Hair Dye, Relaxers, and the Studies Worth Taking Seriously

In October 2022, the National Institutes of Health published a finding that landed harder than anything else in the beauty aisle that year: women who used chemical hair straighteners frequently were about two and a half times more likely to develop uterine cancer than women who never used them. The same analysis found no such association for hair dye, bleach, highlights, or perms.

That's the shape of this entire subject. Some salon fears are folklore. One of them has federal cohort data behind it. They get treated as one anxiety, which is bad for everybody, since it means the dye worriers are stressed for no reason and the straightener finding gets filed under general chemical noise.

What the Sister Study actually found

The straightener result comes from the Sister Study, an NIH cohort following more than 33,000 US women. Published in the Journal of the National Cancer Institute in October 2022, the analysis followed participants for close to 11 years, during which 378 uterine cancers were diagnosed.

Women who reported ever using straightening products were nearly twice as likely to develop uterine cancer, adjusted for other risk factors. Those reporting frequent use, more than four times in the previous year, were roughly two and a half times as likely.

Now the part that responsible coverage includes and viral coverage doesn't. Uterine cancer is not common, so the absolute numbers stay small. The researchers estimated 1.64 percent of women who never used straighteners would develop uterine cancer by age 70, against 4.05 percent of frequent users. That's a real difference and also not a coin flip. It's an association drawn from self-reported product use, and the study could not identify which chemicals were responsible.

What makes it hard to dismiss is that it didn't stay alone. In October 2023, researchers at Boston University published findings from the Black Women's Health Study, following nearly 45,000 women since 1997. Among postmenopausal women, moderate and heavy relaxer use and 20 or more years of use were each associated with roughly 60 to 70 percent higher uterine cancer risk. No association appeared among premenopausal women.

Two large independent cohorts pointing the same direction is the threshold where we stop calling something preliminary.

Why this lands on Black women specifically

Relaxers are marketed overwhelmingly to Black women and girls, often starting in childhood, and used on a repeating schedule for decades. The Sister Study cohort skewed white, which meant its straightener users were a minority of participants; the Boston University work was built specifically to look at the population that uses these products most.

The product testing points the same way. A 2018 Silent Spring Institute study in Environmental Research analyzed 18 hair products marketed to Black women and detected 45 endocrine-disrupting or asthma-associated chemicals across them, with parabens and phthalates in 78 percent of products. Relaxers marketed for children carried the highest levels of chemicals banned in the EU or regulated under California's Proposition 65. And 84 percent of the chemicals the lab found were not listed on the labels.

Add scalp burns, which relaxer users report routinely, and you have a plausible route for more of anything to get in. This is an exposure story about who the products are sold to, not about biology.

There is now enormous litigation attached. The federal hair relaxer cases are consolidated as MDL 3060 in the Northern District of Illinois before Judge Mary Rowland, naming L'Oreal among other manufacturers, and the docket has grown past 11,000 pending cases. No trial has been held and no settlement reached, with the first bellwether trials expected in 2027. A large MDL means many people filed claims. It does not mean anything has been proven in court.

Formaldehyde, keratin treatments, and a rule that keeps slipping

Separate from relaxers, formaldehyde-based smoothing treatments (Brazilian blowouts, keratin treatments) release formaldehyde gas when a flat iron hits the wet product. Formaldehyde is a recognized human carcinogen and a respiratory irritant, and this is one of the few genuinely unambiguous hazards in the category.

The FDA first signaled it would propose banning formaldehyde and formaldehyde-releasing chemicals in hair smoothing and straightening products with a target of October 2023. That date moved to April 2024, then November 2024, then March 2025, then July 2025, then the end of 2025, and the agency missed that one too, prompting letters from members of Congress in early 2026. No proposed rule has been issued.

Nothing legally forces the FDA to hit its own timeline. For now the protection is procedural: ask what's in the smoothing product, and if the salon can't tell you, that's your answer.

The people with the real exposure are behind the chair

IARC evaluated this back in 1993 and drew a distinction worth keeping. Working as a hairdresser or barber was classified as probably carcinogenic to humans, Group 2A, based on consistent excess bladder cancer risk in five large European cohorts of male hairdressers. Personal use of hair colorants landed in Group 3, meaning not classifiable, for lack of adequate evidence.

Same chemicals. Different dose, different duration, and eight hours a day in the same room with them. A stylist performing smoothing treatments in a salon without real ventilation is in a different risk category than the client in the chair, and the client goes home.

Your own box of dye is one of the better-studied products you own

The largest analysis of personal permanent hair dye use came out of the Nurses' Health Study and was published in The BMJ in September 2020. It covered more than 117,000 women followed for 36 years, with over 47,000 cancers recorded.

The headline finding was reassurance: no greater risk of most cancers, and no greater cancer mortality, among women who used permanent dye at home. The authors did report modest increases for a few specific outcomes, including some breast cancer subtypes, ovarian cancer, and basal cell carcinoma, which is why "no risk at all" isn't the right summary either. But across three and a half decades and a very large cohort, the broad pattern was flat.

That's the relative weight to give these things. A finding of nothing much across 117,000 women over 36 years, against a finding of roughly doubled risk in two independent cohorts for a different product category. Both are real results. They point in opposite directions, and treating them as the same worry gets you exactly the wrong precautions.