But first, a note. Recently, the leadership in Florida suggested that they may actively recommend against vaccination for healthy children. I think this is a mistake. The vaccines have been shown to be very safe, and (more below) they show good evidence of protection against serious illness. While it is true that children are at low risk for serious illness, that is also the case for many other diseases for which we vaccinate. In the end, vaccines for children are going to be the choice of individual parents. But the idea of actively pushing against vaccination is counter to public health.
Some of the data below has been used to argue for this move in Florida. Given that, I feel some trepidation in writing about it. However, ultimately making good decisions requires understanding the data, even if it’s sometimes complicated.
With that…
New York study
Study number 1 comes out of New York, where researchers were able to access all the COVID data from children in the state in the 5-to-11 and 12-to-17 age groups. They combine information on vaccination status, COVID-19 test results, and inpatient hospital data. Together, this allows them to estimate the case rates and hospitalization rate among both age groups, and among vaccinated and unvaccinated individuals.
The authors calculate the relative infection rate and, more useful probably, the vaccine efficacy rate, which measures the share of infections or hospitalizations prevented by vaccination.
The paper has both a glass-half-full and glass-half-empty message.
On the glass-half-empty side, vaccine efficacy against any infection declined significantly over time as Omicron took over. This was especially true in the 5-to-11 age group. The graph below shows the estimates of vaccine efficacy over the December 2021 to January 2022 period.
By the end of this period, it looks like very little protection is being provided for the 5 to 11 age group.
The glass-half-full message is that the point estimates still suggest that vaccines are protective against hospitalization. The headline table points to about 50% protection in the 5-to-11 age group by the end of January, versus 75% in the 12-to-17 group. The data looks better overall if you focus on hospitalizations that are for COVID rather than with COVID (which is in the supplementary materials).
People have noted that there is limited statistical precision in the estimates of impacts on hospitalizations. This is true, and in some ways it represents good news. Hospitalizations are really low in these age groups. In the 365,000 children ages 5 to 11 who were vaccinated, there were just six hospitalizations for COVID between January 17 and 30. Among the unvaccinated group (about 1 million children), there were 28 hospitalizations for COVID in that two-week period. The vaccinated rate is lower, but both are small.
However: the combination of what we see here, plus everything else we know about COVID vaccines in other age groups, suggests that the vaccines remain effective against serious illness. Such an outcome is rare, but it’s also not unheard-of.
CDC data
The second study comes out of the CDC. The method here is different — it’s a case-control, test-negative design. Specifically, the agency uses data from about 40,000 ER or urgent-care visits in 10 states. It evaluates people who came in with possible illness and tested for COVID, comparing vaccination status in the people who tested negative with those who tested positive.
Using this approach, the authors find that protection against illness was similar across the whole age range, from 5 to 17. During the Omicron period, recently vaccinated children ages 5 to 11 had about 50% protection against any COVID infection, versus 45% for recently vaccinated 12-to-15-year olds. This represents a considerable drop for the older group from the Delta period (although the authors find that protection was restored in 16- and 17-year-olds by a booster).
For hospitalization, all groups showed high protection (between 73% and 94%) regardless of age or the timing of vaccination. It is worth noting, once again, that the 5-to-11 age group has small hospitalization numbers, so the estimates are statistically imprecise.
Big-picture message, similar to the study from New York: vaccination protection against any infection waned during Omicron, but the vaccines also remained effective against any infection. The main difference with the New York data is that the 5-to-11-year-old group retained similar protection to the older group, and it was generally higher. This difference could be due to variations in methods, or in geographic reach, or something else.
An important bottom line in both of these papers is that serious illness in this group is extremely rare. This is both reassuring and also makes it somewhat difficult to study. It’s easy to see the tremendous value provided by vaccines for older adults in the data; it is simply harder for this group with lower risks.
What does this mean?
The facts here are simple and, I think, not that surprising. It was hard to miss, during the Omicron wave, that vaccination wasn’t as protective against infection, even if it retained a huge value against serious illness, hospitalization, and death. That this would also be true for children seems reasonable.
The question is, then, how to process these facts as a parent. So let me run through a few scenarios in the space of “How should I think about this?”
My 5-to-11-year-old is vaccinated and also had COVID recently. First, I’m sorry to hear it, and I hope they are feeling well now. Second, your child is now quite well-protected, both due to the recency of infection (I wrote more about that here) and the strength of the vaccine/Omicron infection.
My 5-to-11-year-old is not vaccinated, or is partially vaccinated, and had COVID recently. There is increasingly good evidence suggesting that infection-acquired immunity is very strong overall, and also that infection plus one dose of an mRNA vaccine is even better. This means that if your child hasn’t had any vaccine, they may benefit in terms of protection from a shot. If they have had one shot, the marginal value of another may be small after infection.
My 5-to-11-year-old is vaccinated and hasn’t had COVID. For this group, I think the new information here probably presents the biggest change. The degree to which your child is protected against getting COVID at all is less than we thought. They are, however, still well-protected against serious illness, both by the vaccines and by the fact that children are largely low-risk. This calculus may be different if your child is higher-risk or if you’re worried about higher-risk people they are exposed to.
On the plus side for this group: the case rates have come down enormously in recent weeks, so the risks are much lower than they were a few months ago. And there is an expectation that a third dose of the vaccine will be approved for 5-to-11-year-olds, which could provide some additional protection.
I have a child under 5. Will this affect me? Not directly, of course. I do think this news is in broad strokes related to the reason for the delay in the under-5 vaccines; when that was more fully explained, it seemed that the issue was the more limited efficacy of a two-dose series during Omicron. I still expect, though, if the third dose in this group produces a good antibody response, the FDA will approve it. Moderna is also expected to submit filings for this group, perhaps even this week.
Final thoughts
There has been discussion about the fact that these new findings may discourage parents from vaccinating children, either in the 5-to-11 age group or, down the line, the under-5 group. Already, vaccination rates in this group are low and geographically variable.
This seems likely to me, although I am not sure there is much to do about it, or that it necessarily constitutes a mistake. When initially discussing vaccines for kids, I wrote about a number of the reasons that I vaccinated my own children. They included protecting them from the small risk of serious disease (still an excellent reason to do it), making it possible for them to do things that required vaccines, like go to school (still a good reason), and lowering their infection risk to protect their older relatives and anyone at risk who they interact with.
This last piece has less bite now. The decision moves, as many COVID decisions seem to of late, to be more individualistic. There are still good reasons to vaccinate your children — I would do it again in a heartbeat — but it’s also foolish to think that this doesn’t change the calculus to some extent.
On the flip side, a small change in this individual calculus should not be a reason to change the policy of making vaccines widely available and making clear the benefits to families. Ahem, Florida.
