Skip to main content
NCBI home page
American Journal of Epidemiology logoLink to American Journal of Epidemiology
. 2023 Mar 31;192(7):1066–1080. doi: 10.1093/aje/kwad079

Use of Chemical Hair Straighteners and Fecundability in a North American Preconception Cohort

Lauren A Wise ✉, Tanran R Wang, Collette N Ncube, Sharonda M Lovett, Jasmine Abrams, Renée Boynton-Jarrett, Martha R Koenig, Ruth J Geller, Amelia K Wesselink, Chad M Coleman, Elizabeth E Hatch, Tamarra James-Todd
PMCID: PMC10505421  PMID: 37005071

Abstract

Chemical hair straighteners (“relaxers”) are used by millions of North Americans, particularly women of color. Hair relaxers may contain endocrine-disrupting compounds, which can harm fertility. We evaluated the association between hair relaxer use and fecundability among 11,274 participants from Pregnancy Study Online (PRESTO), a North American preconception cohort study. During 2014–2022, participants completed a baseline questionnaire in which they reported their history of relaxer use and completed follow-up questionnaires every 8 weeks for 12 months or until pregnancy, whichever came first. We used multivariable-adjusted proportional probabilities regression models to estimate fecundability ratios (FRs) and 95% confidence intervals (CIs). Relative to never use, fecundability was lower among current (FR = 0.81, 95% CI: 0.64, 1.03) and former (FR = 0.89, 95% CI: 0.81, 0.98) users of hair relaxers. FRs for first use of hair relaxers at ages <10, 10–19, and ≥20 years were 0.73 (95% CI: 0.55, 0.96), 0.93 (95% CI: 0.83, 1.04), and 0.85 (95% CI: 0.74, 0.98), respectively. Fecundability was lowest among those with longer durations of use (≥10 years vs. never: FR = 0.71, 95% CI: 0.54, 0.91) and more frequent use (≥5 times/year vs. never: FR = 0.82, 95% CI: 0.60, 1.11), but associations were nonmonotonic. In this preconception cohort study, use of chemical hair straighteners was associated with slightly reduced fecundability.

Keywords: endocrine-disrupting chemicals, females, fertility, hair straighteners, prospective studies

Abbreviations

BMI

body mass index

CI

confidence interval

FR

fecundability ratio

LMP

last menstrual period

Chemical hair straighteners (“relaxers”) have been used by millions of reproductive-aged North Americans, often for long periods of time (1). The pervasiveness of Eurocentric beauty norms, beliefs, and values as well as targeted marketing of consumer beauty products and race-based discrimination based on hair texture and hairstyle (2, 3) contribute to women of color, particularly Black women, being more likely than White women to use hair relaxers (4, 5). First exposure to relaxers also commonly occurs in childhood among women of color (1, 4–8). Internal exposure to relaxers can occur via dermal absorption and inhalation (9). Relaxers can also cause burns, lesions, and inflammation on the scalp, facilitating entry of chemical ingredients into the body (10–18).

Chemical ingredients in hair relaxers may include sodium or calcium hydroxide, guanidine carbonate, thioglycolic salts, formaldehyde, and hormonally active chemicals such as phthalates, phenols, and parabens (8, 10, 19–22). A 2018 study documented that hair relaxers marketed for children contained chemicals prohibited in cosmetics in the European Union and regulated under California’s Proposition 65 because of their links to reproductive toxicity and cancer (21). This study also demonstrated poor concordance between measured ingredients and product labels, indicating that consumers cannot rely on product labels to avoid harmful chemicals (21). Given that cosmetic products are not subject to premarket approval by the Food and Drug Administration (23), their chemical composition is unclear.

Hair relaxer use has been associated with alterations in reproductive hormones (24, 25) and increased risks of early menarche (26, 27), uterine leiomyomata (28), ovarian cancer (29), uterine cancer (30), and breast cancer in some (31–34) but not all studies (35, 36). To our knowledge, there are no studies of hair relaxer use and fertility outcomes. Exposures to phthalates, phenols, and parabens—endocrine-disrupting chemicals that are found in personal care products—have been associated with infertility (inability to conceive during 12 months of attempt time) and lower ovarian reserve in humans (37–39); however, some studies have indicated null (40, 41) or inverse (42, 43) associations.

In a North American prospective cohort study of participants planning a pregnancy, we investigated prospectively the association between hair relaxer use and fecundability. We examined various aspects of hair relaxer use, and we stratified the data by self-identified race and ethnicity to account for social differences in hair relaxer use.

METHODS

Study population

Pregnancy Study Online (PRESTO) is an ongoing web-based preconception cohort study of pregnancy planners. The study methods have been described in detail elsewhere (44). Briefly, eligible participants identify as female and are aged 21–45 years, residing in the United States or Canada, and not using contraception or fertility treatment at study entry. Participants complete an online baseline questionnaire with items on sociodemographic factors, lifestyle, medical and reproductive histories, and medication use. Participants complete follow-up questionnaires every 8 weeks for up to 12 months to report pregnancy status and update any factors that may have changed over time. Immediately after enrollment, we mail 6 home pregnancy tests to residents of the contiguous United States (45). PRESTO was approved by the Boston University Medical Campus Institutional Review Board, and all participants provide informed consent online.

Assessment of hair relaxer use

Starting in June 2014, we included questions on the baseline questionnaire adapted from the Black Women’s Health Study (28, 46) about use of a “chemical hair straightener (e.g., relaxer),” including history of use (never, current, former, don’t know); age at first use (<10, 10–19, 20–29, ≥30 years), frequency of use per year (1, 2, 3–4, 5–6, ≥7 times), total number of years of use (<1, 1–4, 5–9, 10–14, 15–19, ≥20), number of burns (“a break in the skin, not just tingling”) during their application (never, 1–2, 3–4, 5–9, ≥10 times), and the name of the brand of chemical hair straightener used the longest (Web Figure 1, available at https://doi.org/10.1093/aje/kwad079). Starting in August 2018, we added “chemical perms” to all questions and we asked participants to further specify the products used in the past in the following categories: “chemical hair straightener (e.g., relaxer),” “chemical perm,” “both chemical hair straighteners (e.g., relaxer) and chemical perms,” or “not sure” given that the word “perm” is used variably to mean curling or straightening procedures, and their use may differ across racial groups. The active chemical ingredient in most chemical curling procedures is ammonium thioglycolate or glyceryl monothioglycolate (47–49), whereas the active ingredient in most relaxers is sodium hydroxide (lye-containing) or lithium hydroxide, potassium hydroxide, or guanidine hydroxide (non–lye-containing) (49). Some “no-lye” relaxers contain ammonium thioglycolate, but this chemical is then oxidized with hydrogen peroxide or sodium bromate (49). Finally, in most “no-lye” relaxers sold for home use, the active ingredient is ammonium (bi)sulfite (49).

Assessment of time to pregnancy

We estimated time to pregnancy (TTP) using data from the baseline and follow-up questionnaires. At baseline, participants reported their date of last menstrual period (LMP), usual cycle length (if having regular menstrual cycles), and the number of cycles they had attempted conception. On each follow-up questionnaire, participants reported their most recent LMP date and whether they had become pregnant since the previous questionnaire. More than 96% of participants reported that they planned to use home pregnancy tests to detect pregnancy before seeking confirmation by a clinician. Among those with irregular cycles, defined as those who reported not being able to “predict from one menstrual period to the next about when the next menstrual period would start,” we estimated cycle length based on date of LMP at baseline and prospectively reported LMP dates during follow-up. TTP was calculated as follows: menstrual cycles of attempt at study entry + ((LMP date from most recent follow-up questionnaire − date of baseline questionnaire completion)/usual menstrual cycle length) + 1.

Assessment of covariates

On the baseline questionnaire, participants report their residential address, education, annual household income, use of multivitamins or folate supplements, height, weight, physical activity, cigarette smoking history, intake of alcohol, caffeine, and sugar-sweetened sodas, reproductive and contraceptive history, intercourse frequency, history of sexually transmitted infections, history of physician-diagnosed medical conditions (e.g., depression, anxiety), employment status, average work-hours per week, average sleep duration in the previous month, and whether they were doing anything to improve their chances of conception (e.g., charting menses, ovulation testing, basal body temperature). We calculated body mass index (BMI) using self-reported weight (kilograms) divided by height (meters) squared.

Participants self-identified their race and ethnicity on the baseline questionnaire, and we used these data as a covariate and effect measure modifier in our analyses. The authors view race as a socially and politically constructed variable that shapes many aspects of the lived experience, and, in what Bonilla-Silva describes as the racialized social system, institutions and social systems were designed and operate to the primary benefit of White individuals (50–52). Eurocentricity elevates European phenotype and concepts of beauty, leading to societal pressures to conform to those standards (4, 5), the internalization of European beauty norms, and workplace prohibitions of Afrocentric hairstyles (53–55), contributing to the disproportionate use of relaxers among those with coily hair textures.

Exclusions

From June 2013 through April 2022, 16,093 eligible participants completed the baseline questionnaire. We excluded 160 participants whose baseline LMP was >6 months before study entry and 78 participants with missing or implausible LMP data. We then excluded 3,142 participants who had been trying to achieve pregnancy for >6 cycles at enrollment, to reduce potential for selection bias (e.g., if both hair relaxer use and subfertility influenced participation) and reverse causation bias (e.g., if subfertility caused a change in hair relaxer use). We further excluded 1,259 participants who enrolled prior to the addition of the hair relaxer questions to the questionnaire (before June 2014) and 180 participants who did not answer at least 1 hair relaxer question. Those who did and did not answer at least 1 hair relaxer question were similar with respect to mean age (30 vs. 30 years), alcohol intake (3 vs. 3 drinks/week), and current cigarette smoking (9% vs. 13%), but they differed according to educational attainment (less than a college education: 26% vs. 47%), annual household income (<$50,000: 18% vs. 37%), BMI (mean: 28 vs. 30), identifying as non-Hispanic White (84% vs. 73%), and parity (≥1 birth: 33% vs. 46%). The final analytical population comprised 11,274 participants.

Data analysis

We categorized hair relaxer variables according to their frequency distributions in the analytical sample. Never users of hair relaxers served as the reference group for all comparisons. Participants who reported a history of both hair straightener (relaxer) and perm use (1.6% of cohort) were analyzed as exposed. The 10.5% who reported perm use only (88% of whom identified as non-Hispanic White) were included in the reference group of never users because “perms” may signify products used to curl hair and these products typically contain different active chemical ingredients (e.g., ammonium thioglycolate) than those used to straighten hair (e.g., sodium hydroxide) (47–49). Participants contributed at-risk menstrual cycles to the analysis from study entry until a reported pregnancy (regardless of outcome) or a censoring event (still participating: 1.2%; initiation of fertility treatment: 8.1%; cessation of pregnancy attempts: 2.5%; loss to follow-up: 20.1%; or 12 cycles of attempt time: 9.7%), whichever came first. We censored participants who did not conceive within 12 cycles of attempted conception at 12 cycles because that is the amount of time after which couples typically seek infertility treatment. Participants who were and were not lost to follow-up were similar according to ever use of hair relaxers (12% vs. 10%), mean age (30 vs. 30 years), and alcohol intake (3 vs. 3 drinks/week), but they differed according to educational attainment (less than a college education: 45% vs. 21%), annual household income (<$50,000: 31% vs. 15%), identifying as non-Hispanic White (78% vs. 86%), current cigarette smoking (12% vs. 4%), BMI (mean: 31 vs. 27), and parity (≥1 birth: 38% vs. 32%).

We used life-table methods to compute the percentage of participants that conceived during follow-up, accounting for censoring events. To account for variation in attempt times at study entry (range: 0–6 cycles) and to reduce bias due to left truncation, we analyzed observed cycles only using the Anderson-Gill data structure (56). We used proportional probabilities regression models (57) to estimate fecundability ratios (FRs) and 95% confidence intervals (CIs) for the association between hair relaxer use and fecundability. The FR represents the ratio of fecundability in each exposure category compared with the reference category. Regression models incorporated indicator terms for cycle at risk to account for the expected decline in fecundability in the cohort during follow-up.

We selected potential confounders based on prior literature and the assessment of a causal graph (Web Figure 2). As confounders, we considered potential determinants of subfertility that were also associated with hair relaxer use. Based on these criteria, we constructed 2 sets of multivariable models. The first set adjusted for a reduced set of covariates, including age, education, and race/ethnicity. The second set adjusted for age (<25, 25–29, 30–34, ≥35 years), education (less than a college degree, college degree, graduate school), annual household income (less than $50,000, $50,000–$99,999, $100,000–$149,999, at least $150,000), self-identified race/ethnicity based on US Census categories (non-Hispanic Black, Hispanic/Latina/Latinx, non-Hispanic White, non-Hispanic Asian, non-Hispanic mixed race or other race), BMI (<20, 20–24, 25–29, ≥30), smoking status (never, former, current), marital status (married vs. not married), relationship duration (years), parity (0, 1, ≥2 births), last method of contraception (oral contraceptives, other hormonal contraception, barrier methods, withdrawal/other), intercourse frequency (<1, 1, 2–3, ≥4 times/week), and ever use of perms (yes vs. no).

We used multiple imputation to assign values to missing data on exposures, covariates, and pregnancy status (58). We created 20 imputed data sets with SAS (SAS Institute, Inc., Cary, North Carolina) PROC MI and combined coefficient and standard error estimates from the data sets using SAS PROC MIANALYZE (59). Missingness ranged from <0.1% (race/ethnicity, education) to 3.0% (income); there were no missing values for age. Approximately 15% of participants were lost to follow-up after enrollment. We assigned participants without any follow-up information 1 cycle of observation and multiply imputed their pregnancy status (yes vs. no).

We performed several sensitivity analyses to assess the influence of varied assumptions on our results. We repeated the analyses after excluding participants who reported any history of perm use as a means of assessing the impact of exposure misclassification, as perm use may be interpreted as chemical hair straightening by some participants and because perms may include different harmful chemicals that could lead to attenuated associations (47–49). We also restricted the calendar period to when we asked participants about perm use (in or after August 2018). In addition, we stratified the data by self-identified race and ethnicity (Hispanic/Latina/Latinx, non-Hispanic Black or Black plus another race, and non-Hispanic White) to assess uniformity of results, given that the prevalence and patterns of hair relaxer use differ markedly by race/ethnicity in the cohort. We stratified by BMI (<25, 25–29, ≥30) because we hypothesized that the adverse effects of hair relaxers might be attenuated among those with greater adiposity if some endocrine-disrupting chemicals are sequestered in adipose tissue. For example, galaxolide and linalool are chemicals often found in hair relaxers and they have lipophilic properties (21, 60, 61), which may decrease their bioavailability. Likewise, we stratified by parity (0, ≥1 births) at enrollment because childbearing may be an important route of clearance of some toxicants (62–64). We also stratified the data by age (<30 vs. ≥30 years) because we reasoned that if aging oocytes are more susceptible to the harmful effects of toxicants, hair relaxer use may have a greater impact on older participants. We stratified by pregnancy attempt time at enrollment (<3 vs. 3–6 cycles); if longer pregnancy attempt times at enrollment increase potential for selection bias and reverse causation bias, associations among those with shorter attempt times should be more valid. Finally, to reduce potential for differential exposure misclassification (i.e., changes in hair relaxer use during follow-up that are related to subfertility, resulting in decreased exposure accuracy with increasing pregnancy attempt time), we repeated analyses after truncating follow-up time at 3 cycles after enrollment. We performed all analyses using SAS, version 9.4 (SAS Institute, Inc.) (59).

RESULTS

More than 10% of cohort members reported ever using hair relaxers, with 1.5% reporting current use and 8.6% reporting former use (Table 1). Ever use of hair relaxers was positively associated with older age, lower educational attainment, lower annual household income, higher BMI, residence in the Southern regions of the United States, current cigarette smoking, being unmarried, greater intercourse frequency, and longer attempt times at study enrollment. We observed moderate positive correlations between frequency and duration of hair relaxer use (Spearman correlation coefficient (r) = 0.72), frequency of hair relaxer use and number of burns (r = 0.60), and duration of hair relaxer use and number of burns (r = 0.62).

Table 1.

Baseline Characteristics of 11,274 Participants According to Hair Relaxer Use Reported at Enrollment, Pregnancy Study Online, North America, 2014–2022a

History of HairRelaxer Use Age at First Use, years Duration of Use, years Frequency of Use,No. of Times/Year
Characteristic Current(n = 165) Former(n = 968) Never(n = 10,141) <10(n = 155) 10–19(n = 654) ≥20(n = 324) <1(n = 438) 1–4(n = 308) ≥5(n = 387) 1(n = 615) 2(n = 205) ≥3(n = 313)
Age at baseline, yearsb 30.8 30.7 30.0 30.9 30.2 31.8 30.1 30.9 31.3 30.6 31.1 30.8
Geographic region of residence, %
 US Northeast 22.6 20.1 19.3 16.7 18.3 25.2 21.3 24.5 17.2 20.6 21.3 19.9
 US South 35.5 33.6 22.5 47.4 35.1 26.0 29.0 29.3 43.4 28.4 36.6 42.8
 US Midwest 16.7 19.6 22.8 20.6 20.5 15.1 18.6 16.2 21.7 19.8 16.8 19.4
 US West 13.9 15.6 17.0 11.6 14.9 19.4 15.0 21.2 10.9 17.0 16.8 11.0
 Canada 11.2 11.2 18.3 3.8 11.3 14.3 16.2 8.8 6.9 14.2 8.5 7.0
Education, %
 Less than college 10.2 5.9 5.1 5.0 8.1 3.9 5.8 6.3 8.2 6.0 6.2 8.0
 Some college 30.0 25.8 19.9 37.9 27.7 17.4 24.0 21.8 32.7 24.6 24.3 31.2
 College 25.3 30.3 34.2 22.9 29.7 33.1 34.4 29.0 24.3 32.2 24.4 28.1
 Graduate school 34.6 38.0 40.9 34.2 34.6 45.7 35.8 42.9 34.7 37.2 45.1 32.8
Education, yearsb 15.3 15.6 15.8 15.4 15.5 16.0 15.6 15.7 15.4 15.6 15.7 15.4
Household income in $/year, %
 <50,000 30.8 21.7 17.5 28.2 23.5 20.9 22.5 17.0 28.9 20.8 22.3 27.7
 50,000–99,999 32.5 36.7 36.4 39.0 38.6 29.4 35.7 33.4 38.2 35.6 33.4 38.2
 100,000–149,000 15.6 24.6 27.2 19.0 23.8 25.1 25.6 26.5 18.3 24.5 27.2 19.1
 ≥150,000 21.1 17.0 18.9 13.8 14.0 24.5 16.2 23.1 14.6 19.1 17.1 15.0
Household income, $/yearb 91,073 95,217 100,192 85,331 89,721 106,589 94,396 105,725 85,650 97,832 96,355 87,407
Employed, % 82.8 85.4 86.7 81.5 85.6 86.4 83.7 88.9 83.3 84.5 88.5 83.6
Race/ethnicity, %
 White, non-Hispanic 36.1 53.7 87.9 16.5 50.2 69.3 75.5 61.2 14.1 69.4 49.6 16.5
 Black, non-Hispanic 33.9 21.2 0.5 56.8 25.5 2.5 2.5 8.8 58.7 3.6 21.6 62.5
 Hispanic/Latina/Latinx 15.5 12.1 6.1 11.7 11.5 16.9 12.0 15.4 12.0 13.0 15.8 9.7
 Asian, non-Hispanic 3.9 3.0 1.6 0.8 2.8 4.3 2.7 3.9 2.6 4.8 1.9 0.4
 Mixed race/other race, non-Hispanic 10.7 10.1 3.9 14.2 10.0 7.0 7.3 10.8 12.6 9.3 11.1 10.9
Body mass indexb,c 29.3 29.1 27.9 32.9 29.4 27.3 28.5 27.4 31.3 28.3 29.2 30.9
Cigarette smoking, %
 Current smoker 19.2 12.0 8.3 9.6 15.6 8.8 10.6 13.9 15.7 12.8 10.5 15.3
 Former smoker 11.4 13.3 12.7 12.1 11.9 14.0 15.5 11.2 10.8 13.6 14.3 10.3
 Never smoker 69.4 74.7 79.0 78.2 72.6 77.2 73.9 74.9 73.4 73.6 75.2 74.4
Married, % 80.4 82.6 88.5 78.0 80.7 86.3 87.0 84.3 75.2 85.5 79.5 77.5
Nulliparous, % 69.2 64.9 67.0 62.8 63.0 71.6 65.0 69.1 63.3 63.7 68.2 67.2
Last method of contraception, %
 Oral contraceptives 34.8 29.2 30.5 26.8 28.0 33.9 32.5 32.3 25.1 31.8 30.3 26.7
 Other hormonal methods 10.4 6.4 5.1 8.1 7.3 5.9 4.4 6.6 10.2 5.6 7.8 9.3
 Barrier methods 28.0 40.8 43.8 37.6 39.7 38.8 41.4 36.1 37.4 40.7 39.4 34.4
 Natural methods 26.9 23.6 20.5 27.5 25.0 21.4 21.7 25.1 27.4 22.0 22.5 29.7
Intercourse frequency ≥4 times/week, % 21.8 19.3 14.7 19.0 20.2 18.6 18.6 18.6 22.7 18.9 18.7 22.0
Cycles of attempt time at study entryb 2.5 2.1 2.0 2.5 2.3 1.8 2.0 2.1 2.6 2.0 2.2 2.6
Doing something to improve chances of conception, % 77.4 76.0 79.7 73.5 75.9 77.5 78.0 77.4 72.4 79.5 74.9 70.4
Ever use of permsd, % 17.5 25.3 20.5 29.2 29.1 10.0 19.7 38.5 17.2 28.1 28.0 12.6
Open in a new tab

a Standardized to age distribution of cohort at baseline.

b Values are expressed as mean.

c Weight (kg)/height (m)2.

d Perm users who also used relaxers are included in hair relaxer use columns. Data available after August 2018.

Participants who identified as non-Hispanic Black, Hispanic/Latina/Latinx, or mixed race were more likely to be former or current users of hair relaxers (Table 1). The prevalence of current and former use was disproportionately higher among non-Hispanic Black participants than all other racial/ethnic groups. Specifically, the prevalence of current use was 18% among non-Hispanic Black participants, 3% among Hispanic participants, 3% among non-Hispanic Asian participants, 3% among mixed race/other race participants, and 1% among non-Hispanic White participants. The prevalence of former use was 66% among non-Hispanic Black participants, 15% among Hispanic participants, 15% among non-Hispanic Asian participants, 19% among mixed race/other race participants, and 6% among non-Hispanic White participants. Among ever users, non-Hispanic Black participants started using hair relaxers earlier in life (>56% reported first use before age 10 years vs. 1%–17% among other racial/ethnic groups) and reported more burns, longer durations of use, and higher frequency of use.

In the final models, relative to never use, fecundability was slightly lower among current users (FR = 0.81, 95% CI: 0.64, 1.03) and former users (FR = 0.89, 95% CI: 0.81, 0.98) of hair relaxers. Covariates that resulted in the largest change in the effect estimate (unadjusted vs. adjusted models) for current hair relaxer use and fecundability were education (12%) and income (13%). FRs for first use of hair relaxers at ages <10, 10–19, and ≥20 years (vs. never use) were 0.73 (95% CI: 0.55, 0.96), 0.93 (95% CI: 0.83, 1.04), and 0.85 (95% CI: 0.74, 0.98), respectively (Table 2). The FR for having ever been burned ≥5 times (vs. never use) during applications was 0.81 (95% CI: 0.58, 1.12), and the association with number of burns was monotonic. Fecundability was lowest among those with longer durations of use (≥10 years vs. never: FR = 0.71, 95% CI: 0.54, 0.91) and more frequent use (≥5 times/year vs. never: FR = 0.82, 95% CI: 0.60, 1.11), but associations were nonmonotonic. We did not observe any clear patterns in fecundability when cross-classifying frequency and duration variables, although participants who used hair relaxers for the longest periods of time, regardless of frequency, tended to have the lowest fecundability among ever users. Results based on longest-used brand among ever users did not reveal any appreciable differences in association by brand, although numbers were limited (Table 2; Web Table 1). Results were generally stronger among current users than former users for age at first use, duration of use, and frequency of use.

Table 2.

Hair Relaxer Use in Relation to Fecundability Among Participants in Pregnancy Study Online, North America, 2014–2022

Former and Current Users Combined Former Users Current Users
Unadjusted Adjusted a Adjusted b Adjusted b Adjusted b
Hair RelaxerUse No. ofPregnancies No. of Cycles FR 95% CI FR 95% CI FR 95% CI No. of Pregnancies FR 95% CI No. of Pregnancies FR 95% CI
History of hair relaxer use
 Neverc 6,045 38,955 1.00 Referent 1.00 Referent 1.00 Referent 6,045 1.00 Referent 6,045 1.00 Referent
 Former 492 3,957 0.82 0.75, 0.90 0.89 0.81, 0.98 0.89 0.81, 0.98
 Current 67 650 0.69 0.55, 0.88 0.80 0.63, 1.02 0.81 0.64, 1.03
Age at first use, years
 <10 54 652 0.58 0.45, 0.75 0.72 0.54, 0.95 0.73 0.55, 0.96 47 0.75 0.56, 1.01 7 0.63 0.31, 1.31
 10–19 320 2,554 0.84 0.75, 0.93 0.92 0.82, 1.04 0.93 0.83, 1.04 289 0.93 0.83, 1.05 31 0.89 0.61, 1.31
 ≥20 185 1,401 0.84 0.73, 0.97 0.85 0.74, 0.98 0.85 0.74, 0.98 156 0.86 0.74, 1.01 29 0.80 0.56, 1.13
No. of times burned during application
 0 418 3,130 0.86 0.78, 0.95 0.89 0.81, 0.97 0.89 0.81, 0.98 377 0.90 0.81, 0.99 41 0.78 0.58, 1.06
 1–4 97 1,000 0.68 0.56, 0.83 0.86 0.68,1.07 0.87 0.70, 1.08 79 0.86 0.67, 1.09 18 0.94 0.59, 1.50
 ≥5 44 477 0.64 0.48, 0.85 0.81 0.58, 1.12 0.81 0.58, 1.12 36 0.85 0.59, 1.22 8 0.71 0.35, 1.44
Duration of use, years
 <1 243 1,865 0.85 0.75, 0.96 0.86 0.76, 0.97 0.86 0.76, 0.97 235 0.86 0.76, 0.97 8 0.71 0.33, 1.54
 1–4 169 1,143 0.97 0.84, 1.11 1.00 0.87, 1.16 1.01 0.88, 1.17 149 1.02 0.88, 1.19 20 0.99 0.65, 1.50
 5–9 58 530 0.69 0.53, 0.90 0.77 0.58, 1.01 0.77 0.58, 1.02 45 0.76 0.55, 1.05 13 0.86 0.52, 1.42
 ≥10 89 1,069 0.58 0.48, 0.72 0.69 0.53, 0.90 0.71 0.54, 0.91 63 0.75 0.55, 1.02 26 0.68 0.45, 1.04
Frequency of use, per year
 1 329 2,523 0.85 0.77, 0.95 0.88 0.79, 0.98 0.87 0.78, 0.97 304 0.87 0.78, 0.97 25 0.82 0.56, 1.20
 2 108 801 0.86 0.72, 1.04 0.91 0.76, 1.10 0.95 0.79, 1.14 98 0.97 0.80, 1.17 10 0.81 0.44, 1.48
 3–4 62 571 0.73 0.57, 0.93 0.87 0.67, 1.13 0.89 0.68, 1.15 44 0.90 0.65, 1.23 18 0.91 0.57, 1.44
 ≥5 60 712 0.60 0.47, 0.77 0.77 0.57, 1.05 0.82 0.60, 1.11 46 0.92 0.64, 1.30 14 0.67 0.37, 1.18
Longest-used brand (ever users)
 Keratin or Brazilian 54 357 0.96 0.75, 1.22 0.94 0.74, 1.20 0.91 0.71, 1.17 45 1.09 0.84, 1.42 9 0.50 0.27, 0.93
 Not Keratin or Brazilian 56 640 0.62 0.48, 0.80 0.76 0.58, 0.99 0.80 0.61, 1.04 43 0.88 0.64, 1.19 13 0.64 0.37, 1.11
 Unknown 449 3,610 0.82 0.75, 0.90 0.88 0.80, 0.97 0.89 0.80, 0.98 404 0.87 0.79, 0.97 45 1.06 0.78, 1.43
Duration (years) × frequency (per year)
 <1 × 1 220 1,701 0.85 0.75, 0.96 0.86 0.76, 0.98 0.85 0.75, 0.97 215 0.86 0.75, 0.98 5 0.60 0.20, 1.83
 1–4 × 1 86 603 0.95 0.78, 1.16 1.00 0.82, 1.22 0.98 0.81, 1.20 75 0.97 0.79, 1.20 11 1.10 0.64, 1.89
 ≥5 × 1 23 219 0.60 0.39, 0.93 0.65 0.42, 1.01 0.65 0.42, 1.00 14 0.62 0.34, 1.12 9 0.68 0.37, 1.27
 <5 × ≥2 106 704 0.95 0.79, 1.14 0.98 0.81, 1.17 1.02 0.85, 1.22 94 1.04 0.86, 1.26 12 0.86 0.49, 1.53
 ≥5 × ≥2 124 1,380 0.62 0.52, 0.74 0.74 0.59, 0.94 0.76 0.60, 0.96 94 0.79 0.60, 1.03 30 0.77 0.52, 1.15
Open in a new tab

Abbreviations: CI, confidence interval; FR, fecundability ratio.

a Adjusted for age, education, and race/ethnicity.

b Adjusted for age, education, household income, race/ethnicity, body mass index, smoking status, marital status, relationship duration, parity, last method of contraception, intercourse frequency, and ever use of perms.

c Those with a history of perm use but no relaxer use are included in “never” relaxer use category.

Associations between hair relaxer use (history and duration) and fecundability were generally similar across strata of race/ethnicity and parity (Table 3). Associations were slightly stronger among those with pregnancy attempt times of 3–6 cycles at study enrollment but were still evident among participants with shorter pregnancy attempt times at study enrollment (<3 cycles). Findings were also relatively uniform across strata of age (Table 3), when we restricted to the first 3 cycles of follow-up (Web Table 2), and when we excluded ever users of perms (Web Table 3). However, associations were attenuated among participants with BMIs of ≥30 relative to BMIs of <25 and 25–29 (Table 3).

Table 3.

Hair Relaxer Use and Fecundability, According to Attempt Time at Entry, Race/Ethnicity, Age, Body Mass Index, and Parity, Among 11,274 Participants in Pregnancy Study Online, North America, 2014–2022

Covariate and History of Hair Relaxer Use No. ofParticipants No. ofPregnancies No. of Cycles Unadjusted Adjusted a
FR 95% CI FR 95% CI
<3 cycles of attempt time at enrollment 7,610
 History of hair relaxer use
  Never 4,579 26,962 1.00 Referent 1.00 Referent
  Former 362 2,663 0.83 0.75, 0.91 0.89 0.80, 0.99
  Current 46 372 0.74 0.56, 0.98 0.84 0.63, 1.11
 Duration of hair relaxer use, years
  <1 182 1,315 0.84 0.73, 0.96 0.84 0.73, 0.97
  1–4 123 750 0.98 0.83, 1.15 1.03 0.87, 1.21
  ≥5 103 970 0.65 0.54, 0.79 0.78 0.61, 0.99
3–6 cycles of attempt time at enrollment 3,664
 History of hair relaxer use
  Never 1,466 11,993 1.00 Referent 1.00 Referent
  Former 130 1,294 0.80 0.67, 0.97 0.90 0.74, 1.10
  Current 21 278 0.61 0.39, 0.96 0.76 0.48, 1.22
 Duration of hair relaxer use, years
  <1 61 550 0.87 0.67, 1.13 0.92 0.70, 1.19
  1–4 46 393 0.94 0.71, 1.26 1.02 0.76, 1.37
  ≥5 44 629 0.55 0.40, 0.77 0.64 0.42, 0.97
Non-Hispanic Black or mixed race including Blackb 425
 History of hair relaxer use
  Never 34 270 1.00 Referent 1.00 Referent
  Former 106 1,082 0.83 0.56, 1.21 0.70 0.45, 1.09
  Current 23 239 0.80 0.47, 1.37 0.67 0.37, 1.21
 Duration of hair relaxer use, years
  <1 11 87 0.94 0.47, 1.87 0.77 0.37, 1.60
  1–4 22 133 1.37 0.82, 2.29 1.13 0.68, 1.89
  ≥5 96 1,101 0.74 0.50, 1.09 0.58 0.37, 0.92
Non-Hispanic Whiteb 9,489
 History of hair relaxer use
  Never 5,403 34,397 1.00 Referent 1.00 Referent
  Former 295 2,119 0.88 0.79, 0.99 0.88 0.79, 0.98
  Current 26 238 0.75 0.52, 1.08 0.78 0.54, 1.13
 Duration of hair relaxer use, years
  <1 194 1,387 0.89 0.78, 1.02 0.88 0.77, 1.01
  1–4 102 740 0.90 0.75, 1.08 0.93 0.78, 1.11
  ≥5 25 230 0.64 0.42, 0.97 0.62 0.40, 0.94
Hispanicb 759
 History of hair relaxer use
  Never 334 2,202 1.00 Referent 1.00 Referent
  Former 58 481 0.84 0.64, 1.10 0.83 0.63, 1.09
  Current 11 137 0.51 0.28, 0.95 0.55 0.29, 1.05
 Duration of hair relaxer use, years
  <1 23 263 0.61 0.40, 0.93 0.60 0.39, 0.92
  1–4 29 169 1.10 0.78, 1.57 1.10 0.77, 1.58
  ≥5 17 186 0.66 0.42, 1.06 0.69 0.43, 1.11
Age <30 years 5,252
 History of hair relaxer use
  Never 2,837 17,814 1.00 Referent 1.00 Referent
  Former 207 1,565 0.85 0.74, 0.97 0.92 0.80, 1.05
  Current 30 252 0.73 0.51, 1.05 0.88 0.61, 1.27
 Duration of hair relaxer use, years
  <1 106 780 0.86 0.71, 1.03 0.86 0.72, 1.03
  1–4 67 417 1.03 0.82, 1.29 1.16 0.92, 1.46
  ≥5 64 620 0.66 0.52, 0.85 0.76 0.56, 1.03
Age ≥30 years 6,022
 History of hair relaxer use
  Never 3,208 21,141 1.00 Referent 1.00 Referent
  Former 285 2,392 0.80 0.72, 0.90 0.88 0.78, 1.00
  Current 37 398 0.67 0.49, 0.91 0.80 0.58, 1.10
 Duration of hair relaxer use, years
  <1 137 1,085 0.84 0.71, 0.99 0.86 0.73, 1.01
  1–4 102 726 0.93 0.78, 1.12 0.95 0.79, 1.14
  ≥5 83 979 0.60 0.48, 0.74 0.74 0.56, 0.99
BMI <25c 4,981
 History of hair relaxer use
  Never 3,034 16,865 1.00 Referent 1.00 Referent
  Former 226 1,529 0.85 0.75, 0.96 0.87 0.77, 1.00
  Current 31 289 0.64 0.46, 0.90 0.72 0.51, 1.01
 Duration of hair relaxer use, years
  <1 116 823 0.81 0.67, 0.96 0.83 0.69, 0.99
  1–4 86 530 0.92 0.76, 1.12 0.98 0.80, 1.19
  ≥5 55 465 0.70 0.54, 0.91 0.72 0.53, 0.97
BMI 25–29c 2,769
 History of hair relaxer use
  Never 1,581 9,557 1.00 Referent 1.00 Referent
  Former 133 981 0.85 0.72, 1.01 1.02 0.85, 1.22
  Current 16 169 0.63 0.39, 1.00 0.90 0.55, 1.47
 Duration of hair relaxer use, years
  <1 68 417 0.97 0.77, 1.22 1.03 0.82, 1.30
  1–4 47 307 0.96 0.74, 1.25 1.13 0.86, 1.48
  ≥5 34 426 0.54 0.39, 0.76 0.69 0.45, 1.06
BMI ≥30c 3,524
 History of hair relaxer use
  Never 1,430 12,533 1.00 Referent 1.00 Referent
  Former 133 1,447 0.80 0.67, 0.95 0.84 0.69, 1.03
  Current 20 192 0.85 0.54, 1.36 0.98 0.61, 1.59
 Duration of hair relaxer use, years
  <1 59 625 0.83 0.64, 1.08 0.79 0.61, 1.02
  1–4 36 306 1.00 0.72, 1.39 1.02 0.73, 1.41
  ≥5 58 708 0.69 0.52, 0.90 0.86 0.56, 1.30
Parous 3,738
 History of hair relaxer use
  Never 2,046 11,504 1.00 Referent 1.00 Referent
  Former 162 1,341 0.71 0.61, 0.83 0.82 0.70, 0.96
  Current 18 202 0.52 0.32, 0.84 0.68 0.41, 1.13
 Duration of hair relaxer use, years
  <1 83 614 0.76 0.62, 0.94 0.79 0.64, 0.97
  1–4 45 306 0.85 0.64, 1.12 0.89 0.68, 1.18
  ≥5 52 623 0.52 0.40, 0.68 0.73 0.52, 1.03
Nulliparous 7,536
 History of hair relaxer use
  Never 3,999 27,451 1.00 Referent 1.00 Referent
  Former 330 2,616 0.88 0.79, 0.98 0.94 0.84, 1.05
  Current 49 448 0.78 0.60, 1.02 0.88 0.67, 1.16
 Duration of hair relaxer use, years
  <1 160 1,251 0.89 0.77, 1.04 0.90 0.78, 1.05
  1–4 124 837 1.03 0.87, 1.21 1.07 0.91, 1.27
  ≥5 95 976 0.68 0.55, 0.83 0.74 0.57, 0.97
Open in a new tab

Abbreviations: BMI, body mass index; CI, confidence interval; FR, fecundability ratio.

a Adjusted for age, education, household income, race/ethnicity, BMI, smoking status, marital status, relationship duration, parity, last method of contraception, intercourse frequency, and ever use of perms.

b 601 participants were excluded in the analyses stratified by race/ethnicity because they did not identify as non-Hispanic White, non-Hispanic Black, or Hispanic.

c Weight (kg)/height (m)2.

The prevalence of current hair relaxer use decreased from 2.3% in June 2014 to July 2018 to 0.6% in August 2018 to April 2022 for the overall study population. In analyses restricted to August 2018 to April 2022, the time period when we asked participants about perm use, we did not observe reduced fecundability among current and past relaxer users; however, we observed inverse (albeit less precise) associations for earlier age at first use, greater number of burns, and longer durations of use, regardless of frequency of use (Web Table 4). Results during this time period were not notably different whether we excluded ever users of perms but not relaxers, or controlled for “ever use of perms” (Web Table 4).

DISCUSSION

In this North American preconception cohort study, current and former use of chemical hair straighteners was associated with reduced fecundability. Inverse associations were also observed for earlier age at first use, duration and frequency of use, and a greater number of burns. Longer duration of use and greater frequency of use were each associated with the lowest fecundability, but when these variables were cross-classified, longer duration of use was more strongly associated with fecundability. Results were relatively uniform across strata of self-identified race/ethnicity, attempt time at cohort entry, age, and parity, but were attenuated among participants with BMIs of ≥30. These results are largely consistent with studies that have shown associations between hair relaxer use and increased risks of other hormone-dependent conditions, including uterine leiomyomata (28), ovarian cancer (29), uterine cancer (30), and breast cancer (31–34).

Study limitations include nondifferential exposure misclassification, as participants were asked to report on an exposure that may have been discontinued several years prior to enrollment. In addition, although we asked about age at first use, duration of use, and current use, we did not specifically ask about time since discontinuation among former users. Confounding by unmeasured factors is also likely, particularly confounding by other chemicals that are highly correlated with hair relaxer use (e.g., other cosmetic use). There were small numbers of current users, which limited our precision to examine characteristics of hair relaxer use in this subgroup. The numbers of participants of color included in this study were also small, albeit with much higher prevalence of hair relaxer use. Although our study collected additional data on the individual brands of chemical relaxers used the longest, these analyses had limited sample size.

While systematic bias in the reporting of hair relaxer use by fecundability was unlikely because of the prospective study design and the reporting of hair relaxer use before pregnancy, users of hair relaxers may have discontinued their use after experiencing longer pregnancy attempt times (e.g., subfertility). To explore this possibility, we stratified our data by pregnancy attempt time at study enrollment. While associations were stronger among those with pregnancy attempt times of 3–6 cycles at study enrollment, associations persisted among participants with shorter pregnancy attempt times at study enrollment (<3 cycles), indicating little evidence for reverse causation. Nondifferential misclassification of hair relaxer use is more likely, and the impact of such error on our results is difficult to predict (65). Although participants reported on their current use of hair relaxers on their baseline questionnaire only, results were relatively similar when we truncated follow-up time at 3 cycles to reduce potential for exposure misclassification of current use.

PRESTO had limited racial and socioeconomic diversity, and the prevalence of exposure in the present cohort is lower than what we might expect to observe in the general population. Cohort retention in PRESTO was typical for a cohort study (~80%) (66), but differential loss to follow-up may increase potential for selection bias. Although we observed little difference between those who did and did not complete follow-up with respect to hair relaxer use, age, and alcohol use, we observed important differences in follow-up according to sociodemographic factors such as education, annual household income, race/ethnicity, cigarette smoking, BMI, and parity. If those lost to follow-up were more likely to be current or former hair relaxer users, and use of hair relaxers is associated with subfertility, our results may underestimate the true association between hair relaxer use and subfertility.

The lack of a clear monotonic association for many exposure variables, as well as the inverse association observed for those in the lowest frequency and duration category, raise questions about the extent to which our results reflect a causal association. Moreover, if causal, the specific mechanisms by which hair relaxers may influence risk are unclear. It is possible that hair relaxer use is a proxy for exposure to other personal care products containing high levels of endocrine-disrupting chemicals (8, 67) and that the hair relaxers themselves are not the causal agents. For instance, it has been documented that US non-Hispanic Black women and girls have higher mean levels of exposure to environmental chemicals (67), including phthalates (68, 69), bisphenol A (70), and common insecticides (71), than their non-Hispanic White counterparts.

To our knowledge, this is the first study of hair relaxer use and fertility. Our results are consistent with previous studies indicating associations of hair relaxer use with alterations in sex steroid concentrations (24, 25) and hormone-dependent conditions such as ovarian cancer (29) and uterine pathologies (28, 30), which may be associated with reduced fertility (72). Limited previous studies have observed associations between hair relaxer use and reproductive hormone concentrations. One study of pregnant participants reported that current use of hair relaxers was associated with reductions in concentrations of progesterone and estriol (24). Another study among postmenopausal participants reported that ever use of hair relaxers was associated with increases in concentrations of estriol, 16-epiestriol, and unconjugated estrogens (25).

Some chemical constituents of relaxers (e.g., phthalates, phenols, and parabens) could contribute to reductions in fecundability. For example, chronic exposure to low-dose bisphenol A has been associated with decreased primordial follicles and estrous cycles in rats (73), with effects persisting into adulthood, consistent with a long-term disruption of reproductive function (73). A study by James-Todd et al. (22) also found that combined chemicals in hair products were able to operate as agonists or antagonists of several hormonal receptors, including the estrogen and progesterone receptors, which facilitate menstrual cycling and could have an impact on fertility through perturbations in signaling required for the thinning of cervical mucus necessary for fertilization. These products contained many of the same ingredients present in hair relaxers and were marketed to Black women. Thus, early age at first exposure to chemicals in hair relaxers could plausibly have lifelong effects on reproductive health. Chemical exposure from relaxers may be higher than from other types of personal care products because relaxers can cause burns, lesions, and inflammation on the scalp, facilitating more direct entry of chemical ingredients into the body (10–18). Indeed, the scalp exhibits greater percutaneous absorption of chemicals than other skin on the body (e.g., abdomen) (74). A study of postmenopausal participants reported positive associations between lifetime history of scalp burns and concentrations of unconjugated estrogens (≥5 scalp burns: 76.9 pmol/L vs. no burns: 64.0 pmol/L) (25). The stronger associations observed in PRESTO among those reporting more burns on the scalp may indicate an elevated internal dose to chemical exposures. Despite evidence from PRESTO in support of the hypothesis that hair relaxer use causes delayed conception, we cannot rule out other explanations for the observed results, including chance, bias, and confounding. Thus, our results should be viewed as hypothesis-generating.

In conclusion, in this North American preconception cohort study, use of chemical hair straighteners was associated with reduced fecundability. This work contributes to a growing body of literature documenting associations between exposure to toxic chemicals found in beauty products and reproductive health. Additional research is needed to investigate the specific mechanisms by which chemical hair straighteners influence reproductive health. Moreover, given that women of color often have increased exposure to toxic chemicals found in beauty products, this is an important area of focus for environmental justice and health equity researchers and advocates.

Supplementary Material

Web_Material_kwad079
Click here for additional data file. (579.4KB, pdf)

ACKNOWLEDGMENTS

Author affiliations: Department of Epidemiology, Boston University School of Public Health, Boston, Massachusetts, United States (Lauren A. Wise, Tanran R. Wang, Collette N. Ncube, Sharonda M. Lovett, Martha R. Koenig, Ruth J. Geller, Amelia K. Wesselink, Chad M. Coleman, Elizabeth E. Hatch); Department of Social and Behavioral Sciences, Yale University School of Public Health, New Haven, Connecticut, United States (Jasmine Abrams); Department of Pediatrics, Boston University School of Medicine, Boston, Massachusetts, United States (Renée Boynton-Jarrett); Department of Epidemiology, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States (Tamarra James-Todd); and Department of Environmental Health, Harvard T. H. Chan School of Public Health, Boston, Massachusetts, United States (Tamarra James-Todd).

This work was supported by the National Institute of Environmental Health Sciences (grants R01-ES029951 (PI: L.A.W. and E.E.H.) and R01-ES026166 (PI: T.J.T.)), the National Institute on Minority Health and Health Disparities (grant K01-MD013911 (PI: C.N.N.)), and the Eunice Kennedy Shriver National Institute of Child Health and Human Development (grant R01-HD086742 (PI: L.A.W.)).

Analytical code for this analysis is available upon request. PRESTO participants did not give informed consent to share data with external researchers.

The authors would like to acknowledge the assistance of Michael Bairos and Andrea Kuriyama.

Part of this work was presented at the 2018 annual meeting of the American Society for Reproductive Medicine, October 6–10, 2018, Denver, Colorado.

The content is solely the responsibility of the authors and does not necessarily represent the official view of the National Institutes of Health.

Conflict of interest: L.A.W. serves as a fibroid consultant for AbbVie, Inc. She also receives in-kind donations for primary data collection in Pregnancy Study Online (PRESTO) from Swiss Precision Diagnostics (home pregnancy tests) and Kindara.com (fertility apps). The other authors report no conflicts.

REFERENCES

  • 1. de  Sa Dias  TC, Baby  AR, Kaneko  TM, et al.  Relaxing/straightening of Afro-ethnic hair: historical overview. J Cosmet Dermatol.  2007;6(1):2–5. [DOI] [PubMed] [Google Scholar]
  • 2. Lee  MS, Nambudiri  VE. The CROWN act and dermatology: taking a stand against race-based hair discrimination. J Am Acad Dermatol.  2021;84(4):1181–1182. [DOI] [PubMed] [Google Scholar]
  • 3. Mbilishaka  AM, Clemons  K, Hudlin  M, et al.  Don't get it twisted: untangling the psychology of hair discrimination within Black communities. Am J Orthopsychiatry.  2020;90(5):590–599. [DOI] [PubMed] [Google Scholar]
  • 4. McDonald  JA, Llanos  AAM, Morton  T, et al.  The environmental injustice of beauty products: toward clean and equitable beauty. Am J Public Health.  2022;112(1):50–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5. Zota  AR, Shamasunder  B. The environmental injustice of beauty: framing chemical exposures from beauty products as a health disparities concern. Am J Obstet Gynecol.  2017;217(4):418.e1–418.e6. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6. Gaston  SA, James-Todd  T, Harmon  Q, et al.  Chemical/straightening and other hair product usage during childhood, adolescence, and adulthood among African-American women: potential implications for health. J Expo Sci Environ Epidemiol.  2020;30(1):86–96. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7. Gaston  SA, James-Todd  T, Riley  NM, et al.  Hair maintenance and chemical hair product usage as barriers to physical activity in childhood and adulthood among African American women. Int J Environ Res Public Health.  2020;17, 17(24, 24):9254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 8. James-Todd  T, Senie  R, Terry  MB. Racial/ethnic differences in hormonally-active hair product use: a plausible risk factor for health disparities. J Immigr Minor Health.  2012;14(3):506–511. [DOI] [PubMed] [Google Scholar]
  • 9. Ernstoff  AS, Fantke  P, Csiszar  SA, et al.  Multi-pathway exposure modeling of chemicals in cosmetics with application to shampoo. Environ Int.  2016;92-93:87–96. [DOI] [PubMed] [Google Scholar]
  • 10. Hatsbach de Paula  JN, Basilio  FMA, Mulinari-Brenner  FA. Effects of chemical straighteners on the hair shaft and scalp. An Bras Dermatol.  2022;97(2):193–203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11. Halder  RM. Hair and scalp disorders in Blacks. Cutis.  1983;32(4):378–380. [PubMed] [Google Scholar]
  • 12. Scott  DA. Disorders of the hair and scalp in Blacks. Dermatol Clin.  1988;6(3):387–395. [PubMed] [Google Scholar]
  • 13. Dogra  A, Minocha  YC, Kaur  S. Adverse reactions to cosmetics. Indian J Dermatol Venereol Leprol.  2003;69(2):165–167. [PubMed] [Google Scholar]
  • 14. Gendler  E. Adverse reactions to cosmetics. Cutis.  1987;39(6):525–526. [PubMed] [Google Scholar]
  • 15. Menkart  J. An analysis of adverse reactions to cosmetics. Cutis.  1979;24(6):599  662. [PubMed] [Google Scholar]
  • 16. Nigam  PK. Adverse reactions to cosmetics and methods of testing. Indian J Dermatol Venereol Leprol.  2009;75(1):10–18. [DOI] [PubMed] [Google Scholar]
  • 17. Harris  RT. Hair relaxing. Cosmet Toilet.  1979;94:51–56. [Google Scholar]
  • 18. Food and Drug Administration . Seizure Action Initiated Against “Rio Naturalizer” Hair Products (Press Release). Rockville, MD: Food and Drug Administration; 1995:95–103. [Google Scholar]
  • 19. Pierce  JS, Abelmann  A, Spicer  LJ, et al.  Characterization of formaldehyde exposure resulting from the use of four professional hair straightening products. J Occup Environ Hyg.  2011;8(11):686–699. [DOI] [PubMed] [Google Scholar]
  • 20. Myers  SL, Yang  CZ, Bittner  GD, et al.  Estrogenic and anti-estrogenic activity of off-the-shelf hair and skin care products. J Expo Sci Environ Epidemiol.  2015;25(3):271–277. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21. Helm  JS, Nishioka  M, Brody  JG, et al.  Measurement of endocrine disrupting and asthma-associated chemicals in hair products used by Black women. Environ Res.  2018;165:448–458. [DOI] [PubMed] [Google Scholar]
  • 22. James-Todd  T, Connolly  L, Preston  EV, et al.  Hormonal activity in commonly used Black hair care products: evaluating hormone disruption as a plausible contribution to health disparities. J Expo Sci Environ Epidemiol.  2021;31(3):476–486. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23. Food and Drug Administration . Cosmetic labeling manual: summary of regulatory requirements for labeling of cosmetics marketed in the United States. https://www.fda.gov/cosmetics/cosmetics-labeling-regulations/cosmetics-labeling-guide. 2022. Accessed March 30, 2023.
  • 24. Rivera-Nunez  Z, Ashrap  P, Barrett  ES, et al.  Personal care products: demographic characteristics and maternal hormones in pregnant women from Puerto Rico. Environ Res.  2022;206:112376. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25. Geczik  AM, Falk  RT, Xu  X, et al.  Relation of circulating estrogens with hair relaxer and skin lightener use among postmenopausal women in Ghana. J Expo Sci Environ Epidemiol.  2023;33(2):301–310. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26. James-Todd  T, Terry  MB, Rich-Edwards  J, et al.  Childhood hair product use and earlier age at menarche in a racially diverse study population: a pilot study. Ann Epidemiol.  2011;21(6):461–465. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27. McDonald  JA, Tehranifar  P, Flom  JD, et al.  Hair product use, age at menarche and mammographic breast density in multiethnic urban women. Environ Health.  2018;17(1):1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28. Wise  LA, Palmer  JR, Reich  D, Cozier  YC, Rosenberg  L. Hair relaxer use and risk of uterine leiomyomata in African-American women. Am J Epidemiol  2012;175(5):432–440. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29. White  AJ, Sandler  DP, Gaston  SA, et al.  Use of hair products in relation to ovarian cancer risk. Carcinogenesis.  2021;42(9):1189–1195. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30. Chang  CJ, O'Brien  KM, Keil  AP, et al.  Use of straighteners and other hair products and incident uterine cancer. J Natl Cancer Inst.  2022;114(12):1636–1645. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31. Brinton  LA, Figueroa  JD, Ansong  D, et al.  Skin lighteners and hair relaxers as risk factors for breast cancer: results from the Ghana Breast Health Study. Carcinogenesis.  2018;39(4):571–579. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32. Llanos  AAM, Rabkin  A, Bandera  EV, et al.  Hair product use and breast cancer risk among African American and White women. Carcinogenesis.  2017;38(9):883–892. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33. Rao  R, McDonald  JA, Barrett  ES, et al.  Associations of hair dye and relaxer use with breast tumor clinicopathologic features: findings from the Women's Circle of Health Study. Environ Res.  2022;203:111863. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34. Eberle  CE, Sandler  DP, Taylor  KW, et al.  Hair dye and chemical straightener use and breast cancer risk in a large US population of Black and White women. Int J Cancer.  2020;147(2):383–391. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35. Rosenberg  L, Boggs  DA, Adams-Campbell  LL, et al.  Hair relaxers not associated with breast cancer risk: evidence from the Black Women's Health Study. Cancer Epidemiol Biomarkers Prev.  2007;16(5):1035–1037. [DOI] [PubMed] [Google Scholar]
  • 36. Coogan  PF, Rosenberg  L, Palmer  JR, et al.  Hair product use and breast cancer incidence in the Black Women's Health Study. Carcinogenesis.  2021;42(7):924–930. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 37. Thomsen  AM, Riis  AH, Olsen  J, et al.  Female exposure to phthalates and time to pregnancy: a first pregnancy planner study. Hum Reprod.  2017;32(1):232–238. [DOI] [PubMed] [Google Scholar]
  • 38. Messerlian  C, Souter  I, Gaskins  AJ, et al.  Urinary phthalate metabolites and ovarian reserve among women seeking infertility care. Hum Reprod.  2016;31(1):75–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39. Trnka  B, Polan  M, Zigmont  VA. Exposure to di-2-ethylhexyl phthalate (DEHP) and infertility in women, NHANES 2013–2016. Reprod Toxicol.  2021;103:46–50. [DOI] [PubMed] [Google Scholar]
  • 40. Alur  S, Wang  H, Hoeger  K, et al.  Urinary phthalate metabolite concentrations in relation to history of infertility and use of assisted reproductive technology. Fertil Steril.  2015;104(5):1227–1235. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41. Buck Louis  GM, Sundaram  R, Sweeney  AM, et al.  Urinary bisphenol A, phthalates, and couple fecundity: the Longitudinal Investigation of Fertility and the Environment (LIFE) study. Fertil Steril.  2014;101(5):1359–1366. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42. Specht  IO, Bonde  JP, Toft  G, et al.  Serum phthalate levels and time to pregnancy in couples from Greenland, Poland and Ukraine. PLoS One.  2015;10(3):e0120070. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43. Vélez  MP, Arbuckle  TE, Fraser  WD. Female exposure to phenols and phthalates and time to pregnancy: the Maternal-Infant Research on Environmental Chemicals (MIREC) study. Fertil Steril.  2015;103(4):1011–1020.e2. [DOI] [PubMed] [Google Scholar]
  • 44. Wise  LA, Rothman  KJ, Mikkelsen  EM, et al.  Design and conduct of an internet-based preconception cohort study in North America: Pregnancy Study Online. Paediatr Perinat Epidemiol.  2015;29(4):360–371. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45. Wise  LA, Wang  TR, Willis  SK, et al.  Effect of a home pregnancy test intervention on cohort retention and pregnancy detection: a randomized trial. Am J Epidemiol.  2020;189(8):773–778. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46. Rosenberg  L, Wise  LA, Palmer  JR. Hair-relaxer use and risk of preterm birth among African-American women. Ethn Dis.  2005;15(4):768–772. [PubMed] [Google Scholar]
  • 47. Tricoci University of Beauty Culture . The difference between perms and relaxers. https://www.tricociuniversity.edu/beauty-blog/cosmetology-school-the-difference-between-perms-and-relaxers/. Accessed February 15, 2023.
  • 48. Hair Stylist 101 . Perms vs. relaxers—what is the difference? https://hairstylist101.com/perms-vs-relaxers/. Accessed February 15, 2023.
  • 49. Draelos  ZD. Cosmetic Dermatology: Products and Procedure. 3rd ed. Hoboken, NJ: Wiley-Blackwell; 2022. [Google Scholar]
  • 50. Bonilla-Silva  E. Rethinking racism: toward a structural interpretation. Am Sociol Rev.  1997;62(3):465–480. [Google Scholar]
  • 51. Gilroy  P. There Ain't No Black in the Union Jack: the Cultural Politics of Race and Nation. New York, NY: Routledge; 2002. [Google Scholar]
  • 52. Powell  JA. Structural racism: building upon the insights of John Calmore. N C L Rev.  2008;86:791–816. [Google Scholar]
  • 53. Opie  TR, Phillips  KW. Hair penalties: the negative influence of Afrocentric hair on ratings of Black women's dominance and professionalism. Front Psychol.  2015;6:1311. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 54. Collier  DD. Don't get it twisted: why employer hairstyle prohibitions are racially discriminatory. Hastings Race and Poverty Law Journal.  2012;9. [Google Scholar]
  • 55. Donahoo  S, Smith  AD. Controlling the crown: legal efforts to professionalize Black hair. Race and Justice.  2022;12(1):182–203. [Google Scholar]
  • 56. Therneau  TM, Grambsch  PM. Modeling Survival Data: Extending the Cox Model. New York, NY: Springer-Verlag; 2000. [Google Scholar]
  • 57. Weinberg  CR, Wilcox  AJ, Baird  DD. Reduced fecundability in women with prenatal exposure to cigarette smoking. Am J Epidemiol.  1989;129(5):1072–1078. [DOI] [PubMed] [Google Scholar]
  • 58. Zhou  XH, Eckert  GJ, Tierney  WM. Multiple imputation in public health research. Stat Med.  2001;20(9–10):1541–1549. [DOI] [PubMed] [Google Scholar]
  • 59. SAS Institute Inc.  SAS/STAT 9.4 User’s Guide. Cary, NC: SAS Institute; 2014. [Google Scholar]
  • 60. Zhang  X, Yu  Y, Gu  Y, et al.  In vitro determination of transdermal permeation of synthetic musks and estimated dermal uptake through usage of personal care products. Chemosphere.  2017;173:417–424. [DOI] [PubMed] [Google Scholar]
  • 61. Prasanthi  D, Lakshmi  PK. Terpenes: effect of lipophilicity in enhancing transdermal delivery of alfuzosin hydrochloride. J Adv Pharm Technol Res.  2012;3(4):216–223. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 62. Schildroth  S, Wise  LA, Wesselink  AK, et al.  Correlates of non-persistent endocrine disrupting chemical mixtures among reproductive-aged Black women in Detroit. Michigan. Chemosphere.  2022;299:134447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 63. Schildroth  S, Wise  LA, Wesselink  AK, et al.  Correlates of persistent endocrine-disrupting chemical mixtures among reproductive-aged Black women. Environ Sci Technol.  2021;55(20):14000–14014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 64. Wise  LA, Wesselink  AK, Schildroth  S, et al.  Correlates of plasma concentrations of per- and poly-fluoroalkyl substances among reproductive-aged Black women. Environ Res.  2022;203:111860. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 65. Yland  JJ, Wesselink  AK, Lash  TL, et al.  Misconceptions about the direction of bias from nondifferential misclassification. Am J Epidemiol.  2022;191(8):1485–1495. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 66. Hunt  JR, White  E. Retaining and tracking cohort study members. Epidemiol Rev.  1998;20(1):57–70. [DOI] [PubMed] [Google Scholar]
  • 67. Preston  EV, Chan  M, Nozhenko  K, et al.  Socioeconomic and racial/ethnic differences in use of endocrine-disrupting chemical-associated personal care product categories among pregnant women. Environ Res.  2021;198:111212. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 68. Collins  HN, Johnson  PI, Calderon  NM, et al.  Differences in personal care product use by race/ethnicity among women in California: implications for chemical exposures. J Expo Sci Environ Epidemiol.  2023;33(2):292–300. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 69. Silva  MJ, Barr  DB, Reidy  JA, et al.  Urinary levels of seven phthalate metabolites in the U.S. population from the National Health and Nutrition Examination Survey (NHANES) 1999–2000. Environ Health Perspect.  2004;112(3):331–338. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 70. Calafat  AM, Ye  X, Wong  LY, et al.  Exposure of the U.S. population to bisphenol A and 4-tertiary-octylphenol: 2003–2004. Environ Health Perspect.  2008;116(1):39–44. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 71. Barr  DB, Olsson  AO, Wong  LY, et al.  Urinary concentrations of metabolites of pyrethroid insecticides in the general U.S. population: National Health and Nutrition Examination Survey 1999–2002. Environ Health Perspect.  2010;118(6):742–748. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 72. Whynott  RM, Vaught  KCC, Segars  JH. The effect of uterine fibroids on infertility: a systematic review. Semin Reprod Med.  2017;35(6):523–532. [DOI] [PubMed] [Google Scholar]
  • 73. Lopez-Rodriguez  D, Franssen  D, Sevrin  E, et al.  Persistent vs transient alteration of Folliculogenesis and estrous cycle after neonatal vs adult exposure to bisphenolA. Endocrinology.  2019;160(11):2558–2572. [DOI] [PubMed] [Google Scholar]
  • 74. Bronaugh  RL, Maibach  HI. Percutaneous Absorption: Drugs, Cosmetics, Mechanisms, Methodology. 4th ed. New York, NY: Informa Healthcare; 2010. [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Web_Material_kwad079
Click here for additional data file. (579.4KB, pdf)

Articles from American Journal of Epidemiology are provided here courtesy of Oxford University Press

RESOURCES