West Nile virus looked like it was waning as a health threat, with the number of cases dropping each year. Then last summer, it roared back.
The number of people infected with the mosquito-borne illness suddenly spiked in 2012. And Dallas was hit hardest of all.
People showed up in emergency rooms with encephalitis and paralysis, unable to breathe on their own.
Officials scrambled to launch a massive program to spray insecticides across the city. But by the time planes hit the air in August, it was a little too late.
All told, almost 400 people got sick from West Nile virus in Dallas, about half of them with the most serious form of the disease, which can leave people permanently paralyzed. Nineteen people died.
Health officials were baffled. The virus first appeared in the United States in 1999, and though it has spread throughout the continental U.S., outbreaks have subsided over the past decade. In 2010 and 2011, just one person had fallen ill from West Nile in Dallas each year.
"I'm embarrassed to say I and many others thought that West Nile was gone," says Dr. Robert Haley, director of epidemiology at the University of Texas Southwestern Medical Center, and senior author on a study of the outbreak. "Everybody thought it was gone. Even CDC was saying, well, I guess we can move on and do something else."
But by looking closely at the Dallas outbreak, Haley and other scientists say they've found the clues they need to solve the mystery and perhaps save lives in the next outbreak.
Dallas got lucky in one way. Unlike many urban areas, the county had a network of mosquito traps to monitor populations and test for West Nile. In 2012, mosquitoes infected with West Nile started showing up in mosquito traps much earlier than usual, in late May.
Sick people started showing up in emergency rooms much earlier than usual too, in early June.
When Haley and his colleagues looked at weather data for the winter of 2011-2012, they found that it had been unusually mild, with very few days below freezing. That was also true for the winter before the city's second-worst outbreak, in 2006. And the summer of 2012 was rainy, with lots of chances to create the tiny standing puddles West Nile mosquitoes need to reproduce.
"This gives us a new tool for determining in what year we're going to have a big epidemic of West Nile," Haley told Shots. The results were published in JAMA, a journal of the American Medical Association.
Knowing that a mild winter followed by a rainy summer makes for a more dangerous West Nile season could help predict future outbreaks nationwide.
Since there's no vaccine for humans, the only way to prevent West Nile infections is to keep people from being bitten by mosquitoes. That requires knowing when it's going to be a bad mosquito year and when to spray.
Using a formula developed by the federal Centers for Disease Control and Prevention, the Dallas scientists were able to pinpoint when enough mosquitoes are infected to spark an outbreak the West Nile tipping point.
"And it tells us that three to four weeks before you see the human cases occurring," Haley told Shots. "You're able to intervene right when the first infections are being acquired."
Deploying mosquito-control measures at that point, Haley says, could help prevent most West Nile infections.
Last year, Dallas didn't start a concerted spraying effort until mid-August. By then, the outbreak was waning on its own, so the $1.6 million spraying effort was largely wasted. The researchers calculate that if that spraying had started in early July, 110 cases of West Nile could have been prevented, and 12 lives could have been saved.
The Dallas data revealed one more surprise the outbreak affected the rich more than the poor.
By plotting the homes of West Nile victims on a map, the researchers found to their surprise that it was people in the well-off communities of Park Cities and North Dallas who were most likely to become infected. That's unlike many infectious disease outbreaks, which affect the poor.
The well-off neighborhoods may be vulnerable because there are more houses and fewer woodlands; untended swimming pools and heavy sprinkling of lawns may be a factor, too, Haley says.
Many states have had very few West Nile cases in recent years; others face ongoing threats. The virus will continue to pose a significant threat to health in the future, according to a review by CDC researchers also published in JAMA. Here's a glimpse at how the states are faring: