Ugh, what’s the deal with this new study from South Korea about kids transmitting COVID-19?

Families/Kids Infection and Spread Reopening

Does this mean going back to school is more dangerous than we thought?

A: As you’ve noticed by now, headlines are often more sensational than the underlying facts.

This is a large, solidly conducted study but doesn’t move the needle on what we already know: younger children seem less likely to contract and spread SARS-CoV-2, while tweens and teens seem to act more like adults with regards to transmission.

So why the scary headline?

South Korea has been very thorough about contact tracing (props!). In this study, 59,073 contacts of 5,706 COVID-19 index patients were tracked. An “index case” was defined as the first confirmed case within a cluster. Index cases were divided into ten-year age groups (0-9, 10-19, etc), and the number of contacts infected as a fraction of total contacts was calculated.

From these data, kids ages 0-9 made up only 0.5% of the index cases, compared to around 6% of the total population. These children passed on the infection to 5 contacts total, 3/57 household contacts (5.3%) and 2/180 non-household contacts (1.1%).

Both the low number of index cases for ages 0-9 and the low transmission rate is consistent with previous studies that younger children are less likely to become infected and transmit. Yay, GOOD NEWS!

For older kids (10-19), the news was a bit different. They made up only 2.2% of index cases compared to 7% of the population, so still good news there. This was 124 kids out of the sample of 5706 index cases, so not a huge number.

Out of the 231 household contacts of this group, 43 were positive (18.6%). Among the 226 non-household contacts, only 2 were positive (0.9%).

This still doesn’t sound too bad…what’s the big deal? For all age groups overall, the rate of onward transmission to household contacts was 11.8%, contributing to the media interpretation that kids aged 10-19 were *more* likely to transmit than other groups.

Is this the right take home message? Probably not.

First, there were still a relatively small number of onward cases who were infected by the 10-19 age group (43 in total). Because of these small numbers, it is possible that one super-spreading teen infecting a large family could skew the percentage infected. More broadly, the low number of onward infections means the % infected is what we call “imprecise” and likely to be impacted more by random chance. This makes it difficult to conclude that 10-19 years old actually transmit *more* than other age groups.

Also, we should note that kids ages 10-19 transmitted to only 2/226 NON household contacts, HALF the overall rate of 1.9%.

Dr. Alasdair Munro has another interesting take, claiming to have seen unpublished data from the same cohort:

“Almost every single secondary case from a child index case SHARED the initial exposure. Meaning they probably became infected at the same time, the child just developed symptoms first.”

If this is true, the estimated transmission rate for the 10-19 year old’s may be an over-estimate in this study.

Additional caveats to the study:

-It would have been helpful to divide the age group into 10-14 and 15-19 to better understand how risk changes with age.

-The researchers do not know for sure the direction of transmission. The “index” case is the first person who they knew to be infected, but it is possible that this wasn’t the first person infected.

-Schools were largely closed during this time and South Koreans are very vigilant about masks and social distancing. So we can’t say much about how these children would have transmitted with schools open.

-Kids who were identified as index cases were by definition likely to be symptomatic (or there would have been no reason to test them). This means we still don’t know how likely *asymptomatic* children are to pass along the virus, and this may be the MUCH more important question with regards to school re-openings.


Young kids still appear less likely to transmit; older kids act more like adults.

For us this is quite consistent with previous data and doesn’t dramatically change action points for school policy.

See our recent post on factors to consider for school re-openings, including the importance of low levels of community transmission.

These data further bolster the idea that where trade-offs must be made, prioritizing in-person schooling for younger kids may make sense from both a transmission and educational quality standpoint.

The NYTimes summary (*slightly* overstated in our opinion) of the new study.

The original study.

Excellent explainer from Prof. Emily Oster on the South Korea study and recent kids’ data from Texas and Florida.

Link to original FB post.