I live in Marin County, which has a very high incidence of breast cancer. A few years ago, Marin launched a very expensive investigation, asking each Marin woman to fill out a form documenting how often she eats hot dogs and bacon, and whether she lives near power lines. I was unimpressed, although I dutifully entered the information. One doesn’t need to spend tens of thousands of dollars to figure out why Marin has a high incidence of breast cancer. The answers are obvious, and have to do with social and economic issues.
The social issues are that women in Marin delay childbearing (sometimes forever). Aside from the fact that childbearing and nursing seem to have a positive impact on avoiding breast cancer, the same statistics about delayed childbearing also mean that many women have spent years on oral contraceptives, which may have a negative impact on avoiding breast cancer. There are also a lot of Jewish women in Marin (not religiously affiliated, perhaps, but genetically affiliated), and we know that Ashkenazi Jews have a higher incidence of breast cancer.
Probably the biggest factor in Marin’s breast cancer numbers, though, is that this is a wealthy community, with people eating well, exercising often, and getting good medical care. They avoid heart disease, diabetes and other common youthful killers, and live into a nice old age — and old women get breast cancer. The older you are, the more likely you are to get it, so a population with a lot of older ladies (not as many as Miami, admittedly, but still a lot) is going to pop up with a lot of cancer patients. Not only that, but the confluence of affluence and good medical care means that these cases are going to be diagnosed.
Diagnosis matters a lot when it comes to statistical anomalies. I can’t track the story down, but unless I was hallucinating, I swear that several years ago, when I was still listening to NPR, I heard a story about clusters of brain tumors amongst children in a small Canadian town. The usual panic ensued, with expensive testing of water supplies, soil, power lines, food, etc. It finally turned out that the problem was caused by a new imaging machine at the local hospital. In the old days, when children banged their heads (something children do a lot), the doctors diagnosed concussions by observation. In the modern era, doctors stuck the kids into the MRI machine. It was as a byproduct of these concussion tests that the brain tumors starting appearing. In other words, the brain tumors were always there (apparently a lot of us have innocuous, anomalous growths in our brains), but no one had ever seen them before.
I thought of these two stories when I read today in the New York Times that a simple food allergy test is resulting in over-diagnosis of food allergies, with the resulting inconvenience and, in extreme cases, malnutrition:
Doctors say that misdiagnosed food allergies appear to be on the rise, and countless families are needlessly avoiding certain foods and spending hundreds of dollars on costly nonallergenic supplements. In extreme cases, misdiagnosed allergies have put children at risk for malnutrition.
And avoiding food in the mistaken fear of allergy may be making the overall problem worse — by making children more sensitive to certain foods when they finally do eat them.
More than 11 million Americans, including 3 million children, are estimated to have food allergies, most commonly to milk, eggs, peanuts and soy. The prevalence among children has risen 18 percent in the past decade, according to the Centers for Disease Control and Prevention. While the increase appears to be real, so does the increase in misdiagnosis.
The culprit appears to be the widespread use of simple blood tests for antibodies that could signal a reaction to food. The tests have emerged as a quick, convenient alternative to uncomfortable skin testing and time-consuming “food challenge” tests, which measure a child’s reaction to eating certain foods under a doctor’s supervision.
The problem in each of the examples above isn’t the absence of data, it’s too much data, or the misinterpretation (for reasons of ignorance or politics) of existing data. It’s a cheap and easy shot to add here that the same phenomenon seems to be a part of the hysteria about global warming, but I’m going to do it anyway.
Just as the MRI in Canada, by revealing hitherto hidden, but perfectly normal information, caused a brain tumor panic, so too did the recent ability to take the earth’s temperature cause a panic. And just as people were loath to admit that lifestyle and money could affect breast cancer, and instead tried to blame paranoid bugaboos such as water and power lines, so too have greenies been loath to admit that the earth has a cycle, and they are instead trying to blame American capitalism. And just as we’re discovering that the wrong tests yield the wrong data about food allergies, we’re increasingly learning that the warmies have been messing with or misreading the data about the climate.
We all know the line that a little knowledge can be a dangerous thing. It turns out that in our modern, politically driven, somewhat paranoid era, a lot of knowledge, misinterpreted or messed with, can be a damn dangerous thing too.
UPDATE: I can’t resist adding this little story, about hitherto unknown animal species coming to light in the rainforest. We bemoan the decrease in population of animals that we know exist, but we still have the hubris to think we’re aware of all animals. In the cycle of life, species come and species go, and we don’t know everything about all of them.