On June 18, the U.S. Centers for Disease Control and Prevention (CDC) officially recommended Pfizer and Moderna COVID-19 vaccines for all children between the ages of 6 months and 5 years. While the Food and Drug Administration (FDA) is the agency responsible for authorizing emergency use of vaccines, it’s the CDC that crafts subsequent messaging, makes specific recommendations, and prioritizes who can, should, or should not get vaccinated. In her briefing, CDC Director Rochelle Walensky strongly urged all parents of the nearly 20 million American children in this age group to vaccinate them as soon as possible.
For some parents, Walensky’s briefing came as a huge relief. But if polling from May is anything to go by, a larger number of parents likely greeted the recommendation with skepticism. Even before the underwhelming trial results came out, only 18% of surveyed parents reported that they planned to vaccinate their babies and toddlers. Nationally, uptake in minors between the ages of 5 and 11 as of June 22, 2022, was 29% receiving two doses, and 36% receiving one, but vaccine requirements for sports, camps, and other activities likely drove an unknown percentage of vaccination in this age group.
There remains, moreover, no solid consensus among physicians about the importance of vaccinating healthy children against COVID-19. A survey from December 2021 indicates that as many as 30%-40% may not be recommending COVID vaccination for children ages 5 to 17, to say nothing of infants. A recent editorial in The Lancet expressed uncertainty about whether the benefits of vaccinating healthy 5- to 11-year-olds outweigh the risks, especially in those with a history of infection.
The gap between the CDC’s enthusiasm for vaccinating all children against COVID and that of parents and health care providers is unlikely to be bridged by approval under Emergency Use Authorization. Approval for the COVID vaccines in infants and toddlers is based on two trials that used changes in antibody levels as an estimate of efficacy, but did not assess protection from severe disease, hospitalization, or multisystem inflammatory syndrome in children (MIS-C), important outcomes that parents worry about. In a Food and Drug Administration (FDA) meeting on June 28, Pfizer Vice President for Viral Vaccines Kena Swanson even acknowledged that “there is no established correlate” between antibody levels and protection from disease.
In the Pfizer trial, the confidence interval—which shows the possible range of protection level—was alarmingly wide, with the lower bound suggesting the possibility of a 380% increase in the chance of infection after the third dose. Additionally, neither trial met the 50% efficacy requirement established by the FDA for approval of adult COVID vaccines. Peter Marks, the FDA’s top vaccine official, told Congress in May that the efficacy requirement would be lowered for the pediatric vaccine simply because vaccine efficacy against the omicron variant was lower in general.
With rates of severe disease now much lower in children than at the start of the pandemic—due to higher levels of natural immunity and lower rates of severe disease caused by omicron—trials would have needed to enroll hundreds of thousands of children, if not over a million, in order to detect a significant impact of the pediatric vaccine against severe disease. Vaccine companies could have conducted such time-consuming and costly trials, especially if there had been interest in international collaboration. But there was no economic incentive to do so, and every economic incentive not to: Speed, not providing meaningful information to parents and physicians about safety and efficacy, was the priority of U.S. regulatory agencies.
Because Pfizer and Moderna were permitted to seek approval for pediatric COVID vaccines under the emergency use pathway, Moderna only enrolled 6,300 total children in trials (4,700 in the vaccine group and 1,600 in the placebo group), and Pfizer only enrolled 4,526 total (2,750 in the vaccine group and 1,776 in the placebo), with two-thirds dropping out before the third dose. The trials, in other words, enrolled only a fraction of the number of participants that would have been required to determine efficacy against end points like severe disease, hospitalization, and rare adverse events such as myocarditis, which has been linked to COVID vaccination in males in the 12- to 17-year-old age group at a rate of up to 1 in 2,700.
Furthermore, the follow-up time after the second dose of Moderna and the third dose of Pfizer was only 1-3 months. Data from adults show protection against infection is transient, though protection against severe disease so far seems longer lasting. For the Moderna vaccine, efficacy against infection was not statistically significant for children between 6 months and 2 years, according to one of the company’s two analyses. In the Pfizer trial, there was no evidence of efficacy for the first two doses against omicron for this age group; the “effect” seen after the third dose was so uncertain that it is impossible to draw firm conclusions about how well the vaccine worked to prevent cases.
Still more puzzling is the fact that neither Pfizer nor Moderna—despite continued assurances that mRNA vaccines are uniquely flexible, allowing manufacturers to quickly tweak vaccines to match new variants—has released an updated version of their product: The pediatric vaccines now being administered target an outdated variant. In addition, the infant and toddler trials were mostly limited to children who had not been previously infected with COVID (estimates based on blood work showed less than 15% of children enrolled had previously been infected). With 75% of children nationally having already been infected by February 2022, the immune-naive children enrolled in the trial were not representative of their age group at large.
Even in the already troubled context of the last two years, the CDC’s unqualified recommendation to vaccinate every young child against COVID may further contribute to the profound chasm of trust between U.S. citizens and their public health agencies. In January, a Hart poll found that only 44% of respondents said they believe what the CDC says; a March Gallup poll put it at 32%. Evidence of trust slippage can be seen even in highly vaccinated places like Portland, Oregon, where CDC recommendations were for the most part embraced unquestioningly during the pandemic. Despite the CDC’s recommendation that all children 5 and up should receive a booster, as of June 26 only 8.7% of children ages 5-11 in the Portland area are boosted, compared to 3.9% in the entire state of Oregon. (The CDC and American Academy of Pediatrics have not made nationwide data available.)
The general trust deficit is more troubling than skepticism toward this particular vaccine, because it could conceivably drive down uptake of other childhood vaccines that we know are more important to children’s health, such as those against measles, mumps, rubella, diphtheria, polio, and Haemophilus influenza type b (Hib). This is not an alarmist or trivial concern, as vaccinations are one of the most lifesaving medical interventions in human history, rivaled perhaps only by antibiotics. In 1800, 46% of American children did not make it to age 5, and the majority died from what are now vaccine-preventable diseases. The smallpox vaccine alone is estimated to have saved 150 million to 200 million lives. Rates of diseases such as tetanus, rubella, polio, Haemophilus influenza type b (Hib) have declined by 99% since widespread childhood vaccination became commonplace in the 20th century.
It is therefore worth our attention when, for example, a recent letter in the New England Journal of Medicine noted that flu shot uptake has decreased over the pandemic, which the authors suspect may be due to growing vaccine hesitancy in general. The CDC published a study in April showing that childhood vaccination rates fell by only 1% in 2021, a small proportion of the total when spread over 70 million children. But given that many of these vaccines require two or three doses for full coverage, this still translates to several million missing doses, and could threaten herd immunity for diseases such as measles, which require very high percentages of the population to be vaccinated. It is also difficult to separate out the factors behind this drop in coverage, because schools and local clinics—where many low-income children receive vaccines—were closed for much of the last two years. But it is reasonable to at least assume that low trust in the CDC, the agency responsible for making evidence-based recommendations about vaccines, is not helping.
Compare the CDC’s response to vaccine hesitancy during COVID to a similar challenge in the late 1990s and early 2000s: rotavirus. Only a year after Andrew Wakefield’s false claims in 1998 that the MMR vaccines caused autism—leading to one of the most disastrous setbacks for vaccination uptake in history—Wyeth’s RotaShield vaccine was pulled off the market due to evidence it caused a rare and serious intestinal malfunction (intussusception) in babies. The effect of the RotaShield withdrawal so hard on the heels of the Wakefield disaster is hard to isolate, but CDC officials acknowledged that the combined events led to “a particularly turbulent period” for U.S. vaccine programs. Referring to vaccine hesitancy that might result from the RotaShield adverse events, the CDC’s Dr. John Livengood remarked at the time that the CDC “shouldn’t be seen as withholding information right now.”
The original trial for RotaShield had enrolled 10,054 vaccine recipients and 4,633 placebo recipients. During a February 1998 meeting of the CDC’s Advisory Committee on Immunization Practices (the same body that recently met to discuss the pediatric COVID vaccines), an FDA panel member, Dr. Margaret Rennels, noted that more babies in the vaccine group experienced intestinal intussusception than in the placebo group by about 2.5-fold, with a rate of 1/2011 (0.05%) in the vaccine group compared with 1/4633 (0.02%) in the placebo. But because the absolute numbers were small, and the trial was also relatively small, intestinal intussusception did not achieve statistical significance. RotaShield was licensed by the FDA in 1998, widely rolled out, and championed by the CDC in the spring of 1999. Intussusception was not mentioned further, and the issue was buried in a 19-page document where it was listed as a side effect that did not occur significantly more often in vaccinated babies than in the control group.
By summer, however, officials at the CDC grew concerned about a growing number of intussusception reports from the Vaccine Adverse Event Reporting System (VAERS), and were anxious not to lose gains made during the Carter and Clinton administrations in raising general childhood vaccination rates. By the end of President Clinton’s first term, toddler immunization rates had achieved what was then an all-time high, thanks to Vaccines for Children, a program that expanded access to free and low-cost vaccination.
The CDC was also cognizant that Wakefield’s false claims were continuing to spur a growing movement of vaccine hesitancy. As a result, the CDC—then under the direction of Dr. Jeffrey Koplan—immediately launched a large-scale investigation into the RotaShield VAERS reports. The investigation concluded that one additional case of intussusception was attributable to the vaccine for every 5,000-10,000 infants vaccinated—lower than rates of myocarditis due to vaccine injury in COVID-vaccinated adolescent males age 12-17.
RotaShield was pulled off the market that October. To justify the decision to pull a vaccine that was 85% effective at preventing hospitalization from a viral infection that had killed hundreds of thousands of infants worldwide, CDC personnel wrote the following:
At a time when many parents express concerns about the safety of vaccines and vaccine adverse events are the focus of increasing attention by the public, media, and U.S. Congress, the wisdom of recommending a vaccine that causes a severe adverse reaction in an estimated 1 in 10,000 infants must be considered.
The next vaccine against rotavirus—RotaTeq, made by Merck and released in 2004—was only released after the Rotavirus Efficacy and Safety Trial (REST) trial, which was notable for its “[randomized] design, large sample size, detailed execution, continuous safety monitoring, and lengthy duration,” and was undertaken in direct response to the perceived failures of the RotaShield trial. The authors of a paper describing its execution wrote, “The design and conduct of this study may serve as a useful tool for planning other future clinical trials, especially those evaluating uncommon adverse events.” The REST trial was conducted in 11 countries at more than 500 study sites and enrolled 70,000 subjects (including over 35,000 infants from the United States), making it one of the largest vaccine clinical trials ever conducted pre-approval. Post-approval, Merck conducted an additional study enrolling more than 85,000 infants.
The obvious drawback of a trial like REST is that it took four years to complete (though today it could almost certainly be completed faster due to advances in recruitment methods). A multiyear trial was simply not an option during COVID, which is why the notably small and short COVID vaccine trials were allowed to serve as the basis for approval under the emergency use provision. But because COVID so rarely causes severe disease in children, and current COVID vaccines do not reliably prevent transmission, especially after a few months, it is difficult to understand how such small trials could be justified without meaningful endpoints for this age group.
Consider the case of rotavirus again. Prior to vaccination, rotavirus was a significant cause of morbidity and mortality in infants in the United States (and still is globally). Until 15 years ago, it was the leading cause of gastrointestinal hospitalization in babies in the United States and, prior to rotavirus vaccines, caused an estimated 50,000-70,000 hospitalizations per year in infants. Compare this figure with the number of children age 0-4 hospitalized with COVID: The CDC places the cumulative total during the entire pandemic at approximately 130 in 100,000, or about 26,000 children. The CDC estimates that during omicron, at least 14% of COVID hospitalizations for children ages 6 months to 4 years were incidental (meaning the need for hospitalization was due to something other than COVID itself), though this is likely an underestimate, as 63% of current COVID hospitalizations in the U.K. for all ages are “incidental.” Thus, at the time rotavirus vaccines were being trialed, there were 2-4 times more hospitalizations for rotavirus in this age group than there have been for COVID since the pandemic began. (The CDC estimates the death rate from COVID in 6-month- to 4-year-olds to be 86 per year, compared with 20-60 per year from rotavirus, but the COVID estimate does not separate out deaths primarily due to another cause, nor does it adjust for the reduction in severity associated with omicron for children in this age group.)
The rotavirus experience taught the CDC a hard-earned lesson: Speaking in absolutes about vaccine safety and efficacy regardless of trial standards can backfire. In nearly every dimension by which trial data are measured—proper endpoints, size, rigorous randomization, and other factors—the RotaShield trial was far more robust than the Pfizer and Moderna infant and toddler COVID vaccine trials. Furthermore, if the identification of safety signals is not quickly acknowledged, it becomes even harder to recover trust. More and more Americans are wondering, for example, why Canada and several European countries have advised against the Moderna vaccine for people under 30 due to myocarditis risks, while the U.S. government still won’t even acknowledge the higher risk of myocarditis.
Clinical trial data expert and Tablet contributor Dr. Vinay Prasad has pointed out many times that “expedited pathways do not always benefit people, but they always benefit companies.” This might help explain why no other country in the world has started vaccinating infants against COVID, and only a handful have vaccinated toddlers. (In addition to the United States, the only countries vaccinating 2- to 3-year-olds against COVID right now are Cuba, China, Argentina, Bahrain, Venezuela, Colombia, Hong Kong, and Chile, none of which are using mRNA vaccines.) It is perhaps especially damning that no other country collaborated with the United States on the mRNA COVID-19 vaccine trials for infants and toddlers, which could have quickly enabled enough trial participation to study effects of the vaccines against severe disease, as was done in the RotaTeq trial. Tellingly, the Danish minister of health recently claimed that it was a “mistake” to vaccinate children under 16 against COVID at all, saying, “we’ve gotten smarter and would not recommend the same today.”
In June, the CDC had the chance to help rebuild public trust: In the absence of trials and data that would have met the gold standard for scientific rigor, the CDC could have made a softer recommendation based on the data it does have. It could have been honest about the trials’ shortcomings and what these data do and do not show. It could have told the public that the data are preliminary, do not establish efficacy against severe disease or long COVID, and do not rule out the possibility of a rare adverse event. Perhaps it could have recommended COVID vaccines for high-risk children, and remained cautious about the benefits for healthy children who have already had COVID infections. The CDC and FDA together could have insisted that blanket approval and recommendations would only come after a properly conducted vaccination trial—one that would give pediatricians and public health officials the confidence to make the evidence-based recommendations parents are seeking.
In 1999, the CDC, working closely with the FDA, took such steps to shore up parents’ confidence in their recommendations. After the RotaShield withdrawal, the FDA requested that future trials of any rotavirus vaccine enroll at least 60,000 children. This level of accountability and collaboration between the two agencies responsible for vaccines in the United States resulted in the delivery of a widely trusted vaccine against a virus that posed a similar or greater danger to young children than COVID-19. This level of accountability was what the American public reasonably expected of its public health agencies two decades ago. It’s not too much to expect today.
Leslie Bienen is Professor of Public Health of the OHSU-Portland State School of Public Health, a veterinarian, and a mother of two.
Tracy Beth Høeg, M.D., Ph.D. is an epidemiologist currently doing COVID-19 vaccine research with the Florida Department of Health and a physician in private practice in California. She is a Danish American double citizen and mother of four.