TORONTO - Researchers suggest there may be a link between estrogen from oral contraceptives that has found its way into the environment and rising rates of prostate cancer among men around the world.

In a study in the online publication BMJ Open, researchers at Princess Margaret Hospital in Toronto looked at the percentage of women using the pill, intrauterine devices, condoms and vaginal barrier contraceptives in 87 countries, then examined the incidence and deaths from prostate cancer.

"Looking at these percentages, we find a strong correlation between female use of oral contraceptives at a population level and both new cases of prostate cancer and mortality from prostate cancer," said lead author Dr. David Margel, a urologist and fellow in uro-oncology.

"This was not found among other contraceptive modes," he said. "We also checked the percentage use of intrauterine devices or condoms or vaginal barriers and the same relation was not found."

The research team used data from the International Agency for Research on Cancer and the UN World Contraceptive Use report to determine rates of prostate cancer and associated deaths as well as the proportion of women using common methods of contraception in 2007.

Margel said estrogen in birth-control pills is excreted in the urine and gets into the environment, particularly into water, and scientific evidence suggests that low levels may cause cancer, including prostate cancer.

"What we found was that in countries where the oral contraceptive was used more often, prostate cancer had a greater incidence," said Margel. But he stressed there may be many factors involved, and teasing out the effect of pill-based estrogen alone would take much more research.

"This study does not establish cause and effect ... This is a very, very preliminary finding and we're not telling everybody to quit the pill. But further research needs to be done and it's an interesting finding."

While the amount of estrogen excreted by any single individual is extremely small, "when millions of women are doing it and for a long period of time, it may cause low environmental estrogen levels," Margel explained.

"We think further research is needed to explore both oral contraceptives, but also other estrogenic compounds that may contaminate our environment and may cause and increase the incidence and mortality from prostate cancer."

Estrogen and estrogen-like chemicals are found in all manner of commercial and cosmetic products, among them pesticides. Studies have shown that male farmers exposed to pesticides that contain high doses have a higher risk of prostate cancer compared with the general population.

Fe de Leon, a researcher at the Canadian Environmental Law Association, called the study relevant because it's known that compounds known as endocrine disruptors are increasingly present in low doses in the water supply.

Endocrine disruptors are chemicals that can interfere with the body’s hormonal system and produce adverse reproductive, neurological and immune-system effects. Among these compounds are polychlorinated biphenyls, or PCBs; DDT and other pesticides; and plasticizers such as bisphenol A, or BPA. They are found in such everyday products as plastic bottles, metal food cans, detergents, flame retardants, toys and cosmetics.

"We're now seeing these types of chemicals being detected in our waters," she said. "If you look at the context of the Great Lakes ... pharmaceuticals, including the pill and other chemicals like everyday Aspirins that people use, are being detected in low levels.

"What does this mean? We don't know in terms of the long term ... And if these chemicals are ending up in our water systems, one of the things that we could say is we can't count on our waste-water treatment plants to take those chemicals out."

De Leon said society can't ignore the fact that estrogen and estrogen-mimicking substances may have a significant impact during an individual's development and could eventually lead to breast and prostate cancers.

"But it's very hard to make that distinction. It's hard to pinpoint which chemical's responsible for a particular health endpoint," she said. "It certainly warrants further investigation."

The editors of BMJ Open also added their own cautions, writing that the research "is an ecological study and thus has significant limitations with respect to causal inference. It must be considered hypothesis-generating and thought-provoking."

In future studies, Margel and colleagues want to test drinking water for levels of estrogen and to look for estrogenic compounds in both malignant and non-cancerous prostate tissue.