Is fragrance in personal care products harmful?

What "fragrance" can hide

In the US, a company may list a scent blend as just "fragrance" because the formula counts as a trade secret, so the label cannot tell you which chemicals are inside. FDA also notes that some products labelled "unscented" contain fragrance ingredients¹. The EU requires named fragrance allergens on labels and has expanded that list to more than 80. The new list applies to newly marketed products from 31 July 2026 and to everything on shelves by 31 July 2028³.

Testing shows why disclosure matters. In an analysis of 213 products, the highest concentrations and most detected chemicals were in fragranced items (perfume, air fresheners, dryer sheets) and in sunscreens, and many of the chemicals were not on the label⁹.

Hazard versus exposure

• Allergens. Limonene and linalool are common "natural" fragrance ingredients that become more allergenic as they oxidize in air. Among 5,773 dermatitis patients in Sweden, 7.0% reacted to oxidized linalool and 5.1% to oxidized limonene⁵. These were clinic patients, so the figures overstate what the average person faces.
• Phthalates. Diethyl phthalate (DEP) is used in some fragrances. Women who used perfume had about three times the urinary level of its breakdown product, MEP, of women who did not⁶. Skin absorption is real: about 6% of DEP applied to the whole body in a cream appeared in urine as MEP¹⁰. But that test used a 2% cream over the entire body every day, far more than a spritz of perfume, and it found no short-term effect on reproductive or thyroid hormones¹¹. FDA says it knows of no human health risk from DEP as currently used¹.
• Synthetic musks. In cell studies, polycyclic musks were weakly estrogen-like and blocked androgen and progesterone signals, yet showed no obvious estrogen effect in young rats or zebrafish⁸.

What to do

Start with products that stay on your skin, since they are in contact longest. Pick "fragrance-free", then read the ingredient list for "fragrance" or "parfum". Rinse-off products spend less time on skin, so swap them last. For the broader picture of personal care chemicals, see Hidden Chemicals in Your Daily Routine and Phthalates.

Allergy: the best-documented harm

The cleanest numbers come from a cross-sectional study in Germany, Italy, the Netherlands, Portugal and Sweden. Researchers interviewed 12,377 adults aged 18 to 74 and patch tested a random 3,119. Fragrance mix I was positive in 2.6% (95% CI 2.1 to 3.2) and fragrance mix II in 1.9%. A positive patch test shows sensitization, not necessarily disease, so the authors also gave a conservative figure that required a clinically relevant history: 1.9% (1.5 to 2.5), with about twice as many women affected as men⁷. A meta-analysis of 28 general-population studies (20,107 people) put fragrance mix I at 3.5% (2.1 to 5.4), second only to nickel at 11.4%¹².

Oxidized terpenes are a newer concern. A retrospective Swedish series tested 5,773 patients with dermatitis from 2013 to 2020. Allergy to oxidized linalool (7.0%) and oxidized limonene (5.1%) rose over time, and more than 95% of positives appeared by day 3 or 4. Patients with these allergies were younger than those allergic to other fragrances, and mostly female. The authors said clinical relevance must be clarified before routine screening is recommended⁵. Because these are referral-clinic patients, the rates are not population rates.

Breathing and headaches: surveys, not trials

An online survey of 1,137 US adults found that 26.8% reported a diagnosis of asthma or an asthma-like condition. Of those, 64.3% reported adverse effects from fragranced products, including respiratory problems (43.3%), migraine (28.2%) and asthma attacks (27.9%). Asthmatics were far likelier than others to report effects (prevalence odds ratio 5.76; 95% CI 4.34 to 7.64)⁴. The data are self-reported, unblinded and cannot separate fragrance from cleaning products, smoke or other triggers.

Hormones: exposure is clear, harm is not

In a two-week, single-blind study, 26 men applied a plain cream to the whole body daily. In week two the cream contained 2% each of DEP, DBP and butyl paraben. Serum MEP peaked a few hours later at a mean of 1,001 micrograms per litre. Reproductive and thyroid hormones showed only minor differences that did not track exposure¹¹. Urine collected from the same men showed that 5.79% of the applied DEP came out as MEP¹⁰.

Everyday use involves much smaller amounts. In 337 women who had recently been pregnant, perfume users had MEP 2.92 times higher (95% CI 2.20 to 3.89)⁶. In the HERMOSA study, 100 girls switched to products labelled free of target chemicals for three days. MEP fell 27.4% (95% CI 13.2 to 39.3), methyl and propyl paraben fell about 44% and 45%, and triclosan and benzophenone-3 fell about a third². Ethyl and butyl parabens rose unexpectedly. The design was before-and-after rather than controlled, and no health outcome was measured.

Product testing found 55 compounds across 213 products, and some products free of well-known phthalates contained less-studied ones, suggesting substitution⁹. Laboratory work shows polycyclic musks have very weak estrogenic activity that depends on cell and receptor type, plus anti-estrogenic, anti-androgenic and anti-progestagenic activity⁸.

Where the evidence is thin or conflicting

No study we found links ordinary fragrance use to a hormone-related disease. Urinary markers show exposure, not damage. The musk data come from cells and rodents. FDA's position is that DEP in fragrance has no known human risk as currently used¹. Patch-test studies can overcount relevance, and survey studies can mix up fragrance with other triggers.

Bottom line

Confidence 3 of 5. Fragrance allergy is real, common enough to matter, and well measured. The case that normal fragrance use harms hormones or lungs is unproven. Choosing fragrance-free leave-on products is a low-cost precaution, not a necessity for most people. For scented air products, see candles and air fresheners.