* * *
State health departments collectively receive thousands of cancer cluster reports from citizens each year, but of more than 400 investigations since 1990, only three—including Woburn’s—found a link between an environmental contaminant and a cluster. Epidemiologists have long doubted the value of such investigations. In a seminal speech to the 1989 National Conference on Clustering of Health Events, Kenneth Rothman, founder of the journal Epidemiology, announced that, “With very few exceptions, there is little scientific or public health purpose to investigate individual disease clusters at all.” Since scientists don’t know what causes most cancers, linking a cluster of cases to the same origin often involves guesswork. (Linking two different kinds, such as leukemia and brain cancer, is even more challenging.) As a result, cluster investigations are thought to be fraught with error. Some scientists even dismiss clusters as matters of happenstance, or, as Atul Gawande wrote for The New Yorker in 1999, in “The Cancer-Cluster Myth,” “the result of almost irresistible errors in perception”—of the human tendency to see coincidence as cause for alarm.
When alerted to a suspicious rise in cancer cases, epidemiologists must first determine if the numbers are indeed unusually high, or “statistically significant.” If they are, investigators then catalog the things that distinguish the sick from the healthy and ask if any of those variables could have caused the disease. These comparisons require more statistical tests, and to achieve a significant result, a large sample size. Often, clusters are so small that studies result in false negatives, finding no significant link between a toxic substance and disease even when one could exist.
“It’s a huge source of frustration for everyone, scientists and communities alike,” said Rachel Morello-Frosch, a professor of public health and environmental science at the University of California-Berkeley. “By the time you see a relationship between environmental exposure and a health concern”—when enough cases have piled up to draw the link with any certainty—“you’re basically doing a body count.”
Fallon’s cluster was an exception. A nurse reported it to state health officials in July 2000 after only five cases had emerged. At that point, officials decided it was worth investigating. They drew their study plot along county lines, including a population more than twice the size of Fallon’s. Then more cases appeared, all of them in Fallon or close by. After 11 diagnoses, officials determined there was a one-in-232 million chance that the cluster was coincidental. Odder still was how quickly it arose, with diagnoses only weeks apart, as though the children had caught an infectious disease. The cluster looked like something that could be stopped. It also looked, one researcher told me, like “the perfect opportunity to determine the environmental cause of pediatric leukemia.”
Other scientists had more modest expectations. Despite its significance, the cluster was small. Still, most agree that there had never been a case quite like Fallon’s: An outbreak in an isolated patch of desert, its origins confined to a certain space and span of time, and in a community willing, at least at first, to undergo scientific scrutiny.
Fallon’s potential as a test case prompted the Nevada State Health Division to dedicate its epidemiologist, Randall Todd, to it almost full-time. For guidance, the agency assembled a panel of experts, including Thomas Sinks, associate director for science at the National Center for Environmental Health, and Malcolm Smith, head of pediatrics for the National Cancer Institute. Early in 2001, when it was clear the state could not tackle the problem alone, the Centers for Disease Control and Prevention agreed to investigate—the first time it had probed a cancer cluster in 20 years. The CDC was encouraged, in part, by new technology. Prior investigations relied heavily on interviews to find common exposures. Now, investigators could detect trace contaminants in blood, urine, and fat at levels lower than before. They planned to gather samples from leukemia patients, their families, and healthy residents and compare them to each other and to national averages. The Nevada Department of Environmental Protection and the Agency for Toxic Substances and Disease Registry, meanwhile, would search for sources of toxic exposure.
National politicians took interest, including senators Clinton and Harry Reid, a Nevada Democrat. At a hearing in Fallon before the investigation began, Clinton ventured that “advances in determining how to prevent cancer could really be attributed to the extraordinary response in this community.” For that, she said, “I think the entire country and maybe even the world will be grateful to Fallon.”
These were lofty words, aimed less at parents, whose desperation had little to do with scientific advancement, than at residents who might have seen the scrutiny as too great a sacrifice. Still, they reflected a real and collective hopefulness at the time that answers would be found—that the investigation would yield something more than an enduring sense of uncertainty.
* * *
From downtown Fallon, it is not far in any direction before alfalfa turns to sand and rabbitbrush. The desert is everywhere. In the spring, ditch water takes on the color of the earth. The water comes from the Carson River, dammed in 1915. Before that, Fallon was a string of ranches tied to the river’s scanty offering. The town has never counted on water, but it can always count on wind and sand. On windy days, residents say, “Real estate is really moving.” One told me that Fallon is a “sharing” community: “You get a little of your neighbor’s property. They get a little of yours.”