David Wiseman Testimony

Wiseman PSI Testimony 042726 Page 1 of 10
Consequences of underestimated statistical safety signals, potential risks and suspected adverse
reactions from VAERS associated with COVID-19 and other vaccines
David M Wiseman, PhD, MRPharmS
“Unmasked: How Biden Health Officials Purposely Turned a Blind Eye
Toward COVID-19 Vaccine Safety Signals”
Permanent Subcommittee on Investigations
of the U.S. Senate Committee on
Homeland Security and Governmental Affairs
April 29, 2026
Synechion, Inc., Dallas, TX 75252
David.wiseman@synechion.com
1 OPENING STATEMENT
Distinguished Chairman, Ranking Member, and Committee Members. I am honored to discuss critical limitations in
statistical safety signal and potential risk detection in VAERS, touted by CDC as “the nation’s early warning system
for vaccine safety” and credited by former CBER Director Dr. Peter Marks with “detecting both potential and actual
safety issues and informing vaccine policy decisions.” Among limitations undermining detection of statistical safety
signals is the problem of masking.
Certain data challenges often necessitate deploying surrogate methods for signal detection. CDC was required by
VAERS’ COVID-19 Standard Operating Procedures to use the straightforward Proportional Reporting Ratio (PRR)
method, while FDA would employ Empirical Bayesian modeling, requiring sophisticated software such as Oracle’s
Empirica SignalTM.
Masking occurs when statistical signals for one vaccine in a database are obscured by reports associated with other
vaccines. For example, Moderna’s myocarditis signal becomes hidden if its comparison baseline is artificially
inflated by including Pfizer’s myocarditis reports.
In a simple analogy, hemlock and arsenic appear equally toxic when compared separately with a non-toxic baseline
control such as tea. However, if hemlock’s baseline data include arsenic data, hemlock’s signal is drowned out or
masked, appearing innocuous.
As Chairman Johnson recently highlighted, regulators were well aware of the masking problem. Nonetheless,
officials claimed ignorance on how to address masking, and that developing solutions exceeded FDA’s capabilities.
They considered it “problematic to use a brand new, possibly unvalidated tool in the context of an EUA.”
FDA’s failure to use tools already at their disposal challenge these assertions. In their own 2024 paper about tinnitus,
regulators discussed masking, citing a 2022 seminal publication in the leading journal Drug Safety, noteworthy for
three reasons.
Firstly, its authors included Oracle’s Dr. William DuMouchel, developer of the Bayesian method and Empirica Signal
software, along with FDA biostatistician Dr. Ana Szarfman.
Secondly, it detailed a demasking protocol they developed and available within the same Empirica software used
by FDA.
Thirdly, it applied this protocol to unmask hundreds of statistical signals for the COVID-19 vaccines; a result
consistent with my analysis.

Wiseman PSI Testimony 042726 Page 2 of 10
Despite citing this Drug Safety paper and possessing both software and in-house expertise, regulators failed to
implement this solution to correct their analyses for masking.
This omission impeded detection of perhaps hundreds of signals, including myocarditis and ischemic stroke; a
striking failure given the accelerated, population-scale deployment by Operation Warp Speed of a novel and most
complex class of drugs ever - mRNA.
Signals surviving masking warranted, perforce, urgent investigation. Yet in FDA’s 2025 FOIA release, 163 Janssen
signals evoked minimal discussion.
Compounding the problem, CDC considered that its PRR analyses were rendered “basically redundant” by FDA’s
Bayesian work, ignoring its responsibility to generate PRRs, with relatively simple demasking methods. Accordingly,
the United States lacked a fully functional “early warning system” at the time of its greatest need.
Separately, statistical signals were apparently trivialized by a reluctance to use the term “signal” and recognize a
“statistical signal” as a potential safety signal.
This hedging is incompatible with the “enhanced surveillance” required by VAERS’ Standard Operating Procedures,
hindering investigation of potential risks FDA was legally obliged to consider, even without proven causality. Such
shortcomings impugn decisions regarding authorization, approval, and vaccine injury compensation, and the
marginalization of those apparently injured after receiving a COVID-19 vaccine.
The notion that focusing on VAERS’ shortcomings distracts from current public health priorities is mistaken. Not
only did the system fail to perform as advertised, but, as the Drug Safety paper presciently warned —the COVID-
19 vaccine reporting deluge likely created masking effects that now compromise signal detection for other vaccines.
Regulators must swiftly correct masking and other VAERS deficiencies, heeding earlier warnings of their former
colleagues, Drs. Marks and Jernigan: “failure to detect and communicate a serious health issue related to a vaccine
could impact confidence not only in COVID-19 vaccines but in all vaccines.”
Thank you. I welcome questions.

Wiseman PSI Testimony 042726 Page 3 of 10
2 ANNOTATED VERSION
Distinguished Chairman, Ranking Member, and Committee Members. I am honored to discuss critical limitations in
statistical safety signal and potential risk detection in VAERS, (1-3) touted by CDC as “the nation’s early warning
system for vaccine safety” (4-8) and credited by former CBER Director Marks with “detecting both potential and
actual safety issues and informing vaccine policy decisions.” (9)1 2 Among (e.g. (10)) limitations undermining
detection of statistical safety signals (p18/43 of (11)) is the problem of masking.(12) 3
Empirical Bayesian Data Mining and detection of COVID-19 vaccine safety signals
Certain data challenges often necessitate deploying surrogate methods4 for signal detection. CDC was required by
VAERS’ COVID-195 Standard Operating Procedures to use the straightforward Proportional Reporting Ratio (PRR)
method,(13) while FDA would employ Empirical Bayesian modeling,(14,15) requiring sophisticated software such
as Oracle’s Empirica SignalTM.
What is masking and why is it a problem with Empirical Bayesian Data Mining analyses?
Masking occurs when statistical signals for one vaccine in a database are obscured by reports associated with other
vaccines. (12) For example, Moderna’s myocarditis signal becomes hidden if its comparison baseline is artificially
inflated by including Pfizer’s myocarditis reports.
Simplified analogy of how masking could prevent detection of COVID-19 vaccine safety signals
In a simple analogy, hemlock and arsenic appear equally toxic when compared separately with a non-toxic baseline
control such as tea. However, if hemlock’s baseline data include arsenic data, hemlock’s signal is drowned out or
masked, appearing innocuous.
1 Notably, myocarditis for the modRNA vaccines, and blood clotting issues for Janssen. FDA’s Marks and CDC’s Jernigan proffered: “While not
without limitations, VAERS has proven to be vital in detecting both potential and actual safety issues and informing vaccine policy decisions that
protect the health of the American public. That’s a job the CDC and the FDA take seriously. “We know that failure to detect and communicate a
serious health issue related to a vaccine could impact confidence not only in COVID-19 vaccines but in all vaccines. Bottom line: VAERS works
and has a track record that proves it.”
2 These descriptions dispatch the popular canard of VAERS’ alleged nonutility due to passive and lay reporting. Another canard is that VAERS
is unreliable because it does not prove causality. Indeed, raw VAERS data do not prove causality. However, causality may be determined after
further investigation, including the examination of statistical signals generated from VAERS data. Proof of causality is not required for the
issuance (or non-issuance) of an Emergency Use Authorization (EUA) which requires only FDA’s reasonable belief “that the product may be
effective” based on a “totality” standard which is lower than the regular “safe and effective” standard. This standard requires FDA to consider
“known and potential risks,“ even without proving causality. (Food, Drug and Cosmetic Act, Section 564(c) (2)(B)
www.govinfo.gov/content/pkg/COMPS-973/pdf/COMPS-973.pdf) Emphasis added.
3 These methods are termed “Disproportionality Signal Analyses” (DSA). There are other forms of masking possible with DSA. Additionally, since
DSA methods were developed primarily for signal detection involving classical small-molecule drugs with limited pharmacological effects, the
inherent assumptions underlying the use of DSA may be violated for complex products involving several potential toxicants. This would certainly
be true of the modRNA COVID-19 vaccines, for which toxic effects might arise from the modRNA itself, the lipid nanoparticles (and their
components), RNA-lipid adducts, frameshift proteins, dsRNA, residual DNA, and elicited spike protein. Other issues in the conduct of DSA
include the setting of statistical thresholds. These issues are beyond the scope of this presentation.
4 Because the number of people receiving a drug, and the number of doses they have received is often unknown, it is difficult to calculate an
occurrence rate for a particular adverse event. Accordingly, the total number of adverse events reported is used as a surrogate for the number
of doses or people receiving doses of drug.
5 I use the term “vaccine” to refer to all COVID-19 vaccine types. Although for the Pfizer, Moderna and Jannsen products, the term “pro-vaccine”
more accurately describes their character and distinguishes them from conventional vaccines. Unlike conventional vaccines, these pro-vaccines
do not contain target antigens. Rather, they contain the genetic instructions read by a vaccinee’s body to produce the target spike protein antigen.
This is somewhat analogous to the activation of a pro-drug that lacks the desired pharmacologic action, but is converted by the body to an active
form.

Wiseman PSI Testimony 042726 Page 4 of 10
Were federal health officials able to adjust their data analyses to account for masking?
As Chairman Johnson recently highlighted,(2,16,17)6 regulators were well aware of the masking problem. (footnote
89 in (2))7 Nonetheless, officials claimed ignorance on how to address masking, and that developing solutions
exceeded FDA’s capabilities.(p108 in document 4 of (3)) They considering it “problematic to use a brand new,
possibly unvalidated tool in the context of an EUA.” 8 9
6 Chairman Johnson cites David Menschik MD, MPH who, as of January 2021, was Associate Director for Surveillance Informatics, Division of
Epidemiology/Office of Biostatistics and Epidemiology, CBER, FDA (p179 document 3 of 2026 Johnson release). He authored most of the weekly
email EBDM alerts from January 2021 to mid-2022. The recently released documents contain several drafts for limitation language relating to
masking for the paper by Rosenblum et al. This was included in the preprint version, but removed in the Lancet version, along with mention of
EB data mining. Three versions of this language are contained in an email found on p146 of document 9 from the March 2026 Johnson release.
(Bates PSI-HHS-000008252047, in pertinent part)
From: Menschik, David To: Baer, Bethany Subject: FW: [EXTERNAL] 6 month safety review- approval for CDC clearance- please review and
respond. Date: Thu, 23 Sep 2021
“Attached is the new clean manuscript and there is no redline version per below. I created a “blackline version” of the manuscript (comparing
previous clean version with this clean version; attached) and it appears that there are substantive changes that will require re-clearance.
I’m not satisfied with the late changes to the data mining limitation section. Recent version said:
EB data mining has multiple limitations22 including that an absence of a disproportionality alert does not rule out presence of a safety
problem. Additionally, since most reports received during this surveillance period involved COVID-19 vaccines, disproportionality
scores (which are adjusted by year to control for time-dependent confounders, potentially confounding, exposure and outcome
variables) can be muted by COVID-19 vaccine reports contributing substantially to the comparator group, particularly if both mRNA
COVID-19 vaccines are associated with the same adverse event.
New version says:
A limitation of EB data mining22 is low sensitivity; that is, absence of a disproportionality alert does not rule out a possible adverse
event. A new concern with disproportionality scores, which are adjusted by year to control for time-dependent confounders, is that
during the study period most VAERS reports were for COVID-19 vaccinations. As all VAERS reports are used for vaccine-event
comparisons in EB data mining, potential associations with mRNA COVID-19 vaccines plausibly could be missed.
This is inadequate since there are many limitations to data mining and they are only pointing out ‘low sensitivity’ which is not accurate. If you
agree, would advise revising to:
EB data mining has multiple limitations22,23 including that the absence of a disproportionality alert does not rule out a possible
corresponding adverse event. A new concern with disproportionality scores, which are adjusted by year to control for time-dependent
confounders, is that during the study period most VAERS reports were for COVID-19 vaccinations. As all VAERS reports are used for
vaccine-event comparisons in EB data mining, potential associations with mRNA COVID-19 vaccines plausibly could be missed.
I’m ok with the other data mining parts (including slimming down the results section) and deferring on non-data mining parts of the paper (for
which we were not involved).
Any additional edits or thoughts welcome. Thanks, David”
7 Indeed, they were cautioned by FDA’s Dr. David Menschik against “placing excess value on… the absence of specific data mining alerts.”
8 Bates PSI-HHS-000001136460. In pertinent part. The tinnitus paper referred to is Yih et al.
Email from Nair, Narayan (FDA), Friday, October 27, 2023. To: Su, John (CDC), Bazel, Samaneh (FDA). Cc: Shimabukuro, Tom (CDC)
Subject: FDA coauthor for tinnitus paper
“Hi John,
She does bring up a good point. As you know, data mining has all the limitations of passive surveillance as well as others. However, during the
COVID vaccine era there is an additional limitation. Since most reports received involve COVID-19 vaccines, disproportionately scores (which
are adjusted by year to control for time-dependent, potentially confounding, exposure and outcome variables) can be driven towards the null
by COVID-19 vaccine reports contributing substantially to the comparator group. This would could occur in the setting if there was some type
of class-effect (e.g., if both mRNA COVID-19 vaccines are associated with the same adverse event).
We were aware of this limitation before and during the pandemic. There are many data mining tools and there was some discussion about
utilizing a novel tool to adjust for this. However, we thought it would be problematic to use a brand new, possibly unvalidated tool in the context
of an EUA. We ended up using the same EBGM data mining we use for all vaccines and has a long history of use rather than take an
experimental approach. As new non-COVID vaccine reports are added we think this limitation will be mitigated to some degree.
As far as the paper goes there are several options to address this:
• We could report our data mining findings and just acknowledge this as a limitation (this is what we have done in other papers)
• We could not include any data mining findings
• You could develop another tool that would compensate for the greater number of COVID vaccine reports. I am not sure how to do
this but you would need a statistician with DM experience. This would be beyond our capabilities at FDA.”
9 If this is taken to mean that it might not be possible to validate a new tool, this suggestion is untenable in the light of many examples of FDA
accepting the use of unvalidated, unverified, or low--quality data, such as: a) use of immunobridging methods with no establishment of am
immune correlate of protection to authorize booster for new-variant COVID-19 vaccine authorizations (see comments of Dr. Ofer Levy at the
VRBPAC meeting of October 13, 2021; b) At the VRBPAC meeting of October 26 2021, FDA accepted clinical data it had not verified; c) At the
VRBPAC meeting of June 15, 2022, FDA accepted clinical data it considered “preliminary, imprecise, and potentially unstable” () but
nonetheless “based on the totality of evidence [..] we do feel very confident that the evidentiary standard for benefit for EUA has been met
here. “
It is difficult to understand why validating a statistical method should take more than a few weeks, given the pace of Operation Warp speed
that was able to solve far more challenging issues.

Wiseman PSI Testimony 042726 Page 5 of 10
FDA’s failure to use tools already at their disposal challenge these assertions. In their own 2024 paper about
tinnitus,(18) regulators discussed masking,10 citing a 2022 seminal publication (12) in the leading journal Drug
Safety, noteworthy for three reasons.
Firstly, its authors included Oracle’s Dr. William DuMouchel, developer of the Bayesian method (19) and Empirica
Signal software, along with FDA biostatistician Dr. Ana Szarfman.11
Secondly, it detailed a demasking protocol they developed (20) and available within the same Empirica software
used by FDA.
Thirdly, it applied this protocol to unmask hundreds of statistical signals for the COVID-19 vaccines; a result
consistent with my analysis. (21) 12 13
Despite citing this Drug Safety paper and possessing both software and in-house expertise,(22)14 (p56 of Document
i1 in (3)),15 (reviewed in (21)),16 regulators failed to implement this solution to correct their analyses for masking.
Other data mining limitations
This omission impeded detection of perhaps hundreds of signals,(21) including myocarditis 17 (Figure 1) and
ischemic stroke;18 a striking failure given the accelerated, population-scale deployment by Operation Warp Speed
of a novel and most complex class of drugs ever - mRNA.
What is the significance of signals that survive masking?
Signals surviving masking warranted, perforce, urgent investigation. Yet in FDA’s 2025 FOIA release, 163 Janssen
signals evoked minimal discussion. (21)
10 “VAERS is a system used primarily for signal detection. Since COVID-19 vaccines became available, the majority of VAERS reports have
been for COVID-19 vaccines, which is unprecedented compared to pre-pandemic reporting. This might limit signal detection capability for
COVID-19 vaccines using Empirical Bayesian data mining because disproportionality scores may be driven towards the null [35 – citing Harpaz
et al].”
11 Writing in her personal capacity.
12 For the seven events that were the focus of the Drug Safety paper, Harpaz et al. detected, before demasking, Empirical Bayesian Data Mining
safety signals for eight out of a possible 21 vaccine-adverse event pairs, from the global VAERS database. Using US data, FDA only detected
one of these, missing seven out of eight, a statistically significant failure (p=0.02). See Wiseman, 2025.
13 In addition to demasking, Harpaz et al., used a threshold of 1, instead of the arbitrary threshold of 2. Discussion of this also significant issue
with PRR, Bayesian, and related methods, is beyond the scope of this presentation (see Wiseman, 2025).
14 In addition to Dr. Szarfman, other FDA data mining experts include Drs. Travis Canida (current) and John Ihre (former), statisticians who
developed open-source software for Empirical Bayesian Data Mining.
15 Dr. Brendan Day, working on Empirical Bayesian Data Mining analysis for pneumococcal vaccines, described consideration of data from both
the USA and worldwide, similar to that used in the Drug Safety (Harpaz et al.) paper. Although FDA’s disclosed analysis only considered US
reports, both analyses contribute to the totality of evidence on vaccine safety.
16 Although the Empirica Signal software is able to calculate PRR values, it is unknown which software CDC used to calculate PRR values
disclosed under a FOIA request. These analyses were limited in the time period covered, and in the comparisons made (Pfizer and Moderna
were combined and no analyses were provided for Janssen). These analyses did however substantively account for masking by a) combining
the Pfizer and Moderna reports, and b) comparing these combined data against non-COVID vaccines, presumably excluding Janssen data from
the comparator. (see 8.4.2 in Wiseman, 2025) CDC’s analysis also appears to avoid a different kind of masking that would have been problematic
had CDC complied with its own Standard Operating Procedure. (see 10.1.1 in Wiseman, 2025) How these analyses were used to inform
regulators’ decisions has not been disclosed, but it appears that for the most part they were not. (see p144 in Document 8 of Johnson 2026
release)
17 Taking myocarditis as an example, after demasking Harpaz et al. detected a Bayesian safety signal for Pfizer in the dataset around May 28,
2021, and for Moderna, February 18. When adjustment was also made for a separate statistical problem in which the threshold is set too high,
a Pfizer myocarditis signal was evident March 5, and a Moderna signal for February 5, 2021. Compare these dates with the requests for
information from Israel on February 28, 2021 (p14 in Document 2 submitted to the PSI 5/21/25) and FDA and CDC statements (p 153 in
Document 2 submitted to the PSI 5/21/25). (see Figure 1).
https://www.hsgac.senate.gov/subcommittees/investigations/hearings/the-corruption-of-science-and-federal-health-agencies-how-health-
officials-downplayed-and-hid-myocarditis-and-other-adverse-events-associated-with-the-covid-19-vaccines/
Regarding Pfizer's EUA for its teenage dose, ACIP were told on May 17, 2021 over two months after a signal was evident according to Harpaz
et al., that myocarditis did “not differ[ed] from expected baseline rates.” Finally, a myocarditis signal was announced May 27, 2021. (see Wiseman,
2025)
18 CDC’s substantively demasked PRR analysis (see 8.4.2 in Wiseman, 2025) identified signals for ischemic stroke at least as early as July
2022, despite representations made in early 2023, regarding the VSD signal that there were “no excess reports of stroke from VAERS” and that
“Other safety monitoring systems have not observed similar findings.” (see 9.1.3 in Wiseman, 2025)

Wiseman PSI Testimony 042726 Page 6 of 10
Compounding the problem, CDC considered that its PRR analyses were rendered “basically redundant” by FDA’s
Bayesian work, (p144 in Document 8 of (3)),19 ignoring its responsibility to generate PRRs, (11) with relatively simple
demasking methods.(21) Accordingly, the United States lacked a fully functional “early warning system” (4-7) at the
time of its greatest need.
Separately, statistical signals were apparently trivialized 20 by a reluctance to use the term “signal” (p62 12/1/22 in
Document 1 of (3))21 22 and recognize a “statistical signal” as a potential safety signal. (eg p2 of (1,8)) 23
This hedging is incompatible with the “enhanced surveillance” required by VAERS’ Standard Operating
Procedures,(11) hindering investigation of potential risks FDA was legally obliged to consider, even without proven
causality. Such shortcomings impugn decisions 24 regarding authorization, approval, and vaccine injury
compensation, and the marginalization of those apparently injured after receiving a COVID-19 vaccine.
Ongoing implications
The notion that focusing on VAERS’ shortcomings distracts from current public health priorities is mistaken. Not
only did the system fail to perform as advertised, but, as the Drug Safety paper presciently warned (12)25—the
19 In pertinent part: Email from Dr. Tom Shimabukuro to CDC colleagues (CDC), June 23, 2022. To: Martha Sharan, John Su, cc Kristen Nordlund,
Perstephanie Thompson, Paige Marquez.
“Okay, but the main reason we didn't do PRR is b/c FDA EB data mining is the 'gold standard' for disproportionality analysis so we had a better
and more efficient way of doing disproportionality analysis at a time when we were occupied trying to monitor an onslaught of reports. There
really isn't a reason for CDC to do PRR if FDA is conducting EB data mining b/c it's basically redundant. Now that we are further along in the
pandemic and we better understand some of the limitations of FDA's EB data mining for COVID-19 vaccines we are doing some exploratory
work with PRR, but it's still not a major component of our monitoring.”
20 The VAERS Standard Operating Procedure (p18/43) clearly refers to PRR or EBDM findings as safety signals: “MedDRA terms identified as
safety signals due to elevated PRR and/or a statistically significant finding on data mining will be reviewed as appropriate.”
In the context of Investigational New Drugs, the term “Suspected adverse reaction” appears appropriate, per 21 CFR §312.32 (a):
“Suspected adverse reaction means any adverse event for which there is a reasonable possibility that the drug caused the adverse event. For
the purposes of IND safety reporting, “reasonable possibility” means there is evidence to suggest a causal relationship between the drug and
the adverse event. Suspected adverse reaction implies a lesser degree of certainty about causality than adverse reaction, which means any
adverse event caused by a drug.” at www.ecfr.gov/current/title-21/chapter-I/subchapter-D/part-312/subpart-B/section-312.32
21 In pertinent part (emphasis added). From: Nair, Narayan (FDA) December 1, 2022.
To: Moro, Pedro (CDC) Cc: Alimchandani, Meghna (FDA); Zinderman, Craig E (FDA)
Subject: RE: RE: Weekly data mining
“Hi Pedro,
I hope you are well. FYI - we have shifted some of our responsibilities in our Division so David will no longer be responsible for fielding questions
about data mining. Feel free to contact me if questions come up. With regard to Alison’s question, we have not had any disproportionality alerts
from Data Mining for the mRNA COVID-19 vaccines for any new safety concerns (including none for Parsonage Turner Syndrome.)
[…] We generally try and avoid referring to disproportionality/data mining alerts as “signals” or “safety signals”. From a regulatory
perspective the terms signal and/or safety signal have certain connotations and may trigger actions so we try not conflate data mining alerts with
signals.”
22 In meetings held between the React19.org Patient Advocacy group, and FDA officials, frequent mention is made of the inability to detect
signals of interest from VAERS data. For example see:
https://static.therealpetermarks.com/assets/01_04_22_FDA_PeterMarks_FollowUp_transcript.pdf
https://icandecide.org/wp-content/uploads/2025/04/03_03_2022_FDA_transcript.pdf
https://icandecide.org/wp-content/uploads/2025/04/05_25_22_FDA_transcript.pdf
https://icandecide.org/wp-content/uploads/2025/04/React19-FDA-Transcript-12_14_2022.pdf
23 The 2026 paper by Anyalechi et al., co-authored by CDC and FDA scientists, uses the term “SDR’ which it defines as a "Statistic of
disproportional reporting” and qualifies by saying “SDRs are not necessarily safety signals and should be considered hypothesis- generating. "
Two points are noteworthy. Firstly, the acronym “SDR” is used throughout pharmacovigilance literature to mean “signal of disproportionate
reporting.” The term “statistic of disproportional reporting” does not appear. Secondly, in noting that "A strength of our analysis is that VAERS is
a large national database designed to detect early signals of potential vaccine safety concerns," the paper acknowledges that SDRs as a type
of signal.
24 Importantly, the failure to identify potential safety risks, represented by the masked signals means that FDA failed to adequately inform
VRBPAC and ACIP committees, being asked to consider the “totality of scientific evidence” to make COVID-19 vaccine recommendations that
would inform public policy authorization, approval, and vaccine injury compensation decisions. This extended to health providers and ultimately
to American citizens who would be given these products. Because many countries relied on FDA decisions to make their policy decisions, these
failures affect millions of people around the world.
25 “Furthermore, the volume of reporting for COVID-19 vaccines is likely to influence future statistical associations with other new vaccines.
This suggests that masking may become more frequent and should be carefully considered.”

Wiseman PSI Testimony 042726 Page 7 of 10
COVID-19 vaccine reporting deluge likely created masking effects that now compromise signal detection for other
vaccines. (23),26 (24),27.(25),28 (p17 in document 5 of (3))29
Regulators must swiftly correct masking and other VAERS deficiencies, heeding earlier warnings of their former
colleagues, Drs. Marks and Jernigan: “failure to detect and communicate a serious health issue related to a vaccine
could impact confidence not only in COVID-19 vaccines but in all vaccines.” (9)Error! Bookmark not defined.
Further discussion of masking and related issues welcome
Thank you I welcome questions.30
26 “To mitigate the masking effect of COVID-19 vaccines in post-marketing SRS analyses, we recommend utilizing COVID-19-corrected versions
for a more accurate assessment.”
27 “The impact of COVID-19 vaccine ADR reports on current signal detection practices requires further evaluation and solutions to tackle masking
issues in EudraVigilance may need to be developed.”
28 “COVID-19 vaccines caused substantial masking in both databases.” (Dutch - Lareb database, Spanish – FEDRA database)
29 The following email suggests that the earlier concern regarding the mutual masking of the mRNA products, now extends to the masking by
the COVID-19 vaccines of vaccines in general.
Email (in pertinent part) from: Nair, Narayan [FDA]. Sent: March 15, 2024. To: Menschik, David [FDA] Zinderman, Craig [FDA] Baer, Bethany
[FDA], Subject: Data mining question
“I know in the past we have discussed one of the possible limitations of data mining currently is the vast number of VAERS reports from the
COVID vaccines may limit our ability to detect statistical alerts because disproportionality scores may be driven towards the null. Do you know
if there is a public reference that discusses this limitation? I have found some references that discuss general limitations for data mining but not
sure if there is one that talks about how a large volume of reports from a single class of products could masks results.
Narayan Nair, MD
Division Director, Division of Pharmacovigilance, Office of Biostatistics and Pharmacovigilance [CBER, FDA]”
30 Since these highly technical issues are not easily discussed in this sort of forum, I am happy to do so off-line.

Wiseman PSI Testimony 042726 Page 8 of 10
3 FIGURE: 1
Chronology of myocarditis and pericarditis signal generation by different methods, in comparison to key
regulatory events
Each panel contains three columns representing signals generated weekly by Empirical Bayesian Data Mining
(EBDM). The first column of each panel shows that no signals were detected in FDA’s analysis from US VAERS
data using the EB05>2 criteria at a threshold of 2. (1) The second and third columns of each panel are the signals
detected by Harpaz et al.,(12) The second column uses the same criteria as FDA, but for global data reported to
VAERS. The third column uses the version of EBDM that accounts for masking, (ER05>1) at a threshold of 1.
Orange signifies the presence of a signal. No color, indicates no signal detected. Adapted from Wiseman, 2025
(21) Numbers in each cell are case counts.
4 ATTACHMENTS
The following two works are provided as attachments to this document.
Wiseman D. Signal loss by truancy, masking, and filtering, and underestimation of potential risks and suspected
adverse reactions in the Disproportionality Signal Analyses of VAERS data associated with COVID-19 pro-
vaccines. Research Gate 2025. Epub Sept 9 http://doi.org/10.13140/RG.2.2.16568.40961
Filename: Wiseman2025DSASignalLossVAERS-FOIA-090925PUB.pdf
Wiseman D, Gutschi, LM. Great expectations and generous reading of guidelines underestimate potential risk in
oversight of COVID- 19 pro-vaccine quality, safety testing, and manufacturing: 248 questions for FDA.
ResearchGate 2025. Epub Sept 13 http://doi.org/10.13140/RG.2.2.29026.80324
WisemanGutschi2025FDAIntrospection091325.pdf
An interactive version of this work is available at: https://248questions.com/
5 CONFLICT OF INTEREST
The author has acted as an expert witness or consultant in several legal cases related to COVID-19 vaccines.

Wiseman PSI Testimony 042726 Page 9 of 10
6 ACKNOWLEDGEMENT
I am grateful to the many colleagues whose discussions or collaborations have informed much of the knowledge
that is the basis for this testimony, and to the many patients who have shared their stories of being injured after
receiving a COVID-19 vaccine.
7 REFERENCES
1. FDA. FOIA disclosure to ICAN regarding EBDM and PRR VAERS analyses. 2025 Jan 22. at
https://filedev0128.s3.us-east-1.amazonaws.com/ratio-tables/cdc-proportional-reporting-ratio-tables.zip
https://www.fda.gov/media/184988/download.)
2. Johnson R. Letter to The Honorable Robert F. Kennedy, Jr. Re: COVID-19 Vaccine Safety Risk. 2026 March
23. at https://www.ronjohnson.senate.gov/services/files/CA500350-195E-472C-9F26-BE93B290B9D9.)
3. Johnson R. PSI Chairman Johnson Reveals Further Evidence of Biden Administration Downplaying
COVID-19 Vaccine Safety Risk. 2026 March 25. at https://www.ronjohnson.senate.gov/2026/3/psi-chairman-
johnson-reveals-further-evidence-of-biden-administration-downplaying-covid-19-vaccine-safety-risk.)
4. Shimabukuro T. mRNA COVID 19 bivalent booster vaccine safety update: ACIP. 2023 April 19. at
https://www.cdc.gov/vaccines/acip/meetings/downloads/slides-2023-04-19/03-COVID-Shimabukuro-508.pdf.)
5. Shimabukuro T. COVID-19 vaccine safety update: VRBPAC. 2021 Feb 26. at
https://www.fda.gov/media/146269/download.)
6. Shimabukuro T. Update: Thrombosis with thrombocytopenia syndrome (TTS) following COVID-19
vaccination. ACIP. 2021 May 12. at https://archive.cdc.gov/vaccines/acip/meetings/downloads/slides-2021-05-
12/07-COVID-Shimabukuro-508.pdf.)
7. Shimabukuro T. COVID 19 vaccine safety update: Primary series in young children and booster doses in
older children and adults. ACIP. 2022 September 1. at www.cdc.gov/vaccines/acip/meetings/downloads/slides-
2022-09-01/05-COVID-Shimabukuro-508.pdf.)
8. Anyalechi EG, Marquez PL, Rubin MN, et al. Safety of seasonal inactivated influenza vaccines in children
6 months–17 years of age in the Vaccine Adverse Events Reporting System (VAERS) 2018–2023: before, during,
and after the COVID-19 pandemic public health emergency. Vaccine 2026; 81:128569. Epub
http://doi.org/https://doi.org/10.1016/j.vaccine.2026.128569
9. Marks P, Jernigan, D. VAERS: A Critical Part of the National Vaccine Safety System. 2023 Nov 21. at
https://web.archive.org/web/20231122180604/https://www.fda.gov/news-events/fda-voices/vaers-critical-part-
national-vaccine-safety-system.)
10. Benavides A, Seligmann H. More missing age data in VAERS COVID19 injection reports for severe than
mild adverse events in children and women at peak fertility ages. ResearchGate 2023. Epub July
http://doi.org/10.13140/RG.2.2.14464.00007
11. VAERS. CDC. Vaccine Adverse Event Reporting System (VAERS) Standard Operating Procedures for
COVID-19. 2021 Jan 29. at https://www.documentcloud.org/documents/22119655-cdc-vaers-sop-2021/.)
12. Harpaz R, DuMouchel W, Van Manen R, et al. Signaling COVID-19 Vaccine Adverse Events. Drug Saf 2022;
45:765-80. Epub 20220623 http://doi.org/10.1007/s40264-022-01186-z
13. Evans SJ, Waller PC, Davis S. Use of proportional reporting ratios (PRRs) for signal generation from
spontaneous adverse drug reaction reports. Pharmacoepidemiol Drug Saf 2001; 10:483-6. Epub 2002/02/07
http://doi.org/10.1002/pds.677
14. Szarfman A, Tonning JM, Doraiswamy PM. Pharmacovigilance in the 21st century: new systematic tools for
an old problem. Pharmacotherapy 2004; 24:1099-104. Epub http://doi.org/10.1592/phco.24.13.1099.38090
15. Almenoff JS, LaCroix KK, Yuen NA, Fram D, DuMouchel W. Comparative performance of two quantitative
safety signalling methods: implications for use in a pharmacovigilance department. Drug Saf 2006; 29:875-87. Epub
2006/09/15 http://doi.org/10.2165/00002018-200629100-00005
16. Rosenblum HG, Gee J, Liu R, et al. Safety of mRNA vaccines administered during the initial 6 months of
the US COVID-19 vaccination programme: an observational study of reports to the Vaccine Adverse Event
Reporting System and v-safe. Lancet Infect Dis 2022. Epub 2022/03/11 http://doi.org/10.1016/S1473-
3099(22)00054-8
17. Rosenblum HG, Gee JM, Liu R, et al. Safety Monitoring of mRNA Vaccines Administered During the Initial
6 Months of the U.S. COVID-19 Vaccination Program: Reports to Vaccine Adverse Events Reporting System
(VAERS) and v-safe. medRxiv 2021:2021.10.26.21265261. Epub http://doi.org/10.1101/2021.10.26.21265261
18. Yih WK, Duffy J, Su JR, et al. Tinnitus after COVID-19 vaccination: Findings from the vaccine adverse event
reporting system and the vaccine safety datalink. Am J Otolaryngol 2024; 45:104448. Epub 20240730
http://doi.org/10.1016/j.amjoto.2024.104448

Wiseman PSI Testimony 042726 Page 10 of 10
19. Dumouchel W. Bayesian Data Mining in Large Frequency Tables, with an Application to the FDA
Spontaneous Reporting System. The American Statistician 1999; 53:177-90. Epub
http://doi.org/10.1080/00031305.1999.10474456
20. DuMouchel WaRH. Oracle White Paper: Regression-adjusted GPS algorithm (RGPS). 2012. at
https://docs.oracle.com/health-sciences/empirica-signal-811/ESIUG/RegressionAdjusted_GPS_Algorithm.pdf.
2012.)
21. Wiseman D. Signal loss by truancy, masking, and filtering, and underestimation of potential risks and
suspected adverse reactions in the Disproportionality Signal Analyses of VAERS data associated with COVID-19
pro-vaccines. Research Gate 2025. Epub Sept 9 http://doi.org/10.13140/RG.2.2.16568.40961
22. Canida T, Ihrie, J. openEBGM: An R Implementation of the Gamma-Poisson Shrinker Data Mining Model.
The R Journal 2017; 9:499-519. Epub http://doi.org/10.32614/RJ-2017-063
23. Montes-Grajales D, Garcia-Serna R, Mestres J. Impact of the COVID-19 pandemic on the spontaneous
reporting and signal detection of adverse drug events. Sci Rep 2023; 13:18817. Epub 20231101
http://doi.org/10.1038/s41598-023-46275-w
24. Micallef B, Dogné JM, Sultana J, et al. An Exploratory Study of the Impact of COVID-19 Vaccine
Spontaneous Reporting on Masking Signal Detection in EudraVigilance. Drug Saf 2023; 46:1089-103. Epub
20230914 http://doi.org/10.1007/s40264-023-01346-9
25. Rachwal O, Gutiérrez-Lobón M, Cueto NS, et al. Evaluating COVID-19 Vaccine Masking and Unmasking
Methods in Two National Pharmacovigilance Databases. Drug Saf 2026. Epub 20260107
http://doi.org/10.1007/s40264-025-01644-4