When we talk about the return to school in the fall, there's a lot of emphasis on making sure that community transmission is 'low'. If we have fewer cases in the community, there's less chance we'll introduce #COVID19 into our schools, where we could see outbreaks. /1

I've been asked 'how low is low enough' and it's a hard question to answer. Zero cases is ideal but probably not practical in most places. /2

@ambilinski @aciaranello @ADPaltiel @MeaganCFitz have come up with a really useful way to think about how the number of cases in the community influences the risk of COVID cases entering our schools. You can explore their approach here: https://art-bd.shinyapps.io/school_entry/  /3

Using the rates of reported cases in your location of interest and the characteristics of an individual school (the number of students and staff), you can estimate the probability that someone infected with COVID in the community would show up at school. /4

For instance, based on the average number of cases in Ottawa in the past week (~22 per day) and their age distribution (30% in those aged <20 years), there's an ~35% chance of one or more case appearing in a school with 400 students and 60 staff. /5

For Toronto, which has reported lower rates recently, we'd estimate an ~9% chance of seeing one or more COVID cases arriving at that same school. /6

These estimates assume that we'll find some of these cases before they come to school, because they have symptoms recognized as COVID and will stay at home if they feel sick. If we don't find these cases, the risk of cases being introduced into the school setting goes up. /7

We can also estimate the average number of cases we'd expect to arrive at a school over the course of a month. For our pretend school in Toronto, it's less than 1 case. For the same school in Ottawa it's about 2 cases. /8

For the Ottawa example, if the number of reported cases increased to 100 per day, the probability of having a case arrive at our school increases to 85% and we'd expect to see 8 or 9 cases each month. /9

Remember that these are estimates of the risk of cases showing up at school, based on what's happening in the broader community. Whether these cases lead to further spread within a school depends on the public health and infection control measures that are in place. /10

To use the whack-a-mole analogy, it's easier to keep COVID under control in schools if there are fewer moles to whack. /11

Back to the initial question of 'how much community transmission is too much to re-open schools safely'? My answer is, 'it depends'...on your confidence in schools to control/prevent spread, and your own risk tolerance. But let's get cases as low as possible before the fall. /end