Safety of COVID-19 Vaccination in United States Children Ages 5 to 11 Years

Centers for Disease Control (CDC) established v-safe (https://vsafe.cdc.gov) to monitor health effects after COVID-19 vaccination. Parents and guardians can enroll children in v-safe after any dose. Health surveys are sent to parents or guardians via text messages that link to web-based surveys on days 0 to 7 after vaccination; then weekly through 6 weeks after vaccination; and then 3, 6, and 12 months after vaccination. Health surveys sent in the first week include questions about injection site and systemic reactions and health impacts; injection site and systemic reactions can be further described by severity as mild, moderate, or severe (Supplemental Information). CDC’s v-safe call center contacts a parent or guardian when a report indicates that a child received medical care after vaccination and encourages completion of a VAERS report, if indicated.

VAERS is a United States spontaneous reporting system comanaged by CDC and FDA that accepts reports of adverse events after vaccination from any source, including health care providers, vaccine manufacturers, and members of the public.⁴  Under COVID-19 vaccine EUAs, vaccination providers are required to report adverse events, including hospitalizations and deaths temporally associated with vaccination.⁵  Symptoms, signs, and diagnostic findings in VAERS reports are assigned Medical Dictionary for Regulatory Activities (MedDRA) preferred terms.⁶  More than 1 MedDRA preferred term may be assigned to a report; these terms do not necessarily correspond to medically verified diagnoses. VAERS reports are classified as serious using the US federal regulatory definition: if hospitalization, prolongation of hospitalization, life-threatening illness, permanent disability, congenital anomaly or birth defect, or death are reported.⁷  VAERS staff follow-up on serious reports to obtain additional information, including medical records. Death certificates and autopsy reports are obtained for reports of death, if available. CDC and FDA physicians assessed all serious reports, including reports of death, to form a clinical impression based on available information; each report was only assigned 1 impression on the basis of consensus opinion.

Reports to VAERS of multisystem inflammatory syndrome in children (MIS-C) after receipt of BNT-162b2 vaccine were adjudicated using methods similar to those previously described, which included input from CDC’s Clinical Immunization Safety Assessment Project (Supplemental Information).⁸  The Brighton Collaboration case definitions for anaphylaxis and seizure were used to identify and classify reports coded for these conditions (Supplemental Information).^9,10  Reports to VAERS of myocarditis after receipt of BNT-162b2 vaccine were identified by searching for selected MedDRA preferred terms (Supplemental Information), as well as reported symptom text indicating “myocarditis” or “increased/abnormal troponins.” CDC attempted to collect records and confirm information from health care providers to determine if the CDC myocarditis case definition (Supplemental Information) was met.⁸ 

CDC’s VSD uses rapid cycle analysis (RCA) to conduct near real-time sequential monitoring for prespecified adverse events.¹¹  RCA monitoring of COVID-19 vaccines began in December 2020, and when the BNT-162b2 EUA was extended to include children ages 5 to 11 years, monitoring began in this age group. The participating VSD sites included: Denver Health, Health-Partners, Kaiser Permanente (Colorado, Northern California, Northwest, Southern California, and Washington) and Marshfield Clinic. A description of COVID-19 vaccine surveillance has been published.¹²  In addition to the outcomes being tested sequentially, we monitored for records of acute respiratory distress syndrome, anaphylaxis, MIS-C, and narcolepsy, conditions for which comparator groups were not available. We reviewed medical records and adjudicated cases of anaphylaxis, MIS-C, and myocarditis (Supplemental Information) to verify diagnosis and assess timing of symptom onset.

It is possible for persons to be included in >1 safety monitoring system; for example, an adverse event identified in VSD may also be reported to VAERS. We are not able to identify or remove possible duplications because of deidentification procedures used by the systems. However, the 3 monitoring systems include surveillance for different outcomes and multiple reports of adverse outcomes in a single individual would not affect the interpretation of findings from any of the systems.

Data collected from v-safe and VAERS from November 3, 2021 to February 27, 2022 and from VSD from October 31, 2021 to February 26, 2022, among children ages 5 to 11 years who received BNT-162b2 COVID-19 vaccine were analyzed.

We described reactions and health impacts reported in v-safe health surveys for children who had at least 1 health check-in survey on days 0 to 7 by frequency and reaction severity.

We described nonserious and serious reports to VAERS by MedDRA preferred terms and clinical impressions, respectively. Reporting rates for myocarditis were calculated for VAERS reports that met the CDC case definition for myocarditis (Supplemental Information), by using doses administered as the denominator, and stratified by sex and dose number.³ 

We compared outcome rates in VSD data during a risk interval 1 to 21 days after vaccination to rates on the same calendar days among persons who were then in a comparison interval 22 to 42 days after vaccination.¹²  Vaccinees contributed person-time to each analysis in a 21 day risk interval after dose 1 or 2; they later contributed person-time as comparators in an interval 22 to 42 days after the most recent dose. We used Poisson regression to estimate rate ratio and 95% confidence interval (CI) adjusted for age, sex, calendar day, site, and race and ethnicity groups. We conducted weekly 1-sided sequential tests of the null hypothesis that vaccination did not affect risk during a 1- to 21-day risk interval, using a 1-sided P < .0061 to account for 1 year of planned weekly analyses. Because of a lack of appropriate comparators, we conducted descriptive analyses only for anaphylaxis and MIS-C. SAS software (version 9.4; SAS Institute) was used to conduct all analyses; Lucidchart was used to create the flowchart.

These surveillance activities were reviewed by CDC and conducted consistent with applicable federal law and CDC policy (45 C.F.R. part 46.102(l)(2), 21 C.F.R. part 56; 42 U.S.C. §241(d); 5 U.S.C. §552a; 44 U.S.C. §3501 et seq.); VSD RCA surveillance was approved by the institutional review boards of each participating site with a waiver of informed consent.

There were 48 795 children ages 5 to 11 years who received BNT-162b2 vaccination enrolled in v-safe after dose 1 from November 3, 2021 to February 27, 2022; the median age was 8 years, 24 243 (49.7%) were girls (Table 1), and information for dose 2 was available for 39 416. Most registrants reported BNT-162b2 vaccine was administered alone, without other vaccinations (96.4% and 99.3% for dose 1 and 2, respectively). Seasonal influenza vaccine was the most frequently simultaneously administered (Supplemental Table 10).

Injection site reactions (Table 2) were more frequently reported in the week after dose 2 (22 396; 56.8%) than dose 1 (26 778; 54.9%); likewise, systemic reactions were more frequently reported in the week after dose 2 (16 161; 41.0%) than dose 1 (17 214; 35.3%). For both doses, reactions were most frequently reported the day after vaccination. Injection site pain, fatigue, headache, fever, and myalgia were among the most frequently reported reactions and were mostly mild to moderate in severity (Table 3).

Health impacts, including inability to perform normal daily activities, inability to attend school, and receipt of medical care, were more frequently reported in the week after dose 2 than dose 1. Health impacts were most frequently reported the day after vaccination. ∼7.6% of parents reported that their child was unable to perform normal daily activities on the day after receipt of dose 2, 9.0% reported their child was unable to attend school, and 0.3% reported their child received medical care. In the weeks after dose 1 and dose 2, a visit to an outpatient clinic visit was the most common type of care sought (328; 0.7% and 235; 0.6%, respectively). Hospitalization in the week after vaccination was reported for 10 (0.02%) and 6 (0.02%) children after dose 1 and 2, respectively. Staff members from v-safe’s call center attempted to contact the parents or guardians of all hospitalized children; 5 said the report was made in error or hospitalization was unrelated to vaccination. Information regarding reason for hospitalization was available for 8 children and included appendicitis (2), respiratory infection (2), vomiting and dehydration (1), acute febrile illness (1), MIS-C (1) and retropharyngeal cellulitis (1).

From November 3, 2021 to February 27, 2022, VAERS received and processed 7578 reports of adverse events for children ages 5 to 11 years who received BNT-162b2 vaccine; the median age was 8 years, and 3558 (47.0%) reports were for girls (Table 4). Most reports indicated BNT-162b2 vaccine was administered alone (7403; 97.7%); seasonal influenza vaccine was the most frequently simultaneously administered vaccine (148; 2.0%) (Supplemental Table 11).

Most VAERS reports were classified as nonserious (7379; 97.4%) (Table 5). Among the most commonly reported nonserious events were those related to an administration error, including product preparation issue (1434; 19.4%), incorrect dose administered (1361; 18.4%), product storage error (417; 5.7%), underdose (388; 5.3%), and expired product administered (246; 3.3%). Each report could include several terms related to an administration error; for example, the terms “incorrect dose administered” and “underdose” could both appear in a report. Terms related to an administration error were often accompanied by a term specifying “no adverse event” (1651; 22.4%). Other frequently reported nonserious events included those related to syncope (371; 5.0%), including dizziness (387; 5.2%), systemic reactions known to be associated with the vaccine (fever [541; 7.3%], vomiting [541; 7.3%], headache [465; 6.3%], nausea [330; 4.5%]), urticaria [312; 4.2%], and rash [311; 4.2%]).

Among the 194 serious reports to VAERS, the most common clinical impressions included MIS-C (26; 13.4%), seizure (21; 10.8%), myocarditis (19; 9.7%), appendicitis (13; 6.7%), and allergic reaction (8; 4.1%). Of the 26 reports identified as possible MIS-C, 23 have been reviewed and adjudicated; 3 were under review. Of the 23 adjudicated, 21 met the CDC MIS-C case definition (Fig 1), 19 had laboratory evidence of past or recent severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection (Supplemental Information). All were discharged from the hospital. There were 126 reports of seizure with symptom onset on days 0 to 21 after vaccination, of which 21 were serious. Of these 21, 15 met the Brighton Collaboration case definition; 10 confirmed seizure cases were new-onset seizures, including 2 patients with potential evolving seizure disorders who had evidence of seizures before vaccination. There were 7 preliminary reports of anaphylaxis with symptom onset during days 0 to 7 after vaccination (Table 6); 2 were serious. Of these 7 reports, 2 met the Brighton Collaboration case definition, 1 did not meet the case definition, 1 remains under review, and 3 lacked enough information to confirm anaphylaxis. Assuming the unconfirmed reports met the Brighton Collaboration case definition, the estimated reporting rate of anaphylaxis would be 0.30 cases per million doses administered.

Forty-five preliminary reports of myocarditis with symptom onset during days 0 to 21 after vaccination were identified. On review, 23 were classified as reports of myocarditis; 19 met the regulatory definition of serious (Table 5), of which 17 (37.8%) were verified by health care provider interview or medical record review to meet the CDC case definition for myocarditis. Among the 17 verified cases, 12 were male, and 15 were hospitalized. As of April 22, 2022, 13 had recovered symptomatically and 2 were recovering. When using the generally accepted risk interval of 0 to 7 days postvaccination, the reporting rate for verified vaccine-associated myocarditis after BNT-162b2 was greater among boys than girls and greater after dose 2 (Table 7). Among boys, the reporting rate of verified myocarditis during days 0 to 7 after dose 2 was 2.2 per 1 million doses administered.

VAERS received 4 reports of death after vaccination during the analytic period. Two occurred in girls, ages 5 and 6 years, who had complex medical histories, including autonomic instability and frequent admissions to pediatric intensive care; autopsy was not performed for either decedent. One death occurred in a girl aged 7 years; on autopsy, evidence of influenza infection was documented. One death occurred in an 8-year-old boy, 8 days after dose 2. He experienced nausea and vomiting the night before his death; an autopsy was performed and results are pending. The available information for these reports did not support a causal association between vaccination and death for any of the decedents.

From October 31, 2021 to February 26, 2022, 726 820 doses of BNT-162b2 vaccine (384 905 dose 1 341  915 dose 2) were administered in a cohort of 889 263 children ages 5 to 11 years enrolled in the participating sites. Forty-three percent of children in this age group received at least 1 vaccine dose; 38% received 2 doses. For dose 1, the median age was 8 years and 49.3% were girls (Table 8). Inactivated influenza vaccines were the type most commonly administered together with BNT-162b2 vaccine (5.6%) (Supplemental Table 12). The number of cases of prespecified outcomes during the 21-day risk interval ranged from 0 for 13 of 19 outcomes (eg, Kawasaki disease) to 40 for appendicitis (1 per 1000 person-years) (Table 9). None of the analyses comparing outcomes in the risk versus comparison intervals approached the criterion for a statistical signal (1-sided P < .0061).

Seven potential cases of myocarditis were identified among children ages 5 to 11 years during days 1 to 21 after vaccination. After medical record review and adjudication, 4 of the 7 cases were confirmed as myocarditis; 2 occurred during the 7 days after dose 2 vaccination. Three cases that were not confirmed had a history of myocarditis, congenital heart defect, or chest pain without laboratory findings consistent with myocarditis. Three cases of anaphylaxis were identified in electronic health records during days 0 to 1 after vaccination after dose 1. After medical record review and adjudication, 2 met the Brighton Collaboration case definition and 1 did not (5.2 [95% CI, 0.6–18.7] cases per million dose 1 vaccinations and 2.78 [95% CI, 0.3–9.9] per million total doses). Five potential cases of MIS-C were identified and 4 were confirmed after medical record review and adjudication. All confirmed cases had both documented COVID-19 infection and COVID-19 vaccination before a diagnosis of MIS-C.

We describe safety findings from 3 surveillance systems, v-safe, VAERS, and VSD, during a period when approximately 16 million doses of BNT-162b2 vaccine were administered to United States children ages 5 to 11 years.³  Known and expected systemic reactions were less frequently reported for children ages 5 to 11 years (35.3% in the week after dose 1 and 41.0% in the week after dose 2) than for older children (48.9% in the week after dose 1 and 63.4% in the week after dose 2).^1,2,13  It is possible that systemic reactions are less frequently reported for younger children because they receive a smaller amount of mRNA (10 μg) than persons aged ≥12 years (30 μg).¹  Administration errors were among the most frequently reported events to VAERS, perhaps because of confusion about the recommended dosage for this age group. Importantly, most reports classified as administration errors did not include any mention of an adverse health event.

Similar to systemic reactions, health impacts were more frequently reported to v-safe in the week after dose 2 than dose 1. In the week after dose 2, inability to complete daily activities was less frequently reported for children ages 5 to 11 years (7.8%) than for adolescents ages 12 to 15 years (23.1%).¹³  Inability to attend school and receipt of medical care in the week after dose 2 were more frequently reported for children ages 5 to 11 years (10.9% and 1.2%, respectively) than for adolescents ages 12 to 15 years (6.1% and 0.8%).¹³  Hospitalization in the week after vaccination was uncommon among both age groups; among the 16 children ages 5 to 11 years who were reportedly hospitalized, information was available for 8 and did not indicate a causal association with vaccination.¹³ 

There were 2 confirmed cases of anaphylaxis in the VSD, for a rate of 5.2 per million dose 1 vaccinations and 2.7 per million total doses administered. These findings are consistent with an estimated rate of confirmed anaphylaxis among persons aged ≥12 years of 5.0 (95% CI, 3.5–6.9) per million BNT-162b2 total doses.¹⁴ 

VAERS accepts all reports regardless of biological or clinical plausibility for a causal association between a reported outcome and vaccination.⁷  Four deaths after BNT-162b2 vaccination were reported to VAERS; no evidence of a causal association between vaccination and death was found. VAERS received 26 reports of MIS-C and of those adjudicated, nearly all had evidence of past or recent SARS-CoV-2 infection. Similarly, in the VSD, all confirmed cases of MIS-C after vaccination had documented COVID-19 infection before a diagnosis of MIS-C. The results of enhanced surveillance for MIS-C in children aged 5 to 11 years who had onset within 90 days of their last COVID-19 vaccine dose is ongoing; a more detailed report is forthcoming. Among persons aged 12 to 20 years, the overall reporting rate for MIS-C after COVID-19 vaccination was 1.0 case per million persons who received ≥1 dose and most had evidence of past or recent SARS-CoV-2 infection. The potential contribution of vaccination, if any, to these illnesses is unknown.⁸ 

Myocarditis is a rare adverse event associated with mRNA-based COVID-19 vaccination.^15,16  The reporting rate of verified myocarditis during days 0 to 7 after dose 2 was lower among boys ages 5 to 11 years (2.2 per 1 million doses administered) than boys ages 12 to 15 years (45.7 per 1 million doses administered).¹⁷  In weekly sequential analyses of VSD data, no signal for an increased risk of myocarditis after vaccination was found.

Simultaneous vaccination with BNT-162b2 and other vaccines among children ages 5 to 11 years was not common in our data. Current CDC clinical guidance states that COVID-19 vaccines may be administered without regard to timing of other vaccines, including simultaneous administration of COVID-19 vaccine and other vaccines.¹⁸ 

The findings in this report are subject to several limitations. Data from each of these surveillance systems may not be representative of the vaccinated United States population. V-safe is a new, voluntary system and may be more likely to include data from vaccinated persons who experience an adverse event than those who do not. VAERS is subject to reporting biases, especially underreporting of nonserious events. VSD includes insured populations with ready access to health care. The nature of the passive surveillance data from VAERS means findings from this system generally cannot be used to establish causality.⁴  Data from each surveillance system are preliminary. Data collection, including review of myocarditis reports, is ongoing in each system. Our priority was to summarize the most salient available safety data and make it available for timely risk-benefit assessments regarding use of BNT-162b2 vaccine in US children ages 5 to 11 years.

The Advisory Committee on Immunization Practices and the American Academy of Pediatrics currently recommend BNT-162b2 vaccination for children ages 5 to 11 years to prevent COVID-19 and its potentially serious complications.^19,20  However, as of March 7, 2022, only ∼26% of children ages 5 to 11 years have received 2 doses of COVID-19 vaccine.³  The updated safety findings presented here, collected during the administration of ∼16 million doses of BNT-162b2 vaccine to children ages 5 to 11 years, are consistent with previous findings for this age group and demonstrate a favorable safety profile. Myocarditis after BNT-162b2 vaccination for children ages 5 to 11 years was rarely reported and reporting rates were lower than for older children. Serious allergic reactions were rare and occurred at a rate similar to that observed in other age groups. Systemic reactions are relatively common among children ages 5 to 11 years but usually are clinically mild and resolve quickly.

Karen Broder, Jonathan Duffy, Michael McNeil, Satoshi Kamidani, Isaac McCullum, Thomas G. Boyce, the Clinical Immunization Safety Assessment Project team.

BNT-162b2

Pfizer-BioNTech mRNA COVID-19 vaccine

CDC

Centers for Disease Control and Prevention

CI

confidence interval

COVID-19

coronavirus disease 2019

EUA

Emergency Use Authorization

FDA

Food and Drug Administration

MedDRA

Medical Dictionary for Regulatory Activities

MIS-C

multisystem inflammatory syndrome in children

mRNA

messenger RNA

RCA

rapid cycle analysis

RR

rate ratios

SARS-CoV-2

severe acute respiratory syndrome coronavirus 2

VAERS

Vaccine Adverse Events Reporting System

VSD

Vaccine Safety Datalink