The Editors in
Even with vaccines widely available and better data available about COVID-19, there are still some...
Answers to Anti-School Arguments
In an effort to disregard the overwhelmingly evidence — including Sweden, which has 1.8 million children attending school normally with no social distancing or masks and had zero pediatric deaths and no evidence of pediatric transmission and infection rates among teachers no higher than other occupations — and 22 other countries that have had open schools for months with no adverse consequences for children or community spread, anti-schoolers have advanced four arguments, often stated without explanation as if they are conversation-enders. Each is deeply flawed.
The Korean government report hyped by NYT and other anti-schoolers actually showed 97 percent of households with the virus had it brought in by an adult, and most of the other 3% were *shared initial exposures* between adult and child.
An updated report on the same dataset published in the British Medical Journal on August 7 confirmed this. They updated the earlier – widely publicized – report with a major revision. They found that all but one of the instances previously identified as being infected by a child were in fact infected at the same time as the child in the initial exposure.
“A total of 107 paediatric COVID-19 index cases and their 248 household members were identified. The median age of paediatric COVID-19 index cases was 15 years (IQR 10–17 years); 35.7% and 23.0% were 16–18 years and 13–15 years, respectively. On average, 4.3 household contacts were followed per paediatric COVID-19 index case (range 1–67), and the average number of monitored days was 10.9 days (median 14 days). Nearly half (48.4%, 101/248) of the household contacts were aged 30–49 years, followed by 31.5% aged 0–19 years (78/248) and 18.1% aged 50–69 years (45/248). Of those, 41 were confirmed to have COVID-19: 40 were assessed to have the same exposure as the paediatric COVID-19 index cases. Of those, one pair of paediatric COVID-19 index case-secondary case were identified, giving an SAR of 0.5% (95% CI 0.0% to 2.6%).”
The one child who did infect another child was a 16 year old who infected a 14 year old sibling. Both parents were negative. They found zero confirmed instances of a child infecting an adult in Korea.
This widely cited paper was published in JAMA and purported to show that young children are infectious – not by demonstrating transmission, but by measuring viral load. Of course, even the presence of a high viral load would not, by itself, indicated children are infectious given the overwhelming empirical evidence of limited pediatric transmission. It would only mean that some other factor besides low viral load must explain the phenomenon.
This same viral load approach was previously attempted by German virologist Christian Drosten in a sloppy conclusionary paper that also received wide sensational media coverage – before being dramatically revised because its math actually showed the opposite of its central conclusions. The German medical societies unanimously rejected the approach of lab tests of viral load as a proxy for pediatric infectiousness in the face of empirical data, and German schools proceeded to open without incident.
The Chicago study has several features that make it irrelevant to the schools debate:
Israel’s initial infection wave was mostly limited to its ultraorthodox communities. It has an experienced a larger wave through secular society after reopening, and some anti-school advocates have seized on this as a reason to disregard the 22 countries that opened schools with no increase in community transmission and assert that, after all, schools drive transmission.
The timeline in Israel does not support the theory that schools caused the increase. As the chart shows, mobility levels rose as schools opened but infections stayed low and continued to fall. Only after large events surrounding the Shavuot holiday and the resumption of weddings and other large events did infections accelerate. In many other countries that continued to limit large events while schools were opened, there was – no increase in community transmission.
Additionally, when testing was conducted in the schools the infections rates were lower than the base rate in Israel, that is teachers and students were less likely to be infected than society at large. That is inconsistent with the idea that schools were drivers of transmission, rather than that people infected elsewhere also went to school.
Base: 59k/9M (1/157)
Teachers: 691/170k (1/246)
Students: 1335/1.7M (1/1273)
Among students and teachers in Israel in the May outbreak study that was published in Eurosurveillance, there were zero hospitalizations recorded and zero deaths. There was a total of one emergency room visit.
Out of thousands of camps that ran this summer, there was one identified as having a signficant number of infections. It was in Georgia, a state going through peak transmission, and there is no way to determine whether infections occurred before, during, or after the camp. There were zero hospitalizations the infections were overwhelmingly asymptomatic or mild. This seems like a very strong pro-schools pattern of facts.
The exception is one camp in north Georgia that the CDC reported on in a published paper, which was frantically touted by the media as a harbinger of doom for school reopening. They found that hundreds of kids and counselors tested positive, which resulted in the closure of the camp a few days after it opened.
But have you read anything in the media about a single one being hospitalized or this outbreak setting off another cluster elsewhere? The Atlanta Journal-Constitution, which later identified the camp as “Camp High Harbor” in Rabun County, reported that “state health officials said they were not aware of any hospitalizations or fatalities stemming from the outbreak.” All you will read about are asymptomatic cases, sore throats, and mild fevers. These happen in schools across the country all the time, especially in flu season. But in a country of 330 million, you will find cases like this even in the summer.
Also, based on the information provided in the CDC paper, it’s impossible to know who passed COVID to whom. Thus, without evidence to the contrary, we would have to assume that this follows the pattern of every other study indicating children get the virus from adults. We therefore must presume that the counselors passed COVID to the kids and not the other way around or that outbreak this originated from outside the camp.
The counselors were at the facility for four days of training beginning on June 17 before the campers arrived and likely spread it before the campers came, because the camp was shut down shortly afterwards. Campers didn’t arrive until June 21, the first symptoms were recognized on June 23, and the camp started closing the very next day through June 27. It’s hard to see how the spread occurred among the younger kids and not the older staff.
It’s not as if we found any camp in the country that was opened throughout the past six weeks where there was a sudden outbreak that could be traced from child to child. In fact, because we have not found such a case well into August, it’s quite clear this outbreak of a camp that was opened for just two days in June was baked into the cake before the kids arrived.
As the CDC concedes, “Given the increasing incidence of COVID-19 in Georgia in June and July, some cases might have resulted from transmission occurring before or after camp attendance.”
Similarly, Professor Francois Balloux, director of the UGL Genetics Institute in London, noted:
The fact that we still, months later, have only the one example of a camp with a significant number of infections, makes the most reasonable interpretation of the Georgia camp event is that children were not driving transmission. That is consistent with the overwhelming evidence from all over the world.