Download EWG's Healthy Living AppClose
Sign up to receive email updates, action alerts, healthy eating tips, promotions to support our work and more from EWG!
Sunscreen is a unique body care product: consumers are directed to apply a thick coat over large areas of the body and reapply frequently. Thus, ingredients in sunscreen should not be irritating or cause skin allergies, and should be able to withstand powerful UV radiation without losing their effectiveness or forming potentially harmful breakdown products. People can potentially inhale ingredients in sunscreen sprays and ingest some of the ingredients they apply to their lips, so ingredients must not be harmful to lungs or internal organs. Further, sunscreens commonly include ingredients that act as “penetration enhancers” and help the product adhere to skin. As a result, many sunscreen chemicals are absorbed into the body and can be measured in blood, breast milk and urine samples.
Active ingredients in sunscreens come in two forms, mineral and chemical filters. Each uses a different mechanism for protecting skin and maintaining stability in sunlight. The most common sunscreens on the market contain chemical filters. These products typically include a combination of two to six of the following active ingredients: oxybenzone, avobenzone, octisalate, octocrylene, homosalate and octinoxate. Mineral sunscreens use zinc oxide and/or titanium dioxide. A handful of products combine zinc oxide with chemical filters.
Laboratory studies indicate that some chemical UV filters may mimic hormones, and physicians report sunscreen-related skin allergies, which raises important questions about unintended human health consequences from frequent sunscreen application.
The Food and Drug Administration has not reviewed evidence of potential hazards of sunscreen filters – instead it grandfathered in ingredients used in the late 1970s when it began to consider sunscreen safety. The Danish EPA recently reviewed the safety of active ingredients in sunscreen and concluded that most ingredients lacked information to ensure their safety (Danish EPA 2015). Sixteen of the 19 ingredients studied had no information about their potential to cause cancer. And while the published studies suggest that several chemical filters interact with human sex or thyroid hormones, none of the ingredients had sufficient information to determine the potential risks to humans from hormone disruption.
EWG has reviewed the existing data about human exposure and toxicity for the nine most commonly used sunscreen chemicals. The most worrisome is oxybenzone, which was added to nearly 65 percent of the non-mineral sunscreens in EWG’s 2018 sunscreen database. Oxybenzone can cause allergic skin reactions (Rodriguez 2006). In laboratory studies it is a weak estrogen and has potent anti-androgenic effects (Krause 2012, Ghazipura 2017).
The Centers for Disease Control and Prevention routinely detects oxybenzone in more than 96 percent of the American population. Study participants who reported using sunscreen have higher oxybenzone exposures (Zamoiski 2015). Investigators at University of California, Berkeley, recently reported a dramatic drop in teen girls’ exposure to oxybenzone in cosmetics when they switched from their usual products to replacements that did not contain this chemical (Harley 2016).
In a recent evaluation of CDC-collected exposure data for American children, researchers found that adolescent boys with higher oxybenzone measurements had significantly lower total testosterone levels (Scinicariello 2016). The study did not find a similar effect in younger boys or females. The researchers cautioned that their results are a single-day snapshot instead of a controlled study of the effect of multi-day exposures.
Three other studies reported statistically significant associations between oxybenzone exposure during pregnancy and birth outcomes. One reported shorter pregnancies in women gestating male babies, two reported higher birth weights in baby boys, and one found lower birth weights in daughters (Ghazipura 2017).
Intentional dosing studies in people are rare. In one study, human volunteers applied a lotion with oxybenzone and two other sunscreen ingredients. Researchers reported a minor but statistically significant decrease in testosterone in men, accompanied by a minor increase in inhibin B, another male sex hormone (Janjua 2004). The researchers concluded these differences were normal variations and not attributed to sunscreen exposure, but critics argue that the exposures were too short to be conclusive (Krause 2012).
Given the pervasiveness of oxybenzone exposures, further study is needed to evaluate the association between oxybenzone and hormone disruption in children and adults.
EWG recommends that consumers avoid sunscreens with oxybenzone. But sunscreen users are exposed to other active ingredients as well. Margaret Schlumpf of the University of Zurich detected oxybenzone and four other sunscreen filters in Swiss women’s breast milk, indicating that the developing fetus and newborns may be exposed to these substances (Schlumpf 2008, Schlumpf 2010). She detected at least one sunscreen chemical in 85 percent of milk samples.
Some experts caution that the unintentional exposure to and toxicity of active ingredients erode the benefits of sunscreens (Krause 2012, Schlumpf 2010). But most experts conclude that more sensitive tests are needed to determine whether sunscreen chemical ingredients pose risks to frequent users (Draelos 2010, Gilbert 2013).
This table outlines human exposure and toxicity information for nine FDA-approved sunscreen chemicals. We asked these questions:
|Chemical||EWG Hazard Score||Use in U.S. sunscreens||Skin Penetration||Hormone disruption||Skin Allergy||Other concerns||References|
|UV filters with higher toxicity concerns|
|Oxybenzone||8||Widespread||Detected in nearly every American; found in mother’s milk; 1% to 9% skin penetration in lab studies||Weak estrogen, moderate anti-androgen; associated with altered birth weight in human studies||Relatively high rates of skin allergy||N/A||Janjua 2004, Janjua 2008, Sarveiya 2004, Gonzalez 2006, Rodriguez 2006, Krause 2012, Ghazipura 2017|
|Octinoxate (Octylmethoxycinnamate)||6||Widespread||Found in mothers’ milk; less than 1% skin penetration in human and laboratory studies||Hormone-like activity; reproductive system, thyroid and behavioral alterations in animal studies||Moderate rates of skin allergy||N/A||Krause 2012, Sarveiya 2004, Rodriguez, 2006, Klinubol 2008|
|UV filters with moderate toxicity concerns|
|Homosalate||4||Widespread||Found in mothers’ milk; skin penetration less than 1% in human and laboratory studies||Disrupts estrogen, androgen and progesterone||N/A||Toxic breakdown products||Krause 2012, Sarveiya 2004, SCCNFP 2006|
|Octisalate||3||Widespread; stabilizes avobenzone||Skin penetration in lab studies||N/A||Rarely reported skin allergy||N/A||Walters 1997, Shaw 2006 Singh 2007|
|Octocrylene||3||Widespread||Found in mothers’ milk; skin penetration in lab studies||N/A||Relatively high rates of skin allergy||N/A||Krause 2012, Bryden 2006, Hayden 2005|
|UV filters with lower toxicity concerns|
|Titanium Dioxide||2 (topical use), 6 (powder or spray)||Widespread||No finding of skin penetration||No evidence of hormone disruption||None||Inhalation concerns||Gamer 2006, Nohynek 2007, Wu 2009, Sadrieh 2010, Takeda 2009, Shimizu 2009, Park 2009, IARC 2006b|
|Zinc Oxide||2 (topical use), 4 (powder or spray)||Widespread; excellent UVA protection||Less than 0.01% skin penetration in human volunteers||No evidence of hormone disruption||None||Inhalation concerns||Gulson 2012, Sayes 2007, Nohynek 2007, SCCS 2012|
|Avobenzone||2||Widespread; best UVA protection of chemical filters||Very limited skin penetration||No evidence of hormone disruption||Breakdown product causes relatively high rates of skin allergy||Unstable in sunshine, must be mixed with stabilizers||Klinubol 2008, Bryden 2006, Hayden 2005, Montenegro 2008, Nash 2014|
|Mexoryl SX||2||Uncommon; pending FDA approval; offers good, stable UVA protection||Less than 0.16% skin penetration in human volunteers||No evidence of hormone disruption||Skin allergy is rare||N/A||Benech-Kieffer 2003, Fourtanier2008|
|Six other ingredients approved in the U.S. are rarely used in sunscreens: benzophenone-4, benzophenone-8, menthyl anthranilate, PABA, Padimate O, and trolamine salicylate|
Several common chemical filters appear to be endocrine disruptors. A large number of studies in animals and cells have shown that the chemicals affect reproduction and development by altering reproductive and thyroid hormones, although the evidence is mixed for some studies (Krause 2012). Animal studies report lower sperm counts and sperm abnormalities after exposure to oxybenzone and octinoxate, delayed puberty after exposure to octinoxate, and altered estrous cycling for female mice exposed to oxybenzone. Recently, Danish researchers reported that eight of 13 chemical sunscreen ingredients allowed in the U.S. affected calcium signaling of male sperm cells in laboratory tests, which the researchers suggest could reduce male fertility (Endocrine Society 2016).
As most of the hazard data is generated from animal studies, it is difficult to determine the human health implications of exposure to a mixture of hormone-disrupting ingredients in sunscreen.
In addition to the relationship between oxybenzone and testosterone levels in adolescents, preliminary investigations by a team of researchers at the National Institutes of Health and The State University of New York, Albany, suggest a link between higher concentrations of benzophenones and poorer reproductive success in men seeking assistance at a fertility clinic. Men with greater exposures to benzophenone-2 and/or 4-hydroxyoxybenzone had poorer sperm quality (Louis 2015), and reported that it took longer for their partners to conceive (Buck-Louis 2014). Female exposures to oxybenzone and related chemicals have been linked to increased risk of endometriosis (Kunisue 2012).
Mineral sunscreens are made with zinc oxide and titanium dioxide, usually in the form of nanoparticles.
There is good evidence that little if any zinc or titanium particles penetrate the skin to reach living tissues. Thus, mineral sunscreens tend to rate better than chemical sunscreens in the EWG sunscreen database. However, it is important that manufacturers use forms of minerals that are coated with inert chemicals to reduce photoactivity. If they don’t, users could suffer skin damage. To date, no such problems have been reported.
The FDA should set guidelines and place restrictions on zinc and titanium in sunscreens to minimize the risks to sunscreen users and maximize these products’ sun protection. Our detailed analysis of nanoparticles in sunscreens is available here.
The FDA must also take a close look at the so-called inactive ingredients in sunscreens. These typically make up 50 to 70 percent of a sunscreen product
One ingredient in particular is a cause for concern: the preservative methylisothiazolinone. Methylisothiazolinone is used alone or in mixtures with a related chemical preservative called methylchloroisothiazolinone. The American Contact Dermatitis Society named methylisothiazolinone its “allergen of the year” in 2013. This year, EWG has found methylisothiazolinone listed on the labels of 44 sunscreens – a decline from 94 products in 2017.
Laboratory studies indicate that methylisothiazolinone is a skin sensitizer or allergen. Over the past several years, physicians have reported serious cases of skin allergies, most notably in children exposed to methylisothiazolinone, from baby wipes and other products meant to be left on the skin (Chang 2014). In a study published in 2014, researchers at Baylor University surveyed the ingredients in 152 children’s body care products labeled “hypoallergenic” and found methylisothiazolinone in 30 of them (Schlichte 2014). Three of the sunscreens in this year’s database are marketed for use on babies.
In 2015, researchers from 15 clinics in the U.S. and Canada reported an increase in methylisothiazolinone allergies in patients. The researchers concluded that they had documented “the beginning of the epidemic of sensitivity to methyliosthiazolinones in North America” (Warshaw 2015).
That methylisothiazolinone has become relatively common in sunscreen is a matter of concern because sunscreen users are likely to be exposed to significant concentrations of this chemical. The products that contain it are intended to be applied to large portions of the body and to be reapplied often.
In March 2015, the European Scientific Committee on Consumer Safety concluded that no concentration of the chemical could be considered safe in leave-on cosmetic products (EU SCCS 2014).
But methylisothiazolinone is still allowed in U.S. products. In 2014 the Cosmetics Ingredient Review expert panel – an independent, cosmetics-industry-funded body the American cosmetics industry pays to advise it on the safety of cosmetics ingredients – told the industry that methylisothiazolinone was safe for use in body care products as long as manufacturers come up with formulations that wouldn’t cause allergic reactions (CIR 2014). Since the FDA has little legal power to regulate cosmetics ingredient safety, it has authorized the cosmetics industry to police itself through this CIR panel. The body’s recommendations are not legally binding on any company. In several decades, it has declared only 12 ingredients or chemical groups to be unsafe (CIR 2017).
EWG recommends that the FDA launch a more thorough investigation of the safety of all ingredients currently in sunscreens to ensure that none of them damage skin or cause other toxic effects in consumers.
Take EWG's Sunscreen Guide with you! Download the Healthy Living App Today
Take EWG's Sunscreen Guide with you! Download the Healthy Living App Today