Economist Emily Oster and the Massachusetts retrospective state-wide cohort study
The centerpiece study that the CDC used as evidence for changing the guidelines for social distancing at schools was a retrospective state-wide cohort study conducted in Massachusetts and published on March 10, 2021, a week before Dr. Walensky’s subcommittee testimony. The title of the paper is “Effectiveness of three versus six feet of physical distancing for controlling the spread of COVID-19 among primary and secondary students and staff.”
The name of Emily Oster, Ph.D., listed as one of the authors of the accepted manuscript to the Clinical Infectious Diseases journal, should raise many red flags in readers’ minds. The journal cites her for her COVID-19 School Response Dashboard. The Harvard-graduated economist is currently a professor at Brown University, also author of books on pregnancy and parenting. She has been a staunch advocate for school reopening from day one and has utilized her influence with Ivy League academia to weigh in on the health science for which she, as an economist, is unqualified. The mainstream media has provided her a wide venue for her opinions.
In her blog from May 4, 2020, she wrote: “Full disclosure, I spend about 73 percent of my time obsessing about whether there will be summer camp in Rhode Island. I send borderline appropriate emails to people I work with on virus response, nagging them about this.”
Rachel Cohen, a journalist and former American Prospect writing fellow, who wrote a scathing rebuttal on Oster’s engagement on the issue of schools and the COVID-19 pandemic, states upfront that Oster does not have a background in public health and has “downplayed negative research findings that complicate the picture, and amplified studies that experts say were weak.”
Early in her foray on this issue, Oster had concluded that “infections among kids is simply very unlikely.” She added, “It’s not that they are infected and don’t know it, it seems like they are just not infected very often.” She had utterly ignored the cautionary statements raised by infectious disease specialists and epidemiologists early in the pandemic when lockdowns had been instituted and children, who were rarely tested, were stuck at home.
In late August, Oster announced she would create, with the assistance of school superintendents and principals associations, a systematic data collection and reporting on COVID-19 in schools. After two weeks of data collection, she announced that only 0.23 percent of students and 0.49 percent of teachers had become infected, making her case that schools are not “super spreader events.”
Yet, as Cohen noted, Oster’s database of 550 public and private schools, where more than 200 of these were fully remote, accounting for just over one percent of all student populations was noted to be an “extremely small and unrepresentative sample of schools.” After a deeper dive into Oster’s school database, Cohen remarked, “There was not a single urban traditional public school reporting data across 27 states in her dataset, including from Florida, Texas, New York, Minnesota, Arkansas, and Mississippi.” Oster had also remarked to Cohen that those voluntarily reporting tended to be schools from higher income and suburban districts.
In August, Oster also reached out to Rebekah Jones, the geographic information systems data scientist who had been fired from Florida’s Department of Health over her refusal to manipulate the state’s COVID-19 data. Jones had initiated her own national school coronavirus tracker possessing data from almost 4,000 school districts. She agreed to allow Oster “full and free access.”
According to Rachel Cohen’s report, Jones said of the meeting, “But she [Oster] basically decided to just pick what data she wanted, not what’s available. It’s offensive to researchers, when you see something so unabashedly unscientific, and when the opportunity to do something scientific was there. You can’t just have one point in a state and claim you have a grasp on what is occurring for a wide spectrum of school types and incomes, but that is literally what she does.”
It is with these introductory remarks that we can begin a review of the Massachusetts study.
The 16-week study commenced on September 24, when public schools had just opened, and lasted until January 27, encompassing a time when the winter surge began a rapid acceleration in cases of COVID-19, reaching its peak around the new year. In total, data was obtained retrospectively from 251 school districts across the state. Sixty-four percent of school districts opened with “low on-campus enrollment,” meaning less than 80 percent and more than five percent of the student body was in attendance—a wide range for which the exact details were left unspecified.
Forty-eight school districts had a policy that allowed for three-foot social distancing, while 194 school districts had stated they followed a six-foot distancing criterion. Student and staff case counts were based on a laboratory-confirmed diagnosis of SARS-CoV-2. No systematic or weekly testing of asymptomatic, pre-symptomatic or mild disease was conducted. The study concluded that student and staff case rates for the two distances were similar.
It needs mentioning that the title of the paper completely misrepresents what was actually studied and measured. The authors state in the body of their manuscript that they were evaluating “the impact of three feet versus six feet of physical distancing policies in school settings.” In fact, they could not ascertain if schools abided by or enforced such policies. As the Salt Lake County, Utah, study revealed, despite mandating a six-foot physical distancing, desks were on average three feet apart. Yet, the casual reader would assume actual distances were being evaluated.
Knowing that they would be held accountable for this sleight of hand, the authors wrote, “However, it is unlikely that cases were differentially missed in districts with three versus six feet, mitigating the impact of this limitation on our main study finding. … We also did not have detailed contact tracing data available, and so we were not able to determine if cases in students were due to transmissions that happened within the school environment or independent introductions from cases acquired in the community.”
They go on to add further non-sequiturs. “It is possible that districts that officially allowed three feet or more of distancing between students ultimately succeed in attaining more distance between students, and our methods were only able to capture official policy, not the real-world implementation of the policy.” One could turn this sentence around by restating, “It is possible that districts that officially allowed six feet or more of distance between students ultimately failed,” and still be completely valid. In support of the alternative conclusion, one must only turn to the Salt Lake County, Utah study mentioned above.
That the paper passed the reviewers with such assertions—which had significant implications for national school policy—is incredible and reeks of political considerations that run rampant through the paper. The study is being used to justify the reopening of schools at full capacity in-person instruction, meaning the bioburden of disease spreading throughout schools will be much higher. Glaringly, as it has come to be understood, the spread of disease indoors is through aerosol transmission, meaning that neither three nor six feet can be considered safe.
In this regard, probably the most crucial finding that Dr. Walensky failed to disclose in her congressional testimony was that during the study period, the incidence of cases among students and staff was “highly correlated with community rates.” In other words, cases in schools were climbing along with those in the community, regardless of ascribing to a three-foot or six-foot policy. As the table borrowed from the Massachusetts study shows, rates rose by eight- to 18-fold, essentially negating their safety conclusions. Both three feet and six feet were unsafe!
Yet, once more, they attempt to dismiss these essential findings by asserting an unproven counterfactual. They wrote: “The finding of the strong correlation between community incidence and incidence in schools does, not, however, imply that there is increased transmission in schools when community disease prevalence is high, nor community metrics should dictate school opening/closing policies.” By their admission, cited above, no surveillance or contact tracing was done to appreciate the relationship between school and community incidences.
Conclusion: The SARS-CoV-2 virus, schools, children and community transmission
The cited studies lack sufficient evidentiary weight to make any meaningful conclusions about the claim schools are safe for reopening. The authors go on to dig their own graves when they write: “During the study period, active surveillance programs were rare, and thus we were not able to identify asymptomatic cases that may have resulted from in school transmission, or to measure the effectiveness of this intervention as a tool for controlling SARS-CoV-2 spread in school settings. Additionally, we were not able to measure the impact of physical distancing stratified by school type or age group.”
On the contrary, they have shown that social distancing policies are challenging to enforce, and the inevitable slippage leads to a rise in cases both within schools and in the community with which they are inextricably connected. It also highlights the critical need for a prospective national surveillance initiative to determine the relationship between schools and communities during the pandemic. There has also grown a compelling need to revamp schools’ material conditions to assure air ventilation capacity at schools is optimized to provide a safe environment for students and staff.
On March 5, 2021, around the time the CDC was promoting the reduction of physical distancing requirements in schools, the agency published a small and barely mentioned study looking at the presence of antibodies (seroprevalence) among children and adolescents in Mississippi from May to September of 2020.
The number of COVID-19 cases reported in this age group had reached 8,993 by August 31, 2020. PCR testing was carried out to detect and confirm infection.
The study found, based on blood tests drawn from a random sampling of the child and adolescent population, seroprevalence had risen from 2.5 percent to 16.3 percent during the study period. These seroprevalence estimates placed the number of individuals infected in this age group at 113,842, with a confidence interval of 90,096 to 153,652. This means the actual rate was 11 to 18 times higher than what symptomatic testing had found.
A study conducted in Virginia, published to the medRxiv.org preprint server for health sciences on January 31, 2021, looked at the number of children infected with COVID-19. Over 10 weeks during the first half of the pandemic, researchers found that children had antibodies to the coronavirus at twice the adults’ rate. More than 65 percent never showed any symptoms. Dr. Rebecca Levorson, Division Director of Pediatric Infectious Diseases at Inova Children’s Hospital, said, “I think children have been a silent bearer of infection. They have more mild symptoms so they may not go and get tested because they don’t have a high-grade fever. So, we just really didn’t know that children had COVID to such an extent.”
Yet, the implications of the CDC’s Mississippi study and others were never raised by the director or congressional representatives during the March 17 hearing, when there was ample time to let all of the science weigh in on the issue of school reopening.
In a recent Independent Scientific Advocacy Group (ISAG) webinar discussing a “better way forward toward the elimination of COVID-19,” Dr. Deepti Gurdasani, a clinical epidemiologist and statistical geneticist at the Queen Mary University of London, highlighted data from the Office of National Statistics that conducted regional surveillance every two weeks throughout the pandemic in England.
The data clearly shows that the primary, secondary and college trends dominate in the decline and rise with school closures and openings. The peaks in December are due to the spread of the B.1.1.7 variant, which necessitated the strictest lockdown across the UK. Primary schools were allowed attendance with a subsequent rise in this age group while all others are declining.
In the Canadian province of Ontario, the city of Toronto is facing a brutal third wave of the COVID-19 pandemic with ICUs rapidly filling. On Thursday, Ontario Premier Doug Ford announced that he was placing the entire province under restrictions for four weeks. During a press brief, he said, “I know pulling the emergency brake will be difficult on many people across the province, but we must try and prevent more people from getting infected and overwhelming hospitals. Our vaccine rollout is steadily increasing, and I encourage everyone who is eligible to get vaccinated. This is our best protection against this deadly virus.”
In a recent Tweet, Dr. Eric Feigl-Ding, an epidemiologist and health economist, said, “School closures have been extensively studied. We know they slow or stop pandemics.” He then depicted an epidemiologic curve separating cases at schools and in the community, demonstrating indelibly that cases in school precede any subsequent rise in the community.
Compared to those being promoted by the CDC, these findings have important implications for what a genuinely scientific policy against the SARS-CoV-2 pandemic should entail. The three-foot physical distancing guidelines will ensure more students are packed into the classrooms, further fueling the pandemic underway in Michigan and the Northeast of the US.
Since evidence for asymptomatic transmission has become better understood, using symptom-based studies to endorse changing guidelines can no longer be considered valid, making the Massachusetts study flawed and its findings and the CDC guidelines circumspect. Teachers, parents, and the working class, be warned and take action accordingly.