5 during the latter half of 2020 to
vaccine (63% efficacy), some NPIs or other population-level beha-
represent a second wave. Following the introduction of vaccination
vioural changes may need to remain to control the epidemic (Fig. 1c).
in January 2021, we set Rt to increase in 10 fixed steps. Each step
In the presence of vaccine hesitancy, lifting NPIs and relying
representing the lifting of NPIs. The time for each step increase was
on vaccine-induced immunity for control is predicted to lead to
determined by estimating when vaccination coverage had reached
periodic outbreaks determined by the duration of naturally
levels such that the herd immunity threshold due to vaccine
induced immunity (Fig. 1b, d). For a high efficacy vaccine, daily
immunity was reached. At the end of the vaccination period, Rt
deaths per million at the peak of the first outbreak are projected
remained at a value such that the herd immunity threshold was
to be 11.5 (10.1–13.2) times higher than under the ideal scenario
maintained, given final vaccination coverage and vaccine efficacy
(Fig. 1b). This translates to a cumulative impact of 532 (457–612)
against infection.
more deaths per million population in the two years after
To estimate the coverage needed for each Rt step, the following
vaccination begins. In our results, fewer deaths are projected for a
herd immunity threshold equation was used:
vaccine of moderate efficacy compared to a higher efficacy
1 1 vaccine. This is partly due prolonged NPIs being required to
Coverage ¼ 1 maintain herd immunity where efficacy is lower, resulting in an
Rt efficacy
outbreak that is more spread out and resulting in a lower final Rt
When analysing the impact of lifting NPIs, the Rt profile compared to the high vaccine efficacy simulations. For a
following the introduction of vaccination was generated based on moderate efficacy vaccine, the cumulative impact of vaccine
an ideal scenario for vaccination uptake. Conversely, when hesitancy is projected to lead to 456 (416–504) extra deaths per
evaluating the degree to which NPIs would need to remain in million population.
place, the Rt profile after the introduction of vaccination was set These adverse impacts of vaccine hesitancy on transmission,
up based on vaccine coverage due to vaccine hesitancy. symptomatic disease, hospitalisations and deaths affect vaccinated
as well as unvaccinated individuals because of imperfect vaccine
Scenarios. We consider two potential scenarios for vaccine cov- efficacy (Fig. 2). Under the vaccine hesitancy scenario, the resulting
erage target per age group: an ideal scenario 2020 with 20 cases. A lower vaccination coverage is projected to lead to a 16.7% and
simulation was run for each vaccine coverage scenario for both 30.4% increase in hospitalisations in the vaccinated population for
adult-only vaccination campaign and vaccination campaign the high and moderate vaccine efficacy profile, respectively, and a
including children. As an output for each simulation, we esti- 9.4% and 27.2% increase in deaths in the vaccinated population,
mated the number of deaths and hospitalisations associated with compared to an ideal vaccination scenario (Fig. 2).
COVID-19 over the 2-year period from 1 January 2021 to 31
December 2022. Relaxation of NPIs. As an alternative way to assess the impact
To generate country-specific simulations, we parameterise the of vaccine hesitancy on the pandemic, we evaluated the degree
model with data on the population size and age distribution of the to which other NPIs would need to remain in place given the
country (https://population.un.org/wpp/) and representative con- real-time achieved vaccine coverage in order to prevent further
tact matrices obtained from a systematic review of social contact epidemics (i.e. maintain herd immunity threshold, Fig. 3). For
surveys through the socialmixR package (https://github.com/ the high efficacy vaccine, under the ideal scenario, we predict
sbfnk/socialmixr). The model was then fitted to reported daily that NPIs could be fully lifted by the end of 2021 whilst keeping
cases and deaths up to 31 December 2020 by varying three transmission under control (Fig. S3). However, under the vac-
parameters - the start date of the epidemic, the initial R0 and the cine hesitancy scenario, limited NPIs or other behavioural
COMMUNICATIONS MEDICINE | (2022)2:14 | https://doi.org/10.1038/s43856-022-00075-x | www.nature.com/commsmed 3
ARTICLE COMMUNICATIONS MEDICINE | https://doi.org/10.1038/s43856-022-00075-x
Fig. 1 Projected COVID-19 dynamics given vaccine hesitancy. Panels a, b show a high vaccine efficacy (94% against infection, 98% against
hospitalisation and death), panels c, d moderate vaccine efficacy (63% against infection, 85% against hospitalisation and death). Panels a and c show the
reproductive number Rt profile, which represents the level of NPI stringency, with lower numbers indicating higher stringency. In this illustrative example,
we assume that a first wave of transmission occurred at the beginning of 2020 with the assumed value of Ro: 3. This was followed by NPIs leading to a
reduction in Rt to 1, followed by an Rt of 1.5 as NPIs are lifted leading to a second wave of transmission in the latter half of 2020. After vaccination is
introduced at the beginning of 2021, NPIs in all scenarios are lifted according to a schedule based on coverage under the ideal scenario (no vaccine
hesitancy, 95% of individuals 15 years plus are vaccinated). Panels b and d show projected deaths per million under vaccine hesitancy scenarios: adults-
only vaccination (orange), vaccination including children (purple). Continuous lines represent simulations of median vaccine coverage per age group, while
dashed lines represent simulation of 10% and 90% quantiles. For the ideal scenario black line represents adults-only vaccination and green line represents
ideal scenario when children vaccination is considered. In each scenario, final vaccination coverage per age group and deaths vary according to vaccine
hesitancy. Vertical dashed lines indicate the vaccination rollout period in the ideal scenario.
modifications might need to remain in place, with Rt having to years after vaccination begins (Fig. S4). For a moderate vaccine
stay below 2.05 (1.96–2.14) to prevent further epidemics, this efficacy, higher NPIs stringency at the end of vaccine rollout
represents a 32% reduction of the assumed R0 of 3. A difference entails later outbreaks, which do not take place during the two
of ~35% in the effective reproductive number could represent years after vaccination begins, resulting in similar results for the
the closure of educational institutions or limiting interaction ideal and vaccine hesitancy scenario when including the vacci-
between households to achieve control of the epidemic;17 both nation of children (Fig. 1d and S4). Including children in vaccine
of which are not sustainable or desirable. rollout leads to higher vaccine coverage that compensates for
vaccine hesitancy levels in adults. This is evident when evaluating
the degree to which other NPIs would need to remain in place in
Vaccination of children. As current vaccination rollout plan of order to maintain the herd immunity threshold based on vaccine-
adults continues swiftly in most high-income countries, public acquired immunity levels. For a high efficacy vaccine, in a vaccine
health authorities are now looking to include children into their hesitancy scenario Rt levels can increase up to 2.5 (Fig. 3b), ~20%
vaccination campaigns while results of COVID-19 vaccine effi- more than for adult-only vaccination rollout. This increase entails
cacy in children become available18. To evaluate the impact of milder NPIs at the end of vaccination campaign.
including children in vaccination rollouts, we model all scenarios
with a longer vaccination campaign, which allowed individuals
above 5 years old to get vaccinated, assuming vaccine hesitancy Country-specific simulations. Our illustrative examples above are
for 5–17 years old the same levels reported for 18–24 years old10. comparable to the waves of COVID-19 outbreaks in Europe.
If children are included in vaccine rollout, our results illustrate However, vaccine hesitancy varies between countries. To evaluate
that in a scenario with vaccine hesitancy daily deaths per million the impact of these variations, we chose three European countries
at the peak of the first outbreak could be reduced by 56% with different vaccine acceptance views: France, Germany and the
(51–60%) for a vaccine with high efficacy (Fig. 1b). Which implies United Kingdom (UK) (Fig. 4b). For each country, we fit the
a total reduction of 272 (242–346) deaths per million in the two pandemic trajectory to country-specific data up to vaccination
4 COMMUNICATIONS MEDICINE | (2022)2:14 | https://doi.org/10.1038/s43856-022-00075-x | www.nature.com/commsmed
ARTICLE COMMUNICATIONS MEDICINE | https://doi.org/10.1038/s43856-022-00075-x
a b
3.0 3.0
2.5 2.5
Rt 2.0 Rt 2.0
1.5 1.5
1.0 1.0
Jun−20Dec−20Jun−21Dec−21Jun−22Dec−22 Jun−20Dec−20Jun−21Dec−21Jun−22Dec−22
Time(days) Time(days)
Fig. 3 Stringency of NPIs required to control the epidemic under different vaccine hesitancy scenarios. Panel a shows Rt profiles for an adults-only
vaccination campaign. Panel b shows Rt profiles for a vaccination campaign including children. Reproductive number profiles are estimated to keep the herd
immunity threshold such that epidemic impact is the same for each scenario as in the ideal scenario. A lower reproductive number corresponds to more
stringent NPIs. Continuous lines represent profiles for a high efficacy vaccine and dashed lines represent profiles for a moderate efficacy vaccine. Vertical
dotted lines show the period of vaccination in the ideal scenario.
Country fitting showed a higher initial Rt compared to our varying12,21. Thirdly, we have assumed vaccination rate remains
illustrative example. These values are consistent with those constant over the vaccination period. However, vaccination
estimated for other European countries, where initial Rt values logistics depend on multidisciplinary factors27 and both vaccine
have been estimated as high as ~4.5, which may be due to pos- availably and uptake can be dynamic. Finally, our model does
sible under-ascertainment in deaths in early periods of the not account for immune escape from the vaccine due to new
pandemics1. It is still unknown how transmission levels will variants arising. Whilst second-generation vaccines will likely
develop in the long term as more transmissible variants are become available to address this issue, it is currently unclear
emerging and NPIs behaviour may persist after the pandemic. whether some of the high levels of vaccine uptake observed in
Here we have assumed a staged release of NPIs with a step-wise early vaccine rollouts would be sustained in subsequent booster
increase of Rt, representing governments’ easing of restrictions. programmes.
This step function is a simplification to illustrate the process of Getting vaccinated is an individual choice, but these individual
balancing the relaxation of NPIs whilst continuing to suppress choices have population wide effects that are likely to challenge
transmission. Nonetheless, the evaluation approaches introduced current efforts to control COVID-19. Our findings suggest that
in this study can be adjusted to include complex Rt dynamics as vaccine hesitancy may have a substantial impact on the pandemic
more information on COVID-19 transmissibility evolution trajectory, deaths and hospitalisation. To prevent such adverse
become available. outcomes, NPIs would need to stay in place longer, or possibly
Our analysis necessarily makes many simplifying assump- indefinitely, resulting in high economic and social costs28,29.
tions, and it is important to note that the future trajectory of the Reducing vaccine hesitancy is therefore an important public health
epidemic will depend on the complex interactions between priority. Interventions that aim to build trust, for example with
vaccination uptake, behaviour and government interventions. community-based public education or via positive role-models, are
First, we have assumed homogenous mixing between vaccine- proven efficacious approaches to address hesitancy30. There is an
hesitant individuals. However, as has been seen for other dis- ongoing debate about vaccine passports as a condition to travel, or a
eases, COVID-19 vaccine hesitancy is heterogenous and clus- vaccination requirement for employees31. Such interventions may
tered within population subgroups24. Transmission is more be effective because they incentivize individuals to get vaccinated,
likely to be sustained within clusters with low vaccine but they are controversial in libertarian democracies because they
coverage25,26 and therefore future outbreaks may be limited to curtail personal freedom and individual choice about medical
these sub-populations. Secondly, we have modelled hesitancy treatments. The alternative will be to accept some level of disease,
levels constant over the time frame analysed; yet, self-reported hospitalisation and deaths given the level of vaccine coverage
attitudes to COVID-19 vaccines are changing over time9,10 as achieved whilst allowing NPIs to be lifted, given that NPIs are not a
the perceived risk for both disease and vaccines keeps sustainable long-term method for control.
6 COMMUNICATIONS MEDICINE | (2022)2:14 | https://doi.org/10.1038/s43856-022-00075-x | www.nature.com/commsmed
COMMUNICATIONS MEDICINE | https://doi.org/10.1038/s43856-022-00075-x ARTICLE
a France Germany U.K.
10.0
7.5
Cumulative deaths ratio
5.0 High
2.5
0.0
10.0
7.5
Moderate
5.0
2.5
0.0
0−5
5−10
10−15 0−5
5−10
10−15 0−5
5−10
10−15
15−20
20−25
25−30 15−20
20−25
25−30 15−20
20−25
25−30
30−35
35−40
40−45 30−35
35−40
40−45 30−35
35−40
40−45
45−50
50−55
55−60 45−50
50−55
55−60 45−50
50−55
55−60
60−65
65−70
70−75 60−65
65−70
70−75 60−65
65−70
70−75
75−80
80+ 75−80
80+ 75−80
80+
Age Group
b c 5
U.K. 4
Country
3
Germany Rt
2
1
France
0
0.0 0.2 0.4 0.6 Jun−20 Dec−20 Jun−21 Dec−21 Jun−22 Dec−22
Vaccine uptake Time(days)
Country France Germany U.K Efficacy High Moderate
Fig. 4 Impact of vaccine hesitancy for three European countries. a Cumulative death ratios per age group compared to the ideal vaccine uptake scenario,
by country and vaccine efficacy profile. The ratio compares cumulative deaths projected over a 2-year period after vaccination starts for two scenarios: an
ideal scenario, where 95% of the population older than 15 years gets vaccinated and a vaccine hesitancy scenario, where coverage for people over 15 years
old is based on vaccine acceptance from b. b Reported vaccine acceptance per age group in France, Germany and the United Kingdom reproduced from
Jones et al10. Values show median vaccine coverage and bars show 10–90% quantiles obtained by running the model at the quantiles from the data.
c Reproductive number profile for country-specific simulations. Profiles, before vaccination begins, are taken from model fittings to country-specific data
(https://mrc-ide.github.io/global-lmic-reports/). After vaccination starts, NPIs are lifted based on an ideal vaccination coverage over time. Reproductive
number is set to increase in ten steps from the value at the beginning of vaccination to an average initial reproductive number. Continuous lines show
profiles for a high efficacy vaccine. Dotted lines show profiles for a moderate efficacy vaccine.
Data availability References
All data used in this study are from publicly available sources at the links provided in the 1. Flaxman, S. et al. Estimating the effects of non-pharmaceutical interventions
main text and references. Vaccine hesitancy surveys are from the Imperial College London on COVID-19 in Europe. Nature 584, 257–261 (2020).
YouGov Covid 19 Behaviour Tracker Data Hub (https://github.com/YouGov-Data/covid- 2. Polack, F. P. et al. Safety and efficacy of the BNT162b2 mRNA Covid-19
19-tracker). For ease of reproducibility of our results, the dataset is also stored in our Vaccine. New Engl. J. Med. 383, 2603–2615 (2020).
associated publicly available Github repository32 so that the modelling outputs can be 3. Voysey, M. et al. Safety and efficacy of the ChAdOx1 nCoV-19 vaccine
reproduced without further data manipulation. Demographic information is from the (AZD1222) against SARS-CoV-2: an interim analysis of four randomised
United Nations Population prospects https://population.un.org/wpp/). Mobility data from controlled trials in Brazil, South Africa, and the UK. The Lancet 397, 99–111
Google (https://www.google.com/covid19/mobility). And model fittings to country-specific (2021).
data are from https://mrc-ide.github.io/global-lmic-reports/results. Source data can be found 4. Baden, L. R. et al. Efficacy and Safety of the mRNA-1273 SARS-CoV-2
in the Supplementary Data file. Vaccine. New Engl. J. Med. 384, 403–416 (2020).
5. Logunov, D. Y. et al. Safety and efficacy of an rAd26 and rAd5 vector-based
heterologous prime-boost COVID-19 vaccine: an interim analysis of a
Code availability randomised controlled phase 3 trial in Russia. The Lancet https://doi.org/
Analyses were carried out in R 4.0.2. Code for the transmission model and analysis is
10.1016/S0140-6736(21)00234-8 (2021).
available on GitHub32. COVID-19 vaccination model code is available at https://
6. Hitchings, M. D. T. et al. Effectiveness of CoronaVac in the setting of high
github.com/mrc-ide/nimue15.
SARS-CoV-2 P.1 variant transmission in Brazil: a test-negative case-control
study. medRxiv https://doi.org/10.1101/2021.04.07.21255081 (2021).
Received: 8 June 2021; Accepted: 18 January 2022; 7. Hogan, A. B. et al. Within-country age-based prioritisation, global allocation,
and public health impact of a vaccine against SARS-CoV-2: a mathematical
modelling analysis. Vaccine https://doi.org/10.1016/j.vaccine.2021.04.002
(2021).
COMMUNICATIONS MEDICINE | (2022)2:14 | https://doi.org/10.1038/s43856-022-00075-x | www.nature.com/commsmed 7
Exhibit H
Open Access Review
Article DOI: 10.7759/cureus.43282
Unveiling the Impact of COVID-19 Vaccines: A
Meta-Analysis of Survival Rates Among Patients
Review began 08/01/2023
in the United States Based on Vaccination Status
Review ended 08/08/2023
Published 08/10/2023 Anderson E. Ikeokwu 1 , Rebecca Lawrence 1 , Egbaoghene D. Osieme 1 , Khalifa M. Gidado 2 , Cullen Guy 3 ,
Oladejo Dolapo 1
© Copyright 2023
Ikeokwu et al. This is an open access
article distributed under the terms of the 1. College of Medicine, Richmond Gabriel University, Kingstown, VCT 2. School of Medicine, All Saints University
Creative Commons Attribution License CC- Dominica, Roseau, DMA 3. College of Medicine, Washington University of Health and Science, San Pedro, BLZ
BY 4.0., which permits unrestricted use,
distribution, and reproduction in any
medium, provided the original author and
Corresponding author: Anderson E. Ikeokwu, ikeokwu.anderson@gmail.com
source are credited.
Abstract
The COVID-19 pandemic caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) has
resulted in a significant number of cases and deaths worldwide. Vaccination is the most effective preventive
measure against the disease. This study aimed to assess the mortality rates of COVID-19 patients in the
United States and the effectiveness of Pfizer (Pfizer, NY, USA), Moderna (Moderna, MA, USA), and Janssen
(Johnson & Johnson, NJ, USA) vaccines in preventing mortality.
A systematic review and meta-analysis were conducted following the Preferred Reporting Items for
Systematic reviews and Meta-Analyses (PRISMA-2020) guidelines. Eligible studies reporting on the
effectiveness of COVID-19 vaccines on patient outcomes were included. The search was performed in
PubMed, Cochrane, and Google Scholar databases. The data were extracted, and risk ratios (RR) were
calculated for mortality outcomes. The analysis was performed using Review Manager software, and bias
assessments were conducted using the Joanna Briggs Institute (JBI) Meta-Analysis tools.
A total of seven studies with 21,618,297 COVID-19 patients were included in the meta-analysis. The odds
ratio (OR) for mortality among unvaccinated patients compared to vaccinated patients was 2.46 (95% CI:
1.71-3.53), indicating that unvaccinated patients were 2.46 times more likely to die from COVID-19.
The findings of this study support the effectiveness of COVID-19 vaccination in reducing mortality among
infected individuals. Unvaccinated patients had a significantly higher risk of mortality compared to
vaccinated patients. Vaccination remains a crucial strategy to mitigate the severity of the disease and reduce
mortality rates. Efforts should be made to address vaccine hesitancy and ensure widespread vaccine
coverage.
Categories: Internal Medicine, Infectious Disease, Public Health
Keywords: covid-19 vaccination, sars-cov-2, systematic review and meta analysis, corona virus disease 2019 (covid-
19), mortality, north america, covid-19
Introduction And Background
Ever since the outbreak of the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in Wuhan,
China, in December 2019, the world has witnessed close to 760 million confirmed cases of COVID-19
infection, and as a result close to seven million deaths as of May 2023. The WHO statistics tabulate 115
million confirmed cases and as a result 1.5 million deaths in the North American Region inclusive of the
United States, Canada, and Mexico [1].
At this time, there continue to be new variants of the COVID-19 virus with increasing complications and
mortality rates, thereby forcing researchers to find innovative ways to combat the disease’s lethality
therapeutically and with preventative efforts. Vaccination continues to be the most accessible and safest
method to prevent future reinfections and improves clinical outcomes in the case of hospitalization.
The US Coronavirus vaccine tracker states that 81% of the population has received at least one dose of the
vaccine, 70% received two doses and are considered fully vaccinated whereas an additional 34% of the
population has received at least one booster dose [2]. However, a sizeable portion of the public is still
reluctant to get the vaccine due to concerns about safety, emergency authorization of these vaccines,
mistrust in their public health systems, or misplaced complacency [3]. Therefore, we conducted a systematic
review and meta-analysis to examine mortality rates of American patients infected with COVID-19 and the
effectiveness of the following vaccines namely Pfizer (Pfizer, NY, USA), Moderna (Moderna, MA, USA) and
Janssen (Johnson & Johnson, NJ, USA) available to the public.
Review
How to cite this article
Ikeokwu A E, Lawrence R, Osieme E D, et al. (August 10, 2023) Unveiling the Impact of COVID-19 Vaccines: A Meta-Analysis of Survival Rates
Among Patients in the United States Based on Vaccination Status. Cureus 15(8): e43282. DOI 10.7759/cureus.43282
Materials and methods
Study Design
To evaluate the acceptance rate of COVID-19 vaccination, a meta-analysis was performed on a collection of
studies. The assessment adhered to the guidelines set forth by the Preferred Reporting Items for Systematic
reviews and Meta-Analyses (PRISMA-2020) [4] to review the relevant articles. As the analysis solely utilized
published data, no ethics review or approval was necessary.
Eligibility Criteria
The criteria for inclusion, include studies that report on the effectiveness of COVID-19 vaccination on
patient outcomes with COVID-19 infection. The criteria included studies after the availability of COVID-19
vaccines. All types of COVID-19 vaccines utilized in the United States were included in this review.
Population (P): We included studies with cross-sectional, case-control, cohort designs and randomized
controlled trials of any age published in English from 2020 to July 10, 2022 from the United States. Case
series/reports, conference papers, proceedings, articles available only in abstract form, editorial reviews,
letters of communication, commentaries, systematic reviews, and qualitative studies were excluded. Articles
in languages other than English or study areas not in the United States were excluded.
Intervention (I): We included all types of COVID-19 vaccines utilized in the United States in this review.
Comparison (C): We included studies that compared the patients into two groups according to their
vaccination status. Individuals who received at least one dose of any COVID-19 vaccine were placed in the
“vaccinated group”; individuals who did not receive any vaccine dose were placed in the “non-vaccinated
group.”
Outcomes (O): Our primary outcome measures mortality due to COVID-19 infection.
Information Sources
A systematic search was conducted on April 21, 2022, utilizing three databases: PubMed, Cochrane, and
Google Scholar. To identify additional relevant studies, a “snowball” search strategy was employed by
examining the reference lists of publications eligible for full-text review and screening studies that cited
them using Google Scholar. The database search was further updated on July 7, 2022, while the snowball and
additional searches were conducted on July 8, 2022.
Search Strategy
The search was done using the generic free-text search terms developed based on the study, Patient-
Intervention-Comparison-Outcome (PICO) model to define the clinical question to aid in finding clinically
relevant evidence in the literature. P = “COVID-19” AND “UNITED STATES,” I = “COVID-19 VACCINE,” C =
“VACCINATION STATUS” OR “VACCINATED” AND “UNVACCINATED,” O = “MORTALITY.” In order to
encompass all possible and relevant studies, a broad range of search terms was utilized. All studies published
between 2020 and July 10, 2022 were gathered to determine their suitability for inclusion in this study. The
search was limited to full-text articles written in English. To identify any additional studies that met the
inclusion criteria, the reference lists of the included citations were carefully examined.
Selection Process
Our search strategy yielded a collection of records that were exported to Rayyan Intelligent Systematic
Review software (Rayyan System Inc., MA, USA) [5]. This software helps ensure data integrity by removing
all duplicate articles. The initial examination of the titles and abstracts of the first 100 records was
conducted independently by two researchers (AI and RL). Any disparities encountered were discussed until a
consensus was reached. Subsequently, the researchers worked in pairs to evaluate the titles and abstracts of
all retrieved articles. In the event of discordance, a consensus on which articles to review in the full text was
achieved through discussion. If necessary, a third researcher (EO) was consulted for assistance in making the
final decision. Afterward, the full-text articles were individually reviewed for inclusion by both researchers
(AI and RL). Again, any differences in opinion regarding inclusion or exclusion were resolved through
discussion. The search methodology employed is depicted in the PRISMA flow chart (Figure 1), which
illustrates the included studies as well as those excluded along with the reasons for exclusion. The reasons
for exclusion included: Reason 1: absence of comparable groups (i.e., vaccinated vs. unvaccinated), Reason
2: unavailability of the complete text, and Reason 3: lack of relevance to the research question,
encompassing insufficient data on patient health outcomes.
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 2 of 12
FIGURE 1: PRISMA flow diagram showcasing the inclusion criteria of
studies found eligible in the meta-analysis
PRISMA: Preferred Reporting Items for Systematic reviews and Meta-Analyses
Data Collection Process
We designed a data extraction form based, on which two review authors (AI and RL) used to extract data
from eligible studies. Extracted data were compared, with any discrepancies being resolved through
discussion. The data was entered into Review Manager (RevMan 2014) Version 5.3 (Cochrane, Copenhagen,
Denmark) [6], double-checking this for accuracy.
Data Outcomes
The data included the first author, the year the study was published, the study's location, its design, the
setting, the characteristics of the COVID-19 patients who participated in the trial and their various
comorbidities, the number of doses, the sample size, the proportion, and information needed to assess the
effect estimates. Death from COVID-19 infection was the specific outcome measure that was recorded for
the meta-analysis. Mortality from SARS-CoV-2 was defined as death within 28 days of first testing positive
for SARS-CoV-2 via PCR test [7]. The effectiveness of a COVID-19 vaccine was referred to in this study as to
how well the vaccine works in preventing COVID-19 infection or reducing the severity of the disease among
vaccinated individuals. It was typically measured in this study by comparing the rates of mortality between
vaccinated and unvaccinated patients with COVID-19 infection.
Effect Measures
The effectiveness of COVID-19 vaccination on patient outcomes with COVID-19 infection was reported in
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 3 of 12
pooled estimate proportion with a 95% confidence interval. We analyzed dichotomous outcomes by
calculating the odds ratio (OR) of a patient outcome (i.e., mortality) for each study.
Synthesis Methods
The analysis was performed with the software RevMan 2014. A generic inverse variance with a random-
effects model was applied to pool the proportion of the studies’ data. The heterogeneity was assessed by I2
statistic and p-value. If the p-value is < 0.05 or I2 > 50%, the assumption of homogeneity was rejected, and a
random-effects model was adopted.
Study Risk of Bias Assessment
The risk of bias assessment was assessed using the Risk of Bias tool 2.0 (RoB 2.0) (Cochrane, London, United
Kingdom) to assess the risk of bias for each of the included observational studies [8]. The evaluation of data
quality was conducted using the Joanna Briggs Institute (J.B.I.) to critically appraise the studies included in
the meta-analysis. The meta-analysis encompassed cross-sectional, case-control, cohort studies, and
randomized clinical trials [9]. The risk of bias in the observational studies (case-control and cohort) was
evaluated using nine criteria [9]: (1) appropriateness of the sample frame, (2) appropriateness of the sampled
study participants, (3) adequacy of the sample size, (4) description of the study subjects and setting, (5)
justification of sample size, power description, or variance and effect estimates, (6) valid methods for
identifying the condition, (7) standardized and reliable measurement of the condition, (8) appropriateness
of statistical analysis, and (9) adequacy of the response rate. The risk assessment criteria were categorized as
“yes,” “no,” “unclear,” or “not available.” A score of one (1) was assigned for “yes” responses, while a score of
zero (0) was given for the remaining categories. The risk of bias was considered low when the total score
exceeded 70%, moderate when it ranged from 50% to 69%, and high when it fell between 0% and 49% [9].
Two authors independently performed the bias assessments.
Rating Evidence of Quality
We used the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) approach to
rate the quality of evidence the clinical outcome (mortality from COVID-19 infection), as high, moderate,
low or very low [10]. The assessment included judgments addressing the risk of
bias, imprecision, inconsistency, indirectness, and publication bias low [10]. If there were serious concerns
in any of these domains (for instance, in risk of bias), we rated the quality of the evidence low [10]. The
GRADEpro Guideline Development Tool (GDT) software (McMaster University, ON, Canada and Evidence
Prime, Kraków, Poland) was utilized to rate the quality of evidence [11].
Results
We identified 250 published papers in database searching. Across all of these papers, there were 21,618,297
COVID-19 patients. A total of 240 articles from PubMed and 10 from the Cochrane database were identified
from the initial search. Following duplicate removal, 167 articles were excluded in accordance with the
inclusion and exclusion criteria. We finally selected seven articles for the meta-analysis (Table 1).
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 4 of 12
Study Study Type Total number of Outcomes
Author (Year) Journal name Vaccine type
Area Design patients analyzed
Naleway Retrospective HR, MV, CC, Pfizer, Moderna,
USA MMWR (CDC) 482,464
(2021) [12] Cohort ICU, MR Janssen (J&J)
Johnson Retrospective
USA MMWR (CDC) 9,678,557 CC, MR Unspecified
(2022) [13] Cohort
Danza (2022) Cross- HR, MV, ICU, Pfizer, Moderna,
USA MMWR (CDC) 422,966
[14] sectional MR, Janssen (J&J)
Olson (2022) The New England Journal HR, VT, ICU,
USA Case-control 1,222 Pfizer
[15] of Medicine MR
Griffin (2021) Cross- HR, MV, CC, Pfizer, Moderna,
USA MMWR (CDC) 43,127
[16] sectional ICU, MR Janssen (J&J)
Tenforde
USA Case-control Jama Network 1,983 HR, MR, Pfizer, Moderna
(2022) [17]
Retrospective Pfizer, Moderna,
Xu (2021) [18] USA MMWR (CDC) 10,987,919 MR
Cohort Janssen (J&J)
TABLE 1: Sample size of selected studies and their characteristics
MMWR: Morbidity and Mortality Weekly Report, CDC: Centers for Disease Control and Prevention, HR: Hospitalization rate, MV: Mechanical ventilation,
CC: COVID-19 cases, ICU: Intensive care unit, MR: Mortality rate, VT: Tidal volume
In a review examining the effectiveness of COVID-19 vaccination on patient outcomes with COVID-19
infection, the authors included a table presenting for each included study the citation, study design, country,
sample size, median age, male: female and ethnicity distribution of vaccinated and unvaccinated patients,
patient comorbidities, and type of COVID-19 vaccine used of various studies have been elaborated in Tables
2-6. In this analysis, mortality in various studies is considered a clinical outcome in patients with COVID-19
infections.
Total number of N (%) N (%) Age Range of
Author (Year) N (%) Female N (%) Male
patients Vaccinated Unvaccinated patients
Naleway (2021) 251,552 230,552
482,464 344,848 (71.5) 137,616 (28.5) 18-75
[12] (52.1) (47.8)
Johnson (2022) 2,866,517
9,678,557 6,812,040 (70.4) - - 18-65+
[13] (29.6)
Danza (2022) 184,134
422,966 281,038 (66.4) 141,928 (33.6) 224,173 (53) 18-80+
[14] (43.5)
Olson (2022) [15] 1,222 345 (28.2) 868 (71.8) - - 12-18
Griffin (2021) [16] 43,127 12,326 (28.6) 30,801 (71.4) 21,743 (50.4) 20,425 (47.4) 16-80+
Tenforde (2022)
1,983 314 (15.8) 1,669 (84.2) 969 (48.9) 1,014 (51.1) 18-65+
[17]
6,398,361 5,946,533 5,041,385
Xu (2021) [18] 10,987,919 4,589,557 (41.8) 12-85+
(58.2) (54.1) (45.9)
TABLE 2: Summary of demographics data
N: Number of patients vaccinated/unvaccinated, %: Percentage of patients vaccinated/unvaccinated in relation to the total number of patients
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 5 of 12
N (%) N (%) Mean/Median N (%) N (%) N (%) N (%) N (%) Native N (%) Native Hawaiian/pacific N (%) Multiple
Author (Year) N (%) Male
Vaccinated Female age White Asian Black Hispanic American islander races/others/unknown
Na eway 187,711 156,960 242,110 22,828 8,224
344,848 50 - 1,2880 (0.4) 1,931 (0.6) 68,475 (19.9)
(2021) [12] (54.5) (45.5) (70.2) (6.6) (2.4)
Johnson
2,866,517 - - - - - - - - -
(2022) [13]
Danza (2022) 154,791 117,971 46,612 26,384 15,991
281,038 36 - 530 (0.2) 2,348 (0.8) 40,538 (14.4)
[14] (55.1) (42) (16.6) (9.4) (5.7)
O son (2022)
345 - - 16 143 (41.4) - 68 (19.7) 94 (27.2) - - 49 (14.2)
[15]
Gr ff n (2021) 6,271 5,908 3,718 1,009
12,326 36 819 (6.6) 3,961 (32.1) 19 (0.2) 49 (0.4) 2,447 (19.9)
[16] (50.9) (47.9) (30.2) (8.2)
Tenforde
314 138 (44) 176 (56) 67 201 (64) - 55 (17.5) 44 (14) - - 14 (4.5)
(2021) [17]
3,448,362 2,949,999 2,778,730 633,212 341,189 1,409,187
Xu (2021) [18] 6,398,361 - - - 880,523 (13.8)
(53.9) (46.1) (43.4) (10) (5.3) (22)
TABLE 3: Summary of demographics data for vaccinated patients
N: Number of patients vaccinated/unvaccinated, %: Percentage of patients vaccinated/unvaccinated in relation to the total number of patients
N (%) N (%) Mean/Median N (%) N (%) N (%) N (%) N (%) Native N (%) Native Hawaiian/pacific N (%) Multiple
Author (Year) N (%) Male
Unvaccinated Female age White Asian Black Hispanic American islander races/others/unknown
Na eway 63,841 73,592 83,474 3,930 4,851
137,616 37 - 588 (0.4) 1,021 (0.7) 43,752 (31.8)
(2021) [12] (46.4) (53.5) (60.7) (2.9) (3.5)
Johnson
6,812,040 - - - - - - - - - -
(2022) [13]
Danza (2022) 69,382 66,163 20,529 7,451 12,319
141,928 35 - 342 (0.2) 1,429 (1) 19,214 (1305)
[14] (48.9) (46.6) (14.5) (5.2) (8.7)
O son (2022) 197
868 - - 15 358 (41.2) - 191 (22) - - 122 (14)
[15] (22.7)
Gr ff n (2021) 15,472 14,517 5,620 4,755 10,183
30,801 32 961 (3.1) 51 (0.2) 161 (0.5) 8,551 (27.8)
[16] (50.2) (47.1) (18.2) (15.4) (33.1)
Tenforde 453
1,669 831 (49.8) 838 (50.2) 53 717 (43) - 381 (22.8) - - 118 (7.1)
(2021) [17] (27.1)
2,498,171 2,091,386 1,982,834 633,212 262,766 1,201,784
Xu (2021) [18] 4,589,557 - - - 508,961 (11.1)
(54.4) (45.6) (43.2) (13.8) (5.7) (26.2)
TABLE 4: Summary of demographics data for unvaccinated patients
N: Number of patients vaccinated/unvaccinated, %: Percentage of patients vaccinated/unvaccinated in relation to the total number of patients
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 6 of 12
N (%) N (%) N (%)
Author Total number N (%) chronic N (%) N (%) Chronic lung
cardiovascular Immunodeficiency Neuromuscular/Neurological
(Year) vaccinated kidney disease Diabetes disease
disease disorder disorder
Naleway
344,848 32 (0.009) 24 (0.007) 24 (0.007) - - -
(2021) [12]
Johnson
2,866,517 - - - - - 10 (0.0003)
(2022) [13]
Danza
281,038 - - - - - -
(2022) [14]
Olson (2022)
345 - 28 (8.1) 81 (23.5) 27 (7.8) - -
[15]
Griffin (2021)
12,326 - - - - - -
[16]
Tenforde
314 - 112 (35.7) 100 (31.8) 236 (75.2) 128 (40.8) -
(2021) [17]
Xu (2021)
6,398,361 - - - - - -
[18]
TABLE 5: Summary of patient comorbidities for vaccinated patients
N: Number of patients vaccinated/unvaccinated, %: Percentage of patients vaccinated/unvaccinated in relation to the total number of patient
N (%) N (%)
Author N (%) N (%) chronic kidney N (%) N (%) Chronic lung N (%) cardiovascular
Immunodeficiency Neuromuscular/Neurological
(Year) Unvaccinated disease Diabetes disease disease
disorder disorder
Naleway
137,616 37 (0.03) 98 (0.07) 22 (0.02) - - -
(2021) [12]
Johnson
6,812,040 - - - - - 15 (0.0002)
(2022) [13]
Danza (2022)
141,928 - - - - - -
[14]
Olson (2022)
868 - 72 (8.3) 241 (27.8) 69 (8) - -
[15]
Griffin (2021)
30,801 - - - - - -
[16]
Tenforde
1,669 - 425 (25.5) 327 (19.6) 814 (48.8) 191 (11.4) -
(2021) [17]
Xu (2021)
4,589,557 - - - - - -
[18]
TABLE 6: Summary of patient comorbidities for unvaccinated patients
N: Number of patients vaccinated/unvaccinated, %: Percentage of patients vaccinated/unvaccinated in relation to the total number of patients
Risk of Bias Assessment
In terms of overall risk bias, the risk of bias was low. There were concerns about the uncertain risk of bias in
two out of the nine criteria for all seven studies included. These two criteria were justification of sample size
and adequacy of response rate. All of the studies did not report enough data to justify the sample size or
assess the adequacy of the response rate. Regarding the adequacy of sample size, one study [15] was at high
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 7 of 12
risk of bias. A summary of these assessments is provided in Figure 2.
FIGURE 2: Risk bias assessment
Analysis of Mortality From COVID-19 by Vaccination Status
The patient outcomes in COVID-19 patients were compared between those who received the COVID-19
vaccine and those who did not. In the seven studies analyzed, a total of 139,485 patients were reported to
have died from COVID-19 infection. The OR of COVID-19 mortality between patients with COVID-19
vaccination versus patient without COVID-19 vaccination was 2.46 with a 95% CI ranging from 1.71 to 3.53.
The result was statistically significant which indicates that unvaccinated patients with COVID-19 infection
are 2.46 times more likely to die from COVID-19 infection compared to those who are vaccinated with
COVID-19 infection (p < 0.0001). A heterogeneity test was done with results of I2 = 100%, p = <0.00001
(Figure 3).
FIGURE 3: Mortality from COVID-19 infection by vaccination status
The figure displays for each study included in the meta-analysis a summary of statistics (number of events and
sample size) for the unvaccinated and vaccinated groups, the Odds Ratio (OR) and its 95% Confidence Interval
(CI), heterogeneity, and test for overall effect for the dichotomous outcome mortality from COVID-19 Infection [12-
18].
GRADE Summary of Findings
Evidence for mortality from COVID-19 infection by vaccination status was available from seven
observational studies included a total of 21,618,297 patients (Figure 4). After rating down one level for study
design, we considered the evidence to be low-quality for an observational study design. These observational
studies suggest that COVID-19 vaccination may substantially reduce mortality (OR 2.46, 95% CI 1.71 to
3.53; low-quality evidence) (Figure 3).
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 8 of 12
FIGURE 4: GRADE summary of findings
GRADE: Grading of Recommendations Assessment, Development, and Evaluation
Discussion
This study evaluated the effectiveness of COVID-19 vaccination on patients with COVID-19 infection in the
United States. In this evaluation, several outcomes were analyzed among individuals within the age range of
12-95 years using the following study design types: retrospective cohort, cross-sectional, randomized
control trial and case-control study. Analysis of these studies reveals that COVID-19 vaccination confers a
certain level of protection against poor outcomes in COVID-19-infected individuals.
The pandemic COVID-19 has had a global impact on mortality and morbidity. Vaccination has been linked
to a considerable decrease in the number of symptomatic COVID-19 infections in adults as well as improved
protection against severe disease [19-21]. Patients who were fully vaccinated were less likely to develop
critical illness and require intensive care and were thus discharged faster [22-25]. Inadequate immunity in
unvaccinated patients, combined with the growing prevalence of the delta variation, resulted in greater
illness and fatality rates [26,27]. As the severity of the disease worsens, mortality rises dramatically [15,16].
Comorbidity risk influences both illness progression and mortality [28,29]. In this study, vaccinated
individuals had considerably reduced mortality than unvaccinated patients.
In line with the majority of the articles used for this meta-analysis study, we noted that unvaccinated
patients infected with COVID-19 are 2.46 times more likely to die from the COVID-19 infection compared to
those that are vaccinated but infected with the virus. Our study is also in support of previous studies such as
Tenforde, where it was noted that among patients hospitalized with COVID-19, the outcomes of death or
invasive mechanical ventilation were associated with a lower likelihood when fully or partially vaccinated
[17]. Moreso, in tandem with the study of Xu et al., it was noted that in a cohort of 6.4 million COVID-19
vaccines and 4.6 million demographically similar unvaccinated persons, recipients of the Pfizer-BioNTech,
Moderna, or Jensen vaccines had lower non-COVID-19 mortality risks compared to the unvaccinated
comparison group [18]. They also noted that there is no increased risk for mortality among COVID-19
vaccine recipients and this finding reinforces the safety profile of currently approved COVID-19 vaccines in
the United States.
The risk of mortality among patients with COVID-19 infection is influenced by their sociodemographic
characteristics, with an increased risk observed among those who are unvaccinated. In a recent study that
compared mortality, recovery rates, and disease severity between men and women using a random-effects
meta-analysis [30], the analysis found that male patients have a higher risk of mortality and a lower chance
of recovery compared to female patients. Additionally, male patients were more likely to present with a
severe form of COVID-19. The male-to-female ratio for cases was 1:0.9 [29]. This study however showed that
there was no striking difference between males and females regarding disease susceptibility. The study
however showed that the course of COVID-19 is more severe in men, but the vaccine may improve the
prognosis in men, as fully vaccinated patients had a significantly higher mean age than unvaccinated and
under-vaccinated patients [30].
Men may be more susceptible to COVID-19 due to differences in innate immunity, steroid hormones, and
sex chromosomal characteristics [30,31]. Females may be at an advantage due to increased TLR7 and CD4+
cell expression, which ensures better elimination of viruses [31]. Male patients infected with COVID-19 have
a 61% greater likelihood of dying from the infection than their female counterparts [31]. Cytokine storms
have been observed as more frequently occurring in men, leading to multi-organ failure and death. Men had
poorer rates of recovery and have longer viral RNA shedding for SARS-CoV-2, implying a slower recovery.
Hormonal variables may also have an impact on disease phenotype and severity [28,32,33].
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 9 of 12
Age is a prognostic factor in determining the risk of mortality in patients with COVID-19 infections. This
study included 486 patients with COVID-19, with 54.3% of them being unvaccinated. The median age was 53
years for unvaccinated and partially vaccinated patients and 62 years for fully vaccinated patients [28].
Critical illness was more prevalent in unvaccinated or partially vaccinated patients, and older age, higher
disease severity, higher comorbidity index, and not being fully vaccinated were factors associated with
higher mortality. The study highlights the importance of vaccination in reducing the severity of the disease
and mortality, particularly in older patients with comorbidities [28].
It has been suggested that racial health disparities have contributed to an increased risk of mortality from
COVID-19 infection. A systematic review and meta-analysis by Pal et al. showed that Native American men
had the highest mortality risk [34]. Studies have also reported higher mortality rates among Black people,
but this study found a similar risk of mortality among Black men compared to White men. These
discrepancies between studies also may be due to different timing of sampling and trends in COVID-19
infection among different racial identity groups [35].
In contrast, another study examined the characteristics and outcomes of COVID-19 patients in California,
Oregon, and Washington across different races/ethnicities [36]. The study found that Hispanic patients were
disproportionately affected and had increased odds of hospital mortality. Other minority races/ethnicities
were not significantly associated with increased mortality [36].
Limitations
There are several key limitations to our study that should be mentioned, such as the fact that the risks of
COVID-19 infection are not the same for everyone, therefore the chance of exposure may influence the
likelihood of COVID-19 vaccine acceptance and coverage. Finally, we were unable to independently assess
the preventative impact of single doses against double and booster doses, as well as independently assess
the effectiveness of the specific vaccines approved in the United States against distinct virus strains and
clinical outcomes. Possible explanations include a lack of consistency in vaccine schedules and availability
in the United States.
Conclusions
The meta-analysis study reviewed here provides evidence that COVID-19 vaccination confers a certain level
of protection against poor outcomes in individuals infected with the virus. The study found that
unvaccinated patients with COVID-19 are 2.46 times more likely to die from the infection compared to those
that are vaccinated. Additionally, the study highlights the importance of vaccination in reducing the severity
of the disease and mortality, particularly in older patients with comorbidities. Based on the findings of this
study, it is recommended that individuals receive the COVID-19 vaccine as a means of protecting
themselves against severe disease and mortality associated with COVID-19 infection.
Governments and health organizations should continue to encourage and facilitate vaccination efforts,
particularly amongst high-risk populations such as the elderly and those with underlying health conditions.
Efforts should also be made to address health disparities in access to and uptake of COVID-19 vaccines to
ensure equitable distribution and protection for all populations. The study also emphasizes the need for data
collection, improving access to testing, and the need for active intervention earlier in the disease course in
addition to culturally appropriate public health messaging. The report also emphasizes the necessity of
racial equity in vaccination distribution as well as the need for diversity in clinical trials to guarantee the
safety and effectiveness of vaccines and therapies.
Additional Information
Disclosures
Conflicts of interest: In compliance with the ICMJE uniform disclosure form, all authors declare the
following: Payment/services info: All authors have declared that no financial support was received from
any organization for the submitted work. Financial relationships: All authors have declared that they have
no financial relationships at present or within the previous three years with any organizations that might
have an interest in the submitted work. Other relationships: All authors have declared that there are no
other relationships or activities that could appear to have influenced the submitted work.
References
1. World Health Organization (WHO) Coronavirus (COVID-19) dashboard . (2023). Accessed: May 15, 2023:
https://COVID19.who.int/.
2. US Coronavirus vaccine tracker. (2023). Accessed: May 15, 2023: https://usafacts.org/visualizations/covid-
vaccine-tracker-states.
3. Troiano G, Nardi A: Vaccine hesitancy in the era of COVID-19 . Public Health. 2021, 194:245-51.
10.1016/j.puhe.2021.02.025
4. Page MJ, Moher D, Bossuyt PM, et al.: PRISMA 2020 explanation and elaboration: updated guidance and
exemplars for reporting systematic reviews. BMJ. 2021, 372:n160. 10.1136/bmj.n160
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 10 of 12
5. Ouzzani M, Hammady H, Fedorowicz Z, Elmagarmid A: Rayyan-a web and mobile app for systematic
reviews. Syst Rev. 2016, 5:210. 10.1186/s13643-016-0384-4
6. The Nordic Cochrane Centre: Review Manager (RevMan) Version 5.3 . The Nordic Cochrane Centre, The
Cochrane Collaboration, Copenhagen; 2014.
7. Finelli L, Gupta V, Petigara T, Yu K, Bauer KA, Puzniak LA: Mortality among US patients hospitalized with
SARS-CoV-2 infection in 2020. JAMA Netw Open. 2021, 4:e216556. 10.1001/jamanetworkopen.2021.6556
8. Sterne JA, Savović J, Page MJ, et al.: RoB 2: a revised tool for assessing risk of bias in randomised trials . BMJ.
2019, 366:l4898. 10.1136/bmj.l4898
9. Ma LL, Wang YY, Yang ZH, Huang D, Weng H, Zeng XT: Methodological quality (risk of bias) assessment
tools for primary and secondary medical studies: what are they and which is better?. Mil Med Res. 2020, 7:7.
10.1186/s40779-020-00238-8
10. Schünemann H, Brożek J, Guyatt G, et al.: GRADE Handbook for grading the quality of evidence and the
strength of recommendations using the GRADE approach. The GRADE Working Group, 2013. (October
2013). Accessed: 8 August 2023: https://gdt.gradepro.org/app/handbook/handbook.html.
11. GRADEpro GDT: GRADEpro Guideline Development Tool [Software]. McMaster University and Evidence
Prime.. (2022). Accessed: June 23, 2023: http://gradepro.org.
12. Naleway AL, Groom HC, Crawford PM, et al.: Incidence of SARS-CoV-2 infection, emergency department
visits, and hospitalizations because of COVID-19 among persons aged ≥12 years, by COVID-19 vaccination
status - Oregon and Washington, July 4-September 25, 2021. MMWR Morb Mortal Wkly Rep. 2021, 70:1608-
12. 10.15585/mmwr.mm7046a4
13. Johnson AG, Amin AB, Ali AR, et al.: COVID-19 incidence and death rates among unvaccinated and fully
vaccinated adults with and without booster doses during periods of Delta and Omicron variant emergence —
25 U.S jurisdictions, April 4-December 25, 2021. MMWR Morb Mortal Wkly Rep. 2022, 71:132-8.
10.15585/mmwr.mm7104e2
14. Danza P, Koo TH, Haddix M, Fisher R, Traub E, OYong K, Balter S: SARS-COV-2 infection and
hospitalization among adults aged ≥18 years, by vaccination status, before and during SARS-COV-2
B.1.1.529 (omicron) variant predominance — Los Angeles County, California, November 7, 2021-January 8,
2022. MMWR Morb Mortal Wkly Rep. 2022, 71:177-81. 10.15585/mmwr.mm7105e1
15. Olson SM, Newhams MM, Halasa NB, et al.: Effectiveness of BNT162B2 vaccine against critical COVID-19 in
adolescents. N Engl J Med. 2022, 386:713-23. 10.1056/NEJMoa2117995
16. Griffin JB, Haddix M, Danza P, et al.: SARS-COV-2 infections and hospitalizations among persons aged ≥16
years, by vaccination status — Los Angeles County, California, May 1 - July 25, 2021. MMWR Morb Mortal
Wkly Rep. 2021, 70:1170-6. 10.15585/mmwr.mm7034e5
17. Tenforde MW, Self WH, Adams K, et al.: Association between mrna vaccination and COVID-19
hospitalization and disease severity. JAMA. 2021, 326:2043-54. 10.1001/jama.2021.19499
18. Xu S, Huang R, Sy LS, et al.: COVID-19 vaccination and non-COVID-19 mortality risk — seven integrated
health care organizations, United States, December 14, 2020-July 31, 2021. MMWR Morb Mortal Wkly Rep.
2021, 70:1520-4. 10.15585/mmwr.mm7043e2
19. The true death toll of COVID-19: estimating global excess mortality . (2021). Accessed: August 8, 2023:
https://www.who.int/data/stories/the-true-death-toll-of-covid-19-estimating-global-excess-mortality.
20. Excess mortality during the Coronavirus pandemic (COVID-19) by Mathieu E, Ritchie H, Rodés-Guirao L, et
al. (2020). Accessed: August 8, 2023: https://ourworldindata.org/excess-mortality-covid.
21. Stokes AC, Lundberg DJ, Elo IT, Hempstead K, Bor J, Preston SH: Assessing the impact of the COVID-19
pandemic on us mortality: a county-level analysis [PREPRINT]. medRxiv. 2021,
10.1101/2020.08.31.20184036
22. Moghadas SM, Vilches TN, Zhang K, et al.: The impact of vaccination on COVID-19 outbreaks in the United
States [PREPRINT]. medRxiv. 2021, 10.1101/2020.11.27.20240051
23. Havers FP, Pham H, Taylor CA, et al.: COVID-19-associated hospitalizations among vaccinated and
unvaccinated adults 18 years or older in 13 US states, January 2021 to April 2022. JAMA Intern Med. 2022,
182:1071-81. 10.1001/jamainternmed.2022.4299
24. Moline HL, Whitaker M, Deng L, et al.: Effectiveness of COVID-19 vaccines in preventing hospitalization
among adults aged ≥65 years - COVID-net, 13 states, February-April 2021. MMWR Morb Mortal Wkly Rep.
2021, 70:1088-93. 10.15585/mmwr.mm7032e3
25. Plumb ID, Feldstein LR, Barkley E, Posner AB, Bregman HS, Hagen MB, Gerhart JL: Effectiveness of COVID-
19 mRNA vaccination in preventing COVID-19-associated hospitalization among adults with previous
SARS-CoV-2 infection - United States, June 2021-February 2022. MMWR Morb Mortal Wkly Rep. 2022,
71:549-55. 10.15585/mmwr.mm7115e2
26. Luo CH, Morris CP, Sachithanandham J, et al.: Infection with the SARS-CoV-2 Delta variant is associated
with higher infectious virus loads compared to the Alpha variant in both unvaccinated and vaccinated
individuals [PREPRINT]. medRxiv. 2021, 10.1101/2021.08.15.21262077
27. Atherstone CJ, Guagliardo SA, Hawksworth A, et al.: COVID-19 epidemiology during Delta variant
dominance period in 45 high-income countries, 2020-2021 [PREPRINT]. Emerg Infect Dis. 2023,
29:10.3201/eid2909.230142
28. Sezen YI, Senoglu S, Karabela SN, et al.: Risk factors and the impact of vaccination on mortality in COVID-
19 patients. Bratisl Lek Listy. 2022, 123:440-3. 10.4149/BLL_2022_068
29. Jabłońska K, Aballéa S, Toumi M: The real-life impact of vaccination on COVID-19 mortality in Europe and
Israel. Public Health. 2021, 198:230-7. 10.1016/j.puhe.2021.07.037
30. Ortolan A, Lorenzin M, Felicetti M, Doria A, Ramonda R: Does gender influence clinical expression and
disease outcomes in COVID-19? A systematic review and meta-analysis. Int J Infect Dis. 2020, 99:496-504.
10.1016/j.ijid.2020.07.076
31. Tufan A, Avanoğlu Güler A, Matucci-Cerinic M: COVID-19, immune system response, hyperinflammation
and repurposing antirheumatic drugs. Turk J Med Sci. 2020, 50:620-32. 10.3906/sag-2004-168
32. Conti P, Younes A: Coronavirus COV-19/SARS-CoV-2 affects women less than men: clinical response to
viral infection. J Biol Regul Homeost Agents. 2020, 34:339-43. 10.23812/Editorial-Conti-3
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 11 of 12
33. Xu K, Chen Y, Yuan J, et al.: Factors associated with prolonged viral RNA shedding in patients with
coronavirus disease 2019 (COVID-19). Clin Infect Dis. 2020, 71:799-806. 10.1093/cid/ciaa351
34. Zhang J, Wang X, Jia X, et al.: Risk factors for disease severity, unimprovement, and mortality in COVID-19
patients in Wuhan, China. Clin Microbiol Infect. 2020, 26:767-72. 10.1016/j.cmi.2020.04.012
35. Pal S, Gangu K, Garg I, et al.: Gender and race-based health disparities in COVID-19 outcomes among
hospitalized patients in the United States: a retrospective analysis of a national sample. Vaccines (Basel).
2022, 10:2036. 10.3390/vaccines10122036
36. Dai CL, Kornilov SA, Roper RT, et al.: Characteristics and factors associated with coronavirus disease 2019
infection, hospitalization, and mortality across race and ethnicity. Clin Infect Dis. 2021, 73:2193-204.
10.1093/cid/ciab154
2023 Ikeokwu et al. Cureus 15(8): e43282. DOI 10.7759/cureus.43282 12 of 12
Exhibit I
Exhibit J
CONFIDENTIAL DOJ_23-CR-158_0000159
Exhibit K
File and source
- File
- gov.uscourts.nyed.495645.58.1.pdf
- Size
- 9,465,699 bytes
- SHA-256
- 5a8473db886321e8e4ded4dfa1e3aa9126229771188d2526fcc1ed3f1116e228
- Original
- archive.org