Smart Restart APS1/ Finally got the full pandemic data file via a FOIA to #Arlington Public Schools. What follows is a historic account of how the pandemic hit schools in the smallest "self-governing" county in the USA; also one of wealthiest (7th) and most densely populated (12th). A 🧵 follows with graphics for 3 years of data, including before/after universal mask protections were removed, right before the BA.2 “You Do You” wave in April/May 2022.
2/ COVID continues to be a significant, systemic challenge that impacts schools, teachers, kids and their families — as well as education. Data can inform policies to disrupt COVID & other airborne illnesses.
3/ Data that follows is from APS. Notes: APS stats lack cases from fall 2020, before Qualtrics system (+100 cases?) Also, APS continuously updates/deletes double records. (Likely +/- 20 records out of more than 9,000 cases in this file.)
APS data: http://tinyurl.com/APSVADashboard.
4/ Here's how @APSVirginia has gotten to more than 9,000 cases, along with dates of policy choices, among them:
❌ Failure to promote masks at height of BA.2 & BA.5 surges
❌ Cut isolation to 5 days (first staff, then kids); allow infectious people to return w/o negative test
❌ Failure to promote masks at height of BA.2 & BA.5 surges
❌ Cut isolation to 5 days (first staff, then kids); allow infectious people to return w/o negative test
5/ We looked at both buildings & programs. Buildings w/most cases:
#Wakefield
#Yorktown
#WashingtonLiberty
#Williamsburg
#Gunston
#Discovery (newer bldg)
#ArlingtonTraditional
#Cardinal (new)
6 of 8 in affluent N. Arlington; Wakefield had a super-spreader event (prom) that led to a particularly massive outbreak, where Yorktown’s overall amount of COVID spread over weeks after masks became optional was higher.
3 had $$ HVAC work that failed to get rooms up to code; code is also inadequate!
6/ We mapped elementary schools in comparison to each other.
#Discovery, #Cardinal, and #Taylor were well above peers during the May "You do you" BA.2 wave.
Anecdotally, after spring break, mask use dropped the most here. VDH listed outbreaks, confirmed 100s of linked cases.
7/ We mapped high schools and middle schools as well. Look at the impact of impact of community behavior in this graph. During the first Omicron wave, prevalence in high schools & middle schools is similar. Thousands got sick over winter break while schools were closed.
8/ But during the "You Do You" BA.2 wave, #Yorktown is well above its peer schools. So is #Williamsburg.
So what was different? Teachers & students told us almost no one wore a mask at Yorktown. (We know that’s "anecdata.")
9/ So essentially, APS schools tracked more closely together during Omicron (when masking was uniform policy) and then there is a major variance when it was "you do you."
System wide, it looks like this, comparing 16 weeks at the start of school last year to this year.
10/ Here's what it looks like, assessing cases from after Virginia's March 1st edict that removed universal mask protections in schools. APS capitulated, sent out a note "masks were optional." Some communities were much slower than others to unmask. Some kids/staff still masking.
11/ +2,000 cases of COVID-19 among all our teachers and staff; some got Long COVID; some decided to quit. This much sickness cannot be good for any workforce.
APS needs strong sick leave policies for all and should cover COVID time off. Or hey, #DoBetterAPS to prevent illness!
12/ Nearly 7,000 student COVID cases now, and counting.
Remember data = confirmed cases, not the true burden, which could be undercounted by 10x or more.
Low income schools may have some degree of undercount due to less access to stuff like $24 rapid antigen tests.
13/ Here's the best breakdown we could do on the grades. Roughly a quarter of ALL students in grades first through fifth grade had COVID last year. The numbers were closer to a fifth for older grades. (We could speculate why.)
Please remember: stats are a big undercount.
Side note: Why are we documenting this in a 🧵? (Click to see graphics.) Because Superintendent Duran acts like COVID in our schools isn’t a problem and rarely talks about air quality.
Schools are seeing a quarter of kids get sick, probably double with uncounted cases; 3,000+ kids are missing a week of learning, some likely getting Long COVID. Also, APS now needs to PLAN for subs to cover for 1,000+ workers getting sick annually.
Data helps to show why the huge PROBLEM must be seen & solved.
14/ Wanted to create a graphic on health equity, but APS only has racial/ethnic data for about 1 of 5 COVID cases. So as far as analyzing the impact of COVID on Black, Indigenous, Latino, Asian, and other minority communities, data just wasn't collected.
So let's at least look at pay grades. Teachers & assistants have a shocking case count. But it's the lack of cases in some areas (cafeteria?) that's alarming — can workers not take time off if sick? Do bosses tell bus drivers to work sick?.
15/ APS staff need adequate paid leave, including COVID+ stay at home days. Otherwise, APS is incentivizing sick people to be at school, which is going to get them more COVID, and it’s going to be a vicious cycle.
There will be lower rates of COVID if APS helps sick/infectious to stay home; and gives people sick days they can use to get vaccinated. APS needs to hire more relief workers to cover for the sick; since it looks like 1,000+ staff COVID cases/year will happen w/o better policies.
16/ Contact tracing. APS largely abandoned telling kids when classmates had exposed them to COVID-19 during the May 2022 "You do you" wave. This left kids to DIY outbreak management. (Who was absent in class today?)
Why not tell a class about a COVID case? Like we do w/ lice?
17/ When you know you were exposed, you can watch for symptoms, get tested.
APS data indicate 31% of staff/32% of kids did NOT have symptoms at the point tested positive.
Data supports sending symptomatic kids home, providing testing access, and communicating about exposures.
18 / APS collected data on what symptoms students & staff had.This graph shows JUST this year (2022) as Omicron and subvariants BA.2 and BA.5, circulated. Coughing, congestion, and fatigue were common. Case notes suggested >40 APS families got diagnosed/treated at a hospital. 😭
19/ Sports: Many pairs/trios of cases on teams on same days. Haven't made a fancy graph, but here are cases by sport. Many sports teams had case clusters. (Et tu H-B ultimate frisbee?)
Biggest clusters:
#WashingtonLiberty boy's basketball (Dec. superspreader)
#Wakefield football
Coaches can instill in players concern for others, a "take care of our team health" mentality; ask sick kids to stay home/get tested.
"Play healthy" should be a policy push.
20/ About those APS buses: There is often COVID-19 riding them. Feel really bad for our bus drivers and attendants who are exposed continuously to germs in their air; but not provided with GOOD masks in a variety of styles.
Rolling down those bus windows seems pretty smart.
21/ Took a look at how often a COVID+ kid was on a morning bus route during the first two months of this school year. The answer was about 27% of the time, based on reported cases. Roll down the windows. Mask up on the bus.
22/ APS collected data about prevalence of infectious kids/staff at school:
In 2022:
3,203 students, 717 employees were on campus (& bus, sports) in 48 hours before testing + for COVID.
So a significant number of these folks were possibly infectious at school; after May 2022, many were walking around during outbreaks w/o masks (especially in North Arlington.)
At least 529 kids, 152 staff members were highly likely contagious: on campus AFTER taking the test that would come back +.
23/ On the bird site, we added another aside, and asked:
@BethanyZSutton
@cdiaztorres240
@PriddyAPS
@MaryKaderaAPS
@ReidForSchools
to get serious about cleaning up the air in our schools; disrupting COVID, flu, RSV, etc.
Hot take: 1,000+ sick staff & 3,000+ sick kids annually will burden this system!
24/ Can you spot the schools having outbreaks of COVID-19?
This is a data table snapshot by week for last school year, and this year, set up by program — not building.
