`Cancer Clusters' Often Unfounded, Researchers Say
CHICAGO - In early 1961, four young children living near St. John Brebeuf Church and elementary school in Niles, Ill., died of leukemia. Alarmed by the deaths, parishioners soon discovered four more children in the neighborhood suffering from the disease.
Fears that something in the school or the neighborhood was making the children ill prompted worried residents to ask federal officials to begin exhaustive medical detective work to find the culprit.
No common cause ever was found. Researchers attributed the high incidence of leukemia among St. John Brebeuf children to chance.
Now, 33 years later, Plum Grove Junior High School in Rolling Meadows, Ill., faces a similar cancer mystery. Teachers at the school in Chicago's northwest suburbs say as many as 20 faculty members have developed cancer since the early '80s and seven have died as a result.
As with the Niles experience, finding a link between those cancer cases may prove elusive. Experts say it is very difficult to show that a "cancer cluster" has a direct, common and identifiable cause.
The Plum Grove inquiry has just begun, and preliminary results won't come until later this month. Although the possibility exists that a common cause might be found, the history of research in such cases argues against it.
Each year, hundreds of communities nationwide seek the help of health officials because they believe they live in the midst of a cancer cluster. And the evidence, notably in the form of seemingly abnormal rates of cancer, is compelling.
But most often, experts say, suspected cancer clusters usually are statistical anomalies.
In fact, the only cancer clusters where a common cause has been clearly demonstrated have been in occupational settings, according to the American Cancer Society.
A good illustration of statistical anomalies can be seen in the tossing of a coin.
If a quarter is flipped in 100 sequences of 10 flips each, most sequences will show "heads" or "tails" occurring about equally.
But according to statistical probability, it is likely that about five of those 100 sequences will show an unusually high or low number of heads or tails.
Likewise, a community might have the bad luck of having far more cases of a particular kind of cancer than national averages would predict.
"Things do, in fact, occur by chance," said Faith Davis, an epidemiologist in the School of Public Health at the University of Illinois at Chicago. "You can see a chance clustering of a rare disease without it having anything to do with the environment they are in."
The Niles case was one of the first cancer clusters to be studied intensively by health authorities.
In 1961, the leukemia rate among children in Niles was 21.3 cases per 100,000. Statewide in Illinois that year the rate was only 4.6 cases per 100,000.
"A good deal of this started in the Niles leukemia study, when I was first breaking into epidemiology," said Clark Heath, who had just gone to work for the national Centers for Disease Control when the agency sent him to Niles. Now he is vice president of the American Cancer Society in charge of epidemiology and surveillance.
Heath teamed up with Robert Hasterlik of the University of Chicago Medical School, who put together a team of state and local health officials and specialists.
The doctors reviewed medical case histories of the sick children, their families, neighbors and schoolmates. They oversaw every environmental test known at the time, testing water sources and looking for possible sources of radiation.
In the end, the investigation cost a lot of money and manpower over two years, and came up with precious little.
But the investigation did provide some reassurance to parents that nothing in the environment in the neighborhood or school was causing leukemia.
"The terror was that parents wondered if they were sending their kids to a school, unwittingly exposing them to an unknown cancer agent in the cafeteria food, water, lights or whatever," said Jim Close. Now the head of Mercy Boys and Girls Homes of Chicago, Close was assigned to St. John Brebeuf when Heath released his report in 1963.
Plum Grove Junior High School was built in 1961.
Last year, after the school underwent a $5 million renovation that included removal of asbestos insulation, several students and staff members began complaining about respiratory ailments. Those reports led to discussions of other health concerns that eventually centered on the number of teachers in the school who had developed cancer.
Kathy Merkelz, the teachers union representative, said 20 teachers at the school had contracted cancer in some form in the last 12 to 14 years. Seven of them have died from their illnesses, she said.
District officials have different numbers. The district says there have been at least nine cases of cancer with five deaths among the approximately 125 teachers, administrators and support staff who have worked at the school since 1982.
According to the union's accounting, about half the 20 cases were breast cancer, and the rest included leukemia and stomach, skin and brain cancer, Merkelz said.
That there is a variety of cancer types casts doubt on the likelihood that the illnesses stem from the same source, namely, something in Plum Grove Junior High School, medical experts say.