2

Data Monitoring and Evaluation

During the COVID public health emergency (PHE), the Centers for Disease Control and Prevention (CDC) Immunization Safety Office (ISO) played a critical role in monitoring and evaluating the safety of vaccines administered under unprecedented conditions. Leveraging existing surveillance systems and rapidly scaling new monitoring platforms, ISO implemented a comprehensive monitoring and evaluation framework (Klein et al., 2021; Shimabukuro et al., 2021a). This chapter outlines ISO’s data-driven efforts to detect, assess, and respond to potential safety concerns in real time, highlighting the integration of information from traditional systems, like the Vaccine Adverse Event Reporting System (VAERS) and Vaccine Safety Datalink (VSD), with enhanced tools, such as V-safe and the Clinical Immunization Safety Assessment (CISA) Project.

The safety monitoring infrastructure described in this report informed public health policy makers throughout the COVID vaccination campaign. Early identification of signals of potential adverse events (AEs), such as thrombosis with thrombocytopenia syndrome (TTS), myocarditis, and Guillain-Barré syndrome (GBS), allowed for timely clinical guidance, reduced preventable morbidity, and informed risk–benefit assessments for distinct populations (Greinacher et al., 2021; Montgomery et al., 2021; See et al., 2021). Using a range of methods—from passive surveillance in VAERS to active, population-based analyses of VSD—policy makers and scientists were able to triangulate signals, confirm or refute potential associations, and refine vaccination recommendations in real time (McNeil et al., 2014). While comprehensive and effective monitoring efforts were evident, the complex nature of the data from the evaluation programs sometimes

challenged the ability to clearly and succinctly communicate findings to the public and health care providers (HCPs), limiting the potential to do even greater good for the public’s health (Salmon et al., 2021).

Vaccine safety research spans a spectrum of study designs that vary in their ability to detect different categories of AEs. Randomized controlled trials and early-phase studies are optimized to reveal common, often immunologically mediated reactogenic events that occur within hours to days (Baden et al., 2021; Polack et al., 2020); however, their modest sample sizes (typically ≤50,000) lack the statistical power to uncover events with incidence below ~1 per 10,000 doses (Black et al., 2009). Large postauthorization observational designs—self-controlled case series, cohort analyses, and population-level data linkage studies—are therefore essential for estimating the incidence of less-common outcomes (1 per 10,000–100,000), such as myocarditis or anaphylaxis (Klein et al., 2021; Petersen et al., 2016). VSD was designed to perform these large postauthorization studies; however, extremely rare events (<1 per 100,000), like TTS or GBS, require even broader, often multinational data pools, case-control networks, and rapid cycle analyses (RCAs) of health care databases that aggregate tens of millions of vaccinated person-years to achieve adequate signal-to-noise ratios (Faksova et al., 2024). Finally, mechanistic bench-to-bedside investigations, although infrequently funded, are critical to elucidate causal pathways once epidemiologic associations are flagged (Das, 2023; Moro et al., 2019). Mechanistic investigations are rare, so the biology underlying most serious, vaccine-associated AEs remains unknown.

An essential component of these vaccine safety evaluations was the ongoing review and oversight provided by the Advisory Committee on Immunization Practices (ACIP) and its COVID-19 Vaccines Safety Technical (VaST) Work Group. Building on foundational vaccine safety studies ranging from preclinical trials to Phase 4 postauthorization evaluations, ACIP and VaST functioned as critical technical groups tasked with independently reviewing data generated by ISO’s postauthorization surveillance systems. VaST regularly assessed emerging safety signals, interpreted complex data from diverse surveillance mechanisms, and communicated these findings during ACIP meetings (Markowitz et al., 2024). The goal of these deliberations was to inform timely updates to vaccine recommendations to provide alignment with the latest safety data and public health needs.

Beyond the immediate pandemic context, the integrated approach to vaccine safety monitoring exemplified by ISO, ACIP, and VaST underscores the importance of a multilayered surveillance program with coordinated and complementary strategies for detecting and addressing both common and extremely rare AEs. The continuous feedback loop—emerging signals

prompting further epidemiologic studies and expert clinical consultation—ensures that vaccine policy decisions remain evidence based and responsive to evolving real-world conditions. Ultimately, the goal of this adaptive, science-driven safety network is to provide a foundation for sustained improvements in vaccine safety oversight and rapid global health responses moving forward.

ISO MONITORING AND EVALUATION/ASSESSMENT SYSTEMS

To monitor the safety of COVID vaccines during the PHE, CDC leveraged a coordinated network of complementary surveillance systems. Each system played a distinct role within the broader safety infrastructure—ranging from early signal detection to in-depth clinical evaluation. In this report, we use “signal detection” to refer to identifying a potential vaccine safety concern based on early surveillance data—typically in passive or high-throughput systems, such as RCAs done by VSD. “Signal evaluation” encompasses the follow-up activities used to further characterize the strength, direction, and potential causal nature of an observed association, including what VSD reports traditionally describe as signal refinement and evaluation. While we recognize that these phases can be conceptually distinct, they often occur along a continuum and are treated as part of an integrated process for assessing vaccine safety. VAERS served as the primary tool for signal detection, flagging potential AEs through standardized analyses of spontaneous reports from throughout the United States. VSD was used for signal detection through standardized RCAs and data-mining approaches, like TreeScan, but also the primary tool for population-based signal evaluation using electronic health records from a set of 11 large, integrated health care systems.¹ In addition to consultative activities, the CISA Project provided technical expertise and clinical guidance to evaluate rare or complex cases that required specialized review. CDC also launched V-safe, a new smartphone-based, active-surveillance platform designed to rapidly collect self-reported, postvaccination health experiences, particularly common local and systemic reactions, supporting a registry of vaccinated individuals. To monitor outcomes among pregnant individuals, the V-safe COVID-19 Vaccine Pregnancy Registry was established as a targeted follow-up system to track maternal and infant outcomes over time. Together, these systems formed a multilayered approach to identifying, assessing, and responding to vaccine safety concerns in near real time.

___________________

¹ VSD has 13 participating sites; 11 provide data, and the remaining two offer subject-matter expertise.

VAERS

VAERS is a national passive surveillance program comanaged by CDC and the Food and Drug Administration (FDA). Established in 1990, VAERS aims to detect early signals of possible vaccine-related AEs (CDC, 2024a). HCPs and vaccine manufacturers are legally mandated to report specific postvaccination events, and patients or caregivers may also submit reports. Although these data are crucial for generating hypotheses on vaccine safety concerns, they are usually insufficient to determine causality (Gee et al., 2024). VAERS may provide reasonable certainty for events such as anaphylaxis within minutes of administration (Shimabukuro et al., 2021a) or previously unrecognized diseases (such as TTS) shortly after vaccination (Shay et al., 2021). However, the determination of rates of AEs attributable to vaccines needs to be performed in data systems that do not rely on selective passive reporting. In addition, for most other events, data collected through active-surveillance systems and controlled epidemiological designs are required to help establish a causal relationship between a vaccine and an AE.

During the PHE, VAERS played a particularly prominent role in safety monitoring. VAERS data were reviewed in near real time. The heightened public interest in COVID vaccine safety led to increased transparency around VAERS data, with CDC and FDA issuing frequent public communications, updating online dashboards, and providing summaries of key findings. When signals like myocarditis and TTS were identified, more detailed investigations were done using both VAERS and VSD and sometimes led to revised vaccination guidance (Gargano et al., 2021; Shay et al., 2021).

VAERS collects and codes all spontaneously reported AEs among the vaccinated. Prespecified AEs of special interest (AESI) were selected for enhanced safety monitoring based on biological plausibility, previous vaccine safety experience, and theoretical concerns related to COVID vaccines. By protocol, medical records and autopsies were requested for AESIs and all serious events (e.g., death) for further clinical investigation. In addition to the prespecified AEs, symptoms identified during this enhanced surveillance were added to VAERS as AESIs only for COVID vaccines (Oliver et al., 2020).

VAERS, as a passive surveillance system, is inherently limited by underreporting, variability in data completeness, and differential reporting across time and populations (Shimabukuro et al., 2015; Varricchio et al., 2004). While some reports may be incomplete or lack medical confirmation, outright fabrication is believed to be rare (CDC, 2024a; HHS, 2024). A more common limitation is preferential or stimulated reporting—where known or suspected AEs are more likely to be reported in the context of a new vaccine rollout, media attention, or scientific publications. In addition, it

lacks control groups of nonvaccinated persons, making it difficult to assess whether an AE is simply consistent with background rates (Varricchio et al., 2004). However, within the broader vaccine safety ecosystem, VAERS serves a vital function in the earliest stages of monitoring by highlighting patterns that warrant deeper study (Shimabukuro et al., 2015).

The VAERS experience during the COVID pandemic underscores its adaptability and value. The required reporting of vaccine denominator information (doses administered), expanded reporting mandates, and real-time analytical methods all contributed to more robust safety surveillance (Zou et al., 2022). While VAERS alone cannot establish causation, its ability to detect safety signals—especially when integrated with other active-surveillance platforms that can further evaluate them—proved critical for rapidly identifying potential concerns informing further investigation and evidence-based policy (Gargano et al., 2021).

VAERS summary-level data are publicly downloadable through CDC WONDER, but case-level narratives, medical records, and personally identifiable information are protected. Investigators seeking detailed, deidentified case-level datasets must submit a formal data request to CDC/FDA, sign a data-use agreement (DUA), and obtain Institutional Review Board (IRB) or other ethics approval. Aggregate surveillance tables and weekly data-mining outputs are posted on the CDC VAERS website and updated dashboards.

Analysis of VAERS data requires methodology that accounts for spontaneous reporting. For example, VAERS investigators have used an estimate called the “proportional reporting ratio” (PRR) to detect AEs that occur more frequently than expected after a vaccine. For COVID vaccines, this involves comparing the proportion of a specific AE among all reported AEs to the proportion of that same AE among AEs reported for other vaccines. If the AE occurs more often with the COVID vaccine, it may signal a safety concern. The PRR can be further analyzed by factors like age group, AE severity, and vaccine type. A potential safety signal is typically defined as a PRR of 2 or higher, a chi-squared value of at least 4 (indicating that it is unlikely to be due to chance), and at least three reports of the AE for that vaccine (Shimabukuro et al., 2015). FDA uses a related method—Empirical Bayesian data mining—to identify signals when the lower bound of the 95 percent confidence interval of the empirical Bayes geometric mean exceeds a predefined threshold. However, both CDC and FDA interpreted these methods with caution during the COVID response. The utility of PRR was limited because of enhanced and widespread reporting for COVID vaccines, which made historical comparisons less meaningful. As a result, most safety assessments relied on reported rates of specific AEs—calculated as the number of reports per doses administered—rather than PRR (Sakaeda et al., 2013). This approach was newly possible due to valuable national tracking of administration. Before the pandemic, crude rates of AEs could

be only approximated based on estimated doses of vaccines administered. However, the federal government instituted additional reporting requirements for COVID vaccines (CDC, 2023; Gee et al., 2021). Tracking for each dose administered included information on recipient age and sex, manufacturer, and dose number (CDC, 2024b; Klein et al., 2021). This enabled precise denominators and allowed for calculating reported rates of AESIs—that is, the number of reported events per number of vaccine doses administered over a specified risk window (e.g., 1 or 21 days) (Shimabukuro et al., 2021a). These reported rates could be stratified by specific vaccine, age, and sex and compared to expectations based on published background rates. Since reporting to VAERS is not complete, reported rates would usually be expected to be less than background rates. Nevertheless, if a statistical signal emerged—via PRR, empirical Bayesian data mining, or observed rates approaching or exceeding background—a clinical review would be triggered that considered seriousness, biological plausibility, and consistency with a known clinical syndrome (Shimabukuro et al., 2015). Data mining and review of serious events and AESIs occurred daily at ISO (Gee et al., 2024). The VAERs team met weekly to review tables and clinical reports and discuss signals. Cumulative tabulations, including frequency of AEs and relative proportions by seriousness, sex, and age, were publicly available weekly (to CDC WONDER,² HHS, and Epi-X³). In parallel, ISO and FDA staff held weekly or ad hoc coordination meetings to review new VAERS data and emerging safety concerns (Anderson, 2020).

VSD

VSD is a collaborative project between CDC and 13 large, integrated U.S. health care organizations. It is the ISO’s flagship active-surveillance platform, absorbing the majority of ISO’s analytic resources (McNeil et al., 2014). Established in 1990, VSD uses electronic health records (EHRs) and health care use administrative data (claims) from millions of individuals receiving routine health care within participating systems and has grown from about 6 million covered members at launch to more than 12 million (CDC, 2024c; Chen et al., 1997; Wallace et al., 2022). Of the 13 sites, 11 provide EHR data. This design allows for more robust analyses than passive reporting systems because it includes well-defined population denominators, unvaccinated control populations, longitudinal patient data, and

___________________

² Centers for Disease Control and Prevention Wide-ranging Online Data for Epidemiologic Research (CDC WONDER) is an online system that provides access to a broad array of public health information and datasets.

³ Epidemic Information Exchange (EPI-X) is CDC’s secure, web-based communication system for sharing preliminary health surveillance and outbreak information with public health officials.

the capacity to perform population-based comparative studies (McNeil et al., 2014; Wallace et al., 2022). VSD’s common data model substantially overlaps with that used by the FDA-funded Sentinel Initiative, and several partner sites contribute data to both programs, facilitating shared analytic code and cross-network validation (FDA, 2017). VSD researchers commonly use epidemiological approaches, including RCA and self-controlled risk interval designs, to detect and evaluate potential vaccine-related safety signals in near real time. Specifically, RCA involves weekly sequential hypothesis testing—often employing maximized sequential probability ratio tests—to compare observed AE counts against expected baselines, enabling VSD to flag potential safety signals within weeks of vaccine administration (Davis, 2013).

During the PHE, VSD expanded its analytical frequency and leveraged its comprehensive electronic data to provide timely assessments of COVID vaccines (CDC, 2024c; Klein, 2021). This included near-real-time monitoring of AESIs—such as myocarditis, TTS, and other events flagged by passive systems—to characterize incidence rates, identify potential risk factors (e.g., age, sex, comorbidities), and compare rates with background rates in unvaccinated or prepandemic populations (Klein, 2021). The close integration of EHR and administrative claims data enabled rapid turnaround for signal detection and evaluation, contributing to prompt public health guidance (CDC, 2024c). While VSD’s focused, population-based approach offers advantages in assessing causality and absolute risks, it is concentrated within certain health care systems, does not capture all geographic or demographic groups (McNeil et al., 2014). Nevertheless, its ability to provide active surveillance at scale and rigorous, real-time data analyses has been particularly valuable during the heightened vaccine safety monitoring requirements of the pandemic (CDC, 2024c).

Because VSD relies on protected EHR and insurance-claims data from participating health systems, individual-level datasets remain behind secure firewalls at each site. External researchers may collaborate through vetted protocol proposals reviewed by the VSD Research Committee; approved projects operate under data-sharing agreements, HIPAA waivers, and IRB approvals, with analyses executed on site and only aggregate results released. Public-facing safety updates and peer-reviewed manuscripts are posted on the CDC VSD webpage.

CISA

CISA is a national network of vaccine safety experts coordinated by CDC and funded through collaborative agreements with academic medical centers. Established in 2001, CISA was designed to provide specialized clinical consultation on complex or severe AEs following immunization (CDC,

2024d). By drawing on multidisciplinary expertise—infectious disease specialists, immunologists, allergists, and epidemiologists—CISA enhances CDC’s capacity to investigate rare, high-impact vaccine safety signals that may not be fully understood through passive or broad-based active surveillance alone (e.g., VAERS or VSD) (Williams et al., 2011).

Throughout its history, CISA has fulfilled two core functions. First, it provides individualized clinical consultations, offering expert, case-by-case evaluations for HCPs managing unusual or severe postvaccination events (Williams et al., 2011). These consultations often involve comprehensive medical-record reviews, direct communication with treating clinicians, and, when necessary, advanced diagnostic testing. Second, CISA conducts mechanistic and observational research, including small cohort studies, case-control analyses, and mechanistic investigations, to explore the biological pathways that may underlie rare AEs following immunization (Williams et al., 2011). During the COVID PHE, these functions were adapted and intensified to support rapid, evidence-based responses to emerging vaccine safety concerns (CDC, 2024d).

During the PHE, CISA rapidly pivoted to address emerging signals tied to novel COVID vaccines. For example, it engaged in detailed case series study of TTS and cerebral venous sinus thrombosis (MacNeil et al., 2021). This single mechanistic study found no evidence that polyethylene glycol (a vaccine component) was responsible for anaphylaxis observed with several COVID vaccines (Zhou et al., 2023). Data collection and dissemination accelerated in response to real-time clinical demands: CISA expanded its consultation footprint, offered expedited reviews, and collaborated more intensively with other CDC surveillance systems (such as V-safe and VAERS) to synthesize findings (CDC, 2024d). These adaptations ensured that frontline clinicians and policy makers received timely, evidence-based guidance on preventing, identifying, and managing severe postvaccination events in an evolving pandemic landscape.

CISA consultation records and mechanistic-study data contain identifiable clinical details and therefore are not publicly released. Qualified investigators can access deidentified analytic files or collaborate on joint analyses after submitting a proposal to the CISA executive committee and obtaining IRB clearance and a CDC DUA. Summaries of consultation trends and key study findings are published in CDC reports, ACIP slide decks, and peer-reviewed journals.

V-safe

V-safe was developed and funded by Oracle Health Services and CDC to monitor vaccine safety in near real time (CDC, 2024e; Myers et al., 2023). Launched in December 2020 specifically to track AEs following

COVID vaccination, V-safe was designed to complement existing passive (e.g., VAERS) and active (e.g., VSD) surveillance systems by capturing self-reported health information directly from vaccine recipients. The system leverages text messaging and secure Web-based surveys to collect data on prespecified common postvaccination symptoms (e.g., fever, injection-site reactions) and more serious events that warrant medical attention (Gee et al., 2021; Myers et al., 2023).

In addition to prespecified items, an open-ended prompt collects free-text comments. Natural language inference models can be used to identify patterns (e.g., identifying “missed period” and “PMS” as menstrual irregularities) not solicited as prespecified symptoms. Upon enrollment, vaccinated individuals receive regular check-ins—initially daily, then transitioning to weekly—to document any new or ongoing symptoms (Hause et al., 2022a). This approach generates a robust stream of data that can be analyzed rapidly for emerging safety signals. Proportions of participants reporting local and systemic reactions and health impacts are tabulated by age, sex and severity. Through unique user links, V-safe also tailors reminders about subsequent vaccine doses, thus maintaining participant engagement throughout a multidose regimen. While originally introduced to serve as a rapid-response tool during the height of the PHE, V-safe underwent incremental refinements to accommodate booster doses, track pediatric vaccination, and expand the types of outcomes assessed. In particular, the COVID-19 Vaccine Pregnancy Registry is an expansion of V-safe (CDC, 2024f). These enhancements also included more targeted symptom queries (e.g., specific rare AEs) and refined protocols for transferring serious or urgent reports into more intensive follow-up systems, like the VAERS or the CISA project. As a result of these iterative improvements, V-safe evolved into one of CDC’s largest and most frequently used active-surveillance platforms (CDC, 2024e; Gee, 2024; Gee et al., 2021).

While V-safe played a crucial role in capturing high-volume, near-real-time data on expected, nonserious postvaccination symptoms (e.g., fatigue, injection-site pain), it was less effective for serious AEs. Its structure—requiring smartphone access and voluntary, ongoing survey participation—limits detection of severe AEs, especially those that are incapacitating (Gee et al., 2021, 2024). Available publications suggest V-safe contributed primarily reactogenicity data rather than AE signals, and few known AESIs appear to have originated from V-safe and triggered follow-up in other systems (Gee et al., 2024; Hause et al., 2022a).

Designed, built, and supported under a donation agreement with Oracle Health Services and the Department of Health and Human Services, V-safe’s public–private partnership raises important sustainability considerations: once emergency-phase federal funding ends, continued functionality will

depend on long-term governance, dedicated appropriations, and ongoing collaboration with private-sector technology partners (Shimabukuro, 2023b).

Raw V-safe responses include contact information and protected health data and are stored on secure CDC servers. Researchers may request deidentified datasets through a formal application process requiring a CDC DUA, evidence of IRB approval, and data-security documentation. High-level statistics on local/systemic reactions, health-impact measures, and participation metrics are released periodically via the CDC V-safe dashboard and in MMWR safety summaries.

COVID-19 Vaccine Pregnancy Registry

CDC established the COVID-19 Vaccine Pregnancy Registry in December 2020, coinciding with the initial rollout of COVID vaccines under Emergency Use Authorization (EUA). Recognizing the exclusion of pregnant individuals from early-phase clinical trials, the registry was developed as a targeted postauthorization surveillance mechanism to actively collect safety data in this high-priority population. It was one of several initiatives supported under the authorities of the Public Readiness and Emergency Preparedness (PREP) Act and COVID vaccine EUA framework, which enabled expedited deployment of safety monitoring infrastructure during the PHE (CDC, 2024f; Gee et al., 2024; Moro et al., 2021).

The registry leveraged V-safe as a primary mechanism for enrollment. Individuals who received a COVID vaccine and reported their pregnancy status through V-safe were contacted and invited to participate. Approximately 23,000 people who reported receiving a COVID vaccine during pregnancy—or within 30 days before conception—were enrolled between December 2020 and June 2021, and about 85 percent consented to medical-record review, yielding a large, well-documented cohort for analyses of maternal and infant outcomes (Madni et al., 2024). This registry functioned as a critical component of the broader postauthorization safety surveillance system coordinated by CDC and FDA to ensure continuous monitoring under EUA. Upon consent, participants were followed prospectively through pregnancy and postpartum. Data collection included self-reported health information, pregnancy outcomes (e.g., spontaneous abortion, stillbirth, gestational age at delivery), and infant outcomes through the first few months. When feasible, registry participation also involved medical-record abstraction to validate outcomes.

During the PHE, the registry protocol was updated to reflect changes in vaccine availability (such as bivalent booster rollout), evolving clinical guidance (timing of vaccination during pregnancy), and emerging questions about maternal–fetal antibody transfer and neonatal protection. These

updates were supported by the CARES Act, which enabled rapid scaling of data-collection infrastructure and digital tools through enhanced federal investment. The PREP Act also authorized emergency measures, including liability protections for vaccine administrators and manufacturers, facilitating operational flexibility for surveillance activities (Shimabukuro et al., 2021b).

Primary data sources for the registry included V-safe reports, structured participant surveys, and medical-record reviews. In select cases, linkage to other CDC-managed surveillance systems, such as the National Vital Statistics System, was conducted to enhance outcome verification (CDC, 2024f; Myers et al., 2023).

Due to the identifiable health information and sensitive nature of pregnancy-related data, access to individual-level registry data is restricted. Researchers and public health partners may request deidentified datasets or analytic summaries through formal CDC DUAs and IRB approvals. Public-facing summaries, including key outcome statistics and updated methodological details, are made available via the CDC COVID-19 Vaccine Pregnancy Registry webpage.

MAJOR METHODOLOGIC STRENGTHS/LIMITATIONS OF FINDINGS/SAFETY ASSESSMENTS FROM EACH OF THE MAJOR SYSTEMS

The strengths and limitations of each vaccine safety surveillance system reflect their underlying design and intended role within the broader monitoring framework. Passive systems are best suited for early signal detection and hypothesis generation, offering national reach and transparency but limited by selection, underreporting, and reporting biases. Active-surveillance systems enable more robust, population-based analyses using EHRs with prospectively recorded information but operate within a defined health care network and require longer timelines for complex evaluations. CISA offered expert case reviews and mechanistic insights, particularly for complex or rare events. V-safe provided rapid, participant-reported data on reactogenicity and short-term outcomes but lacked medical validation. Its pregnancy registry allowed longitudinal tracking of vaccine safety during pregnancy. Like VAERS, V-safe does not include unvaccinated persons that can be used for comparisons. Together, these systems formed a complementary network—each with unique contributions but also important constraints that shaped how and when safety questions could be answered. Table 2-1 presents a comparative snapshot of each platform’s principal advantages and constraints1.

In parallel, a range of other vaccine safety systems—operated by FDA, the Department of Veterans Affairs, the Indian Health Service, CMS,

TABLE 2-2 Description of Non-CDC COVID Vaccine Safety Data Sources Relevant to ISO

NOTES: * Indicates systems enhanced during the COVID response; # indicates systems newly created specifically for COVID vaccine safety monitoring. AE = adverse event; AESI = adverse event of special interest; BEST = Biologics Effectiveness and Safety Initiative; CBER = Center for Biologics Evaluation and Research; CDC = Centers for Disease Control and Prevention; CISA = Clinical Immunization Safety Assessment; CMS = Centers for Medicare & Medicaid Services; EHR = electronic health record; EUA = Emergency Use Authorization; FDA = Food and Drug Administration; IHS = Indian Health Service; LTC = long-term care facility; NIH = National Institutes of Health; VA = Department of Veterans Affairs; VAERS = Vaccine Adverse Event Reporting System; VSD = Vaccine Safety Datalink.

SOURCES: Bardenheier et al., 2021; FDA, 2023, 2024a,c; Shimabukuro and Klein, 2023; Wong et al., 2023.

international partners, and vaccine manufacturers—played a critical role in extending the reach of safety monitoring. These systems contributed complementary datasets, populations, and analytic methods. For example, FDA’s BEST Initiative conducted large-scale postauthorization evaluations; VA and the Indian Health Service (IHS) provided insights on specific federal health care populations; CMS enabled timely assessment of risks in older adults (FDA, 2023; Shimabukuro and Klein, 2023); and international surveillance systems helped identify early signals, such as myocarditis and TTS. CMS data from long-term care facilities were also incorporated to enhance monitoring among older adults and high-risk populations, illustrating the potential of interagency collaboration. Manufacturer-led postauthorization studies, required under FDA agreements, also generated targeted safety data, particularly in populations excluded from initial trials (FDA, 2024c). Table 2-2 describes these non-CDC systems and their relevance to ISO’s safety monitoring efforts during the COVID vaccination program.

Main Safety Findings Across Monitoring Systems

Throughout the COVID vaccination campaign, U.S. safety monitoring systems collectively identified, assessed, and responded to a broad range of AEs following immunization. This surveillance network—comprising VAERS, VSD, CISA, V-safe, and the V-safe Pregnancy Registry—enabled a multifaceted understanding of vaccine safety across diverse populations and conditions. While each system had unique capabilities and limitations, they worked in concert to detect early signals, estimate or validate risks, and inform timely public health guidance.

Many of the events under surveillance were prespecified outcomes, determined before vaccine rollout based on historical vaccine safety concerns, COVID disease complications, or findings from clinical trials. These included anaphylaxis, myocarditis, GBS, and thromboembolic events, such as deep vein thrombosis and pulmonary embolism. For each outcome, CDC and FDA required surveillance programs to apply harmonized case definitions—for example, the CDC working definition for myocarditis, ACIP Tier 1–2 criteria for TTS, and Brighton Collaboration standards for anaphylaxis and GBS—to ensure consistency in case ascertainment across VAERS, VSD, V-safe, and CISA (CDC, 2024g, 2025; Korinthenberg and Sejvar, 2020; Marschner et al., 2023; See, 2021; Sejvar et al., 2011). Prespecification helped prioritize investigations, standardize reporting across systems, and reduce analytic bias. Maximizing the utility of these harmonized definitions depends, in part, on compiling results from safety assessments in a centralized, publicly accessible location. Centralized assessment improves transparency, supports integrated interpretation across systems, and enables

both researchers and the public to understand emerging safety findings in a more holistic and accessible way.

Across systems, several consistent findings emerged. Anaphylaxis was one of the earliest signals detected—flagged by VAERS within a week of vaccine rollout, which recorded ≈0.00047 percent of participants self-reporting severe allergic reactions (rash, swelling, dyspnea) on Day 0–1 after dose 1, a proportion that closely matched VAERS case counts, and then clinically evaluated in depth by CISA (Shimabukuro, 2021a). Myocarditis and pericarditis were observed predominantly in young men and boys following mRNA vaccines, especially after dose 2; most cases were reported to be mild and to resolve quickly (Marschner et al., 2023). GBS and TTS were linked primarily to the Janssen vaccine, prompting federal guidance changes and ultimately its withdrawal from the U.S. market (Hanson et al., 2022; Rosenblum et al., 2021; See et al., 2021). No increased risk of death, including from non-COVID causes, was found in any system (Oliver et al., 2022). Evaluations of pregnancy and reproductive outcomes and pediatric vaccination and vaccine coadministration consistently reaffirmed the strong safety profile of COVID vaccines (Moro et al., 2021).

VAERS played a pivotal role in rapid signal detection. Reports of anaphylaxis, myocarditis, TTS, and GBS were identified through it, contributing to regulatory responses and ACIP recommendations (Abara et al., 2023; MacNeil et al., 2021; Oster et al., 2022). Despite its limitations as a spontaneous-event reporting system, VAERS provided valuable data for contextualizing risk. For instance, Oster et al. (2022) analyzed VAERS reports of myocarditis following mRNA COVID vaccination using national vaccine administration data as the denominator, calculated reporting rates, and identified risk to be highest after the second dose in adolescent boys and young men, informing timely ACIP risk–benefit assessments (Gargano et al., 2021). Later (mid-2021 to December 2022), VAERS analytic studies found no disproportionate increased risk of tinnitus in any COVID vaccines (Yih et al., 2024). VAERS data provided some reassurance about lack of pregnancy complications (Moro et al., 2024), and all-cause mortality (Xu et al., 2021). VAERS summaries confirmed that the vast majority of reported AEs were nonserious (Ceacareanu and Wintrob, 2021). As described, for COVID vaccines, national administration data—stratified by age, sex, and product—enabled calculating reported rates, a capability not typically available in passive surveillance. However, these rates should be interpreted with caution, as VAERS remains subject to underreporting, differential reporting, and stimulated reporting, and reported rates do not reflect incidence.

VSD enabled active, large-scale, population-based surveillance and comparative risk evaluations across millions of vaccine recipients at 11 integrated health care sites (CDC, 2024c). It provided robust evidence on AE risks stratified by age, sex, pregnancy status, and underlying health

conditions. For example, VSD quantified observed rates of myocarditis at ≈137 cases per million second mRNA doses in boys aged 12–15 versus ≈9.3 per million in girls of the same age and GBS at ≈3.1 cases per million Janssen doses—estimates that informed benefit–risk assessments for specific subgroups (Goddard et al., 2022b; Hanson et al., 2022). VSD studies also confirmed no increased risk of miscarriage, preterm birth, or neonatal complications among vaccinated pregnant individuals (Lipkind et al., 2022) and showed no safety concerns when COVID boosters were coadministered with the seasonal influenza vaccine (Kenigsberg et al., 2023a). Mortality analyses revealed lower non-COVID death rates among vaccinated versus unvaccinated members—likely reflecting healthy-vaccinee bias rather than a protective vaccine effect (Xu et al., 2023).

V-safe collected real-time, participant-reported data on common side effects. It played a key role in characterizing mild to moderate reactogenicity, confirming that symptoms like fatigue, fever, and injection-site pain were frequent but brief (CDC, 2024e; Chapin-Bardales et al., 2021). Although it was less suited for assessing rare or serious events, V-safe helped capture signals like menstrual irregularities and provided critical reassurance regarding the overall vaccine tolerability (Wong et al., 2022). Its main analytic limitation was the absence of an unvaccinated comparator group, which made it difficult to determine whether self-reported events occurred above background rates or varied by underlying health status.

CISA contributed detailed clinical insights through expert case reviews and consultations, particularly for complex or high-stakes AEs, like anaphylaxis, myocarditis, and TTS (Williams et al., 2011). Its adjudicated investigations incorporated chart reviews and laboratory diagnostics, adding diagnostic clarity to rare events. However, because CISA focused on referred or severe cases, findings were not generalizable across the broader population (Gee et al., 2024; Williams et al., 2011). Moreover, for certain syndromes—most notably multisystem inflammatory syndrome in adults or children (MIS-A/MIS-C)—CISA’s working definitions required laboratory evidence of recent SARS-CoV-2 infection; while this criterion helped distinguish postinfection pathology from coincidental findings, it also meant that vaccine-associated cases without documented infection could be missed (Cortese et al., 2023).

Together, these coordinated vaccine safety monitoring systems were able to comprehensively evaluate the COVID vaccines and provide estimates of risk for various population subgroups. The interplay of rapid signal detection, in-depth clinical review, and longitudinal population-level analysis enabled a comprehensive evaluation of vaccine safety. These analyses were crucial for informing policy and guiding clinical recommendations, although, as noted, challenges occurred in communicating this knowledge to HCPs and the public, undermining the robustness and excellence of

the scientific work. Moreover, the majority of publications of these findings included language that risks were “rare,” “mostly mild,” and “far outweighed by the benefits of protection” against severe COVID. This reassurance was provided even when formal risk–benefit analyses were not included or referenced. Such vaccine promotion may interfere with the perception of independence of ISO vaccine safety monitoring. A cross-system summary of the major COVID-19 vaccine-safety signals, their principal findings, and surveillance periods is provided in Table 2-3.

TIMELINESS AND COMPLETENESS OF SIGNALS

The scale and urgency of the U.S. COVID vaccination campaign required a responsive, transparent, and scientifically rigorous safety infrastructure. VaST was convened by CDC in October 2020 as an ACIP subcommittee to review vaccine safety data in near real time and advise the full ACIP (Lee, 2021; Markowitz et al., 2024). Meeting as often as weekly during the first 18 months of rollout—and then monthly through 2023—VaST examined findings from VAERS, VSD, V-safe, CISA, and external sources, issuing summary memorandums that informed ACIP votes, CDC Health Alerts, and provider advisories (Rosenblum et al., 2022).

To prioritize surveillance, CDC and FDA jointly published a list of AESIs before mass vaccination began; it drew on historical vaccine risks (e.g., anaphylaxis, GBS), COVID disease complications (e.g., MIS-C), and trial signals (e.g., Bell’s palsy) (Gee et al., 2021; Lee, 2021; Markowitz et al., 2024). Prespecification improved consistency across systems, yet the framework remained adaptive: when VAERS and V-safe unearthed unexpected patterns—such as menstrual cycle changes or tinnitus—those outcomes were added to monitoring protocols and VaST agendas (Rosenblum et al., 2022). The detailed decision-making logic for adding or retiring signals, however, is documented only in internal VaST working papers and has not been made publicly available.

Safety signal evaluations closely tracked each phase of the rollout—and the timeliness of VaST briefings became a key operating metric. In its first quarter (December 2020–March 2021), VaST met twice per week and delivered slide-deck summaries to ACIP within 24–48 hours; these were posted on the ACIP website the same day as the public meeting, creating near-real-time transparency (CDC, 2024h; Shimabukuro et al., 2021a). Anaphylaxis illustrates the cadence: VaST reviewed the first 21 VAERS cases on December 19, 2020—3 days after the Pfizer-BioNTech launch—and ACIP discussed the findings in an emergency session the next morning (CDC, 2020), CDC’s interim clinical guidance and a rapid MMWR followed within 2 weeks (CDC COVID-Response Team et al., 2021; Shimabukuro et al., 2021a).

Early all-cause-mortality data were brought to VaST on January 27, 2021; although VAERS counts suggested no excess deaths, the group immediately commissioned RCAs in VSD and previewed interim results to ACIP on March 1, 2021 (ACIP, 2021b, 2021c). Peer-reviewed VSD papers released in 2021, 2022, and 2024 all corroborated the absence of a vaccine-associated mortality signal (Klein et al., 2021; Xu et al., 2023, 2024).

Overall, VaST’s ability to convene within days, circulate presentations in under 48 hours, and present data important for ACIP decision making—often within hours or days—proved critical for rapid, evidence-based updates to CDC guidance and provider alerts (ACIP, 2021d).

Pregnancy safety was an early and high-priority focus of COVID vaccine monitoring efforts, including through a registry that operated during the early rollout but was discontinued in 2023 (Gee et al., 2024; Madni et al., 2024). Ongoing monitoring of pregnancy outcomes continues through systems such as VSD and VAERS. With clinical trials excluding pregnant individuals, CDC launched the V-safe COVID-19 Vaccine Pregnancy Registry in December 2020 (CDC, 2024f). VaST convened a pregnancy-focused session in February 2021 that reviewed the first reports from V-safe and VAERS (Lee and Hopkins, 2021). Subsequent evaluations across V-safe, VAERS, and VSD consistently found no increased risk of miscarriage, preterm birth, stillbirth, or neonatal complications among the vaccinated (Lipkind et al., 2022; Shimabukuro et al., 2021b; Zauche et al., 2021). In parallel, menstrual changes—though not prespecified—were examined after public concern: VaST reviewed V-safe data showing ~1 percent of female participants reporting menstrual irregularities, and VSD analyses found postmenopausal bleeding to be uncommon (Wong et al., 2022).

For newer or rarer safety signals, timeliness depended on how quickly robust data could be gathered and analyzed. TTS linked to the Janssen vaccine was first identified internally by CDC and FDA on April 9, 2021, reviewed by VaST on April 12, less than 2 months after the EUA, and triggered a nationwide “pause” announced on April 13, 2021, followed by an emergency ACIP meeting on April 14 and updated vaccine recommendations issued on April 23 (FDA, 2021; MacNeil et al., 2021; See et al., 2021; Shay et al., 2021).

Myocarditis associated with mRNA vaccines in young men and boys surfaced in VAERS and VSD in May 2021. VAERS reported rates were close to background, suggesting a possible problem, given that reporting to VAERS is generally incomplete. Accumulating U.S. and international data—along with CISA cardiology consultations—led CDC to issue interim clinical guidance on May 17, 2021 and ACIP to hold a dedicated review on June 23, 2021 (CDC, 2025a; MacNeil et al., 2021; Shimabukuro, 2022).

GBS following Janssen vaccination was first discussed by VaST on June 10, 2021. Elevated VAERS reporting rates were later confirmed in a

VSD RCA released to ACIP on December 16, 2021, and published in 2022, underpinning ACIP’s January 5, 2022, recommendation to preferentially use mRNA vaccines (Alimchandani, 2021; Hanson et al., 2022).

Evaluations of coadministration with influenza vaccines began in autumn 2021; VaST and VSD reviews found no serious safety concerns, although mild increases in systemic reactogenicity were noted (Hause et al., 2022c; Kenigsberg et al., 2023b).

As vaccines expanded to children and boosters, new formulations and recommendations were accompanied by ongoing VaST review. Between November 2020 and April 2023, VaST held regular meetings and presented 22 safety assessments to ACIP, supporting benefit–risk evaluations and informing vaccine policy. In April 2023, its responsibilities transitioned back to the ACIP COVID-19 Vaccines Work Group as part of a return to routine safety assessment procedures.

Bivalent mRNA boosters, introduced in August 2022, were evaluated in a joint VAERS and V-safe analysis of >22 million administered doses that revealed no new safety concerns (Hause et al., 2022b).

In January 2023, VSD detected a transient statistical signal for ischemic stroke in adults ≥65 years after the Pfizer bivalent booster, but it disappeared in updated VSD runs and was not corroborated by VAERS, CMS, or international data (FDA, 2023; Shimabukuro and Klein, 2023).

Additional topics reviewed included pediatric safety after primary and booster doses in children 6 months and older (no new signals) (CDC, 2024i; Hause et al., 2023), tinnitus (no association confirmed in VSD) (Yih et al., 2024), and concurrent COVID/flu vaccination (no serious AESIs, minor uptick in reactogenicity) (Kenigsberg et al., 2023a). Throughout its operation, VaST reviewed data from a range of sources, including VAERS, VSD, V-safe, the pregnancy registry, and other monitoring systems external to CDC, such as Biologics Effectiveness and Safety (BEST), VA, DoD, and IHS. CISA contributed technical consultation and clinical insights.

Although peer-reviewed manuscripts sometimes appeared months later, key safety findings were typically communicated first through MMWR bulletins, ACIP slide decks, and clinician listserv alerts, ensuring that frontline providers and the public received timely guidance while full-scale studies were still underway.

Finally, the timelines for generating and disseminating results varied by system and outcome severity. Preliminary analyses for high-priority signals were typically presented to ACIP within days or weeks—anaphylaxis on December 12, 2020 (ACIP, 2020), TTS on April 23, 2021 (Shimabukuro, 2021b), myocarditis on June 23, 2021 (ACIP, 2021d), and the ischemic-stroke assessment for Pfizer’s bivalent booster on January 26, 2023 (Shimabukuro, 2023a)—well before any peer-reviewed papers appeared. VSD publications on acute outcomes, like myocarditis, typically followed within

6–12 months; more complex evaluations—especially of pregnancy or long-term effects—required 1.5–2.5 years. VAERS descriptive studies appeared sooner, providing early context and reassurance, while CISA’s in-depth clinical investigations, though slower to publish, still informed rapid decision making through real-time expert consultation.

These processes underscore the importance of both structure and flexibility in national vaccine safety evaluation. Prespecified outcomes provided a foundation for proactive monitoring, while system adaptability allowed for investigating unanticipated concerns—ensuring that safety questions were addressed promptly and transparently during an evolving public health crisis.

OPPORTUNITY/ATTENTION COSTS

The ISO research agenda necessitates deliberate tradeoffs in resource allocation—both at the systems level and in the prioritization of specific safety signal investigations. Effective navigation of these tradeoffs, particularly in dynamic public health contexts, such as the COVID vaccine safety investigations, requires procedural readiness and access to integrated information that can support accelerated decision making and translating this knowledge for the population.

At the systems level, ISO’s rapid incorporation of long-term care data (CDC, 2025b) to augment existing surveillance platforms illustrated the benefits of interagency collaboration in generating timely and novel insights—an uncommon but highly effective example of cross-agency coordination. At the level of individual safety signals, the timeliness of VaST report generation, as summarized in Table 2-3, demonstrates responsiveness when signals are actively pursued. However, the absence of communication regarding decisions to defer or decline investigations of certain potential signals (e.g., menstrual irregularities) may have diminished public trust, leaving a vacuum filled by speculation and unmoderated discourse.

Limitations and Opportunities for a Unified Vaccine Safety Infrastructure

The committee was not provided information on ISO’s internal decision-making processes for determining which safety analyses to pursue, nor was it given access to data on how its budget was allocated. Please see Chapter 1 for details. Despite these limitations, internal reviews and independent evaluations have examined ISO’s surveillance architecture, including its operational advantages and inherent constraints.

Although certain avenues for accessing vaccination status data were

expanded during the PHE—for example, through temporary agreements with CMS and national pharmacy chains—the absence of a comprehensive, federally mandated vaccination reporting policy remains a significant obstacle to integrated safety surveillance (U.S. GAO, 2021). Each state, territory, and local jurisdiction governs its own Immunization Information System (IIS), and these vary widely in technical functionality, legal authority, and scope of reporting requirements (CDC, 2024j). While IISs are primarily used for tracking vaccine administration, their data can enhance safety monitoring by providing accurate denominator data for reported rates and supporting linkage to AE reports across systems. Although most jurisdictions responded to CDC’s COVID data-sharing requests—enabled in part by federal control of vaccine supply under EUA—many lacked bidirectional interoperability with hospitals, pharmacies, long-term care facilities, and nontraditional vaccination sites, such as mass clinics and community pop-ups. These settings were central to reaching uninsured and underserved populations yet often fell outside routine health data streams and did not consistently transmit records into IISs or systems like VSD. In some states, data from these sources were submitted via spreadsheets and uploaded manually, leading to time lags, reduced data quality, and missed opportunities to link to safety outcomes.

Although CDC has issued functional standards for core IIS data elements—including patient demographics, lot number, and provider details—implementation is voluntary and varies by jurisdiction (CDC, 2024k). The EUA for COVID vaccines temporarily required providers to report administered doses and AEs to federal authorities (CDC, 2024f), which allowed systems like VAERS to calculate reported rates of adverse events of special interest (AESIs) with greater accuracy. However, these mandates did not extend to other vaccines or persist beyond the PHE. Without a national immunization registry or harmonized legal and technical infrastructure, the United States remains limited in its ability to integrate vaccination records—particularly from uninsured or nontraditional care settings—into a comprehensive safety surveillance framework. Structural challenges—most notably the lack of a coordinated, department-wide strategy and infrastructure for generating and integrating vaccine safety and effectiveness data—continue to limit the efficiency, impact, and scalability of CDC and HHS investments (Bauchau et al., 2023).

Stakeholders, including those providing public input during ACIP meetings (e.g., April 2021, Scott Razen, CUNY), have highlighted the opportunity for transitioning from a patchwork of systems toward a modern, integrated, active, and nationally representative surveillance system (ACIP, 2021a; Razen, 2021). ISO has demonstrated capacity for leveraging partnerships with platforms funded by the National Institutes of Health, CMS, and private-sector entities (e.g., pharmacies) to access data not traditionally

available within VSD (Bauchau et al., 2023; Haendel et al., 2021). This underscores the potential for coinvestment in a unified postauthorization/postmarketing system capable of evaluating both safety and effectiveness of vaccines, devices, and therapeutics. Opportunities for optimization include integrating benefit–risk assessments within a single infrastructure, deduplicating parallel data streams and redundant procurements, and creating a “data sandbox” of deidentified assets to support cross-agency analytics.

Transparent Coordination Across CDC and Non-CDC Systems

Enhanced transparency in the governance of surveillance systems and safety signal workflows is essential. This includes increased public input—particularly from communities with specific concerns—and clear communication regarding system and signal prioritization criteria. Rather than consolidating into a single data source, a modernized approach should preserve multiple, independent input streams while integrating findings and communications into a unified, transparent, and accessible framework. For example, complementary analyses from the passive VAERS and active VSD during the investigation of myocarditis following mRNA vaccination offered both rapid signal detection and structured follow-up analysis, illustrating the value of diverse inputs with coordinated interpretation (Goddard et al., 2022b; Marschner et al., 2023; Shimabukuro, 2022). Ensuring high standards of evidence quality across systems—and synthesizing findings in a coordinated voice for vaccine risk assessment—will strengthen both public trust and scientific rigor in safety assessments.

While pandemic-era adaptations showcased CDC ISO’s ability to leverage partnerships with FDA, CMS, academic centers, and private-sector entities, the absence of a durable, high-level oversight body limited the strategic coordination of safety efforts across federal and state platforms. Improved alignment, via interagency agreements, shared technical standards, and joint prioritization of safety signal evaluation will be essential to optimize resource use, eliminate redundancies, and improve analytic transparency in future PHEs.

The Global Vaccine Data Network (GVDN) played a significant complementary role in the global evaluation of COVID vaccine safety. Despite a slow start due to early funding limitations, GVDN implemented standardized protocols to conduct RCAs and background rate estimation studies across an international network that included New Zealand, Indonesia, Argentina, the African COVID-19 Vaccine Safety Surveillance system (South Africa, Mali, Ghana, Nigeria, Ethiopia, Kenya, Malawi, and Mozambique), Australia (Victoria and New South Wales), Canada (British Columbia and Ontario), Denmark, Finland, the Republic of Korea, Hong Kong, and multiple Vaccine Monitoring Collaboration for Europe sites, including the

United Kingdom and Spain (Valencia and Catalonia). Published studies from the network have provided foundational estimates of background rates for AESIs (Phillips et al., 2023), multinational risk estimates for GBS (Nasreen et al., 2025), and large-scale cohort analyses involving nearly 100 million individuals (Faksova et al., 2024). A signal for acute disseminated encephalomyelitis identified in this cohort was validated in a follow-up study (Morgan et al., 2024). Additional GVDN studies of TTS and myocarditis/pericarditis are forthcoming. While these efforts involved U.S.-based collaborators and contributed important scientific insights, this report focuses on systems primarily developed and operated by CDC, HHS, or directly funded federal partners. Nonetheless, GVDN offers a compelling model for international collaboration, harmonized protocols, and large-scale signal detection.

Technology Investment and Ecosystem Coordination

Optimizing the informational value derived from the surveillance ecosystem will require sustained investments in technological infrastructure and intersystem coordination. Operational costs for individual systems are not disclosed, and clarity is insufficient regarding population-level overlap across CDC’s active-surveillance platforms and those of partner agencies. This lack of transparency impedes efforts to assess resource efficiency, identify gaps in demographic or geographic coverage, and reduce redundancy. Improved interoperability and shared technical standards across platforms—such as those operated by CDC, CMS, FDA, and others—could enhance scalability, accelerate analytic turnaround, and support coordinated responses during future PHEs (CDC, 2024l).

The PHE demonstrated what is possible when legal and technical barriers to data access are lifted. Under the PREP Act and EUAs (see Chapter 1), CDC and its partners were able to access vaccination data from pharmacy chains, long-term care providers, and health insurers; link immunization records with AE data across systems; and generate timely, high-resolution safety signals (CDC, 2024c; FDA, 2024b). These flexibilities enabled calculating reported AE rates with denominators, even for populations often excluded from traditional health system surveillance.

Now that these authorities have expired (Hickey, 2025), many structural constraints have returned. No federal mandate exists for real-time vaccination data reporting outside of emergencies, and state-level IISs remain highly variable in legal authority, technical standards, and bidirectional connectivity. Fragmented governance, inconsistent adoption of interoperability frameworks, and limited mechanisms for accessing data from uninsured individuals or nontraditional vaccination sites reduce the completeness and utility of national safety surveillance. Without sustained legal and technical

infrastructure, the country remains underprepared to replicate this level of surveillance performance outside a declared emergency.

To ensure sustained preparedness, a dedicated cross-agency governance structure is needed to coordinate vaccine safety monitoring activities across CDC, FDA, CMS, NIH, and state public health systems. Such coordination should preserve system independence while enabling shared analytic priorities, streamlined signal evaluation protocols, and unified communication of findings.

Signal Investigation Prioritization

Investment decisions must also address the intensity with which specific safety signals are investigated and updated. While the initial lists are typically grounded in prior evidence and included in surveillance protocols, the criteria and processes by which new signals are nominated, prioritized, or deprioritized remain opaque. In interviews, most CISA researchers described priority-setting as a “black box,” noting limited involvement and advocating for earlier and more transparent collaboration with CDC in shaping research agendas. One researcher called to develop structured mechanisms to incorporate academic expertise and methodological innovation within a coordinated federal framework, rather than relying solely on traditional CDC-led processes (Westat, 2025; Appendix C).

System-specific limitations further constrain signal investigation. For instance, vaccine safety experts reported that VAERS lacked the infrastructure for systematic AE follow-up, and that the initial design of V-safe—intended to support direct response to all reports—was quickly overwhelmed, limiting its utility for sustained monitoring. These constraints diminish the capacity of surveillance systems to support iterative investigation of safety signals. Ensuring the ability to respond to emerging postauthorization data will require strengthened mechanisms to identify and escalate signals meriting longitudinal, methodologically rigorous follow-up beyond the inherent constraints of passive surveillance systems and even VSD’s limited subpopulation stratification (Westat, 2025; Appendix C).

ADDITIONAL UNPUBLISHED, REAL-TIME FINDINGS/SAFETY ASSESSMENTS

VaST and ACIP Presentations

Throughout the COVID vaccine rollout, ACIP—a federal advisory body responsible for developing immunization recommendations—served as the primary venue for public presentation and deliberation of vaccine safety data. CDC established VaST in November 2020 as a rapid-review body composed of independent safety experts. VaST met regularly—sometimes

weekly—to assess preliminary data from multiple surveillance platforms, including VAERS, VSD, and V-safe, and inputs from the CISA network and external sources, such as international partners. Its assessments were used to inform ACIP discussions about benefit–risk balance, especially during time-sensitive decisions related to EUA expansion, booster eligibility, and pediatric vaccination (Gee et al., 2021; Shimabukuro et al., 2021a).

ISO played a foundational role in enabling these assessments. It is responsible for managing and coordinating the federal postauthorization vaccine safety monitoring infrastructure, including data collection, curation, and preliminary analysis across surveillance systems. While ISO does not make recommendations or conduct formal benefit–risk assessments, it supports ACIP and VaST by providing timely, high-quality safety data and technical interpretation. These data are integrated into ACIP’s structured Evidence to Recommendation framework, which weighs safety alongside other domains, such as disease burden, vaccine efficacy, acceptability, and equity. The distinct yet complementary roles of ISO (data generation) and ACIP (policy recommendation) supported efforts to make decisions more transparent, evidence based, and appropriately contextualized.

In addition to prespecified AESIs, VaST and ACIP rapidly reviewed real-time safety signals arising during the vaccination campaign. These included both known concerns—like myocarditis and anaphylaxis—and emerging issues, such as tinnitus and coadministration with influenza vaccines. Table 2-4 summarizes selected signals, highlighting the timelines from first internal review, to public communication (e.g., ACIP meetings or FDA warnings), to peer-reviewed publication; ACIP meetings often provided the earliest public transparency on safety signals—sometimes months ahead of formal studies—reinforcing the importance of this advisory process as a mechanism for real-time communication of vaccine safety data.

CONCLUSION

The COVID PHE placed extraordinary demands on ISO, requiring rapid data collection, evaluation, and communication of emerging vaccine safety signals. The integrated infrastructure deployed—spanning passive systems, like VAERS, active platforms, like VSD and V-safe, expert consultation from CISA, and specialized initiatives, such as the COVID-19 Pregnancy Registry—enabled a scale and depth of vaccine safety surveillance without precedent in U.S. public health enterprises. Taken together, these systems supported real-time risk assessment, informed regulatory and clinical recommendations, and guided programmatic decisions.

However, the pandemic exposed structural challenges that, if addressed, could position ISO to be even more effective in future public health responses. As stated in Chapter 1, the committee is applying guiding principles to its

assessment and recommendations. Its conclusion of ISO’s data monitoring and assessment activities is organized following those principles.

Relevance: ISO’s monitoring activities were often shaped by urgent public health priorities, but the absence of a clearly articulated mission or long-term strategic plan limited external visibility into how those were determined or adjusted. Ensuring the relevance of ISO’s work will depend on formal mechanisms to incorporate input from health professionals, public health stakeholders, and the broader public into planning, prioritization, and communication strategies.

Credibility: The scientific credibility and excellence of ISO’s surveillance outputs was supported by consistent use of robust epidemiological methods and transparent engagement in public forums, such as ACIP meetings. However, variation in risk estimates across platforms and the lack of a centralized, accessible portfolio of risk information, tailored to technical and lay audiences, made it difficult for many to interpret the data consistently. More standardized communication tools—such as plain-language summaries, clearly labeled system-specific findings, and consistent risk metrics—could improve clarity and usability. Additionally, where possible, ISO should adopt standardized risk-reporting formats that incorporate relevant subgroup analyses to facilitate clearer public understanding and comparison of findings.

Improvement and Innovation: The rapid deployment of new tools during the PHE also underscores ISO’s capacity for continuous improvement and innovation. Developing structured processes to evaluate and integrate emerging scientific methods, data technologies, and communication research could support more agile and forward-looking safety surveillance. Sustaining these efforts beyond emergency response will require dedicated resources, coordination across federal systems, and the flexibility to evolve alongside novel vaccine platforms.

Independence: Finally, the integrity of ISO’s work depends on its ability to operate with independence—producing data and evaluations that are scientifically rigorous and insulated from policy or promotional influence. While ISO collaborates across CDC and HHS to inform immunization efforts, independence requires that its analyses and communications remain clearly distinct from vaccine advocacy and policymaking. Ensuring this separation—articulated in ISO’s mission, decision-making, functions and communications—will be essential to maintain clarity of purpose and protect the scientific objectivity of ISO’s work.

The COVID pandemic reinforced the indispensable role of ISO’s surveillance systems in identifying, evaluating, and communicating vaccine risks. As future challenges emerge, strengthening ISO’s capacity through transparent planning, inclusive stakeholder engagement, methodological rigor, and clear independence will be essential for advancing a robust, coordinated vaccine safety infrastructure.

REFERENCES

CDC. 2024c. Vaccine Safety Datalink (VSD). https://www.cdc.gov/vaccine-safety-systems/vsd/ (accessed August, 7, 2025).

CDC. 2024d. Clinical immunization safety assessment (CISA) project. https://www.cdc.gov/vaccine-safety-systems/hcp/cisa/index.html (accessed August 7, 2025).

CDC. 2024e. About V-safe. https://www.cdc.gov/vaccine-safety-systems/v-safe/index.html (accessed September 10, 2025).

CDC. 2024f. V-safe COVID-19 vaccine pregnancy registry. https://www.cdc.gov/vaccine-safety-systems/monitoring/covid-preg-reg.html (accessed August 7, 2025).

CDC. 2024g. Myocarditis and pericarditis after mRNA COVID-19 vaccination. https://www.cdc.gov/vaccines/covid-19/clinical-considerations/myocarditis.html (accessed August 7, 2025).

CDC. 2024h. Advisory Committee on Immunization Practices (ACIP) meeting agendas. https://www.cdc.gov/acip/meetings/agendas.html (accessed August 7, 2025).

CDC. 2024i. Evidence to recommendation framework (EtR) for use of COVID-19 vaccines in the 2024–2025 formula for persons 6 months of age and older. https://www.cdc.gov/acip/evidence-to-recommendations/covid-19-2024-2025-6-months-and-older-etr.html (accessed August 7, 2025).

CDC. 2024j. Immunization information systems (IIS): Policy and legislation. https://www.cdc.gov/iis/policy-legislation/index.html (accessed August 7, 2025).

CDC. 2024k. Immunization information systems (IIS): Core data elements. https://www.cdc.gov/iis/core-data-elements/index.html (accessed August 7, 2025).

CDC. 2024l. CDC data modernization efforts accelerate nation’s ability to detect and rapidly respond to health threats. https://www.cdc.gov/media/releases/2024/p0411-CDC-datamodernization.html (accessed August 7, 2025).

CDC. 2025a. Interim clinical considerations for use of COVID-19 vaccines in the United States. Centers for Disease Control and Prevention.

CDC. 2025b. Long-term care facilities (LTCF) component. https://www.cdc.gov/nhsn/ltc/index.html (accessed August 8, 2025).

CDC and FDA (CDC COVID Response Team and Food and Drug Administration). 2021. Allergic reactions including anaphylaxis after receipt of the first dose of Pfizer-BioNTech COVID-19 vaccine—United States, December 14–23, 2020. MMWR Morb Mortal Wkly Rep 70(2):46–51.

Ceacareanu, A. C., and Z. A. P. Wintrob. 2021. Summary of COVID-19 vaccine-related reports in the vaccine adverse event reporting system. J Res Pharm Pract 10(3):107–113.

Chapin-Bardales, J., T. Myers, J. Gee, D. K. Shay, P. Marquez, J. Baggs, B. Zhang, C. Licata, and T. T. Shimabukuro. 2021. Reactogenicity within 2 weeks after mRNA COVID-19 vaccines: Findings from the CDC V-safe surveillance system. Vaccine 39(48):7066–7073.

Chen, R. T., J. W. Glasser, P. H. Rhodes, R. L. Davis, W. E. Barlow, R. S. Thompson, J. P. Mullooly, S. B. Black, H. R. Shinefield, C. M. Vadheim, S. M. Marcy, J. I. Ward, R. P. Wise, S. G. Wassilak, and S. C. Hadler. 1997. Vaccine safety datalink project: A new tool for improving vaccine safety monitoring in the United States. The Vaccine Safety Datalink Team. Pediatrics 99(6):765–773.

Cortese, M. M., A. W. Taylor, L. J. Akinbami, A. Thames-Allen, A. R. Yousaf, A. P. Campbell, S. A. Maloney, T. A. Harrington, E. G. Anyalechi, D. Munshi, S. Kamidani, C. R. Curtis, D. W. McCormick, M. A. Staat, K. M. Edwards, C. B. Creech, O. Museru, P. Marquez, D. Thompson, J. R. Su, E. P. Schlaudecker, and K. R. Broder. 2023. Surveillance for multisystem inflammatory syndrome in US children aged 5-11 years who received Pfizer-BioNTech COVID-19 vaccine, November 2021 through March 2022. J Infect Dis 228(2):143–148.

Das, M. K. 2023. Adverse events following immunization: The known unknowns and black box: Based on 10th Dr. I. C. Verma Excellence Award for Young Pediatricians delivered as oration on 9th Oct. 2022. Indian J Pediatr 90(8):817–825.

Davis, R. L. 2013. Vaccine safety surveillance systems: Critical elements and lessons learned in the development of the US vaccine safety datalink’s rapid cycle analysis capabilities. Pharmaceutics 5(1):168–178.

Day, B., D. Menschik, D. Thompson, C. Jankosky, J. Su, P. Moro, C. Zinderman, K. Welsh, R. B. Dimova, and N. Nair. 2023. Reporting rates for VAERS death reports following COVID-19 vaccination, December 14, 2020–November 17, 2021. Pharmacoepidemiol Drug Saf 32(7):763–772.

Faksova, K., D. Walsh, Y. Jiang, J. Griffin, A. Phillips, A. Gentile, J. C. Kwong, K. Macartney, M. Naus, Z. Grange, S. Escolano, G. Sepulveda, A. Shetty, A. Pillsbury, C. Sullivan, Z. Naveed, N. Z. Janjua, N. Giglio, J. Perala, S. Nasreen, H. Gidding, P. Hovi, T. Vo, F. Cui, L. Deng, L. Cullen, M. Artama, H. Lu, H. J. Clothier, K. Batty, J. Paynter, H. Petousis-Harris, J. Buttery, S. Black, and A. Hviid. 2024. COVID-19 vaccines and adverse events of special interest: A multinational Global Vaccine Data Network (GVDN) cohort study of 99 million vaccinated individuals. Vaccine 42(9):2200–2211.

FDA (Food and Drug Administration). 2017. Final assessment of the FDA sentinel initiative. Food and Drug Administration.

FDA. 2021. FDA and CDC lift recommended pause on Johnson & Johnson (Janssen) COVID-19 vaccine use following thorough safety review. https://www.fda.gov/newsevents/press-announcements/fda-and-cdc-lift-recommended-pause-johnson-johnson-janssen-covid-19-vaccine-use-following-thorough (accessed August 7, 2025).

FDA. 2023. CDC and FDA identify a preliminary COVID-19 vaccine safety signal for persons aged 65 years and older. https://www.fda.gov/vaccines-blood-biologics/safety-availabilitybiologics/cdc-and-fda-identify-preliminary-covid-19-vaccine-safety-signal-persons-aged-65-years-and-older (accessed September 3, 2025).

FDA. 2024a. CBER biologics effectiveness and safety (BEST) system. https://www.fda.gov/vaccines-blood-biologics/safety-availability-biologics/cber-biologics-effectiveness-andsafety-best-system (accessed August 7, 2025).

FDA. 2024b. Emergency use authorization. https://www.fda.gov/emergency-preparedness-and-response/mcm-legal-regulatory-and-policy-framework/emergency-use-authorization (accessed August 7, 2025).

FDA. 2024c. Postmarketing requirements and commitments: Introduction. https://www.fda.gov/drugs/guidance-compliance-regulatory-information/postmarketing-requirements-and-commitments-introduction (accessed August 7, 2025).

Frontera, J. A., A. A. Tamborska, M. F. Doheim, D. Garcia-Azorin, H. Gezegen, A. Guekht, A. H. K. Yusof Khan, M. Santacatterina, J. Sejvar, K. T. Thakur, E. Westenberg, A. S. Winkler, E. Beghi, and Contributors from the Global COVID-19 Neuro Research Coalition. 2022. Neurological events reported after COVID-19 vaccines: An analysis of VAERS. Ann Neurol 91(6):756–771.

Gargano, J. W., M. Wallace, S. C. Hadler, G. Langley, J. R. Su, M. E. Oster, K. R. Broder, J. Gee, E. Weintraub, T. Shimabukuro, H. M. Scobie, D. Moulia, L. E. Markowitz, M. Wharton, V. V. McNally, J. R. Romero, H. K. Talbot, G. M. Lee, M. F. Daley, and S. E. Oliver. 2021. Use of mRNA COVID-19 vaccine after reports of myocarditis among vaccine recipients: Update from the advisory committee on immunization practices—United States, June 2021. MMWR Morb Mortal Wkly Rep 70(27):977–982.

Gee, J. 2024. CDC response to vaccine safety needs for the U.S. COVID-19 vaccination program—an overview of vaccine safety systems. Presentation to the Committee to Review the Centers for Disease Control and Prevention’s COVID-19 Vaccine Safety Research and Communications, Washington, DC, August 7.

Gee, J., P. Marquez, J. Su, G. M. Calvert, R. Liu, T. Myers, N. Nair, S. Martin, T. Clark, L. Markowitz, N. Lindsey, B. Zhang, C. Licata, A. Jazwa, M. Sotir, and T. Shimabukuro. 2021. First month of COVID-19 vaccine safety monitoring—United States, December 14, 2020–January 13, 2021. MMWR Morb Mortal Wkly Rep 70(8):283–288.

Gee, J., T. T. Shimabukuro, J. R. Su, D. Shay, M. Ryan, S. V. Basavaraju, K. R. Broder, M. Clark, C. Buddy Creech, F. Cunningham, K. Goddard, H. Guy, K. M. Edwards, R. Forshee, T. Hamburger, A. M. Hause, N. P. Klein, I. Kracalik, C. Lamer, D. A. Loran, M. M. McNeil, J. Montgomery, P. Moro, T. R. Myers, C. Olson, M. E. Oster, A. J. Sharma, R. Schupbach, E. Weintraub, B. Whitehead, and S. Anderson. 2024. Overview of U.S. COVID-19 vaccine safety surveillance systems. Vaccine 42 Suppl 3:125748.

Goddard, K., K. E. Hanson, N. Lewis, E. Weintraub, B. Fireman, and N. P. Klein. 2022a. Incidence of myocarditis/pericarditis following mRNA COVID-19 vaccination among children and younger adults in the United States. Ann Intern Med 175(12):1169–1771.

Goddard, K., N. Lewis, B. Fireman, E. Weintraub, T. Shimabukuro, O. Zerbo, T. G. Boyce, M. E. Oster, K. E. Hanson, J. G. Donahue, P. Ross, A. Naleway, J. C. Nelson, B. Lewin, J. M. Glanz, J. T. B. Williams, E. O. Kharbanda, W. Katherine Yih, and N. P. Klein. 2022b. Risk of myocarditis and pericarditis following BNT162b2 and mRNA-1273 COVID-19 vaccination. Vaccine 40(35):5153–5159.

Greinacher, A., T. Thiele, T. E. Warkentin, K. Weisser, P. A. Kyrle, and S. Eichinger. 2021. Thrombotic thrombocytopenia after ChAdOx1 nCOV-19 vaccination. N Engl J Med 384(22):2092–2101.

Haendel, M. A., C. G. Chute, T. D. Bennett, D. A. Eichmann, J. Guinney, W. A. Kibbe, P. R. O. Payne, E. R. Pfaff, P. N. Robinson, J. H. Saltz, H. Spratt, C. Suver, J. Wilbanks, A. B. Wilcox, A. E. Williams, C. Wu, C. Blacketer, R. L. Bradford, J. J. Cimino, M. Clark, E. W. Colmenares, P. A. Francis, D. Gabriel, A. Graves, R. Hemadri, S. S. Hong, G. Hripscak, D. Jiao, J. G. Klann, K. Kostka, A. M. Lee, H. P. Lehmann, L. Lingrey, R. T. Miller, M. Morris, S. N. Murphy, K. Natarajan, M. B. Palchuk, U. Sheikh, H. Solbrig, S. Visweswaran, A. Walden, K. M. Walters, G. M. Weber, X. T. Zhang, R. L. Zhu, B. Amor, A. T. Girvin, A. Manna, N. Qureshi, M. G. Kurilla, S. G. Michael, L. M. Portilla, J. L. Rutter, C. P. Austin, K. R. Gersing, and N. C. Consortium. 2021. The National COVID Cohort Collaborative (N3C): Rationale, design, infrastructure, and deployment. J Am Med Inform Assoc 28(3):427–443.

Hanson, K. E., K. Goddard, N. Lewis, B. Fireman, T. R. Myers, N. Bakshi, E. Weintraub, J. G. Donahue, J. C. Nelson, S. Xu, J. M. Glanz, J. T. B. Williams, J. D. Alpern, and N. P. Klein. 2022. Incidence of Guillain-Barre Syndrome after COVID-19 vaccination in the vaccine safety datalink. JAMA Netw Open 5(4):e228879.

Hause, A. M., J. Baggs, P. Marquez, T. R. Myers, J. R. Su, P. G. Blanc, J. A. Gwira Baumblatt, E. J. Woo, J. Gee, T. T. Shimabukuro, and D. K. Shay. 2022a. Safety monitoring of COVID-19 vaccine booster doses among adults—United States, September 22, 2021–February 6, 2022. MMWR Morb Mortal Wkly Rep 71(7):249–254.

Hause, A. M., J. Gee, T. Johnson, A. Jazwa, P. Marquez, E. Miller, J. Su, T. T. Shimabukuro, and D. K. Shay. 2021. Anxiety-related adverse event clusters after Janssen COVID-19 vaccination - five U.S. mass vaccination sites, April 2021. MMWR Morb Mortal Wkly Rep 70(18):685–688.

Hause, A. M., P. Marquez, B. Zhang, T. R. Myers, J. Gee, J. R. Su, P. G. Blanc, A. Thomas, D. Thompson, T. T. Shimabukuro, and D. K. Shay. 2022b. Safety monitoring of bivalent COVID-19 mRNA vaccine booster doses among persons aged >/=12 years—United States, August 31–October 23, 2022. MMWR Morb Mortal Wkly Rep 71(44):1401–1406.

Hause, A. M., P. Marquez, B. Zhang, J. R. Su, T. R. Myers, J. Gee, S. S. Panchanathan, D. Thompson, T. T. Shimabukuro, and D. K. Shay. 2023. Safety monitoring of bivalent COVID-19 mRNA vaccine booster doses among children aged 5–11 years—United States, October 12–January 1, 2023. MMWR Morb Mortal Wkly Rep 72(2):39–43.

Hause, A. M., B. Zhang, X. Yue, P. Marquez, T. R. Myers, C. Parker, J. Gee, J. Su, T. T. Shimabukuro, and D. K. Shay. 2022c. Reactogenicity of simultaneous COVID-19 mRNA booster and influenza vaccination in the US. JAMA Netw Open 5(7):e2222241.

HHS (Department of Health and Human Services). 2024. VAERS data use guide. https://vaers.hhs.gov/data/dataguide.html (accessed August 7, 2025).

Hickey, K. 2025. The PREP Act and COVID-19, part 2: The PREP Act declaration for COVID-19 countermeasures. Library of Congress, Congressional Research Service. https://www.congress.gov/crs-product/LSB10730 (accessed August 7, 2025).

Kachikis, A., J. A. Englund, M. Singleton, I. Covelli, A. L. Drake, and L. O. Eckert. 2021. Short-term reactions among pregnant and lactating individuals in the first wave of the COVID-19 vaccine rollout. JAMA Netw Open 4(8):e2121310.

Katherine Yih, W., M. F. Daley, J. Duffy, B. Fireman, D. McClure, J. Nelson, L. Qian, N. Smith, G. Vazquez-Benitez, E. Weintraub, J. T. B. Williams, S. Xu, and J. C. Maro. 2023. Tree-based data mining for safety assessment of first COVID-19 booster doses in the vaccine safety datalink. Vaccine 41(2):460–466.

Kenigsberg, T. A., K. Goddard, K. E. Hanson, N. Lewis, N. Klein, S. A. Irving, A. L. Naleway, B. Crane, T. L. Kauffman, S. Xu, M. F. Daley, L. P. Hurley, R. Kaiser, L. A. Jackson, A. Jazwa, and E. S. Weintraub. 2023a. Simultaneous administration of mRNA COVID-19 bivalent booster and influenza vaccines. Vaccine 41(39):5678–5682.

Kenigsberg, T. A., K. E. Hanson, N. P. Klein, O. Zerbo, K. Goddard, S. Xu, W. K. Yih, S. A. Irving, L. P. Hurley, J. M. Glanz, R. Kaiser, L. A. Jackson, and E. S. Weintraub. 2023b. Safety of simultaneous vaccination with COVID-19 vaccines in the Vaccine Safety Datalink. Vaccine 41(32):4658–4665.

Klein, N. 2021. Rapid cycle analysis to monitor the safety of COVID-19 vaccines in near real-time within the vaccine safety datalink: Myocarditis and anaphylaxis. Paper read at Vaccine Safety Datalink – Immunization Safety Office, CDC: Atlanta, GA.

Klein, N. P., N. Lewis, K. Goddard, B. Fireman, O. Zerbo, K. E. Hanson, J. G. Donahue, E. O. Kharbanda, A. Naleway, J. C. Nelson, S. Xu, W. K. Yih, J. M. Glanz, J. T. B. Williams, S. J. Hambidge, B. J. Lewin, T. T. Shimabukuro, F. DeStefano, and E. S. Weintraub. 2021. Surveillance for adverse events after COVID-19 mRNA vaccination. JAMA 326(14):1390–1399.

Korinthenberg, R., and J. J. Sejvar. 2020. The brighton collaboration case definition: Comparison in a retrospective and prospective cohort of children with Guillain-Barre Syndrome. J Peripher Nerv Syst 25(4):344–349.

Lee, G. M. 2021. COVID-19 vaccine safety technical (vast) subgroup. Atlanta, GA. https://stacks.cdc.gov/view/cdc/107716 (accessed January 23, 2025).

Lee, G. M., and R. Hopkins. 2021. COVID-19 vaccine safety technical (vast) subgroup: Discussion and interpretation, Atlanta, GA.

Lipkind, H. S., G. Vazquez-Benitez, M. DeSilva, K. K. Vesco, C. Ackerman-Banks, J. Zhu, T. G. Boyce, M. F. Daley, C. C. Fuller, D. Getahun, S. A. Irving, L. A. Jackson, J. T. B. Williams, O. Zerbo, M. M. McNeil, C. K. Olson, E. Weintraub, and E. O. Kharbanda. 2022. Receipt of COVID-19 vaccine during pregnancy and preterm or small-for-gestational-age at birth-eight integrated health care organizations, United States, December 15, 2020–July 22, 2021. MMWR Morb Mortal Wkly Rep 71(1):26–30.

MacNeil, J. R., J. R. Su, K. R. Broder, A. Y. Guh, J. W. Gargano, M. Wallace, S. C. Hadler, H. M. Scobie, A. E. Blain, D. Moulia, M. F. Daley, V. V. McNally, J. R. Romero, H. K. Talbot, G. M. Lee, B. P. Bell, and S. E. Oliver. 2021. Updated recommendations from the Advisory Committee on Immunization Practices for use of the Janssen (Johnson & Johnson) COVID-19 vaccine after reports of thrombosis with thrombocytopenia syndrome among vaccine recipients—United States, April 2021. MMWR Morb Mortal Wkly Rep 70(17):651–656.

Madni, S. A., A. J. Sharma, L. H. Zauche, A. V. Waters, J. F. Nahabedian, 3rd, T. Johnson, C. K. Olson, and CDC COVID-19 Vaccine Pregnancy Registry Work Group. 2024. CDC COVID-19 vaccine pregnancy registry: Design, data collection, response rates, and cohort description. Vaccine 42(7):1469–1477.

Markowitz, L. E., R. H. Hopkins, Jr., K. R. Broder, G. M. Lee, K. M. Edwards, M. F. Daley, L. A. Jackson, J. C. Nelson, L. E. Riley, V. V. McNally, R. Schechter, P. N. Whitley-Williams, F. Cunningham, M. Clark, M. Ryan, K. M. Farizo, H. L. Wong, J. Kelman, T. Beresnev, V. Marshall, D. K. Shay, J. Gee, J. Woo, M. M. McNeil, J. R. Su, T. T. Shimabukuro, M. Wharton, and H. Keipp Talbot. 2024. COVID-19 vaccine safety technical (vast) work group: Enhancing vaccine safety monitoring during the pandemic. Vaccine 42(Suppl 3):125549.

Marschner, C. A., K. E. Shaw, F. S. Tijmes, M. Fronza, S. Khullar, M. A. Seidman, P. Thavendiranathan, J. A. Udell, R. M. Wald, and K. Hanneman. 2023. Myocarditis following COVID-19 vaccination. Heart Fail Clin 19(2):251–264.

McNeil, M. M., J. Gee, E. S. Weintraub, E. A. Belongia, G. M. Lee, J. M. Glanz, J. D. Nordin, N. P. Klein, R. Baxter, A. L. Naleway, L. A. Jackson, S. B. Omer, S. J. Jacobsen, and F. DeStefano. 2014. The vaccine safety datalink: Successes and challenges monitoring vaccine safety. Vaccine 32(42):5390–5398.

Montgomery, J., M. Ryan, R. Engler, D. Hoffman, B. McClenathan, L. Collins, D. Loran, D. Hrncir, K. Herring, M. Platzer, N. Adams, A. Sanou, and L. T. Cooper, Jr. 2021. Myocarditis following immunization with mRNA COVID-19 vaccines in members of the US military. JAMA Cardiol 6(10):1202–1206.

Morgan, H. J., H. J. Clothier, G. Sepulveda Kattan, J. H. Boyd, and J. P. Buttery. 2024. Acute disseminated encephalomyelitis and transverse myelitis following COVID-19 vaccination: A self-controlled case series analysis. Vaccine 42(9):2212–2219.

Moro, P. L., G. Carlock, N. Fifadara, T. Habenicht, B. Zhang, P. Strid, and P. Marquez. 2024. Safety monitoring of bivalent mRNA COVID-19 vaccine among pregnant persons in the vaccine adverse event reporting system—United States, September 1, 2022–March 31, 2023. Vaccine 42(9):2380–2384.

Moro, P. L., P. Haber, and M. M. McNeil. 2019. Challenges in evaluating post-licensure vaccine safety: Observations from the Centers for Disease Control and Prevention. Expert Rev Vaccines 18(10):1091–1101.

Moro, P. L., L. Panagiotakopoulos, T. Oduyebo, C. K. Olson, and T. Myers. 2021. Monitoring the safety of COVID-19 vaccines in pregnancy in the US. Hum Vaccin Immunother 17(12):4705–4713.

Myers, T. R., P. L. Marquez, J. M. Gee, A. M. Hause, L. Panagiotakopoulos, B. Zhang, I. McCullum, C. Licata, C. K. Olson, S. Rahman, S. B. Kennedy, M. Cardozo, C. R. Patel, L. Maxwell, J. R. Kallman, D. K. Shay, and T. T. Shimabukuro. 2023. The V-safe after vaccination health checker: Active vaccine safety monitoring during CDC’s COVID-19 pandemic response. Vaccine 41(7):1310–1318.

Nasreen, S., Y. Jiang, H. Lu, A. Lee, C. L. Cutland, A. Gentile, N. Giglio, K. Macartney, L. Deng, B. Liu, N. Sonneveld, K. Bellamy, H. J. Clothier, G. Sepulveda Kattan, M. Naus, Z. Naveed, N. Z. Janjua, L. Nguyen, A. Hviid, E. Poukka, J. Perala, T. Leino, L. A. Chandra, J. A. Thobari, B. J. Park, N. K. Choi, N. Y. Jeong, S. A. Madhi, F. Villalobos, M. Solorzano, C. A. Bissacco, J. J. Carreras-Martinez, E. Correcher-Martinez, A. Urchueguia-Fornes, D. Roy, A. Yeomans, T. Aurelius, K. Morton, G. Di Mauro, M. C. Sturkenboom, J. J. Sejvar, K. A. Top, K. Batty, L. Ghebreab, J. B. Griffin, H. Petousis-Harris, J. Buttery, S. Black, and J. C. Kwong. 2025. Risk of Guillain-Barre Syndrome after COVID-19 vaccination or SARS-CoV-2 infection: A multinational self-controlled case series study. Vaccine 60:127291.

Oliver, S. E., J. W. Gargano, M. Marin, M. Wallace, K. G. Curran, M. Chamberland, N. McClung, D. Campos-Outcalt, R. L. Morgan, S. Mbaeyi, J. R. Romero, H. K. Talbot, G. M. Lee, B. P. Bell, and K. Dooling. 2020. The Advisory Committee on Immunization Practices’ interim recommendation for use of Pfizer-BioNTech COVID-19 vaccine—United States, December 2020. MMWR Morb Mortal Wkly Rep 69(50):1922–1924.

Oliver, S. E., M. Wallace, I. See, S. Mbaeyi, M. Godfrey, S. C. Hadler, T. C. Jatlaoui, E. Twentyman, M. M. Hughes, A. K. Rao, A. Fiore, J. R. Su, K. R. Broder, T. Shimabukuro, A. Lale, D. K. Shay, L. E. Markowitz, M. Wharton, B. P. Bell, O. Brooks, V. McNally, G. M. Lee, H. K. Talbot, and M. F. Daley. 2022. Use of the Janssen (Johnson & Johnson) COVID-19 vaccine: Updated interim recommendations from the Advisory Committee on Immunization Practices - United States, December 2021. MMWR Morb Mortal Wkly Rep 71(3):90–95.

Oster, M. E., D. K. Shay, J. R. Su, J. Gee, C. B. Creech, K. R. Broder, K. Edwards, J. H. Soslow, J. M. Dendy, E. Schlaudecker, S. M. Lang, E. D. Barnett, F. L. Ruberg, M. J. Smith, M. J. Campbell, R. D. Lopes, L. S. Sperling, J. A. Baumblatt, D. L. Thompson, P. L. Marquez, P. Strid, J. Woo, R. Pugsley, S. Reagan-Steiner, F. DeStefano, and T. T. Shimabukuro. 2022. Myocarditis cases reported after mRNA-based COVID-19 vaccination in the US from December 2020 to August 2021. JAMA 327(4):331–340.

Petersen, I., I. Douglas, and H. Whitaker. 2016. Self controlled case series methods: An alternative to standard epidemiological study designs. BMJ 354:i4515.

Phillips, A., Y. Jiang, D. Walsh, N. Andrews, M. Artama, H. Clothier, L. Cullen, L. Deng, S. Escolano, A. Gentile, G. Gidding, N. Giglio, T. Junker, W. Huang, N. Janjua, J. Kwong, J. Li, S. Nasreen, M. Naus, Z. Naveed, A. Pillsbury, J. Stowe, T. Vo, J. Buttery, H. Petousis-Harris, S. Black, and A. Hviid. 2023. Background rates of adverse events of special interest for COVID-19 vaccines: A Multinational Global Vaccine Data Network (GVDN) analysis. Vaccine 41(42):6227–6238.

Polack, F. P., S. J. Thomas, N. Kitchin, J. Absalon, A. Gurtman, S. Lockhart, J. L. Perez, G. Perez Marc, E. D. Moreira, C. Zerbini, R. Bailey, K. A. Swanson, S. Roychoudhury, K. Koury, P. Li, W. V. Kalina, D. Cooper, R. W. Frenck, Jr., L. L. Hammitt, O. Tureci, H. Nell, A. Schaefer, S. Unal, D. B. Tresnan, S. Mather, P. R. Dormitzer, U. Sahin, K. U. Jansen, W. C. Gruber, and C4591001 Clinical Trial Group. 2020. Safety and efficacy of the BNT162b2 mRNA COVID-19 vaccine. N Engl J Med 383(27):2603–2615.

Razen, S. 2021. Public comment at CDC ACIP April 23, 2021 meeting (starting at 30:35). https://www.youtube.com/watch?v=j5KqOLNpTt4&list=PLvrp9iOILTQb6D9e1YZWpbUvzfptNMKx2&index=51 (accessed August 25, 2025).

Rosenblum, H. G., J. Gee, R. Liu, P. L. Marquez, B. Zhang, P. Strid, W. E. Abara, M. M. McNeil, T. R. Myers, A. M. Hause, J. R. Su, L. E. Markowitz, T. T. Shimabukuro, and D. K. Shay. 2022. 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 22(6):802–812.

Rosenblum, H. G., S. C. Hadler, D. Moulia, T. T. Shimabukuro, J. R. Su, N. K. Tepper, K. C. Ess, E. J. Woo, A. Mba-Jonas, M. Alimchandani, N. Nair, N. P. Klein, K. E. Hanson, L. E. Markowitz, M. Wharton, V. V. McNally, J. R. Romero, H. K. Talbot, G. M. Lee, M. F. Daley, S. A. Mbaeyi, and S. E. Oliver. 2021. Use of COVID-19 vaccines after reports of adverse events among adult recipients of Janssen (Johnson & Johnson) and mRNA COVID-19 vaccines (Pfizer-BioNTech and Moderna): Update from the Advisory Committee on Immunization Practices—United States, July 2021. MMWR Morb Mortal Wkly Rep 70(32):1094–1099.

Sakaeda, T., A. Tamon, K. Kadoyama, and Y. Okuno. 2013. Data mining of the public version of the FDA adverse event reporting system. Int J Med Sci 10(7):796–803.

Salmon, D., D. J. Opel, M. Z. Dudley, J. Brewer, and R. Breiman. 2021. Reflections on governance, communication, and equity: Challenges and opportunities in COVID-19 vaccination. Health Aff (Millwood) 40(3):419–425.

See, I. 2021. Updates on thrombosis with thrombocytopenia syndrome (TTS). Paper read at Advisory Committee on Immunization Practices (ACIP) meeting, December 16, Atlanta, GA.

See, I., A. Lale, P. Marquez, M. B. Streiff, A. P. Wheeler, N. K. Tepper, E. J. Woo, K. R. Broder, K. M. Edwards, R. Gallego, A. I. Geller, K. A. Jackson, S. Sharma, K. R. Talaat, E. B. Walter, I. J. Akpan, T. L. Ortel, V. C. Urrutia, S. C. Walker, J. C. Yui, T. T. Shimabukuro, A. Mba-Jonas, J. R. Su, and D. K. Shay. 2022. Case series of thrombosis with thrombocytopenia syndrome after COVID-19 vaccination—United States, December 2020 to August 2021. Ann Intern Med 175(4):513–522.

See, I., J. R. Su, A. Lale, E. J. Woo, A. Y. Guh, T. T. Shimabukuro, M. B. Streiff, A. K. Rao, A. P. Wheeler, S. F. Beavers, A. P. Durbin, K. Edwards, E. Miller, T. A. Harrington, A. Mba-Jonas, N. Nair, D. T. Nguyen, K. R. Talaat, V. C. Urrutia, S. C. Walker, C. B. Creech, T. A. Clark, F. DeStefano, and K. R. Broder. 2021. US case reports of cerebral venous sinus thrombosis with thrombocytopenia after Ad26.COV2.S vaccination, March 2 to April 21, 2021. JAMA 325(24):2448–2456.

Sejvar, J. J., K. S. Kohl, J. Gidudu, A. Amato, N. Bakshi, R. Baxter, D. R. Burwen, D. R. Cornblath, J. Cleerbout, K. M. Edwards, U. Heininger, R. Hughes, N. Khuri-Bulos, R. Korinthenberg, B. J. Law, U. Munro, H. C. Maltezou, P. Nell, J. Oleske, R. Sparks, P. Velentgas, P. Vermeer, M. Wiznitzer, and Brighton Collaboration GBS Working Group. 2011. Guillain-Barre Syndrome and Fisher Syndrome: Case definitions and guidelines for collection, analysis, and presentation of immunization safety data. Vaccine 29(3):599–612.

Shay, D. K., J. Gee, J. R. Su, T. R. Myers, P. Marquez, R. Liu, B. Zhang, C. Licata, T. A. Clark, and T. T. Shimabukuro. 2021. Safety monitoring of the Janssen (Johnson & Johnson) COVID-19 vaccine—United States, March–April 2021. MMWR Morb Mortal Wkly Rep 70(18):680–684.

Shimabukuro, T. 2021a. Allergic reactions including anaphylaxis after receipt of the first dose of Moderna COVID-19 vaccine—United States, December 21, 2020–January 10, 2021. Am J Transplant 21(3):1326–1331.

Shimabukuro, T. T. 2021b. Thrombosis with thrombocytopenia syndrome (TTS) following Janssen COVID-19 vaccine, April 23. Atlanta, GA. https://stacks.cdc.gov/view/cdc/107508 (accessed August 25, 2025).

Shimabukuro, T. 2022. Update on myocarditis following mRNA COVID-19 vaccination. Paper read at Vaccines and Related Biological Products Advisory Committee (VRBPAC), June 14, Silver Spring, MD.

Shimabukuro, T. T. 2023a. COVID-19 mRNA bivalent booster vaccine safety. Paper read at Advisory Committee on Immunization Practices (ACIP) meeting, February 24, Atlanta, GA.

Shimabukuro, T. T. 2023b. Update: V-safe after vaccination health checker. Paper read at Advisory Committee on Immunization Practices (ACIP) meeting, April 19, Atlanta, GA.

Shimabukuro, T. T., M. Cole, and J. R. Su. 2021a. Reports of anaphylaxis after receipt of mRNA COVID-19 vaccines in the US—December 14, 2020–January 18, 2021. JAMA 325(11):1101–1102.

Shimabukuro, T. T., S. Y. Kim, T. R. Myers, P. L. Moro, T. Oduyebo, L. Panagiotakopoulos, P. L. Marquez, C. K. Olson, R. Liu, K. T. Chang, S. R. Ellington, V. K. Burkel, A. N. Smoots, C. J. Green, C. Licata, B. C. Zhang, M. Alimchandani, A. Mba-Jonas, S. W. Martin, J. M. Gee, D. M. Meaney-Delman, and CDC V-safe COVID-19 Pregnancy Registry Team. 2021b. Preliminary findings of mRNA COVID-19 vaccine safety in pregnant persons. N Engl J Med 384(24):2273–2282.

Shimabukuro, T. T., and N. Klein. 2023. COVID-19 mRNA bivalent booster vaccine safety. Paper read at Vaccines and Related Biological Products Advisory Committee meeting, January 26, Silver Spring, MD.

Shimabukuro, T. T., M. Nguyen, D. Martin, and F. DeStefano. 2015. Safety monitoring in the Vaccine Adverse Event Reporting System (VAERS). Vaccine 33(36):4398–4405.

U.S. GAO (Government Accountability Office). 2021. Covid-19: Critical vaccine distribution, supply chain, program integrity, and other challenges require focused federal attention. U.S. Government Accountability Office.

Varricchio, F., J. Iskander, F. Destefano, R. Ball, R. Pless, M. M. Braun, and R. T. Chen. 2004. Understanding vaccine safety information from the Vaccine Adverse Event Reporting System. Pediatr Infect Dis J 23(4):287–294.

Wallace, M., D. Moulia, A. E. Blain, E. K. Ricketts, F. S. Minhaj, R. Link-Gelles, K. G. Curran, S. C. Hadler, A. Asif, M. Godfrey, E. Hall, A. Fiore, S. Meyer, J. R. Su, E. Weintraub, M. E. Oster, T. T. Shimabukuro, D. Campos-Outcalt, R. L. Morgan, B. P. Bell, O. Brooks, H. K. Talbot, G. M. Lee, M. F. Daley, and S. E. Oliver. 2022. The Advisory Committee on Immunization Practices’ recommendation for use of Moderna COVID-19 vaccine in adults aged >/=18 years and considerations for extended intervals for administration of primary series doses of mRNA COVID-19 vaccines—United States, February 2022. MMWR Morb Mortal Wkly Rep 71(11):416–421.

Williams, S. E., N. P. Klein, N. Halsey, C. L. Dekker, R. P. Baxter, C. D. Marchant, P. S. LaRussa, R. C. Sparks, J. I. Tokars, B. A. Pahud, L. Aukes, K. Jakob, S. Coronel, H. Choi, B. A. Slade, and K. M. Edwards. 2011. Overview of the clinical consult case review of adverse events following immunization: Clinical Immunization Safety Assessment (CISA) network 2004–2009. Vaccine 29(40):6920–6927.

Wong, H. L., E. Tworkoski, C. Ke Zhou, M. Hu, D. Thompson, B. Lufkin, R. Do, L. Feinberg, Y. Chillarige, R. Dimova, P. C. Lloyd, T. MaCurdy, R. A. Forshee, J. A. Kelman, A. Shoaibi, and S. A. Anderson. 2023. Surveillance of COVID-19 vaccine safety among elderly persons aged 65 years and older. Vaccine 41(2):532–539.

Wong, K. K., C. M. Heilig, A. Hause, T. R. Myers, C. K. Olson, J. Gee, P. Marquez, P. Strid, and D. K. Shay. 2022. Menstrual irregularities and vaginal bleeding after COVID-19 vaccination reported to V-safe active surveillance, USA in December, 2020–January, 2022: An observational cohort study. Lancet Digit Health 4(9):e667–e675.

Woo, E. J., A. Mba-Jonas, R. B. Dimova, M. Alimchandani, C. E. Zinderman, and N. Nair. 2021. Association of receipt of the Ad26.COV2.S COVID-19 vaccine with presumptive Guillain-Barre Syndrome, February–July 2021. JAMA 326(16):1606–1613.

Xu, S., R. Huang, L. S. Sy, S. C. Glenn, D. S. Ryan, K. Morrissette, D. K. Shay, G. Vazquez-Benitez, J. M. Glanz, N. P. Klein, D. McClure, E. G. Liles, E. S. Weintraub, H. F. Tseng, and L. Qian. 2021. 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 70(43):1520–1524.

Xu, S., R. Huang, L. S. Sy, V. Hong, S. C. Glenn, D. S. Ryan, K. Morrissette, G. Vazquez-Benitez, J. M. Glanz, N. P. Klein, B. Fireman, D. McClure, E. G. Liles, E. S. Weintraub, H. F. Tseng, and L. Qian. 2023. A safety study evaluating non-change to COVID-19 mortality risk following COVID-19 vaccination. Vaccine 41(3):844–854.

Xu, S., L. S. Sy, V. Hong, P. Farrington, S. C. Glenn, D. S. Ryan, A. M. Shirley, B. J. Lewin, H. F. Tseng, G. Vazquez-Benitez, J. M. Glanz, B. Fireman, D. L. McClure, L. P. Hurley, O. Yu, M. Wernecke, N. Smith, E. S. Weintraub, and L. Qian. 2024. Mortality risk after COVID-19 vaccination: A self-controlled case series study. Vaccine 42(7):1731–1737.

Yih, W. K., M. F. Daley, J. Duffy, B. Fireman, D. McClure, J. Nelson, L. Qian, N. Smith, G. Vazquez-Benitez, E. Weintraub, J. T. B. Williams, S. Xu, and J. C. Maro. 2023. A broad assessment of COVID-19 vaccine safety using tree-based data-mining in the vaccine safety datalink. Vaccine 41(3):826–835.

Yih, W. K., J. Duffy, J. R. Su, S. Bazel, B. Fireman, L. Hurley, J. C. Maro, P. Marquez, P. Moro, N. Nair, J. Nelson, N. Smith, M. Sundaram, G. Vasquez-Benitez, E. Weintraub, S. Xu, and T. Shimabukuro. 2024. Tinnitus after COVID-19 vaccination: Findings from the Vaccine Adverse Event Reporting System and the Vaccine Safety Datalink. Am J Otolaryngol 45(6):104448.

Yousaf, A. R., M. M. Cortese, A. W. Taylor, K. R. Broder, M. E. Oster, J. M. Wong, A. Y. Guh, D. W. McCormick, S. Kamidani, E. P. Schlaudecker, K. M. Edwards, C. B. Creech, M. A. Staat, E. D. Belay, P. Marquez, J. R. Su, M. B. Salzman, D. Thompson, A. P. Campbell, and MIS-C Investigation Authorship Group. 2022. Reported cases of multisystem inflammatory syndrome in children aged 12–20 years in the USA who received a COVID-19 vaccine, December, 2020, through August, 2021: A surveillance investigation. Lancet Child Adolesc Health 6(5):303–312.

Zauche, L. H., B. Wallace, A. N. Smoots, C. K. Olson, T. Oduyebo, S. Y. Kim, E. E. Petersen, J. Ju, J. Beauregard, A. J. Wilcox, C. E. Rose, D. M. Meaney-Delman, S. R. Ellington, for the CDC V-safe COVID-19 Pregnancy Registry Team. 2021. Receipt of mRNA COVID-19 vaccines and risk of spontaneous abortion. N Engl J Med 385(16):1533–1535.

Zhou, Z. H., M. M. Cortese, J. L. Fang, R. Wood, D. S. Hummell, K. A. Risma, A. E. Norton, M. KuKuruga, S. Kirshner, R. L. Rabin, C. Agarabi, M. A. Staat, N. Halasa, R. E. Ware, A. Stahl, M. McMahon, P. Browning, P. Maniatis, S. Bolcen, K. M. Edwards, J. R. Su, S. Dharmarajan, R. Forshee, K. R. Broder, S. Anderson, and S. Kozlowski. 2023. Evaluation of association of anti-PEG antibodies with anaphylaxis after mRNA COVID-19 vaccination. Vaccine 41(28):4183–4189.

Zou, C., X. Xue, and J. Qian. 2022. Characteristics and comparison of adverse events of coronavirus disease 2019 vaccines reported to the United States Vaccine Adverse Event Reporting System between 14 December 2020 and 8 October 2021. Front Med (Lausanne) 9:826327.

This page intentionally left blank.