In This Issue >>

Disease Forecasting

Disease Forecasting by Jim Duffy, Page 3

Viruses on the Move

Greg Glass, PhD, worked backward to find his way to disease forecasting. In 1993, a colleague aware of Glass's expertise tracking viruses spread by rats asked him to look at an outbreak of hantavirus pulmonary syndrome (HPS) in the southwestern United States' Four Corners region (where New Mexico, Colorado, Utah and Arizona meet). Rare but potentially deadly, HPS is caused by a virus that makes its way to humans through air they breathe near rodent droppings, urine or saliva.

Trained as an ecologist, Glass looked first at the big environmental picture. "Basically, all infectious diseases are two or more populations—humans and pathogens—coming together," he says. "The pathogens might be in mice or in monkeys, whatever. If you find where things overlap, you find where the action is."

Countless factors—weather fluctuations, vegetation patterns, insect abundance, predator populations and the like—can drive shifts in mice populations. Glass, a professor of Molecular Microbiology and Immunology, needed to identify which ones were both in common among and exclusive to the outbreak's hotspots.

This involved extensive fieldwork, of course, but Glass's specialty is using satellite imagery to chase epidemiological mysteries. This was still regarded as a newfangled tool; not long before, Glass had applied for a grant to study rodents linked to Argentine hemorrhagic fever.

"Nobody thought it would work," he says. "One of the reviewers' comments was, 'I don't know how they expect to see mice with a satellite.'"

Not all reactions were so laughably uninformed. With Lyme disease erupting in the northeastern United States, public health officials in Maryland developed an interest in Glass's technique.

"It was quite successful," Glass says. "We took a straightforward epi approach, and by the time we got done, we'd discovered that people living in high-risk areas identified with the satellite were 20 times more likely to get Lyme disease than people living somewhere else."

In the Four Corners, the HPS outbreak at first stumped even researchers who had studied the region's rodents for decades. They were unable to differentiate in the field between sites where people were getting sick and sites where they weren't.

Satellite images told a different story. They helped Glass track the 1993 outbreak to a chain of natural events ecologists term a "trophic cascade." This particular cascade commenced with heavy 1992 rains associated with an El Niño climate variation, spurring plant growth in certain pockets of the landscape, which in turn boosted insect numbers. The next year, mice numbers were booming.

"As far as the mice are concerned, this was Garden of Eden stuff," Glass says.

The avian flu strain known as H5N1 has more than once leapt from Asia's poultry population to infect humans. Could it lead to a pandemic as deadly as the 1918 flu?

Under ordinary conditions, most Four Corners mice live at elevations higher than humans. But the population explosion of 1993 sent them into new territories and lower elevations.

"It's the reverse of the story Nathan [Wolfe] has to tell," Glass says. "What he's found is people moving into places where retroviruses are. Here, the virus is on the move, going to where people are."

It was after solving this mystery that Glass began to work backward. Had his retrospective discoveries provided him with enough information to travel back in time and predict the outbreak?

Though still a work in progress, the strategy shows great promise. Backed by a five-year NIH International Collaborations in Infectious Disease grant, Glass is embarking on a new project with colleagues in New Mexico and Chile. He is taking the satellite technique to Chile, where hantavirus is a much bigger problem than it is in the United States.

Meanwhile, Glass continues to develop predictive maps for the Four Corners. On his computer screen, the map for last summer shows one lone danger zone, a jagged stretch of yellow across a deep blue backdrop. Earlier, Glass had reported this finding in an e-mail to three colleagues in the Southwest.

"We told them the only place that's likely to light up this year is northeast Arizona," he says. "It was about two days later that I got an e-mail from a fourth person saying, 'Have you tracked any mice up in northeast Arizona? We're getting some cases up there.'

"That's the perverse aspect of all this," Glass continues. "At the moment, being right means somebody's getting sick, and that's a bad deal. But on the other hand, it makes you feel like maybe we're really onto something."

<< Previous Page | Next Page >>

Support JHSPH

The Johns Hopkins Bloomberg School of Public Health strives every day to keep millions of people around the world safe from injury or illness.

Invest in
Public Health >>