The data Friedman refers to come from a 2008 survey by the Centers for Disease Control and Prevention that examined urine samples from more than 2,500 people selected as a representative sample of the U.S. population. Researchers detected oxybenzone (sometimes called benzophenone-3) — a key ingredient in many chemical sunscreens — in 97 percent of the samples. The findings suggest that almost all Americans have absorbed oxybenzone into their bodies, but it doesn't clarify where the compound came from (oxybenzone is also used in cosmetics) or how it may affect health.
Scientists have seen hints of what oxybenzone — and two other common sunscreen ingredients, octocrylene and octyl methoxycinnamate — can do in other studies, and the results have raised some concern. In a small 2006 study that tested these UV filters on a skin model made from human cells, chemist Kerry Hanson of the University of California at Riverside found that the filters break down in UV light, losing their ability to protect the skin and ultimately generating more free radicals — molecules that can "steal" electrons from cells, damaging them in the process — in the skin than if there were no sunscreen applied at all. Free radical formation is one of the ways that a sunburn damages the skin, and Hanson found that degraded UV filters can amplify the effect.
"Your skin naturally makes free radicals, but it also has natural antioxidants [molecules that protect the body from free radicals] that balance out the free radical load," says Hanson, who has consulted with sunscreen companies interested in improving their products. "But the three UV filters we tested show more free radical generation than would naturally occur."
A Swiss study published in 2001 found that oxybenzone and octyl methoxycinnamate could also behave like estrogens, causing changes in uterine weight in rats that were fed very high doses — much higher than a person would absorb using sunscreen every day. But a 2004 study failed to find detectable hormone changes in people using sunscreen, so its hormone-disrupting potential remains unclear.