It isn’t age. It’s how long you wait to get tested.

Kids are getting COVID, and with the highly-infectious Delta variant still surging in the US, I have a lot of concern about outbreaks once all the children are back in school.

Of course, while some kids get very sick from COVID-19, most do not, thankfully.  But kids tend to live with adults and so a key question is how often infected kids pass COVID-19 to other family members.  And this week, appearing in JAMA Pediatrics, we get the most granular data yet answering that question.

The study comes out of Ontario – where a unified health system allowed researchers to track all known positive COVID cases on a household-by-household basis. The timeframe is important here – data ranged from June to December of 2020 – so we are pre-vaccine and pre-delta variant. The researchers found 6,280 pediatric index cases – that’s where the first case in the household occurred in someone 17 or younger.  They then checked how many people in the household tested positive for COVID within the next 14 days.

Let me start with the topline results – 27% of households with a pediatric index case had at least one other household member become infected. In any house where transmission occurred, on average 2 additional family members were infected.

This is interesting – an attack rate that is a bit lower than might be expected given the lack of vaccines during the study period. But we also see that when anyone in the house got a secondary infection, often more than one person got a secondary infection, which supports the idea that some people are just more infectious than others.

And the researchers were able to quantify that.  For example, kids age 0-3 were more likely to transmit to a household member than those aged 14-17. This could be due to biological factors, but frankly I suspect it’s just much easier for a 14-year old to isolate himself from the rest of the family compared to a 3-year old. And overall, the difference wasn’t huge. The attack rate was about 31% if a toddler was the index case, 27% if it was a teenager.

One risk factor for family transmission really stuck out though – the testing delay. This is the length of time between symptoms and COVID test. There was a strong, dose-response relationship between testing delay and the rate of household infection – the longer people had symptoms before getting tested, the more likely they were to transmit to a family member.

Those who never developed symptoms were unlikely to transmit, by the way – but it still happened about 9% of the time.

We’re not in the same world as when this study was conducted. We can get vaccinated, which should significantly reduce transmission and mitigate the effects when transmission occurs. But we also have the delta variant – which is much more contagious than the coronavirus circulating at the end of 2020. Because of Delta, we are likely to see a much greater rate of transmission among unvaccinated kids as schools open and the household attack rate may be higher.

But rapid testing may be the key here. By identifying sick kids early – as soon as they develop symptoms, protective measures can be taken and secondary infections reduced. I’m picking up a few rapid tests along with my first-day-of-school shopping. Might be worth adding them to your list too.

A version of this commentary first appeared on medscape.com.