Closing schools could stop the flu

In June 2009, the new H1N1 flu strain was spreading like wildfire in western Canada, just as it was in dozens of countries around the world. But within a few weeks, the flames were nearly out, and a new study pinpoints a possible reason: summer vacation.

On June 12, high schools in the province of Alberta let out for the summer. On June 19, the middle schools finished, followed by the elementary schools on June 26. Researchers from McMaster University compared those dates to the incidence of new H1N1 cases in Alberta, and using a complex statistical analysis, estimated that closing schools reduced flu transmission among school children by more than 50%.

That, in turn, reduced transmission in the population at large. The findings, published in the Annals of Internal Medicine, support the idea that closing schools could reduce or slow down a dangerous outbreak of influenza.

Many public health experts believe that closing schools could reduce the spread of the flu, but hard evidence has been scarce, said study author Dr. David Earn.

“There’s been a lot of work to see if there’s an effect [from school closure] at all. Everyone thinks it ought to work, but it’s one thing to think that, and another to demonstrate that it does.”

The results, he wrote in the paper’s conclusion, are “compelling evidence.”

The finding is important but not surprising, says Dennis Chao, a researcher in the Vaccine and Infectious Disease Division at the Fred Hutchinson Cancer Research Center.

“If you know anyone with little kids, they get sick all the time, and kids get the flu more than adults. If you interrupt transmission in the highest transmission group, that’s a good way to stop the flu,” says Chao, who has also studied the effects of school closure.

Indeed, the Canadian study is not the first to measure those effects. Medical historian Howard Markel found that closing schools – along with other measures, like banning public gatherings – slowed flu transmission during the deadly pandemic of 1918. A handful of studies have found that seasonal flu transmission slows during the winter school break, and another study, published last year, found that H1N1 cases in England dropped after the start of the 2009 summer break.

What makes the Canadian study powerful is the detailed quality of the data on which it’s built.

While the United States and most other countries stopped doing laboratory confirmations of suspected H1N1 cases as soon as it became clear the strain was no widespread killer, Alberta continued lab testing through October 2009, providing a definitive measure of changes over time.

Also, unlike schools in the U.S., schools in Alberta follow a uniform schedule: all students in a given age group finish school on the same date. That makes it easier to see the effect of school attendance, or non-attendance, on transmission rates.

The most recent guidance from the Centers for Disease Control and Prevention, developed in 2007, says school closure is not recommended when the illness is similar to seasonal flu, but says local officials should consider it an option when the flu strain is causing more serious disease.

The CDC is working to update its basic planning document, but adjustments are a delicate and politically sensitive task. Even if the benefits are potentially huge, closing schools for an extended period may come with significant burdens – for example, parents missing work, increased childcare needs and disruptions to education. Other challenges include finding ways to feed children who depend on school lunches.

“When you expect a huge case fatality rate, like what we first thought we saw in Mexico City [in 2009], then you just do it,” says Chao, speaking of school closure. “But for a normal flu, it’s probably not worth it.”