McFarland Cases : Child Cancer Cluster Poses Puzzle

McFARLAND, Calif.--Ask a few scientists when they will know precisely why at least 13 children have come down with cancer in this tiny farming town in the San Joaquin Valley, and sooner or later most will say the answer may well be never.

Few cancer clusters have ever been explained, they say. And this one is especially puzzling. Some suggest McFarland may even be a fluke--the extremely rare case, like flipping a coin and having it consistently come up tails.

But put the same question to the politicians, union organizers and celebrities who have flocked to McFarland in recent months. They will say bluntly what others suspect but cannot prove: Agricultural chemicals, especially pesticides, are somehow to blame.

Contaminated Water

Perhaps chemicals contaminated the water in the past, they say. Maybe babies played in yards inadvertently dusted with pesticides. If scientists have found no evidence to support those theories, some residents and activists say, they should look a little harder.

Now the certainties of politicians and the uncertainties of science have torn this little town into angry pieces--factions adhering to one absolute or another, hungering for the kind of clear answer that the state’s own scientists increasingly suspect may never emerge.

McFarland, a place that calls itself “the heartbeat of agriculture,” has become the unhappy symbol of public fears about pesticide use--a challenge to public health officials that some say is not only scientific but also peculiarly personal.

“The way science works, you eliminate this, you eliminate this, you eliminate that,” said Dr. Raymond Neutra of the state Department of Health Services. “And scientists always feel very good when we are able to say, ‘Well, we really have shown you it’s not (this).’ ”

‘I’ve Eliminated the Butler’

“But then people say, ‘But you haven’t told me what it is !’ ” said Neutra, who until recently headed the McFarland investigation. “It’s like a murder case: ‘I’ve eliminated the butler. Don’t you feel good?’ But (the next question is), ‘Well, who did it?’ ”

McFarland, population circa 6,200, straddles California 99, the long zipper that runs the length of the Central Valley, passing places with names like Weedpatch and Pumpkin Center and aging billboards for U-Pick pistachios.

The town, 25 miles north of Bakersfield, is little more than a dozen blocks wide or long, afloat on an ocean of cotton, almonds, kiwi and grapes. It has one dentist, three doctors and daily stops by the Greyhound bus. The town is mostly Latino; agriculture is the main employer.

Word of the cancers first got around McFarland in early 1984. Connie Rosales had begun counting cases. In addition to her teen-aged son, Randy, some nine other children had been diagnosed with cancer within a few years. That was three to four times the expected rate.

Since then, more cases have been diagnosed. Some parents put the total at 17, with eight children dead. The state, which counts only those children diagnosed while living within the boundaries of the town, puts the total at 13, with six children dead.

They range from toddlers to football players; five in one neighborhood, the rest scattered through town. In addition to the cancers, officials say the rates of infant deaths, fetal deaths and low birth weight babies doubled and tripled in 1981-1983 over previous years.

“I don’t know whether you’ve ever seen a child die before, but we went through this eight times,” Rosales said one morning, in her house in the subdivision that has been hardest hit. “Chemotherapy in an adult is bad enough. It absolutely tortures children.”

In early 1984, Rosales notified the Kern County Health Department. The following year, she and other parents wrote to the Legislature. Public hearings were held in McFarland in 1985 and reporters began covering the story. McFarland became a national issue.

Now Jesse Jackson and Cesar Chavez march in McFarland. The United Farm Workers president spent 36 days fasting this summer, drawing attention to pesticide risks. Movie stars, Canadian labor leaders and various Kennedys turned out to support Chavez and call attention to hazards some of them say are symbolized by McFarland.

County and state investigators have tested the town’s water, air and soil. They’ve measured radio frequency waves from the Voice of America transmitter north of town. They’ve examined pipes, household chemicals, auto emissions, radiation, insulation, ventilation.

They say there is no problem with the water (though they cannot say whether there was in the past). They say the air, the soil, the houses, the pipes appear to present no obvious threat. If something specific causes the cancers, they say, they have not found it.

‘Science Just Way Behind’

“I think the science of epidemiology needs a real kick in the rear end,” complained Rosales. “ . . . We’re getting into problems all over the country and it’s obvious there are environmental causations. The science itself is just way behind.”

Indeed, the history of cancer cluster investigations is less than illustrious. Few have ended in pinpointing a specific cause. The most successful involve clusters in which the exposure was direct and measurable, and the cancers were of a single type.

For example, a cluster of rare liver cancers in Kentucky was traced to workers’ exposure to vinyl chloride while working in a tire plant. A cluster of vaginal cancers in Boston was traced to the anti-miscarriage drug, DES, used by the victims’ mothers during pregnancy.

But investigations of environmental clusters are even more difficult. One of the few solved was a leukemia cluster in Woburn, Mass. Even there, scientists say new cases occurred after the town’s contaminated drinking wells were closed, raising some doubt about the finding.

Most of the dozens of apparent clusters reported every year in California alone turn out, upon close scrutiny, not to be clusters. And even some statistically significant clusters may occur simply by chance, according to the laws of probability.

What are the odds of a cluster like McFarland’s occuring? Official estimates have ranged from 1 in 100 to 1 in 1,000. Some scientists say such a cluster could even conceivably occur in several dozen similarly sized towns or census tracts in California.

“Given what we know about how often childhood cancer occurs, some places in the U.S. are going to show increases over a period of time, at random, by chance,” said Dr. Matthew Zak, an epidemiologist with the federal Centers for Disease Control and a scientific adviser on the McFarland case. “It’s very possible that in McFarland that’s what happened.”

The McFarland case presents its own peculiar difficulties.

Childhood cancer is in many ways a mystery, even more so than adult cancer, which is also poorly understood. Without knowing what causes it, scientists say, it is difficult to know where in McFarland to look for explanations.

The variety of cancers in McFarland is also puzzling--liver, lymph nodes, bone, eye, adrenal, kidney. Could one cause lead to so many varieties? Some of the state’s scientific advisers say they doubt that so many types could have come from a single agent.

Agricultural chemicals, too, are not well understood. More is known about their effects on animals than their effects on people. And state records on what pesticides have been used, and where and when, turn out to be incomplete and sometimes even illegible.

Finally, although the McFarland cancer cases are clearly excessive, scientists say there are fewer there than one would use in a solid, scientific study that looks for causes by comparing the lives of cancer patients to those of a control group of healthy residents.

“I want to emphasize, the state’s inability to identify a single cause is not a result of bad science,” said Robert W. Haile, an epidemiologist with the UCLA Cancer Center and a member of the state’s advisory board on McFarland.

“There are probably multiple steps that lead to cancer and multiple factors that act together,” Haile said. “So it makes it very difficult to tease out exactly what those factors are when you don’t have large numbers.”

Haile, like many others, believes pesticides must be closely scrutinized.

“My opinion is there is enough experimental and anecdotal evidence suggesting that some pesticides may have carcinogenic effects that I think it warrants careful study,” he said. “ . . . We need money, much more money, to be put into the study of pesticides.”

People in McFarland share his suspicions.

“If pesticides kill bugs, then they probably have some effect on us,” mused Ronald G. Huebert, the amiable, silver-haired superintendent of the McFarland Unified School District. “Because we’re kind of a big bug.”

The McFarland cluster fell initially to the Kern County Health Department, an agency more familiar with outbreaks of salmonella and sexually transmitted diseases than cancer. Officials admit frankly that the case far exceeded the capacities of a county health department.

The agency had no epidemiologist at the time. There was one nurse who had taken a short course in the subject. The department had no experience in chronic disease epidemiology. The water quality program had no toxicologist on staff. Nevertheless, the department set about in 1984 investigating the cluster, beginning with a census of childhood cancer cases. Though there were informal reports of some 23, the study focused on the 10 diagnosed between 1975 and 1985 and living in McFarland at the time of diagnosis.

The county collected water samples from homes, wells and control sites. It tested them for approximately 100 chemicals. Officials scraped soil samples from yards, parks, playgrounds and water runoff sumps and tested them, according to the county, for some 80 chemicals.

Sanitarians also inspected the houses of the 10 cases and four control homes for asbestos, formaldehyde, household chemicals and other sources of indoor pollution. County air pollution officials later measured ambient levels of carbon monoxide from the highway traffic.

They found nothing, county officials say, that could account for the cancers.

Next, the state Department of Health Services did a case-control study, comparing the cancer victims to a healthy control group. Parents were questioned at length about their children’s medical histories, habits and hobbies, as well as their own lives and jobs.

The state also looked into pesticide use around McFarland, relying on the reports applicators and farmers must file on certain pesticides. They concentrated on 1980 and 1981--the period when, they figure, something in the environment could have caused the cluster.

Initially, four pesticides caught their attention: They had been used heavily during the potentially critical 1980-81 window. One member of the scientific advisory panel strongly suspects those chemicals; but Neutra said further study of their effects suggests they probably could not be the cause.

A second striking finding was that 80% of the fathers of children with cancer, compared to 45% of the fathers of healthy controls, had worked in the fields during the period between shortly before their child’s conception and his or her diagnosis.

To pursue that lead, health officials intend to re-interview the fathers. Sanitarians will tour their workplaces, looking for clues as to how they or their children might have been exposed to chemicals at work that might have been transported home.

The state is also expanding the study: It is collecting data on childhood cancers in all four counties of the southern San Joaquin Valley. If the McFarland cluster is too small to pinpoint a precise cause, officials hope they might find it by looking further afield.

A finding of a high childhood cancer rate regionwide might suggest a problem endemic to farming communities, perhaps related to chemicals, investigators say. Additional cases would then be added to the study group for a larger, maybe more revealing, case-control study.

A finding of no excess cancers in the four-county area, on the other hand, could be something of a dead end.

“I’m not sure what one would have to do next, if the problem appears to be unique to McFarland,” said Haile of UCLA. “Launch even more exhaustive studies in McFarland in hopes of finding something, knowing full well that the odds are against you?”

Meanwhile, the protracted uncertainty has shattered the town. Connie Rosales, among others, believes the county and state, at least initially, dragged their feet. Rosales said she is “not conspiracy-minded.” But she does not underestimate the power of agriculture, property values and status quo in the public debate.

Though her son survived, she said she has been left with little. She said she is broke, her marriage has ended and she is unable to get work--a fact she traces in part to her bitter feud with the farm workers’ union over what she charges is its exploitation of the cancer issue.

The union, she alleges, has used McFarland’s misery to raise national support for its boycott of non-union grapes--a cause she does not support. At the same time, she said the union has offered little financial or medical help to the children and their families.

Some of the parents have left McFarland. Others say they would go but they cannot afford to make the move. Some have felt ostracized by the community, haunted by the loss of their children, obsessed with preserving the health of those who remain.

“Our goal is to never have this happen again,” said Rosales, who with other parents is suing various agencies, officials and chemical manufacturers.

Active in Community

On the other side of town, Arturo Munoz lays considerable blame with the media. They are killing the town, he bitterly complains. They have singled out McFarland, branding it with a problem that, if it exists, exists throughout the San Joaquin Valley, he said.

Munoz, 66, came to McFarland 45 years ago as one of the first Mexican residents. He has raised seven children, served on the City Council, been active in his church. The reports about the cancer are a blight on his high hopes for McFarland’s future.

He wants the press and politicians to leave health officials alone and let them explain the cancers. If there is a problem, Munoz is confident that government and science can fix it. Most of all, Munoz wants the issue resolved and his town left in peace.

What irks Ronald Huebert, the schools superintendent, is the steady ebb and flow of politicians, union organizers and others--people who Huebert says announce their presence with a press release. He calls them “human tornadoes”: They blow in, do their damage and leave.

Their charges have bred distrust towards officials, Huebert says. They have encouraged conspiracy theories and blame. So in the absence of answers from science, Huebert is turning to faith--a community-wide prayer meeting next weekend called “Hands Around McFarland.”

Was he saying McFarland’s problems will be solved by faith, not science?

“Our Creator creates the scientists,” Huebert said. “So I would say a little mutual work there.”

Meanwhile, the state is continuing its search.

“I think there is no doubt that a lot of people in that community are under a lot of stress,” said Dr. Lynn Goldman, who recently took over the case from Neutra. “I feel it’s incumbent on us to at least be able to establish for them that it’s OK for them to continue living there, that they can raise their children there.”

Both Goldman and Neutra say that as time passes and they turn up no clear cause, it looks increasingly as though the McFarland cluster is a chance occurrence. As Goldman put it, “More likely than not, what we’re looking at is a kind of random aggregation of childhood cancers.”

And chance, some point out, can be an unsettling answer.

“It’s sort of a last-resort explanation,” said Zak, the federal epidemiologist. “It really doesn’t explain anything. All it says is it’s possible it could occur; it doesn’t explain why it occurred in McFarland. Maybe, in fact, we’ll never know.”

COMPARING McFARLAND’S CANCER RATES These are figures to compare childhood cancer rates in McFarland with those of other Kern County towns and with selected regions in the United States. Rates are computed per 100,000 children per year. In the Kern County figures, childhood is defined as infancy. In the national figures, it is defined as infancy to 19. McFarland Est. Pop. Aged 0 to 20: 2,475.0 Confirmed Cases 1975-85: 10.0 Rate/100,000 per year: 36.7 Delano Est. Pop. Aged 0 to 20: 6,615.0 Confirmed Cases 1975-85: 6.0 Rate/100,000 per year: 8.2 Kern County Est. rate: 11.5 to 14.7 Source: Kern County Health Dept. HOW CANCER RATES COMPARE ELSEWHERE There is no accurate figure for the national childhood cancer rate. An estimated national rate of 14 per 100,000 children per year is based on averaging known regional rates. San Francisco Rate/100,000 per year in children aged 0-19 in 1978-81: 16.3 Source: the San Francisco metropolitan tumor registry New Mexico Rate/100,000 per year in children aged 0-19 in 1978-81: 11.6 Source: the New Mexico state tumor registry Utah Rate/100,000 per year in children aged 0-19 in 1978-81: 14.8 Source: the Utah state tumor registry