A Look at DESE's Cited Sources: Results of a Critical Look at the Initial Reopening Guidance References

Note: I've been doing a lot more work around education safety lately, but moved it away from this blog. I founded Massachusetts Education Safety Advocacy Group (MESAG) and you can find up-to-date research on the website, Twitter, and Facebook. Please follow, like, and share!

I read and analyzed all of the sources cited in the DESE document released 6/25. You can see my write-ups of all of the sources here.

Here's what I found:

The report has 84 citations of 53 unique sources:

• 8 of the 53 sources are public-facing CDC web pages
• 2 of the 53 sources are state school reopening plans (California and Oregon)
• 4 of the 53 sources are from government sources in non-USA countries. The reports are from Australia, Denmark, France, and Hong Kong.
• 2 of the 53 sources are WHO (World Health Organization) web pages.
• 1 of the 53 sources is a public-facing guidance document from a Toronto Children's Hospital
• 3 of the 53 sources are Massachusetts public-facing state web pages.
• 4 of the 53 sources are news articles
• 18 of the 53 sources are peer-reviewed scientific journal articles.
• 5 of the 53 sources are pre-print (not yet peer reviewed) scientific journal manuscripts
• 6 of the 53 sources were scientific journal articles/manuscripts for which peer review status was unclear

Many of the studies cited have an extremely small sample size.

• This study looked at 13 family clusters in Israel.
• This study looked at 39 households in Switzerland. 
• This study looked at 11 people, only one child, in the French Alps.
• This study looked at 18 COVID-19 cases in schools in New South Wales, Australia.
• This study focused on 6 COVID-19 cases in schools in Ireland.

Several locations where studies took place or for which guidelines were written have much lower COVID-19 infection rates than Massachusetts.

• Oregon's infection rate per 100,000 people is over 9 times lower than that of Massachusetts.
• California's infection rate per 100,000 people is less than half of that of Massachusetts.
• France's infection rate per 100,000 is over 6 times lower than Massachusetts.
• New South Wales, Australia's infection rate per 100,000 is over 37 times lower than that of Massachusetts.

Additionally, some of the sources did not prove the claim that DESE attempted to make with them. Of 81 claims made: 

• 7 are not supported by the source cited
• 59 claims are supported by the source cited
• 4 claims have partial support
• 1 claim is somewhat supported
• 4 claims have limited support 
• 4 claims have extremely limited support

Here are some of the unsupported claims (all quotes from DESE guidance):

• "In NYC, in households with at least one COVID-19 case, prevalence of infection for children 5-≤18 was 31.9% vs. overall prevalence 52.5%."
• "Whereas for COVID-19 it appears children are less likely to be infected with and to transmit COVID-19, this is not the case for influenza, where children are frequent transmitters."
• "As noted in previous guidance, temperature checks are not recommended as screening for all students due to the high likelihood of potential false positive and false negative results"
• "Finally, this guidance is for fall reopening and is predicated on the Commonwealth continuing to progress through the phases of reopening with low COVID-19 public health metrics"
• "Mask breaks should occur throughout the day."
• Schools do not appear to have played a major role in COVID-19 transmission. In a review of COVID clusters, only 4% (8 of 210) involved school transmission."
•  "If infected, it appears children may be less likely to infect others with COVID-19."

It appears that DESE cherry-picked data that supported a return to school, rather than looking critically at all available evidence to make a science-based decision about school safety. Inconsistencies in the report, such as inconsistent citation formatting and claims which do not align with the geographical facts of a study, raise concern as to the accuracy of the report. Additionally, the fact that none of the medical professionals involved are named raises questions as to why they do not want their work recognized.

The DESE report also leaves out important research on adult COVID-19 infection risk and transmission data. For schools to open for in-person schooling, numerous adults will need to be in school buildings. There was a recent outbreak in a school in North Carolina amongst the staff. To ignore the data on adult transmission when adults must be present for schools to re-open, is irresponsible. I will write a separate post with some studies about COVID-19 in adults, and school-related COVID-19 cases.

I urge you to contact state level education officials to urge them to reconsider this guidance in light of the dangers to school staff safety. 

• Massachusetts Board of Education: boe@doe.mass.edu
• Massachusetts Secretary of Education James Peyser: eoe@massmail.state.ma.us
• Massachusetts Return to School Working Group: RTSWG@mass.gov
• Massachusetts Governor Charlie Baker contact form or constituent.services@state.ma.us

A Look at DESE's Cited Sources: Footnote #83 - How can airborne transmission of COVID-19 indoors be minimised?

DESE Claim: "Consider ways to increase facility ventilation (e.g., open windows through fall, perform an HVAC inspection)"

Cited Source: Morawska, Lidia, et al. "How can airborne transmission of COVID-19 indoors be minimised?." Environment International (2020): 105832.

Note: no link provided in DESE report citation, I found the article here using Google.

Peer Reviewed?: yes

Study Methodology: unclear

Notes: There may be risk of airborne transmission. "Ventilation airborne protection measures which already exist can be easily enhanced at a relatively low cost to reduce the number of infections and consequently to save lives." This paper gives recommendations of different ways to increase ventilation.

Supports DESE Claim?: Yes

A Look at DESE's Cited Sources: Footnote #82 - Stronger Together A Guidebook for the Safe Reopening of California's Public Schools

DESE Claim: "Where physical distancing is difficult to implement (e.g., office space, reception desks), consider installing barriers or changing the configuration to support student/staff health and safety."

Cited Source: California Department of Education, Stronger Together

A Guidebook for the Safe Reopening of California's Public Schools. (2020, June 8). Available at https://www.cde.ca.gov/ls/he/hn/documents/strongertogether.pdf.

Peer Reviewed?: No - government guidance

Study Methodology: N/A - government document

Notes: This is the California plan to reopen schools. "Install physical barriers, such as sneeze guards and partitions, at point of sale and other areas where maintaining physical distance of 6 feet is difficult." There are provisions for illness, busing, and most other aspects of reopening. "If physical distancing between workspaces or between employees and students and visitors is not possible, add physical barriers that cannot be moved to separate workspaces." California's plan seems to be based primarily on a hybrid model to support physical distancing.

Supports DESE Claim?: Yes

A Look at DESE's Cited Sources: Footnote #80 - How To Clean and Disinfect Schools To Help Slow the Spread of Flu

DESE Claim: "Please refer to the federal guidance related to cleaning of facilities for more information regarding appropriate cleaning supplies, protocols, and frequency (e.g., wear appropriate protection such as gloves, wash hands often, follow instructions on all cleaning products, handle waste properly)."

Cited Source: CDC, How To Clean and Disinfect Schools To Help Slow the Spread of Flu. (2018, July 31). Available at https://www.cdc.gov/flu/school/cleaning.htm

Peer Reviewed?: no - public health guidance on government website.

Study Methodology: N/A - public health guidance

Notes: This page is flu specific, and does not suggest cleaning for COVID-19. 

Supports DESE Claim?: N/A - this is the federal guidance mentioned that schools should refer to.

A Look at DESE's Cited Sources: Footnote #79 - Cleaning and Disinfecting Your Facility

DESE Claim: "Please refer to the federal guidance related to cleaning of facilities for more information regarding appropriate cleaning supplies, protocols, and frequency (e.g., wear appropriate protection such as gloves, wash hands often, follow instructions on all cleaning products, handle waste properly)."

Cited Source: CDC, Cleaning and Disinfecting Your Facility. (2020, April 14). Available at https://www.cdc.gov/coronavirus/2019-ncov/community/disinfecting-building-facility.html

Peer Reviewed?: No - public health guidance

Study Methodology: N/A - public health informational website

Notes: This page contains recommendations on cleaning and disinfecting. It suggests wearing gloves to disinfect, using soap and water before disinfectant, and other measures. There are specific measures for cleaning the facility if someone is sick.

Supports DESE Claim?: N/A - footnote is a reference for where to find federal guidance to be used for cleaning

A Look at DESE's Cited Sources: Footnote #78 - How Denmark got its children back to school so soon after lockdown

DESE Claim: "Prepare for frequent cleaning and sanitization of facilities and surfaces, especially high-touch surfaces (e.g., doorknobs, hand rails)."

Cited Source: How Denmark got its children back to school so soon after lockdown (2020, May 28). Available at https://www.thelocal.com/20200528/how-denmark-got-its-children-back-to-school

Peer Reviewed?: no - news article

Study Methodology: N/A news article

Notes: This article explains the steps being taken as schools reopen in Denmark. "At kindergarten, my four year-old can’t hug or hold hands with her friends but adults can comfort them with cuddles whenever needed." Masks are not common in Denmark, according to the article. "New hygiene guidelines stated that children should wash their hands at least every two hours. Surfaces also needed to be cleaned twice a day." 

Supports DESE Claim?: Yes

A Look at DESE's Cited Sources: Footnote #75 - COVID-19 in Children and the Dynamics of Infection in Families

DESE Claim: "In Switzerland, a study of household clusters found that only 8% had a child as the index case. In nearly 80% of the cases, the child got COVID-19 from an adult family member"

Cited Source: Posfay Barbe, C., Wagner, N., Gauthey, M., Moussaoui, D., Loevy, N., Diana, A., & L'Huillier, A. (2020). COVID-19 in Children and the Dynamics of Infection in Families. Pediatrics, e20201576. Available at https://pediatrics.aappublications.org/content/early/2020/05/22/peds.2020-1576

Peer Reviewed?: yes

Study Methodology: "chart reviews were used to retrieve clinical data and parents were called for patients and household contacts (HHC) follow-up"

Notes: This is a study of COVID-19 cases in Geneva, Switzerland. "Adult HHC [household contacts] were suspected or confirmed with COVID-19 prior to the study child in 79% (31/39) of cases. In only 8% (3/39) of households did the study child develop symptoms prior to any other HHC". There were "confinement measures" in place at the time, though it is unclear if school closures were a part of those measures. 

Supports DESE Claim?: Yes

A Look at DESE's Cited Source: Footnote #72 - The Role of Children in the Dynamics of Intra Family Coronavirus 2019 Spread in Densely Populated Area

DESE Claim: "A study in Israel found that children 5-17 were 61% less likely to have positive COVID-19 tests compared with adults in the same household."

Cited Source: Somekh, E., Gleyzer, A., Heller, E., Lopian, M., Kashani-Ligumski, L., Czeiger, S., ... & Stein, M. (2020). The Role of Children in the Dynamics of Intra Family Coronavirus 2019 Spread in Densely Populated Area. The Pediatric Infectious Disease Journal. Available at https://journals.lww.com/pidj/Abstract/9000/The_Role_of_Children_in_the_Dynamics_of_Intra.96128.aspx

Peer Reviewed?: unclear (DESE linked source is not published article)

Study Methodology: "Clusters of infections within families living in Bnei Brak were identified and investigated."

Notes: There were 13 family clusters investigated. "Children 5–17 years of age were about 61% and children 0–4 years were 47% less likely to have positive PCR results compared with adults older than 18 years of age residing in the same household." Notably, Bnei Brak has a very high population of young people. "Our results do not necessarily indicate that reopening school is safe regarding the potential of re-emergence of Covid-19 spread since various rates of transmission are expected to occur also in school (as occurred in a family set up) and therefore reopening school should be practiced with caution and with close monitoring of Covid-19 spread." (emphasis is my addition). 

Supports DESE Claim?: Yes (but the fact that DESE doesn't account for the author's statement that results don't mean school is safe is alarming) 

A Look at DESE's Cited Sources: Footnote #70 - New York State Coronavirus 2019 Response Team, COVID-19 Testing, Epidemic Features, Hospital Outcomes, and Household Prevalence, New York State—March 2020

DESE Claim: "In NYC, in households with at least one COVID-19 case, prevalence of infection for children 5-≤18 was 31.9% vs. overall prevalence 52.5%."

Cited Source: Eli S Rosenberg, Elizabeth M Dufort, Debra S Blog, …, New York State Coronavirus 2019 Response Team, COVID-19 Testing, Epidemic Features, Hospital Outcomes, and Household Prevalence, New York State—March 2020, Clinical Infectious Diseases, https://doi.org/10.1093/cid/ciaa549

Peer Reviewed?: unclear - "accepted manuscript"

Study Methodology: "epidemiological investigations and hospital record-linkage"

Notes: This is a study of household infection, not overall infection rates. It looks at infections of people in New York State outside of New York City. A total of 138 children ages 5 to <18  with a household member who had COVID-19 were tested, and 44 tested positive. 

Supports DESE Claim?: No - this is not a New York City study. The data is from New York State excluding NYC.

A Look at DESE's Cited Sources: Footnote #69 - Age specificity of cases and attack rate of novel coronavirus disease (COVID-19)

DESE Claim: "In one study from Japan, 7.2% of exposed male children ages 0-19 and 3.8% of exposed female children tested positive for COVID-19, compared to 22.2% of exposed males ages 20-59 and 21.9% of exposed females ages 20-59"

Cited Source: Mizumoto, K., Omori, R., & Nishiura, H. Age specificity of cases and attack rate of novel coronavirus disease (COVID-19). medRxiv 2020. Available at https://www.medrxiv.org/content/10.1101/2020.03.09.20033142v1

Peer Reviewed?: No - pre print letter to the editor

Study Methodology: Analysis of public health data

Notes: The study examined domestically acquired cases of COVID-19 in Japan. "What our short analysis shows is that children are less likely to be diagnosed as cases, and moreover, the risk of disease given exposure among children appears to be low." 

Supports DESE Claim?: Yes.

A Look at DESE's Cited Sources: Footnote #68 and #74 - Household secondary attack rate of COVID-19 and associated determinants

DESE Claim 1 (footnote #68): "In another study from China, exposed children less than 19 years of age became infected at a rate of 5.3%, vs. 13.7% for 20-59 and 17.7% for 60+."

DESE Claim 2 (footnote #74): "In a study from China, only 5% of household clusters were found to have a child <20 as the index patient"

Cited Source: Jing, Q. L., Liu, M. J., Yuan, J., Zhang, Z. B., Zhang, A. R., Dean, N. E., ... & Lu, Y. (2020). Household secondary attack rate of COVID-19 and associated determinants. medRxiv. Available at https://www.medrxiv.org/content/10.1101/2020.04.11.20056010v1

Peer Reviewed?: No - preprint

Study Methodology: " Based on a comprehensive contact-tracing dataset from Guangzhou, we estimated

both the population-level effective reproductive number and individual-level secondary attack

rate (SAR) in the household setting."

Notes: This study studies transmission in the household setting. "Case finding and isolation alone may be inadequate to contain the pandemic and need to be used in conjunction with heightened restriction of human movement as implemented in Guangzhou." There were fewer household secondary cases in young people.

Supports DESE Claim 1?: Yes, for household exposed children.

Supports DESE Claim 2?: Yes

A Look at DESE's Cited Sources: Footnote #64 - Suppression of COVID-19 outbreak in the municipality of Vo, Italy

DESE Claim: "After an outbreak in Italy, no children under 10 were infected and children 11-20 were infected at half the overall rate."

Cited Source: Lavezzo, E., Franchin, E., Ciavarella, C., Cuomo-Dannenburg, G., Barzon, L., Del Vecchio, C., ... & Abate, D. (2020). Suppression of COVID-19 outbreak in the municipality of Vo, Italy. medRxiv. https://www.medrxiv.org/content/10.1101/2020.04.17.20053157v1

Peer Reviewed: No - preprint

Study Methodology: "We collected information on the demography, clinical presentation, hospitalization, contact network and presence of SARS-CoV-2 infection in nasopharyngeal swabs"

Notes: The municipality was locked down upon the first death from COVID-19. A large percentage of the population was tested for SARS-CoV-2. 73 cases (out of 2,812 subjects) were found. Of those, none were kids under 10. The positive rate for people 11-20 was 1.2%, compared to an overall infection rate of 2.6%.

Supports DESE Claim?: Yes. In this particular outbreak, there were not any infected children under 10, and a lower infection rate in people 11-20.

A Look at DESE's Cited Sources: Footnote #63 - Coronavirus Disease-19: The First 7,755 Cases in the Republic of Korea

DESE Claim: "In a South Korea study, children under 20 only accounted for 6.2% of all positive cases."

Cited Source: Coronavirus Disease-19: The First 7,755 Cases in the Republic of Korea. Osong Public Health Res Perspect. 2020;11(2):85-90. Published online April 30, 2020. DOI: https://doi.org/10.24171/j.phrp.2020.11.2.05

Peer Reviewed?: unclear

Study Methodology: Epidemiological investigation of public health data

Notes: This is a summary of the demographic characteristics of the cases in Korea. 

Supports DESE Claim?: Yes

A Look at DESE's Cited Sources: Footnote #56 - Massachusetts Child and Youth Serving Programs Reopen Approach: Minimum Requirements for Health and Safety

DESE Claim: "This list focuses on establishing processes and communication structures; future guidance will have more details about concrete operational planning."

Cited Source: For example, see: Massachusetts Department of Early Education and Care, Massachusetts Child and Youth Serving Programs Reopen Approach: Minimum Requirements for Health and Safety (pg. 6-8) https://eeclead.force.com/resource/1592325278000/Min_Req

Peer Reviewed?: no - government document

Study Methodology: N/A - policy document

Notes: The cited "example" source is a Massachusetts Early Education and Care reopening document. It includes much more specific detail than the DESE document. 

Supports DESE Claim?: N/A - cited source is an exemplar

A Look at DESE's Cited Sources: Footnote #55 - Thinking Outside The Box: Adaptive Reuse For Education

DESE Claim: "Schools should consider engaging community partners to find spaces outside the school (e.g., libraries, community centers) to set up additional classrooms"

Cited Source: Spaces4learning, Thinking Outside The Box: Adaptive Reuse For Education (November, 2019). Available at https://spaces4learning.com/Articles/2019/11/01/Adaptive-Reuse-For-Education.aspx?Page=2

Peer Reviewed?: No.

Study Methodology: N/A - news style article 

Notes: Pre-COVID pandemic (November 1, 2019) article. The article is about reuse of non-school space for school facilities needs. It discusses a specific case of a big box retail store being repurposed into district offices and an early childhood center. Notably, both cases discussed required substantial construction and renovations, which may not be practical in the two months until school reopens.

Supports DESE Claim?: Yes

A Look at DESE's Cited Sources: Footnote #54 - Ready Schools, Safe Learners, Guidance for School Year 2020-21 (June 10, 2020)

DESE Claim: "Utilizing external facility spaces: Schools should consider engaging community partners to find spaces outside the school"

Cited Source: Oregon Department of Education, Ready Schools, Safe Learners, Guidance for School Year 2020-21 (June 10, 2020). Available at https://www.oregon.gov/ode/students-and-family/healthsafety/documents/ready%20schools%20safe%20learners%202020-21%20guidance.pdf

Peer Reviewed?: No

Study Methodology: N/A - school opening guidance from the state of Oregon.

Notes: This source is Oregon's school reopening plan. As of June 26, 2020 Oregon has 7,083 cases of COVID-19 in a population of 4.218 million, a rate of 168 cases per 100,000 people. Massachusetts has 108,443 COVID-19 cases in a population of 6.893 million, a rate of 1,573 cases per 100,000 people. In other words, this guidance document is intended for a state with a case rate nearly 10 times lower than the one in Massachusetts. Oregon released their guidance on June 10. Page 13 of the plan does include the suggestion of considering other spaces. "Consider utilizing outdoor spaces, common areas, and other buildings in planning." The guidance also recommends/requires masks, physical distancing, cohorting, and hand washing. 

Supports DESE Claim?: Partial - supports use of other buildings, does not mention community partners

A Look at DESE's Cited Sources: Footnote #53 - Seasonal incidence of symptomatic influenza in the United States

DESE Claim: "Whereas for COVID-19 it appears children are less likely to be infected with and to transmit COVID-19, this is not the case for influenza, where children are frequent transmitters."

Cited Source: Tokars, J. I., Olsen, S. J., & Reed, C. (2018). Seasonal incidence of symptomatic influenza in the United States. Clinical Infectious Diseases, 66(10), 1511-1518. Available at https://academic.oup.com/cid/article/66/10/1511/4682599

Peer Reviewed?: Yes

Study Methodology: Literature review and statistical analysis

Notes: This is a 2018 (pre-COVID) study of the flu. The incidence of the flu can only be estimated because there are so many cases. Kids were estimated to have the flu more than adults. No claims about transmission to others are included.

Supports DESE Claim?: No. The study does not discuss transmission rates of the flu, but does indicate that children are more likely to have the flu.

A Look at DESE's Cited Sources: Footnote #52 - Distribution of influenza virus types by age using case-based global surveillance data from twenty-nine countries, 1999-2014

DESE Claim: "Whereas for COVID-19 it appears children are less likely to be infected with and to transmit COVID-19, this is not the case for influenza, where children are frequent transmitters."

Cited Source: Caini, S., Spreeuwenberg, P., Kusznierz, G. F., Rudi, J. M., Owen, R., Pennington, K., ... & Njouom, R. (2018). Distribution of influenza virus types by age using case-based global surveillance data from twenty-nine countries, 1999-2014. BMC infectious diseases, 18(1), 1-10. Available at https://bmcinfectdis.biomedcentral.com/articles/10.1186/s12879-018-3181-y

Peer Reviewed?: yes

Study Methodology: "random-effects meta-analysis models" and "meta-regression and sub-group analyses"

Notes: Influenza infection rate study. "young children are the age group most affected by influenza in relation to their size in the country’s population" 

Supports DESE Claim?: Partial (influenza part of the claim only)

A Look at DESE's Cited Sources: Footnote #51 - Determining the dynamics of influenza transmission by age

DESE Claim: "Whereas for COVID-19 it appears children are less likely to be infected with and to transmit COVID-19, this is not the case for influenza, where children are frequent transmitters."

Cited Source: White, L. F., Archer, B., & Pagano, M. (2014). Determining the dynamics of influenza transmission by age. Emerging themes in epidemiology, 11(1), 4. Available at https://link.springer.com/article/10.1186/1742-7622-11-4

Peer Reviewed?: Yes

Study Methodology: data analysis

Notes: This study looks at transmission patterns of H1N1 influenza. Children spread influenza more than other age groups.

Supports DESE Claim?: Partial (influenza part of the claim only)

A Look at DESE's Cited Sources: Footnote #50 - On the relative role of different age groups in influenza epidemics

DESE Claim: "Whereas for COVID-19 it appears children are less likely to be infected with and to transmit COVID-19, this is not the case for influenza, where children are frequent transmitters."

Cited Source: Worby, C. J., Chaves, S. S., Wallinga, J., Lipsitch, M., Finelli, L., & Goldstein, E. (2015). On the relative role of different age groups in influenza epidemics. Epidemics, 13, 10-16. Available at https://www.sciencedirect.com/science/article/pii/S1755436515000511

Peer Reviewed?: Yes

Study Methodology: Mathematical modeling and simulation

Notes: This is a 2015 (pre-COVID-19) article. It looks at which age groups are most important in the spread of influenza. Children were more likely to spread the flu. The study had "consistent evidence for a leading role of school age children in propagating the major influenza A epidemics."

Supports DESE Claim?: Partial (influenza part of the claim only)