Month: October 2020

COVID-19 cases in Wallonia schools

In Wallonia (Southern part of Belgium), universities are already back to only giving online classes, schools will be closed two additional days after the Autumn holidays (so November 2-11), and secondary schools (12-18 years-old children) will be virtual for the 3 days before the Autumn holidays (so October 28-30). The reason? The exploding number of COVID-19 cases in schools.

In Wallonia, education is in the hands of the French-speaking Community (along with Brussels) but its statistics department doesn’t seem to provide public data on COVID-19. For that, we have to look at ONE (roughly: “Office for births and infancy“) that communicate weekly numbers of cases and quarantines in children in schools via press releases (forcing us to parse PDFs but it’s better than no data).

So far, the students in secondary school (12-18 years old) are the worst hit with a total of 6,258 positive cases since September 2020 (I’m writing this on October 27), followed by teachers and other personnel (total: 2,497 positive cases).

Is it a lot? Consider this: for the week ending on October 18, incidence in primary school (6-11 years old) is 365 / 100,000, incidence in secondary school (12-17 years old) is 1,117 / 100,000 while the average incidence over the last 14 days in the whole Belgian population is 1,289 / 100,000 (epidemiological bulletin of Oct. 26). Adolescents are therefore a driver of the incidence.

But one can see on the charts below that all age categories are exponentially seeing new cases:

Unfortunately, when you read the press releases, you realize that these numbers are minima. Indeed, the situation is actually worse but there are several reasons why numbers are not completely reported:

  • Health services in schools are not staffed to face a pandemic, they were not prepared and now some personnel also got the virus.
  • As a consequence, data is not completely transmitted to ONE since mid-October (it’s apparently worse for quarantine data, not shown here: at least 21% of cases don’t have data associated with potential follow-up quarantine in the last (7th) report).
  • Since October 1st, protocols (quarantine decreased at 7 days, definitions of close contact, etc.) changed.
  • Children below 6 years are exceptionally tested.
  • Children between 6 and 12 years (primary school) are tested only if they meet some conditions (symptoms, contacts in the family, or if 2 cases in the class).
  • It seems there are issues with reporting in students 18+ (“écoles supérieures“).
  • Universities are not reported in this count.
  • For adults (here: 18+ students, teachers and personnel), Belgium is back at testing only symptomatic patients since October 19, 2020.

So the additional days of holidays and making a few additional days of virtual school for secondary students is meant to try to break transmission of COVID-19 in schools.

Talking about transmission, it seems there is a kind of exploration on sources of infection in the ONE reports. It is not reported systematically nor in a similar way but the source of infection for reported cases is the school (close contact with a student, a teacher or a personnel) in 16-20% of cases.

I really hope this extended holidays will reduce transmission. It seems the younger a child is, the less symptoms he/she’ll display, it therefore seems ok for them to get the disease. But children remain important transmission vectors and we don’t want them to transmit the disease to more vulnerable groups of the population, like grand-parents but also adults and children with co-morbidities or immune diseases. Let’s not add a COVID-19 burden to the usual disease associated with winter (like flu).

To be continued …

As usual, you’ll find other graphs on my page about COVID-19 in Belgium (and figures above are updated with new data as they appear) and the data, code and figures are on Github (including the AVIQ one in this post).

COVID-19 clusters in Belgium

Recently (I’m writing this on October 20), the (new) Belgian government decided to apply more stringent prophylaxis measures to contain COVID-19. One of the controversial measure is to close bars and restaurants for a month.

Unfortunately, in a way, at approximately the same time, AVIQ released its latest poll on COVID-19 clusters in Wallonia (AVIQ is the Walloon agency for well-being, health, handicap and family). I wrote it was unfortunate because I read and heard several people who criticized the closing of bars and restaurants by citing this poll. But this poll cannot answer in favor or against this closure; it doesn’t look at that …

Here are the results:

From the meager press release, here is what we can reconstruct … AVIQ looked at the 5,043 COVID-19 clusters in Wallonia so far and went to interview one or several patients from these clusters (AVIQ defines a cluster as a place where there are 2 or more confirmed COVID-19 cases). The question was, more or less, where did you go before getting COVID-19? (in French: “collectivités que les personnes covid-19 positives ont déclaré avoir fréquentées“).

From there, nearly 84% of clusters were families, far ahead from schools (4%), companies/bars/restaurants (3%), and other places (note schools are still open in Belgium, except universities starting today).

First, bars and restaurants are amalgamated with companies (where home working was encouraged). One cannot easily disentangle them, unfortunately. Then all places are linked and the virus didn’t suddenly appear in the family – but one is more inclined to remember it’s in the family because it is close to dear people (spouse, children, parents, …). Also, there is the potential recollection bias (a classical limitation of interviews), interviewees willing to please the interviewer or simply not willing to disclose behaviors that may be frowned upon. A recent example of this was when the previous Belgian Prime Minister announced she was positive:

This tweet was quickly put in perspective with a plenary meeting of Mrs Wilmès party where the recommended precautions were not all followed:

Well, back to our clusters … My last points for this AVIQ poll is that unfortunately there is no more details than this. We don’t know much about the methodology, it was minimally put in context and there was little caution against wild interpretations (just a “[Ces données] restent toutefois parcellaires compte tenu de ce qu’elles sont déclaratives et tributaires des délais de testing“).

On the other side of Belgium, Zorg en Gezondheid (~AVIQ in the Flemish Region) did a similar poll but gave a bit more details about how they did it and provided more explanations in the results. For instance, they started by asking the index patient where he/she thinks he/she was contaminated: in the chart below, most patients didn’t know (“onbekend” – at least it was an option) but family (“gezin“) and workplace (“werk“) are respectively second and third in the places where they think they most likely got infected (but quite behind “unknown”).

What is interesting is that Zorg en Gezondheid then asked in which social places were these patients before self-isolating. And then we see (below) than most mention bars (“cafés“), restaurants, sports and then only the rather vague “public activities”. It is striking to note that none of these activities are related to school (maybe they only interviewed adults?).

And again, as it was mentioned elsewhere, these are interesting results but it doesn’t show the contagiousness or risk of contamination of these places.

For that, you’ll need serious tracing studies following knows outbreaks. But that’s another story …

To be continued …

As usual, you’ll find other graphs on my page about COVID-19 in Belgium (and figures above are updated with new data as they appear) and the data, code and figures are on Github (including the AVIQ one in this post).