Only 20% of Sunscreens Pass EWG's 2026 Safety Audit
The Environmental Working Group's 20th annual sunscreen guide flagged 550 of 2,784 products as both safe and effective, with mineral filters dominating the recommended list and bemotrizinol named the first meaningful chemistry shift in two decades.
Key Takeaways
- EWG analyzed 2,784 sunscreens for its 2026 guide; 550 (about 20%) met both safety and efficacy criteria.
- 497 of the 550 recommended products are mineral-based (zinc oxide, titanium dioxide); only 53 are chemical-based with limited concerning ingredients.
- Oxybenzone use in US sunscreens dropped from 70% to 5% over 19 years; retinyl palmitate dropped from 40% in 2010 to 3% today.
- FDA's 2019 proposed rule identified 12 of 16 chemical UV filters as lacking the safety data needed to be classified GRASE, with no final order issued.
The Environmental Working Group released its 20th annual Guide to Sunscreens on May 19, 2026, finding that only 550 of 2,784 products analyzed, roughly 20%, deliver both safe and effective UV protection. Mineral filters dominate the recommended list. Of 550 cleared products, 497 are predominantly zinc oxide or titanium dioxide; only 53 chemical formulations made the cut.
The guide arrives during a regulatory standoff. In 2019, the FDA issued a proposed rule asking manufacturers to submit safety data on 12 of the 16 active chemical UV filters then in commercial use. Seven years later, no final order has been published, and those 12 filters remain on store shelves under the agency's pre-2019 framework.
## The Twenty-Year Shift Toward Mineral Chemistry
The 2026 report captures a structural shift that has been building since the first EWG guide in 2007. Oxybenzone, a benzophenone-class chemical filter once present in roughly 70% of US sunscreens, now appears in 5% of products. Retinyl palmitate, a vitamin A ester that photo-degrades into reactive byproducts under UV exposure, has dropped from 40% of formulations in 2010 to 3% in 2026, according to Alexa Friedman, senior scientist at EWG.
Two regulatory events accelerated the shift. A 2020 randomized clinical trial published in JAMA by Matta and colleagues at the FDA's Division of Applied Regulatory Science found that all six tested chemical filters (avobenzone, oxybenzone, octocrylene, homosalate, octisalate, and octinoxate) exceeded the agency's 0.5 ng/mL plasma safety threshold after a single day of application. Homosalate and oxybenzone remained above threshold for more than two weeks after use stopped. Hawaii, Key West, the US Virgin Islands, and Thailand have separately banned oxybenzone-containing sunscreens for coral-reef-protection reasons, with the European Union now regulating homosalate as a suspected endocrine disrupter.
Mineral filters sit on the skin surface and scatter UV photons. Chemical filters absorb UV energy and dissipate it as heat, a process that consumes filter molecules and depends on stable absorption spectra across UVA-1 (340–400 nm), UVA-2 (320–340 nm), and UVB (290–320 nm) wavelengths. The mechanism difference is why mineral filters are the default for sensitized skin and post-procedure use.
## Why Do EWG-Tested Sunscreens Underperform Their Labels?
A 2022 EWG laboratory study published in Photodermatology, Photoimmunology & Photomedicine tested 51 US sunscreens with labeled SPF values between 15 and 110. On average, products delivered only 24% of the labeled UVA protection and 42–59% of the labeled UVB protection. The gap reflects two issues: US SPF testing measures UVB-driven erythema rather than UVA-driven dermal damage, and chemical filters degrade in sunlight if not paired with photostabilizers like octocrylene or bis-ethylhexyloxyphenol methoxyphenyl triazine.
For consumers, the consequence is silent. UVB exposure produces a visible sunburn within hours. UVA exposure drives photoaging, hyperpigmentation, and immune suppression on a timescale of years, with no acute signal that protection has failed. A product labeled SPF 50 that delivers SPF 25 in practice will still prevent same-day sunburn while permitting most of the long-wavelength UVA dose the consumer believed they were blocking.
## When Will the Chemistry Actually Change?
The first meaningful US filter approval in roughly 20 years is now in the pipeline. On December 11, 2025, the FDA issued a proposed administrative order to amend OTC Monograph M020 to add bemotrizinol, known internationally as Tinosorb S, as a GRASE active ingredient at concentrations up to 6%. The agency's review found broad-spectrum UVA and UVB coverage, low systemic absorption, and a low incidence of skin irritation. The public comment window closed January 26, 2026, and the agency has not committed to a final-order timeline.
Friedman, speaking to CNN about the proposed approval, said bemotrizinol delivers adequate UVA protection, shows minimal skin absorption, and carries the deepest safety dataset of any UV filter studied to date. She described it as the first innovation change in sunscreen chemistry in 20 years.
Until bemotrizinol clears final approval, consumers in the United States are working with a UV filter palette finalized in 1999. The 53 chemical sunscreens cleared by EWG this year are formulations that managed to deliver adequate UVA coverage and avoid the most-scrutinized ingredients using that legacy palette. The 497 mineral options on the recommended list reflect the only filter category with new formulation chemistry, mostly refinements to zinc oxide particle size, dispersion, and tint matching, that has continued to evolve through the regulatory freeze.
For readers selecting a daily product, the EWG 2026 data supports three operational rules: prioritize broad-spectrum mineral formulations for daily wear, treat label SPF as the upper bound rather than the floor of actual performance, and apply at the AAD-recommended dose of one ounce (a shot glass) to exposed skin with reapplication every two hours of UV exposure.
More on bemotrizinol's mechanism and the 25-year regulatory backstory.
A zinc-concentration-ranked look at the mineral filter category EWG cleared.
Why SPF labels diverge from real-world UV transmission.
External references:
Environmental Working Group 2026 Guide to Sunscreens: executive summary and product database.
Matta et al., JAMA 2020: FDA randomized clinical trial of systemic chemical-filter absorption.
Andrews et al., Photodermatol Photoimmunol Photomed 2022: laboratory testing of US-market sunscreen labels.