1 of 19

COVID Surveillance Proposal

Belmont School Committee Meeting

September 22, 2020

Jamal Saeh, Kate Jeffrey, Larry Schmidt

We have no conflict of interest to declare with the proposed plan, or recommendation for the selection of the specific lab.

Jamal Saeh, PhD is the Executive Director and Global Program Leader at a local pharmaceutical company overseeing the research and development of candidate drugs in oncology focused on epigenetics and cell death pathways. He is a Belmont parent to two BHS students.

Kate Jeffrey, PhD is an immunologist at Massachusetts General Hospital and Harvard Medical School. She is an Associate Professor of medicine, a John Lawrence MGH Research Scholar 2020-2025, faculty member of Harvard Immunology and Harvard Virology and an associate member of the Broad Institute. She is a Belmont parent to a first grader.

Lawrence Schmidt, PhD is the Victor J. Menezes Career Development Professor of Finance at the Massachusetts Institute of Technology’s Sloan School of Management, where he conducts research and teaches courses on topics at the intersection of finance and macroeconomics. He is a Belmont parent of three children.

2 of 19

What is surveillance testing?

Diagnostic testing looks for occurrence of COVID-19 at the individual level when there is a reason to believe an individual may be infected

i.e symptomatic

Surveillance testing looks for infection within a population or community, and can be used for making health management decisions at the population level

i.e asymptomatic

3 of 19

BPS has effective risk mitigation strategies in place to support hybrid

DESE metrics support opening Belmont schools in Hybrid mode now

In the past 14 days, 1 new case was registered in Belmont

Experts agree for BPS-like hybrid model the risk is very low for students (0.4%) and teachers 0.6% (Glanz et al., Cohen et al)

Proactive COVID-19 surveillance reduces risk and provides opportunity to expedite and increase in-person learning

Saeh et al Belmontonian

4 of 19

Detection of SARS-CoV-2 with PCR

https://covidtestingproject.org/

3 viral genes in parallel and requires 2 of 3 to be positive

false positive rate <3%

5 of 19

Surveillance a tool to enhance/increase in-person learning

Covid testing costs are decreasing

Examples of affordable testing options to support surveillance

Urgent Care PCR $160/person (nasal swab, available to individuals)
Broad PCR $50/person (nasal swab, used by area private schools and Universities)
Mirimus labs ~$15 per person (pooled saliva, proposed for Belmont)

Emerging technologies:

BINAXnow 15-min $5/subject (EUA; limited to symptomatic subjects, limited availability until Oct), 3M & other $1-5/subject by Oct/Nov (?)

6 of 19

Mirimus SalivaDetect Rapid Test implemented in school setting

Self collection of saliva via a straw. 2 lay people/250 samples to help with logistics/barcodes

7 of 19

Draft Surveillance Testing Plan and Next Steps

All ~4500 BPS students and staff (prior to start of hybrid). Saliva pooled per class

~10% students

~100% teachers

(e.g. Cohort A end of day on Friday)

~10% students

~100% teachers

(e.g. Cohort B end of day on Friday)

Week 1

~$12K

Week 2

~$12K

Week 0 (Baseline)

~$80k

Next Steps:

Individuals/parents in the pool are informed of a suspected positive case
Mirimus identifies positive pool down to 2 subjects in ~24 hours (also will likely eliminate a false positive)
Individual confirms positivity through PCP and informs school
School identifies cohort, initiate contact tracing, and informs individuals to initiate further testing and quarantining if necessary

...

8 of 19

Indicative budget for surveillance scenarios

Scenarios	Frequency	Budget ($M)^	Comment
1	Once Weekly	~$2.8	Testing 100% of BPS population: Logistically challenging Brute Force. Random sampling a scientifically established alternative.
2	Twice weekly	~$5.5
3	Once Weekly	~$0.08*	Boston Public Schools (random 5% teachers)
4	Once weekly	~$0.4	Belmont Hill: 15% students at random & teachers can opt in
5	Once weekly	~$0.5	Baseline testing (all BPS), then 10% of students at random and 100% teachers weekly. Provides infrastructure for more parents to opt in.

^Assumes BPS population ~4542 K-12, cost of test is $15 through academic year, includes cost of deconvolution at $760 per pool, assumption for # deconvolution are scaled for DESE yellow metric (calc in backup).

*Assumes a $160 per test; e.g. Urgent Care to minimize infrastructure build up

Notes on Baseline Testing:

Belmont is Green per DESE. The test positivity rate in Belmont for Covid 19 is 0.51% (Sep 19 2020) based on ~2000 tests conducted over a 14 day period. The baseline testing of BPS ~5000 is likely to reflect the community test positivity rate. As such, it is likely to identify some +ve case. Identifying a +ve cases is an indication that the strategy works and should not be to delay school opening but rather, identify, contact trace, and quarantine the individuals.

Belmont High School: 1200 students, 97 Teachers, support staff unknown
Chenery Middle School: 1150 students, 84 Teachers, 30 support Staff
Burbank Elementary School: 417 Students, Approx 26 Teachers, ? Support Staff
Butler Elementary School: 384 Students, Approx 25 Teachers, ? Support Staff
Wellington Elementary School: 630 Students, approx 39 Teachers, ? Support Staff
Winn Brook Elementary School: 461 Students, Approx 29 Teachers, ? Support Staff

9 of 19

Surveillance Proposal

We recommend daily symptom reporting/temperature checks & surveillance scenario option 5

We recommend a two-stage approach: Use current available lab-based technology, and switch to cheaper Point of Care (PoC) solution once available (Oct/Nov?)

The budget can be significantly reduced/offset by

Funding from CARES act or future COVID relief for K-12
Fundraising drive (some already committed)
Availability of cheaper PoC tests before year’s end
Utilizing existing State and health care infrastructure

10 of 19

Request School Committee

Adopt proposed Option 5 to make hybrid sustainable

With robust surveillance implemented, accelerate timetable for hybrid

Revisit hybrid model to include more in-person learning

11 of 19

Backup

12 of 19

COVID testing aligns with BPS strategy

BPS Strategic Pillars	Background and comments
Safety	•Need a plan should safe and effective vaccine not be available for 2020/21 academic year •Testing=Vaccine •Multiple K-12 schools have initiated an surveillance/tests (e.g. Wellesley, Belmont Hill, Boston Public Schools, NH, UN Int/NYC)
Social/Emotional	•Enabling sustained in school learning key to social emotional and academic success
Equity	•Adverse effects of remote learning widen the equity gap •Exit from BPS is on the rise; defunds schools, and negatively impacts the mission of public education
Education	•Impossible to maintain the rigor of in-person learning without sacrificing quality and content •Kids have fallen behind and will likely continue to fall behind •Belmont schools & students will become less competitive relative to school districts already in more comprehensive hybrid in MA and around the country

13 of 19

Statistical justification for 7.5+% threshold

Input Parameter	Criteria	Rationale
BPS Population	~4500	Teachers and Students
Target Prevalence	0.11%	Yellow DESE
Confidence in prevalence	>90%	Belmont and neighbouring towns Green or better for ~3 months
Assay Sensitivity	>96%	PCR or ab specs
Prob of infection in between testing	<1%	Meyers’ et al risk 1/1000
Desired Confidence Interval (CI) for Prev.	>90%	90% 1 sided 5% acceptable to ensure Prev <0.11% 95% 2-sided, conservative

Sample calculation using online Epitool

Sample size: estimates based on

A. Fodai et al, One Health, 2020

n~700 (95% CI), or 170 (90% CI)

J. Cohen et al white paper suggest testing 70/1000 students, or ~350 tests per week for BPS (~7.5%)

Frequency: Authors suggest surveillance over 14 day period is acceptable; provides flexibility to test each BPS cohort at different days over a 2-week period.

14 of 19

Calculation worksheet

15 of 19

Maximizing education while minimizing COVID risk

In person with countermeasure presents low CUMULATIVE risk

Cohen et al, Institute for Disease Modeling, Seattle, Washington, Aug 13, 2020

BPS-like

Enhanced Hybrid

16 of 19