A month ago, enterovirus D68 (EV-D68) was a mild concern to parents around the country as kids headed back to school. The virus caused fever, runny nose, sneezing, coughing, and body aches in mild cases, and wheezing and difficulty breathing in severe cases. But the virus seemed to be isolated to the Southeast and parts of the Midwest and unlikely to be fatal. A month later, the landscape seems very different.

"We suspect that millions of people have been infected with this virus in the last two months,”  said Mark Pallansch, Director of the Division of Viral Diseases at the Centers for Disease Control and Prevention. "This is by far the largest cluster [of EV-D68] we've seen," said Stephen Morse, professor of epidemiology at Columbia University's Mailman School of Public Health.

Per the CDC, EV-D68 has now been confirmed in 46 out of 50 states. "It's almost sure to soon be the whole continental United States, because it’s spread so well," Morse said.

While panicked school boards call emergency meetings and parents try to figure out how to protect their children, the unfortunate truth is that much about EV-D68 remains a mystery. And it's likely to remain that way.

EV-D68 was, until recently, fairly rare. First discovered in California in 1962, it's one of hundreds of seasonal viruses that pop up every summer and fall. When everyone in the office seems to have a cold, or your kid suddenly begins producing alarming amounts of mucus, it's usually because of a virus similar to EV-D68.

But 2014 has been unique. "What's different this year are the number of children coming to the ER and actually being sick enough to be admitted or end up in the pediatric intensive care unit," said Lisa Miller, division director for Disease Control and Environmental Epidemiology at Colorado's Department of Public Health. "Those numbers were higher than normal for this time of year. That’s what first got people’s attention, how sick these children were." In Colorado, Missouri, California, and other states, children (some of whom have tested positive for EV-D68) have been struck with with polio-like symptoms, including partial paralysis. And in New Jersey, 4-year-old Eli Waller was found dead after falling asleep. The county health commissioner later said it was due to complications from EV-D68.

The problem with viruses like EV-D68 is that, with current technology, we can't do much more than track them as they spread across the country—and even that's difficult.

Currently, testing for EV-D68 can only be done at the CDC or a limited few state agencies. Colorado, for instance, has only been able to send 25 samples to the CDC for testing. The CDC, meanwhile, is struggling to make its way through a backlog. "Right now, a specimen from receipt to result is taking 72 hours to process," Pallansch said. "And there's a limit to how many of those we can test in a day, probably about 100 or so."

The CDC is working on a test that should speed up processing to about 24 hours, and allow  their labs to process several hundred samples a day. But that's just a fraction of the millions of people the CDC estimates have contracted EV-D68. It's simply unfeasible, given current technology, to test every person who complains of respiratory distress for EV-D68. And even if there was a cheap and quick method to detect EV-D68, there isn't much doctors could do—there's no treatment for the virus.

Both Miller and Pallansch believe that EV-D68 is likely on a downward course. Colorado is seeing fewer children in hospital than in mid-September, and the CDC is getting anecdotal information that there's a similar decline in other states. "People should understand that there are a lot of people getting infected and getting common colds," Pallansch said. The children ending up in hospitals are "still a relatively uncommon outcome to this very common infection."

This is not the first time that EV-D68 has hit the U.S. From 2008 to 2010, there were clusters of outbreaks of EV-D68 significant enough for the CDC to note in its Morbidity and Mortality Weekly Report. Morse is hopeful that this outbreak results in more research than was done in 2011—including sequencing the genome of EV-D68, which could help provide clues about why this historic outbreak has occurred—but remains pessimistic that treatment or prevention is likely.

The most likely course for EV-D68 is that it will infect a great number of people this year. Most will barely notice. An unlucky few will get very, very ill. And then the season for this virus will pass, and we will shift our focus elsewhere until the next outbreak.

But there's always the danger in something like EV-D68 suddenly mutating. Because viruses are so prevalent within the population, even a small uptick in the deadliness of a virus can wreak havoc. "The Spanish flu didn’t have a high mortality rate," Morse said. "But it’s the law of large numbers. Even a low mortality rate in a large number of people—half the world was infected with Spanish flu—is going to produce large numbers. Thank goodness it wasn’t Ebola, or none of us would be here.”