A recent study claims to confirm what a lot of people felt they already knew: eating organic can reduce your risk of getting cancer. But a closer look at the details reveals a different story.
Asking how frequently someone chooses organic instead of conventional also doesn’t really tell you anything about how much someone is consuming. Some of the categories in the organic food score are things that you might consume occasionally or in very small quantities, such as condiments, herbal tea, or vitamin supplements. Other categories are daily staples like bread and milk. How can these contribute equally to your organic food score?
If I drink only organic wine and take only organic vitamins, but never eat organic fruits and vegetables, I’d get the same score as someone who ate only organic fruits and vegetables but settled for conventional marmalade and coffee.
The authors wanted to test their hypothesis that reduced exposure to pesticides would be associated with a lower cancer risk. But, in my opinion, the Organic Food Score that they developed is an extremely poor proxy for pesticide exposure.
But perhaps people who buy organic foods more frequently (or at least say that they do) have a lower risk of cancer for reasons that have nothing to do with pesticides. High scores on the Organic Food Score might reflect more disposable income, or greater health consciousness, or beliefs or fears about food and health—all of which may in turn be associated with other healthy habits.
And, in fact, the researchers found that “a higher frequency of organic food consumption was associated with a reduced risk of cancer.” Those with the highest organic food scores had a 25% reduction in cancer risk compared those with the lowest. That was the headline.
But if you look more closely, you’ll find there are some really weird things about this data. If organic food consumption (or pesticide exposure) were associated with cancer risk, you’d expect to see some sort of dose effect: the more organic foods you eat, the lower the risk. And for certain kinds of cancers, this was the case.
For example, with skin cancer, the risk went down as the organic food score went up. But for other types of cancer, including colorectal cancer, prostate cancer, and premenopausal breast cancer, those with organic food scores in the middle of the range had a higher risk than those with the lowest scores.
Here's a graph that might make it easier to see what I’m talking about.
Perhaps the strangest case is for non-Hodgkins lymphoma. Those in the highest category, with average organic food score of 19, had an 86% reduction in risk compared to those in the lowest category, who had an average score of less than 1. But those in the second highest category, whose scores averaged 10, had a 20% increase in risk compared to those with scores of less than 1. To me, that’s a red flag for some funky data.