Document text
Vaccine safety signal detection and evaluation
John Su
Immunization Safety Office
Advisory Committee on Immunization Practices meeting
September 19, 2025
Outline
•Definitions
•Approach to signal detection and evaluation
•Example
-Previous statistical signals for ischemic stroke in the Vaccine Safety Datalink (VSD)
Definition: adverse event following immunization (AEFI)
•Any untoward medical occurrence which follows immunization and which
does not necessarily have a causal relationship with the usage of the
vaccine. The adverse event may be any unfavorable or unintended sign,
abnormal laboratory finding, symptom, or disease.
Notes: “Immunization” as used in these definitions means the usage of a vaccine for the purpose of immunizing individuals. “Usage” i ncludes all processes that occur after a vaccine
product has left the manufacturing/packaging site, i.e. handling, prescribing and administration of the vaccine.
Source: Council for International Organizations of Medical Sciences. Definition and application of terms for vaccine pharmacovigilance: report of CIOMS/WHO Working Group on Vaccine
Pharmacovigilance. Geneva: CIOMS, 2012.
Definition: signal
•Information that arises from one or multiple sources (including
observations and experiments) which suggests a new potentially causal
association, or a new aspect of a known association, between an
intervention and an event or set of related events, either adverse or
beneficial, that is judged to be of sufficient likelihood to justify
verificatory action.
Source: Council for International Organizations of Medical Sciences. Definition and application of terms for vaccine pharmacovigilance: report of CIOMS/WHO Working Group on Vaccine
Pharmacovigilance. Geneva: CIOMS, 2012
Vaccine safety signal pathway
*Adapted from: Practical Aspects of Signal Detection in Pharmacovigilance: Report of CIOMS Working Group VIII –CIOMS . (2010)Signal detection
(and assessment )
Signal Verified Signal Ruled OutHypothesis testing study for adverse event
risk after vaccine
Evaluate biologic mechanisms and prevention strategiesCommunication and education,
regulatory or public health action,
possible further studies
CDC’s Immunization Safety Office Monitors Vaccine Safety
Through Strong, Complementary Systems
VAERS VSD CISA Project V-safe
Systems work together to rapidly detect and assess potential
safety concerns to help inform public health actions 1990 1990 2001 2020
Evaluation following previous statistical signals for
ischemic stroke in the Vaccine Safety Datalink (VSD)
VSD •13 integrated healthcare organizations, covering >15.5
million people per year
•Active monitoring using electronic medical records (EMR)
and chart reviews
•Rapid monitoring for pre -specified events as well as
monitoring for unexpected events
•Can detect and assess safety signals
•Develops innovative methods for monitoring safetyVaccine Safety Datalink (VSD)
Collaborative Model for High -Quality Vaccine Safety Data
•Links vaccination data to health outcomes Vaccine Safety Datalink (VSD)
Vaccination recordsHealth care visits
•Hospital
•Emergency dept.
•ClinicPatient characteristics
Linked by study IDs
Data are linked and kept at each site, not at CDC
•Weekly sequential monitoring as data accrue
•Monitors a limited set of prespecified adverse events of special interest
-Potential cases identified using ICD -10 codes for healthcare encounters
•Designed to detect statistical signals
-A statistical signal is a finding from an analysis where a calculated value (i.e., the
test statistic) exceeds a specified statistical threshold
-A statistical signal does not necessarily represent a vaccine safety problem and requires further assessment before conclusions can be drawn.
•Since 2020, VSD has utilized a vaccinated concurrent comparator methodVaccine Safety Datalink (VSD)
Rapid Cycle Analysis (RCA) surveillance
Vaccinated Concurrent Comparator Method
COVID -19 vaccinated
Oct 30th
COVID -19vaccinated
Sept 20th Nov 3rd
1 21
43 63On each calendar day that an outcome
occurs in a vaccinee (e.g., Nov 3rd), we
compare vaccinees in their risk interval (1 -
21 days) with similar vaccinees in their
comparison interval (e.g., 43 -63 days).
‘Similar’ means that people were in the
same age group, sex, race/ethnicity, and
VSD site.
Inherently adjusts for calendar time.Risk interval 1 -21 days post -vaccination
Comparison interval 43 -63 days post -vaccination
Ref: Klein N, et al. Kaiser Permanente Northern California –
https://jamanetwork.com/journals/jama/fullarticle/2784015Nov 3rd
Vaccine Safety Datalink (VSD)
Statistical signal detection and evaluation
•Statistical signal detection: RCA concurrent comparator analysis
-Intended to be rapid
-Might have residual bias
•Statistical signal evaluation: self -controlled case -series (SCCS) analysis
-Requires waiting for complete accrual of follow -up time
-Less prone to bias
Core list of adverse events of special interest monitored by U.S. federal agencies, 2020 -2023
,
https://www.sciencedirect.com/science/article/pii/S0264410X2400224X?via%3Dihub
History of VSD RCA statistical signals for ischemic
stroke following mRNA COVID -19 vaccines
mRNA vaccine formulaVSD RCA
statistical signal for
ischemic stroke
2020 -21original, primary series No
2021- 22original, booster dose No
2022- 23b ivalent (original and BA.4/BA.5) Yes
2023- 24XBB.1.5 Yes
2024- 25KP. 2 No
VSD RCA ischemic stroke after Pfizer -BioNTech bivalent booster,
age ≥65 years, counts and adjusted rate ratios,
Oct 16, 2022– April 8, 2023
This information was previously presented to the Advisory Committee on Immunization Practices (ACIP) meeting on April 19, 202 3
RR= 1.26 (95% CI 0.99– 1.60)
Red dot represents sequential signal: p- value <0.01 (1- sided)
Initial VSD statistical signal for ischemic stroke
January 13, 2023
May 31, 2023 update
•“Other safety monitoring systems have not observed similar findings.”
•“As time has passed and more safety data have accumulated, the initial
finding has decreased, and scientists believe factors other than vaccination might have contributed to the initial finding.”
•“The current evidence does not support the existence of a safety issue.”
CDC and FDA Identify Preliminary COVID -19 Vaccine Safety Signal for Persons Aged 65 Years and Older | FDA
History of VSD RCA statistical signals for ischemic
stroke following mRNA COVID -19 vaccines
•2022- 2023 (bivalent vaccine)
-Pfizer COVID -19 vaccine, age ≥65 years
•Question about effect of same day administration with influenza vaccine
•2023- 2024 (XBB.1.5 vaccine)
-Moderna COVID -19 vaccine, age ≥65 years
-Pfizer COVID -19 vaccine, age 50 -64 years
•Additional information that was previously pending
-A follow -up VSD analysis using self- controlled case -series ( SCCS) analysis
-FDA’s 2023- 2024 COVID -19 vaccine safety surveillance using Medicare claims
database results
VSD SCCS analysis of ischemic stroke after Pfizer COVID -19
vaccine with or without influenza vaccine, 2022- 2023
No statistically significant increased risks found for either separate or
same day administrationIRR (95% Confidence Interval)
21-d ay risk interval 42-d ay risk interval
A ge groupCO VID -19
vaccination
without
influenza
vaccinationSam e -day
coad ministration
o f CO VID -19 and
influenza
vaccinationsCO VID -19
vaccination
without
influenza
vaccinationSam e -day
coad ministration
o f CO VID -19 and
influenza
vaccinations
PfizerOverall 0.87 (0.58–1.29) 1.14 (0.88–1.47) 0.83 (0.58–1.19) 1.12 (0.91–1.37)
12 - 64 years 0.92 (0.45–1.90) 0.88 (0.50–1.54) 0.93 (0.51–1.70) 1.06 (0.71–1.60)
65+ years 0.85 (0.53–1.37) 1.25 (0.94–1.68) 0.79 (0.51–1.23) 1.15 (0.91–1.45)
VSD SCCS analysis of ischemic stroke after mRNA COVID -19
vaccine with or without influenza vaccine, 2023- 2024
No statistically significant increased risks found for either separate or
same day administrationIRR (95% Confidence Interval)
21-d ay risk interval 42-d ay risk interval
A ge groupCO VID -19
vaccination
without
influenza
vaccinationSam e -day
coad ministration
o f CO VID -19 and
influenza
vaccinationsCO VID -19
vaccination
without
influenza
vaccinationSam e -day
coad ministration
o f CO VID -19 and
influenza
vaccinations
PfizerOverall 1.19 (0.65- 2.17) 0.90 (0.64- 1.28) 1.12 (0.67- 1.89) 1.08 (0.79- 1.46)
12 - 64 years 0.60 (0.08- 4.64) 1.17 (0.62- 2.21) 0.89 (0.24- 3.41) 1.23 (0.69- 2.17)
65+ years 1.34 (0.71- 2.53) 0.83 (0.55- 1.26) 1.20 (0.68- 2.11) 1.06 (0.73- 1.53)
ModernaOverall 2.11 (0.81- 5.53) 1.41 (0.35- 5.76) 1.29 (0.48- 3.46) 1.13 (0.33- 3.93)
12 - 64 years 2.73 (0.51- 14.52) 0.70 (0.05- 10.67) 1.53 (0.30- 7.80) 0.39 (0.01- 10.17)
65+ years 1.94 (0.60- 6.30) 1.96 (0.43- 9.03) 1.22 (0.36- 4.16) 1.56 (0.46- 5.24)
FDA surveillance: Safety Monitoring of Multiple Health Outcomes Following
2023– 2024 COVID -19 Vaccination among Medicare Beneficiaries Aged 65
Years and Older in the United States
•Methods:
-Medicare Fee- for-Service (FFS) claims database from September 2023 to April 2024
-Self-controlled case series design
•Results:
-Approximately 7.6 million Medicare FFS beneficiaries received a 2023 –2024 COVID -19 vaccination
-Non -hemorrhagic stroke or transient ischemic attack
•Pfizer: IRR: 1.03 [99% CI: 0.91, 1.17]
•Moderna: IRR: 0.96 [99% CI: 0.85, 1.10]
•Novavax: IRR: 0.98 [99% CI: 0.19, 5.13]
•Interpretation: “we did not find evidence for an elevation in non- hemorrhagic stroke risk
following vaccination with any of the vaccine brands”
https://www.medrxiv.org/content/10.1101/2025.01.03.25319975v1
Summary
•The Vaccine Safety Datalink (VSD) rapid cycle analysis (RCA) surveillance
monitored a list of pre -specified adverse events of special interest after
COVID -19 vaccines, including ischemic stroke
•The VSD detected statistical signals for ischemic stroke during 2022- 2023
and 2023- 2024
•Signal evaluation using self -controlled case- series analyses in the VSD and
in other data sources have not confirmed an increased risk of ischemic stroke
For more information, contact CDC
1-800- CDC- INFO (232 -4636)
TTY: 1 -888- 232- 6348 cdc.gov
Follow us on X (Twitter) @CDCgov & @CDCEnvironment
The findings and conclusions in this report are those of the authors and do not necessarily represent the official position
of the U. S. Centers for Disease Control and Prevention.