How long do kids actually spread Covid?

Small children with Covid-19 exhale fewer viral particles compared to adults, making them less contagious on a moment-to-moment basis. But Covid-19 is, if you hadn’t noticed, a very contagious pathogen. With adequate time and proximity, anyone can potentially catch the virus from anyone else who is shedding virus, kids included.

So, yes, kids are less likely to spread Covid during a brief exposure, but still can spread the virus. As they say, “two things can be true at the same time.” (Nothing makes people’s brains stop working more effectively than pediatric Covid-19.)

A recent JAMA Pediatrics study contains some important information on how long kids might be contagious—which could mean a future in which isolation periods for Covid-19 might be safely shortened.

Kids shed contagious virus, but not for long.

What we already knew: Kids shed SARS-CoV-2, the virus that causes Covid-19.

What we just learned: This new paper suggests that in the first 3-4 days after a positive PCR test, most (if not all) children create enough virus to be contagious (i.e., swabs had enough virus to grow in a lab in culture) in some degree. To their credit, the researchers visited infected children’s homes 5 times over 10 days, allowing us to follow what happened to these kids’s contagiousness over the course of these illnesses. The researchers found that the average kid was contagious for just 3 days after their first positive test. Now, it’s possible that children were also contagious the day before the test (i.e., if they’d been tested, contagious virus would have been detected), but in the absence of screening protocols, we can’t assume these cases will get picked up much earlier most of the time. By day 5-7, around 90% of kids were no longer contagious. (It’s worth noting that 10% still were, with just a few being contagious for as long as 9-10 days.)

What it means: Shorter isolation periods may end up being safe. The problem is that we can’t say for sure which days people should isolate based on symptoms or the date of a test. It’s not necessarily the first 3-4 days after symptoms or a first positive test (though often it is). That’s why at-home tests that tell you how contagious you are—which we have!—could really be useful in the future. I spend a lot of time worrying that people will do nothing to stop the spread of viruses if we ask them to do too much, like isolating for 7-14 days. Better to ask them to do something important like isolating for as little time as possible, without sacrificing safety. As I have written before, using the brightness of the test line on rapid tests (compared to the control line) may be powerful information that we can leverage in the future in this exact way.

Extra hot take: This study had an interesting finding that nobody seems all that excited to talk about. Vaccinated and boosted kids were apparently (statistically) just as contagious as unvaccinated ones. One caveat, of course: recently vaccinated people have a lower chance of infection, so vaccines do temporarily decrease spread simply by decreasing infections for a time.

But if you look at the data closely, it looks like a subset of vaccinated kids were contagious for a bit longer (to the tune of several days) than any single kid in the unvaccinated group was.

None of the unvaccinated kids were contagious 5 days after their first positive Covid test. A small percent of vaccinated kids were contagious until day 9-10. This could be a result of the study design (i.e., there is really no difference between these groups in real life), or it could be a sign of immunity imprinting—a problem that the new vaccines might help with. Image: JAMA Pediatrics.

There are some possible explanations for this finding, including limitations in the study design—meaning that this piece of data might be meaningless. But for the sake of musing, we’ll assume for the moment that this finding is “real.” If so, immunologic imprinting is an important possibility to consider here. Imprinting occurs when the immune system is trained to remember a certain invader (from exposure to a virus or a vaccine that mimics part of it). If the immune system initially sees the original Wuhan version, but later encounters Omicron virus, the rush of antibodies that will be scrambled when the Omicron infection occurs will remain Wuhan-specific. Those antibodies will work well enough to prevent Omicron from causing severe Covid-19 disease (mission accomplished), but perhaps not well enough to quell the infection as fast as Omicron-specific antibodies could. Meanwhile, the unvaccinated kids (especially those who have never been previously infected), faced a higher risk of severe Covid-19 disease going in (i.e., no immunity at all), but despite starting without the immunologic advantage gained from a previous infection or vaccination (i.e., no prior antibodies lying around), they actually seem to have managed to manufacture Omicron-specific antibodies from scratch fast enough to catch up and actually knock the virus out sooner. The good news is that imprinting may be less of a problem with the new Omicron-only vaccines we now have available. In the past, one of my major concerns on “boosters for all” (i.e., boosting even low-risk people) was imprinting—that is, short-term protection at the expense of long-term vulnerability to repeated infections. It’s possible that some segments of the population were in fact “over-exposed” to the Wuhan spike protein from repeated doses of the original vaccine formulation, leading to more infections later, an idea which our data support. And that same phenomenon may be why vaccinated kids in this JAMA Pediatrics paper were contagious for longer compared to unvaccinated ones (again, I say this with caution, because the study design can't tell us that for certain). Hopefully we’ll get similar data for the new vaccine era. In general, I think it’s a good thing that we’ve gotten away from the original version of the vaccine. Omicron vaccines for an Omicron world.

Questions? Comments? Chime in below!