Flight bans don’t help much this late in a pandemic

In 2006, four epidemiologists — two from the U.K., two from the U.S. — published a remarkable study in the scientific journal Nature on “Strategies for mitigating an influenza panic.” For a lay reader, the most compelling parts are the mesmerizing videos, available as quite-large files in the “Supplementary Information” section below the main article (don’t bother trying to watch them on a mobile phone), that show the spread, peak and decline of a hypothetical pandemic over 140 days in the U.K. and 170 in the U.S.

In the U.S., the outbreak starts as a few red pinpricks on the map, spreads slowly for weeks, then quickly explodes over the whole country. Or, as the authors described it, “early spread is focal around seed infections (typically in urban centres) imported from overseas, but rapidly becomes almost homogenously distributed across the whole population.”

They also offered some findings based on mathematical modeling of what could slow the disease’s spread. The article discusses school closings, household quarantines, medicines and vaccines, but let’s focus here on travel restrictions. They can slow the course of a pandemic, according to the model, but have to be spectacularly effective to slow it much. If travel bans reduced imported infections by 99.9%, they could delay the peak of the pandemic in the U.S. by six weeks, the study found. If they reduced them by 90%, the delay would be less than two weeks.

So travel restrictions can be effective measures against a pandemic — but not forever, and their impact is minor to nonexistent if the disease is already widely present. A brand-new modeling exercise by 16 researchers from Italy and the U.S. that was published last week in the journal Science (both articles discussed here are outside the journals’ usual paywalls) makes some estimates for Covid-19, and found that the Chinese travel quarantine of Wuhan, where the outbreak started, only slowed the progression of the coronavirus epidemic by three to five days within China because so many people infected in Wuhan had already traveled to other Chinese cities.

The authors found that banning travel to and from China, as the U.S. did and airlines effectively did by canceling flights, likely had a bigger impact on the number of infected people in other countries, reducing cases imported from China by 77%. Still, the study also found that even a 90% effective ban would delay the diffusion of the disease by only a couple of weeks. Without major declines in transmission within countries — by canceling large events, school and office closures, and other social-distancing measures — it would “only modestly affect the epidemic trajectory.”

These are the results of mathematical modeling exercises (which a lot of epidemiology consists of these days), not proof of anything. But they do fit with what we’ve seen over the past month. The travel restrictions that President Donald Trump imposed Jan. 31 on China — criticized at the time by the World Health Organization and some Democratic politicians as counterproductive — do seem to have slowed the spread of the coronavirus here relative to other countries such as Iran and Italy. But they didn’t stop it, and the epidemic now seems to be on the same frightening trajectory in the U.S. that it has followed almost everywhere else. Meaning that the restrictions on travel to Europe that Trump announced last night are almost certainly pointless.