By George Johnson
Lay a chessboard on a table. Then grab a handful of rice and let the grains fall and scatter where they may. They won't spread out uniformly with the same number occupying each square. Instead there will be clusters. Now suppose that the chessboard is a map of the United States and the grains are cases of cancer.
Each year about 1.6 million cases of cancer are diagnosed in the United States, and epidemiologists regularly hear from people worried that their town has been plagued with an unusually large visitation. Time after time, the clusters have turned out to be statistical illusions — artifacts of chance.
The Erin Brockovich incident, one of the most famous, is among the many that have been debunked. Hexavalent chromium in the water supply of a small California town was blamed for causing cancer, resulting in a $333 million legal settlement and a movie starring Julia Roberts. But an epidemiological study ultimately showed that the cancer rate was no greater than that of the general population. The rate was actually slightly less.
Of the handful of residential clusters that have not been dismissed as flukes, only two in the United States have been associated, with a great deal of uncertainty, with environmental contaminants. Both involved childhood cancer. One was found in the 1980s in Woburn, Mass. The other was found about a decade later in Toms River, N.J., and is the subject of an absorbing new book by Dan Fagin, a former reporter for Newsday and the director of the science, health and environmental reporting program at New York University. I first read it in manuscript about a year ago, and I've been puzzling over it ever since.
We're not talking about thousands of cancer cases unleashed in a town by industrial poisons. Or hundreds. Cancer clusters occur on a far smaller scale. An early study of Toms River in 1995, as suspicions were beginning to grow, found a total of 56 childhood cancers in the township (population 76,000) over a period of 13 years. Based on figures for the whole state, 43 of those cases might have occurred anyway. They were part of the normal background rate, cancers that may happen for no apparent reason.