Maybe they were just better at school closures 100 years ago?

An emergency manager compiles lessons on closing and reopening schools from the 1918 pandemic.

13 min readJul 10, 2020

School reopening is on everyone’s mind. Many people are writing about the need to open schools as an emergency; others say that reopening is a disaster in the making. Almost nobody is writing about the lessons we have learned about the mechanics of closing and reopening schools safely and transparently.

So, let’s take a crack at that, shall we? As an emergency manager, I compiled stories on what school districts did during the 1918 Influenza Pandemic and how it worked in order to help parents and school officials wrestling with this planning now.

(BTW, all glory to the U Michigan Flu Archive for primary sources for much of this, with some other sources like Markel, Stern and Michalski listed at the bottom)

Fair warning, there are no quick fixes here. Epi tells us that schools shouldn’t reopen before community transmission is looooooow. MUCH LOWER than they are as of this writing in most places across this country (Florida, I’m looking at YOU). So, nothing here applies until case rates go DOWN just as in 1918 schools stayed closed until case rates plummeted.

Deep breath; it’s a deep dive. Let’s go — Warning: SO MANY BULLET POINTS!

First, some background

In 1918, U.S. schools were dealing with the push to end World War I, sell liberty bonds and enlist students into the war effort. Average class sizes were about 30 students (it’s about 16 now, tho I get that seems low) and many schools had only recently moved from one-room schoolhouses to classes by grade. It was crowded in there!

Here’s a key point: Many urban schools had just received a massive influx of Progressive Era resources. School nurse programs in NYC were founded in 1902; hygiene programs, new buildings and clean water supplies came in these years. Municipal infrastructure was in; teacher salaries increased by 34% on average each year from 1917–1920. On the other hand (sigh) the cost of living increased faster, meaning that in real wages teachers made less than they made in 1913.

The massive Progressive funding for buildings and medical care was pretty damn racist, too. A dissertation about the pandemic in Baltimore, for example, found that only property taxes from (much poorer) black households went to black schools — giving those schools an overcrowded baseline with many fewer teachers and none of the medical checks of the white schools.

What’s the plan?

Let’s focus on what we can learn about how school closures and reopenings could work and what schools did in 1918.

We’ll do that through five questions: Did closing schools help and for how long? How were closing/reopening decisions made? What did schools and students do during closures? What happened in cities that didn’t close schools? and What safety practices were in place to lessen outbreaks at schools when they reopened and did they work?

Listen, I’m a public health guy; it goes without saying that Covid-19 is very different that 1918’s influenza. I get it. But THESIS ALERT: Many of the societal mechanics of managing them are similar, and there are some key lessons that 2020 has obviously not learned.

Did closing schools help and for how long?

Howard Markel and team studied flu measures in 43 cities and found that layered distancing measures put in place early and held there for longer both “flattened the curve” of the epidemic and lowered mortality over all.

What does that mean? In 1918, we’re talking about school closure concurrently combined with public gathering bans as the most common combo of measures — implemented in 34 cities (79%) for a median duration of 4 weeks (range, 1–10 weeks). Schools themselves were officially closed a median of 6 weeks (range, 0–15 weeks).

Not everybody closed, and some closed poorly. In fact, Stern and team found four ways that cities managed school closure (quoting here):

Cities that kept schools open and relied heavily on the daily medical inspections of students; (e.g., NYC, Chicago, New Haven)
Cities that closed schools and experienced interagency conflict and low compliance with nonpharmaceutical interventions; (e.g., Baltimore, Minneapolis)
Cities that closed schools and experienced inconsistent and sporadic interagency cooperation and mixed compliance with nonpharmaceutical interventions (e.g., Boston, Portland) ; and
Cities that closed schools and experienced interagency cooperation and high compliance with nonpharmaceutical interventions (e.g., Milwaukee, St. Louis, Rochester).

In 1918, closing schools did not guarantee lower mortality (Looking at you, Philadelphia!). However, cities that closed schools well (Looking at you, St. Louis!) were generally top tier in terms of mortality reduction. Management and communication mattered!

There probably isn’t a way to reopen without causing ANY additional cases. Some cities in 1918 could track and contain school outbreaks; some couldn’t. 14 cities closed schools again after reopening. Kansas City (bless y’all’s heart) ran the cycle three times. But cities that closed early and held on through a peak of cases (whether once, twice or three times) still had lower mortality over all.

When should these decisions be made? Well, closing early is part of what makes school closures helpful. From Mary Battenfield: “St. Louis closed the schools about a day in advance of the epidemic spiking, for 143 days. Pittsburgh closed 7 days after the peak and only for 53 days. And the death rate for the epidemic in St. Louis was roughly one-third as high as in Pittsburgh.”

Big Lesson: If closing is reactive, it’s probably too late. So, when/if schools reopen, they need a trigger for closure that is NOT a confirmed outbreak of Covid-19 in the school.

How were decisions to close and reopen schools handled?

Since it matters how cities managed closures, what can we learn from their decision-making?

Most cities had emergency advisory councils made of residents and community leaders who weighed in on key decisions and helped secure buy-in across the city. NYC had three, as Stern relates, “an Emergency Advisory Committee that included leaders from the American Red Cross and the American Public Health Association; the Women’s Emergency Advisory Committee, which spearheaded volunteer campaigns on the ground; and a nursing committee led by Lillian Wald to coordinate home care”. Transparency!
Cities that implemented school closure/reopening well had clear lines of authority between agencies/levels of government and “existing trust and transparent communication” between health officials and population. Communication!
Here’s an important one: Openings and closing were rarely district wide, but based on children’s age and neighborhood conditions. They were often made school by school. High schools might open first, to make sure practices worked, or schools would reopen based on neighborhood case levels. Teachers and principals had a big voice.
Schools tried to standardized why they were closing to make decisions more rational. Cleveland allowed individuals schools to decide to close or not based on absenteeism, for example. There was debate across the country about whether 5% absenteeism for illness vs 8% was the right closure trigger, rather than confirmed influenza cases.

Big Lesson: The cities that did best had transparent decision-making processes that were highly local and involved the folks actually at risk (e.g. principals and teachers).

From the U. Michigan Influenza Archive — Mrs. Pearl Webster, Redondo Beach School Teacher, Hearing a Recitation

How did school districts handle closures?

Just like today, schools tried to keep learning going, but they also supported the healthcare response and invested in teachers.

LA created mail-in courses for high schools and gave teachers professional development courses.
Teachers and school nurses generally volunteered for health department operations either in hospitals or in door to door canvassing, which gave them immense public respect. Lowell, Ma and Richmond, VA teachers coordinated neighborhood health inspections, for example.
Teachers were always paid, though sometimes attorneys general were needed to enforce that payment. Usually janitorial staff were paid. Contract staff such as bus drivers were rarely paid.
Just like today, students were generally fed because of the loss of school lunch programs.
Many teachers went house to house to hear students recite their lessons from their doorsteps (see the above picture).
Special programs (in Milwaukee, for example) conducted outreach to immigrant and minority students who faced more poverty and disease burden. In places like Denver these involved scapegoating, but many programs were driven by a legitimate desire to help, even though the racism of the school system and other structures had created some of the problem to begin with.
Homework assignments were printed in newspapers. You have to squint, but here’s an example from Albany, New York. High school seniors have been justifying the characterization of Lady Macbeth for a loooong time.

Big Lesson: This sense of everyone working together (e.g. teachers supporting healthcare, communities giving to teaching, newspapers chipping in) was a key piece of what made certain cities successful. Of course, there were limits, as shown with often limited minority outreach and lack of pay to contract staff.

What about cities that didn’t close schools?

Only three cities kept schools open: New York City, Chicago and New Haven. All of them knew they were making a controversial decision; all of them thought they were doing it in the best interests of the children; all of them poured massive money into the effort.

New York City responded early to the outbreak with a number of measures like bans on public gathering and sophisticated quarantine procedures based on their reputation as the best public health system in the world. In fact, during this pandemic, NYC actually regularly inspected EVERY BLOCK for cases: NYC Health “borrowed inspectors from other city agencies to complete block-by-block surveys for influenza cases, they recruited laypeople and nongovernmental organizations to increase their surveillance capacity.” Stern’s account of how New York City handled school days is worth quoting in full:

Children were not allowed to loiter outside the school as usual, waiting for the bell to ring. Instead, they reported immediately to their teachers for an inspection. Teachers looked for the usual signs of upper respiratory condition: runny noses, red eyes, sneezing, or coughing. Children displaying any of these symptoms were moved to an isolation room for a professional examination. If feverish, they were sent home in the company of someone from the health department who determined on the spot if home conditions were amenable to isolation and care. When homes did not meet standards, children were sent to a hospital. The health department required families of the children recovering at home to either have a family physician or use the services of a public health doctor at no charge.

Chicago stayed open but faced high absenteeism of students and teachers as the fall wave went through — on the order of 50% or so at times, which meant at times schools were functionally closed, if not officially so.
New Haven didn’t close in part because of faith in a serum created by Yale from recovered patients’ blood, which they offered to all of their teachers. Can’t find any data on the effects of the serum! New Haven also suffered high rates of teacher and student absenteeism (no surprise, really). The city made returning teachers and students submit to a medical exam before being allowed in the classroom, which created a fiasco — a larger crowd of kids, parents and teachers all smashed together for hours on line.

How’d these cities do?

In Markel’s mortality study, these three cities were all better than average at lowering mortality — all had long term, layered distancing measures which they implemented early and held until few cases remained. NYC was ranked #15 out of 43 cities, with Chicago doing better and New Haven slightly worse.

HOWEVER, a key question is whether reopening caused outbreaks in schools themselves. Here’s one anecdote from New Haven, that shows the danger.

“The [St. Francis Orphan Asylum] had closed its doors to the outside world on October 1 as a way of protecting the 464 children, sisters, and staff. Only the school supervisor was allowed to enter. The gambit worked for two months straight, until, finally, on December 10, a staff member grew ill. From there, the infection exploded. By January 7, there were 424 cases and seven deaths. Of the stricken, 398 were children, prompting Mayor Fitzgerald to issue a broad appeal for help.20 Community physicians rose to the occasion, and the Red Cross rushed volunteer nurses to the scene, some from as far away as Boston and New York.”

Luckily, this outbreak at an orphanage was past the peak of cases and could be contained. In contrast to this, most schools did not see massive outbreaks. A review of Baltimore attendance records during this time, for example, didn’t show unexpected absences post-reopening.

Big Lesson: With other long-term interventions in place, keeping schools open didn’t lead to massive differences in mortality. However, it did cause some mortality, since none of these cities made Stern’s top tier AND it likely moved some of the mortality burden onto school staff. And remember, rigorous measure like block by block inspections and checking out students’ homes were employed that aren’t being contemplated today.

What safety practices were in place to lessen outbreaks at schools once they opened?

Here’s the biggie! Schools that stayed open or reopened were given significant additional resources in order to manage student/teacher staff health. These are based on school systems which, in general, had seen surging Progressive Era funding for nearly a decade (at least in white communities), in new buildings and with talented, driven leadership. When they reopened, many cities strove to copy what NYC had been doing throughout.

Here are some general practices found:

There were a number of last minute changes, first of all. Plans were announced, changed, then shifted just a few days before schools were meant to open.
Schools were massively cleaned beyond what was probably necessary.
Significant additional doctors and nurses were hired in order to give every arriving child and staff member an individual health inspection every morning. In many cases, this included daily surveys for illness in students’ households. Ill children or those with illness in their households were sent home.
Sophisticated health clinics were set up at many schools — some provided primary care directly on premises.
Teachers also had the power to immediately send students home based on symptoms at any time. Fun story in New Jersey: One middle school class brought pepper to school and managed to get sent home several days in a row, until the teacher accidentally snuffed the pepper and figured out what was going on.
Most cities had neighborhood health inspections run by the health department which many teachers participated in. Students with household illness could be quickly investigated by district captains and excluded from school BEFORE they went. School medical corps often reached out directly to homes with illness and connected families to care.
Quarantine violations came with both fines and 30 days in jail, so families did not send kids to school if they thought they might be breaking that law.
Classes entered schools in staggered 15 or 20 minutes shifts to reduce hallway crowding, with each class assigned to one teacher and one room (no changing classes). Classes were kept fully separate and all students were ordered straight home after school.
Many classes were held in the open air; some Sunday Schools followed suit to prevent outbreaks in churches!
Curriculum focused on the pandemic itself, since it was everywhere. High school shop classes constructed beds for hospitals while home economics classes got together to sew gowns and surgical masks for medical workers. Children were often required to bring in a report on influenza cases–both past and present–in the household.
Schools that could (colleges and boarding schools) isolated students on campus with no visitors.
Many students and staff utilized gauze masks successfully. San Francisco actually blew all their fire whistles in celebration once the masks were no longer required.
Attendance naturally surged back once parents were certain the virus had passed. Schools that opened before this point generally had to close again. Once reopening stuck, extra hours were added to some school days and some vacation days were lost, but no school cancelled or shortened the next year’s summer vacation, though it was discussed.

A big caveat to all of this. In her dissertation on Baltimore’s experience, Victoria L. Michalski found that many of these checks were not done in black communities because of personnel shortages. Black doctors and nurses hired to work in black schools were not paid, and black mortality was substantially higher. Progressive Era reforms didn’t reach everywhere; there’s racism in that lack of resources.

Big Lesson: Massive resources and communications were injected into an already robust school system, though minority communities were often excluded. Still, many of the practices considered here go well beyond what school districts are considering for Fall 2020.

Conclusions

This is not a look at whether or when we should reopen schools during Covid-19; it IS about how a set of leaders managed those decisions a century ago. What’s startling is the level of improvisation, transparency and, frankly, money, they were willing to spend to get it right. Remember that some of these schools were only closed for six weeks or so: The medical inspection systems discussed here were invented that quickly.

So much is different now than in 1918. And, to be clear, none of these analyses are able to bring a full lens on racial injustice at the time, which means that we have a lot of unknowns.

However, what strikes me is the sheer amount of new resources, innovative thinking and public discussion around how best to handle these situations. These leaders thought beyond “closing” and “reopening” — have we?