Hair Relaxers and Cancer Risk: What the Evidence Shows

A National Institutes of Health study reported that women who used chemical hair straighteners more than four times in the previous year were more than twice as likely to develop uterine cancer as women who never used them. Here is what it measured, what a larger study in Black women found afterward, and what any of it changes for someone who has relaxed her hair since childhood.

A relaxer straightens by breaking the chemical bonds that hold the hair's shape. Lye relaxers are built on a sodium hydroxide base, no-lye relaxers on a guanidine hydroxide base. That chemistry is the whole of the difference. In a South African half-head study, the no-lye product straightened more effectively and stripped more cystine, one of the amino acids that gives hair its strength. The researchers also judged it safer to use, but they tested five women, which is far too small to settle a safety question.

In the health data, no-lye is a marketing distinction more than a safety one. When researchers examined fibroid risk by formulation type in more than 23,000 Black women, risk was unrelated to which kind was used.

Formaldehyde belongs to a different product category. NIEHS lists some salon hair smoothing and straightening products among the consumer sources of formaldehyde exposure, and the National Toxicology Program has listed formaldehyde as a known human carcinogen since 2011. The specifics matter: the human evidence behind that listing comes from workers with high occupational exposure such as industrial workers and embalmers, and the cancers involved were myeloid leukemia, sinonasal cancer and nasopharyngeal cancer. Uterine cancer is not on that list, and NIEHS notes that a listing does not by itself mean a substance will cause cancer, because amount and duration of exposure matter. Brazil has prohibited formaldehyde and its derivatives in these products.

The study is the NIEHS Sister Study. Researchers followed 33,947 women aged 35 to 74 for an average of 10.9 years and recorded 378 uterine cancers. Ever using straightening products carried a hazard ratio of 1.80 against never using them (95% CI 1.12 to 2.88), and using them more than four times in the previous year carried a hazard ratio of 2.55 (95% CI 1.46 to 4.45). Hair dyes, permanents and body waves showed no association with uterine cancer in the same cohort.

"More than doubled" is the number that traveled. The absolute numbers tell you how to weigh it. NIEHS estimated that 1.64% of women who never used straighteners would develop uterine cancer by age 70, rising to 4.05% among frequent users. The gap is 2.4 percentage points by age 70, which is the horizon the study modeled and not a full lifetime. Expressed as counts, which is our arithmetic on the NIEHS figures rather than a number the study published, that is roughly 1 woman in 61 versus roughly 1 in 25. The researchers' own framing was that uterine cancer is a relatively rare cancer, and that context is part of the finding rather than a softening of it.

About 60% of participants reporting straightener use in the previous year were Black women, and the study did not find the relationship differed by race. Both are true at once: the biology here looks the same for everyone, and the exposure is concentrated, so any real effect lands disproportionately on Black women.

The next question was whether the finding would hold in a cohort of Black women specifically. The Black Women's Health Study answered it in 2023, following 44,798 Black women with an intact uterus from 1997 to 2019 and recording 347 uterine cancers.

The result was weaker. Across the whole cohort, heavy use, defined as 15 or more years at five or more times a year, gave a hazard ratio of 1.18 with a confidence interval of 0.81 to 1.71, which crosses 1 and is not statistically significant. In postmenopausal women the associations were significant but modest: 1.60 for moderate use, 1.64 for heavy use, 1.71 for 20 or more years of use, with lower confidence bounds near 1.01. In premenopausal women the results were null. Two large cohorts pointing the same direction at different magnitudes is what an unsettled question looks like. It is not a debunking and it is not a confirmation of the doubling.

One finding here is actionable today, and it comes from fibroids rather than cancer. In the Black Women's Health Study's earlier work, 23,580 premenopausal women were followed from 1997 to 2009 and 7,146 fibroid cases were confirmed by ultrasound or surgery. Ever using relaxers carried an incidence rate ratio of 1.17 (95% CI 1.06 to 1.30), and risk rose with frequency of use, duration of use, and the number of burns reported, with the burn trend at p below 0.001. Risk was unrelated to type of formulation. The authors framed the burn finding as a hypothesis rather than a conclusion, and no uterine cancer study has tested it.

NIEHS states the proposed mechanism plainly: chemical exposure from straighteners could be more concerning than from other personal care products because of increased absorption through the scalp, which may be worsened by burns and lesions caused by the straighteners themselves. That is a proposed pathway, not a measured one. Separately, a dermatology review found associations with eczema, scaling, pain, burns and scalp inflammation, alongside hair loss and shaft damage. Those harms stand on their own whatever the cancer question turns out to be. A relaxer that burns is not working better.

A systematic review in the Journal of Applied Toxicology reached the opposite conclusion from the cohorts above. All seven of its authors list the same affiliation, Benchmark Risk Group, a private risk-assessment and litigation-support consultancy rather than an academic or government group, and the PubMed record carries no conflict-of-interest statement. That belongs on the table before the findings do.

The review covered eight studies, seven cohort and one case-control, on relaxer use and breast, ovarian and uterine cancer plus fibroids in US Black women. It concluded that the weight of the evidence does not support relaxer use as a risk factor for gynecological and breast cancers in this group, while recording that one of the studies did report an association with fibroids. Its central criticism was that none of the studies characterized the chemical constituents of the relaxers involved.

That criticism is correct, and it is the same gap NIEHS named, which is why the review is worth reading rather than discarding. What it is not is an independent replication. Weigh it alongside the federally funded cohort studies, not over them.

Nothing here supports a change in cancer screening, and the researchers did not propose one. They called for more research to identify which chemicals, if any, are responsible. The honest list of what to do is short.

Reduce burns first. Burns are the one exposure step with a measured dose-response behind them, in the fibroid data, and the damage they do to the scalp is documented independently of any cancer question. Do not let an application stay on past discomfort, and skip a session when the scalp is already irritated or scratched.

Frequency is what the data measured, so frequency is the lever. The Sister Study's risk estimate tracked how often, not whether you had ever used a relaxer. Stretching the interval between applications is a smaller decision than stopping, and it is the one the numbers speak to.

Ask the salon whether a smoothing treatment contains formaldehyde or a formaldehyde releaser, and ask about ventilation. That exposure is inhaled rather than absorbed, and it hits stylists hardest because they are in it all day.

Know the symptom, not the statistic. Uterine cancer is usually found because of bleeding. Our guide to uterine cancer warning signs in Black women covers what to bring to a doctor. If heavy or irregular bleeding is already part of your life, see heavy menstrual bleeding causes and care and fibroids in Black women. If years of relaxers have cost you hair, start with traction alopecia and CCCA.