Sunday, November 8, 2020

Works cited

 I tweeted this out yesterday (starting here, though the thread breaks halfway through), but I know many find paragraphs easier to work with. 


On Friday, after the latest round of the Governor/Secretary/Commissioner's pounding their fists on tables (metaphorically) about getting students back into buildings, the Department of Elementary and Secondary Education issued updated guidance. As the Governor and Commissioner continue to cite what is known from other countries, I was interested to see what they cited as their sources in urging all students, even those in communities where the virus is surging, to go back into buildings for school.

The Department has four footnotes on their recommendations:

Obviously, I am not a doctor, but I am able to read research and review conclusions, and so I did to the pieces here cited, as the argument here is that these are indicating that students even as the virus surges, should be going back into buildings.

The short answer is that none of these pieces say that. A few other general notes:
  • all of the pieces are from August and September, thus they are looking at the spring response to the virus. Any updates on what we know about it since then are thus not included.
  • all of these are from countries that responded very, very differently to how our own country, and yes, even our own state, have responded to the virus. This is particularly important, as you'll see.
  • there are repeated cautions against drawing too broad a conclusion from these pieces
  • and perhaps most importantly, none of these studies draw the conclusion that the Governor and Commissioner have claimed
On to each of the studies:

The Public Health England "SARS-CoV-2 infection and transmission in educational settings:
cross-sectional analysis of clusters and outbreaks in England" study looks particularly at what happened in the UK in the first month after the government lifted the spring lockdown and sent students back into schools. For those who haven't followed that closely, note: attendance in schools was not required and thus, as is noted on page 12, "[o]nly 1.6 million of the 8.9 million students nationally attended any educational setting during the summer mini-term...Additionally, very few secondary schools opened (and those that did, did so with small class sizes) during the summer mini-term." So, no the U.K. did not open all schools or anything of the sort. 
What did they find? The interpretation says "SARS-CoV-2 infections and outbreaks were uncommon in educational settings during the first month after the easing of national lockdown in England." Uncommon, not unknown. The very next sentence says, "The strong correlation with regional SARS-CoV-2 incidence emphasises the importance of controlling community transmission to protect educational settings." On page 7, where they describe process, likewise, they write, "There was a strong correlation (R2 = 0.82, p=0.001) between outbreaks in educational settings and regional COVID-19 incidence" thus again tying what is happening outside the schools to what happens inside schools. Schools are not set apart: higher rates one place correlates with higher rates within the same community.
After they run through much of what we've been able to find out--low known incidence among younger children, higher apparent rates in older children--they note the limitations of what they can conclude, first, by noting limited number of students in school, then:
Settings that opened had stringent social distancing and infection control measures in please and, in addition to school attendance not being mandatory, there were strict protocols for class and bubble sizes, 12 which may not be achievable when schools opening fully in the next academic year (and indeed, updated schools guidance now recognises that bubble size may need to be increased from September to ensure that a full range of activities is feasible).

Massachusetts has of course not set limits on numbers of students together, nor of bubble sizes, and the comments made Friday are frankly in opposition to that.
Because so few secondary schools opened, they write, their conclusions "therefore, are not likely to be generalisable to secondary schools, especially since the risk of infection, disease and transmission is likely to be higher in older than younger children."
And they conclude:

The strong correlation between COVID-19 outbreaks and regional SARS-CoV-2 incidence highlights the importance of controlling the disease in the community to protect the staff and students in educational settings.
The Governor can change the map colors any way he'd like; we still aren't controlling the spread of the disease in our communities.

The study "Stepwise School Opening Online and Off-line and an Impact on the Epidemiology of COVID-19 in the Pediatric Population"looking at data from Korea Disease Control and Prevention examines specifically what happened over May and June ""Online classes started from April 9, and off-line classes started from May 20 to June 8 at four steps in different grades of students." During that time, "[t]here was no sudden increase in pediatric cases after the school opening, and the proportion of pediatric cases remained around 7.0% to 7.1%." The study says:

As of July 11, 45 children from 40 schools and kindergartens were diagnosed with COVID-19 after off-line classes started. More than 11,000 students and staff were tested; only one additional student was found to be infected in the same classroom. Among those 45, 32 (71.1%) patients had available information for the source of infection. Twenty-five (25/45, 55.6%) were infected by the family members.

So that's low, and more than half were infected by family members; there's a really low finding here of cases within the same classroom.
Again, that's great! If kids don't get each other sick, that is good news. 
That does, though, leave out a LOT of people in schools--this is, after all a pediatric study, which does not look at staff members. The study itself also says this:

The proportions of pediatric patients without information on infection sources were higher in older age group (middle and high school students) than in younger age group (kindergarten and elementary school students) (47.6% vs 12.5%, p=0.010). In the younger age group, 79.1% of children were infected by family members, while only 28.6% of adolescents in the older age group were infected by family members (p<0.001).

...which again is to note that elementary and secondary are very different things here. In this study, they could track the elementary cases mostly back to their families; that isn't the case with the secondary school students.
This is part of why several districts have asked that the infection rate data breakout that 0-19 age group. As we entered October, that group was particularly of concern

The third study that DESE cites in their new guidance is this one "Surveillance of COVID-19 school outbreaks, Germany, March to August 2020" via Eurosurveillance. I don't know about you, but my first response to Germany is "they put a scientist in charge of the country." Germany has, of course, dealt aggressively with controlling the virus. That matters: the discussion section of the report notes: "When schools reopened, the incidence of COVID-19 in the general population was low and there was no community transmission."
No community transmission and low general incidence. Did they have cases in schools?
Yes: "Despite the low-incidence period and enhanced hygiene measures implemented in schools, school outbreaks occurred."
Not just single cases: outbreaks of infection.
Interestingly: 
The average number of outbreaks and of cases per outbreak was smaller after schools reopened than before school closure, suggesting that containment measures implemented in schools may have some protective effect.

They go on to discuss how discovering what is going on with children and coronavirus is difficulty because children often are asymptomatic, and thus less often tested. 
Despite what the Governor and the Commission said, though, right here in this report they cited, it says:

 There is some indication that transmission occurred within a school.

Of the school they're discussing, they write:

As the number of student cases of the same grade was 25 in outbreak number 5, it is unlikely that no transmission occurred between students. Moreover, in some outbreaks, more than one grade was affected
It does appear transmission was low, however.
They then discuss the opening of schools in places in the EU and in Israel, which had very different experiences, in part because the procedures under which they opened were very different. This of course matters a great deal, despite the argument from the Commissioner on Friday that "fully back" was a real option for any but a very few. 

The thing I always try to be careful about in citing research is the "however" part. Here's the "however" piece from this German study:
  • "There are some limitations to our analysis. Outbreaks, particularly in primary schools, may have been difficult to detect because the children may have been asymptomatic."
    This is the question over if we really even know if/when children have it, because it is frequently asymptomatic in them; this makes it that much harder to discover if students in schools do have it, are spreading it, and to whom.

  • "Household outbreaks epidemiologically linked to schools are not always reported as linked outbreaks or as outbreaks at all."
    In other words, maybe the child got it at school and someone at home got sick, but it was not reported as traced back to that school.

  • "we did not know in which class a student had been and can therefore not exclude that cases of similar age may had been in parallel classes"
    ...so sometimes, we don't know

  • "as the period of reopening schools coincided with relaxing measures in other settings, it is difficult to assess the impact of school reopening on transmission dynamics within a school."
    This is messy data because there were other things going on at the same time.

And what is their conclusion in this study?
While schools remain open, well-designed evaluations of the preventive measures are needed to assess effectiveness in terms of reducing SARS-CoV-2 transmission and to guide future decision-making during the COVID-19 pandemic. Moreover, school openings should be accompanied by developing surveillance capability and the ability to rapidly test, trace and isolate suspected COVID-19 cases and their contacts.

We don't have that yet. It is only then that they conclude: "To avoid detrimental effects on children, school closures should be applied only cautiously and in combination with other control measures." We haven't done the first piece, as yet, though, so we aren't here yet.

The fourth study cited by DESE in their Friday guidance is the European Centre for Disease Control and Prevention "COVID-19 in children and the role of school settings in COVID-19 transmission"report from August 6.
For this one, I'm going to give you the entire executive summary:

  • "A small proportion (<5%) of overall COVID-19 cases reported in the EU/EEA and the UK are among children (those aged 18 years and under). When diagnosed with COVID-19, children are much less likely to be hospitalised or have fatal outcomes than adults." So, children have been diagnosed less often, and when we know they have it, they get less sick.
  • "Children are more likely to have a mild or asymptomatic infection, meaning that the infection may go undetected or undiagnosed." Thus BUT we do not always know that they have it. 
  • "When symptomatic, children shed virus in similar quantities to adults and can infect others in a similar way to adults. It is unknown how infectious asymptomatic children are." When they are sick, they make other people sick just as adults do, and we don't know how much that happens if they have it but don't have symptoms.
  • "While very few significant outbreaks of COVID-19 in schools have been documented, they do occur, and may be difficult to detect due to the relative lack of symptoms in children." Yes, schools are places where people can get sick, and sometimes we may not know, because children can so often be asymptomatic.
  • "In general, the majority of countries report slightly lower seroprevalence in children than in adult groups, however these differences are small and uncertain. More specialised studies need to be performed with the focus on children to better understand infection and antibody dynamics." I had to look this up. This means kids are less often testing positive, but the differences are small and we don't know enough about this yet and need more data. 
  • "Investigations of cases identified in school settings suggest that child to child transmission in schools is uncommon and not the primary cause of SARS-CoV-2 infection in children whose onset of infection coincides with the period during which they are attending school, particularly in preschools and primary schools." Kids seem not to often get each other sick, and for young children in particular, it seems they are not getting sick from school.
  • "If appropriate physical distancing and hygiene measures are applied, schools are unlikely to be more effective propagating environments than other occupational or leisure settings with similar densities of people." If schools do everything right, they're no more likely than anywhere else to have people get sick from each other.
  • "There is conflicting published evidence on the impact of school closure/re-opening on community transmission levels, although the evidence from contact tracing in schools, and observational data from a number of EU countries suggest that re-opening schools has not been associated with significant increases in community transmission." As of then, it appears that schools did not contribute to community increases in transmission. There has since been evidence otherwise
  • "Available evidence also indicates that closures of childcare and educational institutions are unlikely to be an effective single control measure for community transmission of COVID-19 and such closures would be unlikely to provide significant additional protection of children’s health, since most develop a very mild form of COVID-19, if any." Closing schools alone isn't enough to control the virus, and it seems unlikely to protect children from COVID.
  • "Decisions on control measures in schools and school closures/openings should be consistent with decisions on other physical distancing and public health response measures within the community." Schools are part of a larger group of decisions which should be made consistently. 
This then is the evidence that Governor Baker, Secretary Peyser, and Commissioner Riley are putting forward to argue that schools in our state--a state which by all accounts is seeing another surge, for which the most stringent recent methods the Governor has proposed to control the virus is a 10pm curfew and masks outdoors, which continues to not have enough testing or contact tracing among much else--should be open. 

That isn't what the research that they themselves are citing actually says, though. 

No comments: