TL;DR: The FDA’s current position, last reaffirmed in their Cosmetics Constituent Guidance and the 2022 modernization rules, is that the parabens approved for cosmetic use (methylparaben, ethylparaben, propylparaben, butylparaben) are safe at the concentrations typically used. The original Soni 2005 safety review remains the regulatory reference. The Darbre 2004 breast tissue paper was the cultural trigger for paraben fear and has been criticized for methodology and overinterpretation. A ‘paraben-free’ label in 2025 is almost always a marketing position, not a safety position. The replacement preservatives sometimes have worse evidence than the parabens they replaced.
Quick answer
Parabens are p-hydroxybenzoic acid esters used as antimicrobial preservatives in cosmetics since the 1920s. The FDA does not currently regulate them as restricted ingredients in cosmetics, and the 2022 Modernization of Cosmetics Regulation Act (MoCRA) did not change that. The agency’s published position, citing the Cosmetic Ingredient Review and the Soni 2005 safety assessment, is that methylparaben and ethylparaben at the concentrations commonly used (up to 0.4 percent for a single paraben, up to 0.8 percent for a mixture) are safe. Propyl- and butylparaben have more uncertainty around endocrine effects at higher exposures but are still permitted. The European Union restricts isopropyl-, isobutyl-, phenyl-, benzyl-, and pentylparaben (a 2015 rule). A “paraben-free” label in the US in 2025 is a marketing claim, not a safety claim, and the substitute preservatives are often less well studied than the parabens they replaced.
The reader scenario
You stand in front of a moisturizer with a “PARABEN-FREE” sticker the size of your thumb on the front of the bottle. You assume that means it is safer than the alternative that does contain parabens. You flip it over. The preservative system is phenoxyethanol with ethylhexylglycerin and a touch of caprylyl glycol. You have no idea whether that is better, worse, or different from a paraben system. The brand is hoping you do not ask. The honest answer, after twenty years of moving back and forth on the regulatory question, is that the paraben-free label is a position taken in 2005 by a market that has not seriously revisited it since, and the replacement chemistry it created is patchier than the chemistry it replaced.
What the FDA actually says
The FDA’s Cosmetic Ingredients page on parabens, last meaningfully updated in 2022 in the context of MoCRA, says the following: parabens are permitted in cosmetics, the agency reviews safety data periodically, and the CIR Expert Panel concluded in 2008 (a reaffirmation of Soni 2005) that the commonly used parabens are safe at concentrations up to 0.4 percent for a single paraben and 0.8 percent for mixtures. The agency notes the European Union restrictions on five parabens (the longer-chain and branched ones) and explains it has not adopted the same restrictions because the safety data on the shorter-chain commonly used parabens (methyl, ethyl) does not support a restriction.
This is the FDA’s actual position. It has not changed substantively in the last decade. It is also not what most consumers believe it is. The narrative that “the FDA banned parabens” is wrong. The narrative that “the FDA is reviewing parabens for restriction” is also wrong. The agency reviews data, accepts the CIR’s safety conclusions, and currently permits the chemistry.
What the studies actually show
The literature on paraben safety has three eras.
Era one: the establishment of safety. Soni 2005 is the canonical safety review. He summarized the toxicology, metabolism, and human exposure data on methyl-, ethyl-, propyl-, and butylparaben and concluded that at typical cosmetic concentrations, the no-observed-adverse-effect levels were two to three orders of magnitude above realistic exposures. PMID: 15833376. The CIR Expert Panel reaffirmed in 2008 (PMID: 19101832), reviewing additional data through 2007 and concluding the same.
Era two: the endocrine-disruption alarm. Darbre 2004 reported finding paraben residues in human breast tumor tissue samples. The paper was widely covered in mainstream media and effectively launched the paraben-free marketing category. The paper itself has been heavily criticized: the sample was small (20 women), there was no comparison with healthy breast tissue, the assay had contamination questions (the lab had used parabens in their own consumables), and the conclusion (that topical paraben use was causally linked to breast cancer) did not follow from the data presented (presence does not establish causation, and parabens are absorbed from many sources, including diet). PMID: 14745841. Darbre has continued to publish in this space; the field’s reception of her work has remained mixed.
Era three: the more recent endocrine review. Nowak 2018 is the closest thing to a balanced contemporary look at paraben endocrine effects. They reviewed in-vitro estrogenic potency (parabens are weak estrogen-receptor binders, several orders of magnitude weaker than estradiol), animal data (effects at high doses, generally not at cosmetic-relevant doses), and human epidemiology (limited and conflicting). Their conclusion was that propyl- and butylparaben warrant continued scrutiny at high exposures, that methyl- and ethylparaben at cosmetic concentrations are unlikely to produce meaningful endocrine effects, and that the cumulative-exposure question (food, drugs, cosmetics combined) is the part where the science is genuinely incomplete. PMID: 29596967.
This is, more or less, where the regulatory state of play has been since.
What replaced parabens, and how well they were vetted
When the paraben-free marketing wave hit in the mid-2000s, formulators had to switch preservative systems. The substitutes that filled the gap, in approximate order of market penetration:
Phenoxyethanol. Glycol ether. Effective broad-spectrum preservative at 0.5 to 1 percent. Has its own safety questions, particularly around infant use, with the French ANSES restricting it for under-3 leave-on cosmetics in 2019. Generally considered safe at typical concentrations. Allergy rate is low. Most “paraben-free” creams use this.
Ethylhexylglycerin. Multifunctional. Boosts other preservatives. Mild allergen for some users. Often paired with phenoxyethanol.
Methylisothiazolinone (MIT) and Methylchloroisothiazolinone (MCI). These were the dominant paraben replacement of the late 2000s and early 2010s. They turned out to be one of the most aggressive contact allergens introduced to cosmetics in modern history. The European Union effectively restricted them for leave-on cosmetics in 2017 after a years-long epidemic of contact dermatitis cases. Many users who switched from parabens to “paraben-free” products in this period experienced allergic reactions that parabens had never given them. The CIR also tightened MIT/MCI limits.
Sodium benzoate and potassium sorbate. Food-grade preservatives. Effective in low-pH formulas. Less effective above pH 5. Many “natural” preservative systems rely on them. The narrower pH range means some formulations are less stable.
Benzyl alcohol with dehydroacetic acid. Common in cleaner-label brands. Effective. The benzyl alcohol component is a known allergen for some users.
The pattern that emerges is unflattering. The replacement chemistry sometimes solved a hypothetical problem (long-term endocrine effects of parabens at concentrations far above realistic exposures) by introducing a real problem (acute contact allergy from MIT, infant exposure questions from phenoxyethanol). The trade was not obviously a net good.
Darbre 2012 specifically
The 2012 Darbre paper (and its follow-up work) is the citation that most “parabens are dangerous” arguments rest on. It is worth reading once if only to see what the source of the alarm actually is. The methodology questions raised by other researchers center on: lab contamination, the absence of healthy-tissue controls, the leap from biomarker presence to causation, and the fact that paraben exposure from food is roughly an order of magnitude higher than from cosmetics for most people, which means topical cosmetic parabens cannot account for the systemic exposure observed.
This is not a takedown of Darbre’s hypothesis. The hypothesis (that low-level chronic endocrine-active chemical exposure may have cumulative health effects) is reasonable to investigate. The specific claim that cosmetic paraben use is causally linked to breast cancer does not follow from the data she has published, and the regulatory bodies in both the US and EU have reviewed her work and not adopted it as a basis for policy. The cultural impact of the paper, however, has dramatically outrun its scientific weight.
What I would tell my past self
Stop reading the front of the bottle. The “paraben-free” sticker is telling you what the brand wants you to think about, not what is actually in the product. Read the preservative system on the INCI list and ask whether that system is well-studied, allergen-prone, or appropriate for your skin.
A methylparaben at 0.2 percent in a cream you have used for five years without irritation is not a thing to fear. A methylisothiazolinone in a “paraben-free” cleanser is a much more reasonable thing to be wary of, because the allergy data on it is bad and the alternative reasons to choose it are mostly marketing.
If you have a personal preference for parabens-free formulations on values grounds, that is fine and is your call to make. The position has reasonable arguments behind it about cumulative environmental exposure and precaution. What it does not have, in 2025, is a clean regulatory or epidemiological foundation that makes “paraben-free” synonymous with “safer.”
FAQ
Does the FDA test cosmetics before they go to market?
No. Cosmetics in the US do not require pre-market approval, with the exception of color additives. MoCRA (2022) introduced new requirements for adverse event reporting, registration, and good manufacturing practices, but did not introduce pre-market safety review. Safety is the responsibility of the manufacturer, with FDA enforcement after the fact.
Are parabens banned in Europe?
Five parabens are restricted in the EU: isopropyl-, isobutyl-, phenyl-, benzyl-, and pentylparaben (banned in cosmetics under EC Regulation 1004/2014). Methyl- and ethylparaben are permitted up to 0.4 percent each. Propyl- and butylparaben are permitted up to 0.14 percent. Not banned. Restricted.
Are parabens linked to breast cancer?
The Darbre 2004 paper raised the question. No subsequent epidemiological study has established a causal link. Major cancer research bodies (American Cancer Society, NCI) state that the evidence does not support a causal relationship between cosmetic paraben use and breast cancer. The field is not closed; the strongest current statement is “no evidence of causation, ongoing investigation of endocrine effects at population scale.”
What preservative system should I look for?
For most adult skin, a phenoxyethanol-based system at 0.5 to 1 percent with ethylhexylglycerin is well-tolerated and effective. For very sensitive skin or contact-allergy history, sodium benzoate and potassium sorbate in low-pH formulas, or formulations preserved with multifunctionals like pentylene glycol, are reasonable. Avoid methylisothiazolinone in leave-on products.
Are parabens safe in pregnancy?
The data on methyl- and ethylparaben at cosmetic concentrations does not show pregnancy-specific risks. Propyl- and butylparaben have more uncertainty, and most prenatal guidance suggests preferring methyl/ethyl. The OBGYN literature has not flagged parabens as a class to avoid in pregnancy. Compare this to retinoids, where the avoidance recommendation is firm. The two are not in the same risk category.
Related Elelaf tools
Sources
- Soni MG, Carabin IG, Burdock GA. Safety assessment of esters of p-hydroxybenzoic acid (parabens). Food Chem Toxicol. 2005;43(7):985-1015. PMID: 15833376
- Darbre PD, Aljarrah A, Miller WR, Coldham NG, Sauer MJ, Pope GS. Concentrations of parabens in human breast tumours. J Appl Toxicol. 2004;24(1):5-13. PMID: 14745841
- U.S. Food and Drug Administration. Parabens in Cosmetics. FDA Cosmetic Ingredients. 2022. Available: fda.gov/cosmetics/cosmetic-ingredients/parabens-cosmetics
- Cosmetic Ingredient Review Expert Panel. Final amended report on the safety assessment of methylparaben, ethylparaben, propylparaben, isopropylparaben, butylparaben, isobutylparaben, and benzylparaben as used in cosmetic products. Int J Toxicol. 2008;27 Suppl 4:1-82. PMID: 19101832
- Nowak K, Ratajczak-Wrona W, Górska M, Jabłońska E. Parabens and their effects on the endocrine system. Mol Cell Endocrinol. 2018;474:238-251. PMID: 29596967