pmc88035912065Reprod ToxicolReprod ToxicolReproductive toxicology (Elmsford, N.Y.)0890-62381873-1708417633331322928710.1016/j.reprotox.2026.109207NIHMS2179287ArticlePer- and polyfluoroalkyl substances (PFAS) and menstrual cycle characteristics in midlife women: The Study of Women’s Health Across the NationHarlowSiobán D.aHoodMichelle M.aBrooksRachaelaCalafatAntonia M.bRandolphJohn F.JrcGoldEllen B.dhttp://orcid.org/0000-0001-9981-6250ParkSung Kyunae*Department of Epidemiology, School of Public Health, University of Michigan, Ann Arbor, MI, United StatesDivision of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention, Atlanta, GA, United StatesDepartment of Obstetrics and Gynecology, School of Medicine, University of Michigan, Ann Arbor, MI, United StatesDepartment of Public Health Sciences, University of California, School of Medicine, Davis, CA, United StatesDepartment of Environmental Health Sciences, School of Public Health, University of Michigan, Ann Arbor, MI, United StatesCorrespondence to: Departments of Epidemiology and Environmental Health Sciences, University of Michigan, School of Public Health, 1415 Washington Heights, Ann Arbor, MI 48109-2029, United States. sungkyun@umich.edu (S.K. Park).24520264202626220260262026141109207109207https://creativecommons.org/licenses/by-nc/4.0/This is an open access article under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).Objective:

Evidence suggests exposure to endocrine-disrupting chemicals, including per- and polyfluoroalkyl substances (PFAS), may disrupt ovarian function and influence menstrual bleeding. PFAS have been associated with reproductive aging; however, longitudinal data on PFAS and menstrual cycle characteristics (cycle length, bleeding patterns) are limited.

Methods:

We examined 952 women aged 45–55 years from the Study of Women’s Health Across the Nation, with seven PFAS concentrations measured in serum collected in 1999–2000. Menstrual cycle data were collected prospectively using monthly calendars for two years following PFAS measurement. Menstrual cycle length (MCL) was evaluated using linear quantile mixed models at the 25th, 50th, 75th, and 90th percentiles. Bleed length and binary bleeding outcomes (long and heavy) were analyzed using linear and generalized linear mixed models. Models were adjusted for race/ethnicity, education, physical activity, menopause status, smoking status, body mass index, and baseline menstrual characteristics.

Results:

Higher perfluorooctane sulfonic acid concentrations were associated with shorter short cycles (MCL at the 25th percentile) (−0.52 days; 95% CI, −1.09–0.05) and longer long cycles (MCL at the 90th percentile) (2.77 days; 95% CI, 1.96–3.58). Perfluorononanoic acid was associated with longer short cycles (0.56 days; 95% CI, 0.06–1.05). No PFAS was associated with median MCL. Most PFAS, except 2-(N-methyl-perfluorooctane sulfonamido) acetic acid and 2-(N-ethyl-perfluorooctane sulfonamido) acetic acid, were associated with shorter menses and lower odds of long or heavy bleeding.

Conclusion:

PFAS exposure may be associated with increased menstrual cycle variability and reduced bleeding. Further research in younger cohorts is needed.

Menstrual bleed lengthMenstrual cycle lengthMenstrual cycleMidlife womenPer- and polyfluoroalkyl substances (PFAS)Introduction

Alterations in menstrual characteristics can be an early warning sign of the impact of environmental exposures on ovarian and reproductive function [1–3]. Growing evidence suggests exposure to endocrine disrupting chemicals may alter menstrual bleeding as a result of effects on ovarian function [4,5], including the insecticide dichlorodiphenyltrichloroethane and its metabolite, dichlorodiphenyldichloroethylene [6–9], organophosphates [10], polychlorinated biphenyls [11,12], polybrominated diphenyl ethers [13], dioxin [14], di-2-ethylhexyl phthalate metabolites, and bisphenol A [15].

Given their widespread use, an emerging area of concern is the group of per- and polyfluoroalkyl substances (PFAS), often referred to as “forever chemicals.” PFAS are persistent environmental chemicals present in industrial products, such as some fire-fighting foams, and in many consumer products, including non-stick cookware, food packaging, stain- and water-resistant coating for clothing, furniture, and carpets [16–18]. Long-chain legacy PFAS, such as perfluorooctanoic acid (PFOA) and perfluorooctane sulfonic acid (PFOS), have been phased out of most consumer products in the United States and other regions since the early 2000s, resulting in marked declines in their concentrations in human blood [19,20]. These legacy PFAS may have been replaced by other PFAS in many applications [21]. PFAS have been associated with alterations in ovarian function and accelerated ovarian aging [22–25]. Recent findings from the Study of Women’s Health Across the Nation (SWAN) Multi-Pollutant Study (MPS) have reported positive associations of PFOA and PFOS with follicle stimulating hormone (FSH) and inverse associations of PFOA and perfluorononanoic acid (PFNA) with estradiol (E2) in midlife women during the menopausal transition [26].

Studies on menstrual function are more limited. Most, but not all, studies have reported associations with irregular and longer menstrual cycles [27–31]. However, prior studies have been cross-sectional, relied on recalled information about women’s past menstrual cycles, and lacked precision in measurement of menstrual characteristics. The present study takes advantage of two-years of menstrual calendars prospectively collected in SWAN and timed to the PFAS exposure assessment. We hypothesized that higher serum concentrations of PFAS would be associated with alterations in the distribution in menstrual cycle lengths and in the duration and amount of menstrual bleeding.

Materials and methodsStudy population

The SWAN which began in 1996 is a multi-site, multi-race and -ethnic longitudinal study of the menopause transition. To be eligible at baseline, women had to be 42–52 years old, have an intact uterus, have at least one menstrual period in the previous three months, not be pregnant, and have no use of reproductive hormones in the previous three months. Initially, 3302 premenopausal women were enrolled at seven study sites, including White women at all sites, Black women at four sites (Boston, Pittsburgh, Southeast Michigan, and Chicago), Hispanic women at Newark, Chinese women in Oakland, CA, and Japanese women in Los Angeles. At baseline and subsequent approximately annual follow-up visits, women completed questionnaires about their menstrual, gynecological and medical history, including information on surgeries and exogenous hormone use, and were asked to complete a monthly menstrual calendar from enrollment until two years after their last menstrual period or for up to ten years. The study protocol was approved by institutional review boards, and all women gave written informed consent at all sites.

The SWAN Multi-Pollutant Study (MPS) was initiated to assess exposure to multiple environmental pollutants and to examine their associations with reproductive aging using repository serum and urine samples from the third SWAN follow-up (MPS baseline, 1999–2000) for environmental exposure assessment (n = 2694). Women from Chicago (n = 368) and Newark (n = 278) did not collect urine specimens and thus were not eligible for SWAN-MPS. A total of 648 women with insufficient serum or urine samples were excluded, providing a sample at the MPS baseline of 1400 eligible women from four race and ethnic groups (White, Black, Chinese, and Japanese).

Of the women in the MPS, 1384 completed at least one menstrual calendar. We excluded 157 women who were postmenopausal (natural or surgical) by the second SWAN follow-up (1998–1999). In these analyses, we included menstrual cycles that occurred in the two-year window, from one year before to one year after the MPS measure. Women were required to have completed at least three consecutive monthly calendars while not on hormone therapy (HT) within this two-year window, resulting in a sample of 952 women with eligible menstrual calendar data. One woman was excluded for missing typical flow length at baseline, resulting in a final analytical sample of 951 for models relating to bleed length. An additional four women were missing baseline education level, and 17 women were missing cycle length at baseline, resulting in an analytical sample of 931 for the final adjusted models of cycle length. Additional menstrual cycles were excluded from analyses if they occurred during HT use, HT-washout periods, pregnancy, or if there were gaps in menstrual calendar records, yielding a total of 16,929 cycles in longitudinal analyses of bleed length and 16,606 cycles for cycle length.

Cycle length and bleed length

On the monthly menstrual calendar, women noted each day they experienced any spotting or bleeding and the amount of bleeding, categorized as spotting, bleeding or very heavy bleeding (needing to change sanitary product every 1–2 h for more than four hours during the day). On the last day of the month, women answered questions about their HT use, gynecological procedures that could affect bleeding, cigarette use, and physical exercise. For months with missing HT information, menstrual cycles were coded as untreated if a woman never reported HT use in the study, if the menstrual cycle occurred before the first report of HT use, or if the menstrual cycle occurred in a year where no HT use was reported in the monthly calendars or annual interview. We calculated each woman’s sequence of menstrual cycle lengths and bleed lengths during the two-year window. Menstrual cycle length (MCL) was calculated using World Health Organization bleeding definitions [32] adapted by the ReSTAGE Collaboration for application to perimenopausal women [33]. A menstrual cycle consists of a bleeding episode and a subsequent bleed-free interval of at least three days. We also created two categorical bleeding variables. A long bleeding episode was defined as a bleed length greater than eight days, and a heavy bleeding episode was defined as a three or more days of heavy bleeding [34,35].

PFAS exposure assessment

Concentrations of PFAS in serum collected in 1999–2000 were measured at the Division of Laboratory Sciences, National Center for Environmental Health, Centers for Disease Control and Prevention (CDC). The CDC laboratory’s involvement did not constitute engagement in human-subjects research. Online solid phase extraction-high performance liquid chromatography-isotope dilution-tandem mass spectrometry was used to quantify 11 PFAS, including linear PFOA (n-PFOA), sum of branched PFOA isomers (Sb-PFOA), PFNA, perfluorodecanoic acid (PFDA), perfluoroundecanic acid (PFUnDA), perfluorododecanoic acid (PFDoDA), perfluorohexane sulfonic acid (PFHxS), linear PFOS (n-PFOS), sum of perfluoromethylheptane sulfonate isomers (Sm-PFOS), and two PFOS precursors [36], namely 2-(N-methyl-perfluorooctane sulfonamido) acetic acid (MeFOSAA) and 2-(N-ethyl-perfluorooctane sulfonamido) acetic acid (EtFOSAA) [37]. Total PFOS (hereafter, PFOS) was computed as the sum of n-PFOS and Sm-PFOS. The limit of detection (LOD) for all analytes was 0.1 ng/mL. Sb-PFOA, PFDA, PFUnDA, and PFDoDA had low detection frequencies (<50%) and were excluded from this analysis. Detection frequencies of included PFAS were > 97%. For these PFAS, concentrations below the LOD were substituted with LOD/2. Comprehensive QA/QC procedures were conducted. The coefficient of variation of low- and high-concentration quality controls ranged from 5.9% to 12.1%, depending on the analyte.

Covariates

Self-reported race/ethnicity (White, Black, Chinese, Japanese) and level of education (high school or less vs. some college or more) were assessed at SWAN baseline (1996–1997). Other covariates, including menopause status, smoking status (current versus former or never), body mass index (BMI), and fibroids, were assessed at SWAN Visit 2 (1998–1999), the visit closest to the beginning of the two-year window of cycles included in these analyses. Physical activity was assessed at SWAN baseline and at Visit 3 (1999–2000). We chose to use the Visit 3 data, as it may better reflect physical activity levels during the follow-up period. If a covariate was missing at Visit 2, we used the measure at Visit 3 or the closest prior visit. Physical activity was based on a modified Kaiser Physical Activity Survey [38], adapted from the Baecke physical activity questionnaire [39], and ranged in total score from 3 to 15. BMI was calculated as measured weight in kilograms divided by measured height in meters squared; we categorized BMI as normal weight or underweight (<25 kg/m2), overweight (25–29.9 kg/m2), and obese (≥30 kg/m2). Menopause status was classified as pre-menopausal (menses in last 3 months and no change in predictability), early peri-menopausal (menses in the last 3 months but decreased predictability), late peri-menopausal (menses within the last year but not within the last 3 months), or unknown (due to either HT use or missing Visit 2). Menstrual characteristics at SWAN baseline were considered potential confounders, as menstrual bleeding is an important route of PFAS excretion in women [40–42]. At SWAN baseline, participants self-reported their cycle length (<26 days, 26–32 days, 33–35 days, <35 days, too variable to say) and typical menstrual flow length (1–2 days, 3–7 days, >7 days) during the past year.

Statistical analysis

We calculated means and SDs for continuous variables and frequencies for categorical variables. Due to the right-skewed distribution of the PFAS concentrations, we calculated medians and IQRs for all PFAS. We examined whether PFAS influenced the distribution of MCL, as disruptions to menstrual cyclicity often influence the probability of having both long and short cycles without necessarily altering the median cycle length. Thus, we ran linear quantile mixed models (LQMMs) to examine the associations between PFAS and MCL at the 25th, 50th, 75th, and 90th percentiles. To examine the effects of PFAS on bleeding, we used linear mixed models (LMMs) for bleed length (continuous) and generalized linear mixed models (GLMMs) with a logit link function for binary variables of long bleeding and heavy bleeding. Bleed lengths greater than or equal to 40 days were excluded from the LMMs of bleed length due to being extreme outliers (n = 5). Because of the right-skewed distribution of bleed length, we used robust, empirical standard error estimates in the LMMs.

Models included the two-year window of cycles, beginning with cycles that finished within one year prior to the PFAS measurement at Visit 3 and ending with those that began within one year after Visit 3. All models included a random intercept for each woman, an unstructured covariance, and were adjusted for race/ethnicity, BMI, and menopause status. Models of MCL were additionally adjusted for physical activity, education, and cycle length at SWAN baseline, while bleeding models were additionally adjusted for smoking status and typical menstrual flow duration at SWAN baseline. Covariate selection was based on a priori knowledge and significant associations in bivariate models. Menopause status was dichotomized (late peri vs. pre-menopausal/early peri/unknown due to HT) for models of cycle length but not bleed length due to the differential effect of being early peri versus pre-menopausal on bleed length but not cycle length. Each outcome was regressed on each PFAS separately. PFAS concentrations were log base-2 transformed in all models to capture log-linear dose-response relationships and to allow for interpretations per doubling in PFAS concentration.

We conducted several sensitivity analyses. First, for MCL and bleeding models, we additionally adjusted for the other menstrual characteristics at SWAN baseline (i.e., typical menstrual flow duration for MCL models and cycle length for bleeding models). Second, for bleeding models, we also evaluated the impact of further adjustment for fibroids and prior oral contraceptive use. Third, because menstrual characteristics vary by menopausal status, we performed stratified analyses to evaluate associations separately in pre- and early perimenopausal women. Women in late perimenopause and those with unknown menopausal status due to HT use were excluded from these analyses because of small numbers in these groups.

All analyses were run in SAS 9.4 (SAS Institute, Inc., Cary, North Carolina) except for the LQMMs, which were run in RStudio using the package lqmm (1.5.5) [43,44].

Results

At the time of PFAS exposure assessment, women ranged in age from 45 to 55 years with a mean age of 49 years (Table 1). In the analytical sample, 50.3% of the women were White with other half of the sample including Black (18.0%), Chinese (14.4%), and Japanese women (17.4%). Most women had some college or greater education, were normal weight or underweight, were not current smokers, and did not have fibroids. At the beginning of the analytical window, most women were early peri-menopausal (72.0%), with 22.1% being pre-menopausal and 3.9% being late peri-menopausal; a small portion had unknown status due to HT use or a missed visit (2.0%). Across the 2-year window, individual cycle lengths showed considerable variation (Table 1). The median individual 25th percentile cycle length was 25 days (IQR: 23–27), whereas the 90th percentile was 41 days (IQR: 31–71). The medians (IQRs) of individual bleed length at 25th and 90th percentiles were 5 [4–6] days and 7 [6–9] days, respectively. During the 2-year window, 49.2% and 27.8% of women experienced at least one cycle with a long bleeding episode (> 8 days) and heavy bleeding episode (≥3 days of heavy bleeding), respectively.

The median PFAS concentration was 4.0 ng/mL for n-PFOA; 0.6 ng/mL for PFNA; 1.4 ng/mL for PFHxS; 16.8 ng/mL for n-PFOS; 7.0 for Sm-PFOS, 23.9 ng/mL for PFOS; 1.4 ng/mL for MeFOSAA; and 1.2 ng/mL for EtFOSAA (Table 2).

No PFAS was associated with median MCL (50th percentile) (Table 3). Shorter MCLs at the 25th percentile were associated with doubling of PFOS (−0.52 days; 95% CI, −1.09–0.05); in contrast, PFNA was associated with longer MCLs at the 25th percentile (0.56 days; 95% CI, 0.06–1.05). All PFOS were associated with a longer MCL at the 90th percentiles (1.73 days (95% CI, 0.57–2.90) for n-PFOS; 1.54 days (95% CI, 0.05–3.03) for Sm-PFOS; and 2.77 days (95% CI, 1.96–3.58) for PFOS). The serum concentrations of all other PFAS were not associated with differences in MCL at the 25th, 75th, or 90th percentiles. Additional adjustment for typical flow length at baseline did not alter the results (Table S1).

In adjusted analyses, no significant associations were found with either MeFOSAA or EtFOSAA and any of the bleeding outcomes. Bleed length was inversely associated with a doubling of all other PFAS concentrations except MeFOSAA and EtFOSAA: n-PFOA (−0.29 days; 95% CI, −0.43 to −0.14), PFNA (−0.21 days; 95% CI, −0.34 to −0.09), PFHxS (−0.28 days; 95% CI, −0.37 to −0.18), n-PFOS (−0.28 days; 95% CI, −0.42 to −0.15), Sm-PFOS (−0.20 days; 95% CI, −0.32 to −0.09), and PFOS (−0.29 days; 95% CI, −0.42 to −0.16) (Table 4). Similarly, lower odds of long bleeds were associated with a doubling of all PFAS concentrations except MeFOSAA and EtFOSAA: n-PFOA (OR: 0.75; 95% CI: 0.62–0.89), PFNA (OR: 0.75; 95% CI: 0.63–0.88), PFHxS (OR: 0.72; 95% CI: 0.63–0.82), n-PFOS (OR: 0.68; 95% CI: 0.57–0.82), Sm-PFOS (OR: 0.79; 95% CI: 0.68–0.92), and PFOS (OR: 0.69; 95% CI: 0.57–0.82). Lower odds of heavy bleeding were also associated with a doubling of all PFAS concentrations except MeFOSAA and EtFOSAA: n-PFOA (OR: 0.64; 95% CI: 0.47–0.88), PFNA (OR: 0.64; 95% CI: 0.48–0.85), PFHxS (OR: 0.61; 95% CI: 0.48–0.78), n-PFOS (OR: 0.60; 95% CI: 0.44–0.81), Sm-PFOS (OR: 0.65; 95% CI: 0.50–0.85), and PFOS (OR: 0.58; 95% CI: 0.43–0.79). These associations remain essentially unchanged with further adjustment for baseline cycle length (Table S2), fibroids (Table S3), and oral contraceptive use (Table S4).

In sensitivity analyses stratified by menopausal status, the results were largely consistent between pre- and early perimenopausal women, although some variation was observed in effects by different PFAS and in statistical significance, likely due to smaller sample sizes in each group (Tables S5 and S6).

Discussion

This is among the first longitudinal studies to examine the effect of PFAS on MCL and menstrual bleeding using prospectively collected monthly menstrual calendar data from midlife women undergoing the menopause transition. Several PFAS were associated with the distribution of MCL, with higher PFOS associated with shorter short cycles and longer long cycles, while PFNA was associated with longer short cycles. Higher concentrations of all PFAS except MeFOSAA and EtFOSAA were associated with a shorter duration of menses and lower odds of having a prolonged menses > 8 days or episodes of menses with 3 or more days of very heavy menstrual bleeding. These findings remained robust after additional adjustment for baseline menstrual characteristics and were consistent between premenopausal and early perimenopausal women.

Consistent with most but not all studies, we found that PFAS exposure was associated with increased variance in MCL. The Longitudinal Investigation of Fertility and the Environment (LIFE) Study followed women attempting pregnancy for 12 months [29]. They reported an association between serum PFDA concentrations and 3% longer MCL and between PFOA concentrations and 2% shorter MCL, while PFOS was not associated with MCL. Other previous studies of the association between PFAS and MCL have been cross-sectional, but report associations with menstrual irregularity and long menstrual cycles. In the Inuit-endocrine (INUENDO) three-country cohort, the highest tertile of serum PFOA concentration was associated with a 80% higher odds of having a MCL ≥ 32 days compared to the lowest tertile, while PFOS concentrations were not associated with MCL [28]. In a Chinese study, plasma concentrations of PFOA, PFOS, PFNA and PFHxS were associated with a 30%−80% higher odds of reporting irregular cycles, defined as variation in MCL of > 7 days in the past year, and a 30%−70% higher odds of having MCL > 35 days while no associations were found with perfluorobutane sulfonic acid, perfluoroheptanoic acid, PFDA, PFUnA, PFDoDA or perfluorooctane sulfonamide [30]. In the Danish National Birth Cohort women with higher plasma concentrations of PFOA and PFOS were more likely to report having irregular periods [27]. In a United Kingdom study of women undergoing fertility treatment, higher serum PFOS concentrations were observed among women with irregular versus regular menstrual cycles [45]. A study from the Veneto region in Italy, one of the areas worldwide that is heavily polluted with some PFAS, reported that the frequency of irregular menstrual cycles was significantly higher among highly exposed girls compared to other girls (29.5% versus 21.5%) [46]. In contrast to our findings, a Norwegian study that asked pregnant women to recall characteristics of their menstrual cycles in the year prior to pregnancy reported no association between plasma PFAS concentrations and menstrual regularity or MCL [47]. However, among recent oral contraceptive users, women with long cycles had higher concentrations of PFNA and PFUnDA compared to those with normal length cycles. As oral contraceptives are known to regulate menstrual cycles and reduce menstrual flow, this association may be due to reduced menstruation from recent oral contraceptive use resulting in higher plasma PFAS concentrations [47]. Additionally, a Danish birth cohort study that recruited pregnant women in 1988–1989 and followed their female offspring in 2008–2009 found no associations between prenatal maternal serum concentrations of PFOA and PFOS and MCL of female offspring [48]. The long-term effects of prenatal PFAS exposure on menstrual cycle characteristics, however, remain unclear and warrant further investigation.

Consistent with the fact that menstruation is an important route of elimination for PFAS [40–42], this longitudinal study found that higher serum concentrations of all PFAS–except MeFOSAA and EtFOSAA, which are PFOS precursors with relatively shorter half-lives [36]) and were not associated with any menstrual parameters–were associated with shorter duration of menses and lower odds of having prolonged or heavy bleeding. Notably, these associations were observed even after adjustment for baseline menstrual history that could influence blood PFAS concentrations as well as subsequent menstrual cycles during follow-up. These findings extend results from a prior cross-sectional analysis in the SWAN-MPS that reported higher serum PFAS concentrations in women who had not menstruated in the past three months compared to women who had menstruated [49]. Additionally, in a subset of the SWAN-MPS examining longitudinal trends in serum PFAS concentrations, women menstruating consistently showed lower concentrations of serum n-PFOA, PFNA and Sm-PFOS than women who had not experienced menstrual bleeding since the last visit [50]. The aforementioned Chinese study also found that high plasma concentrations of PFOA, PFOS, PFNA and PFHxS were associated with a 15–60% lower odds of reporting heavy or very heavy menses [30]. The Study of Environment, Lifestyle, and Fibroids also found strong inverse associations between PFHxS (−18.8%), PFOS (−16.4%), PFNA (−10.5%), and PFOA (−10.0%) and intensity of menstrual bleeding when comparing heavy versus light bleeding [51]. Importantly, the present study addresses concerns about reverse causation inherent in cross-sectional studies and supports a link between PFAS exposure and menstrual cycle variability and reduced bleeding, although residual confounding by unmeasured menstrual characteristics may remain.

Without accounting for hormone use, the association between PFAS and menstrual cycle characteristics could be confounded [42]. To mitigate the influence of exogenous HT use, the present study excluded women who reported current use of HT as well as menstrual cycle data occurring during HT use or HT-washout periods.

The precise mechanisms by which PFAS affect the menstrual cycle length and regularity are not yet fully understood, but research suggests that PFAS may act as endocrine disruptors by interfering with ovarian folliculogenesis and steroidogenesis [2,24,42]. In vitro and in vivo studies have documented that PFAS exposure can adversely affect folliculogenesis and steroidogenesis, altering the timing and quality of follicle and oocyte development [2]. PFAS exposure has been linked to impaired proliferation and differentiation of granulosa and theca cells [52,53], both of which are essential for normal follicle development [1]. Specifically, PFAS may disrupt the uptake of cholesterol by theca cells and its conversion to androgens, leading to reduced estradiol production by granulosa cells with resultant decreased endometrial proliferation in the follicular phase of the cycle. Lower total volume of endometrium would manifest as shorter and/or lighter bleeding episodes. Potential molecular mechanisms underlying these effects include activation of peroxisome proliferator-activated receptors, disruption of intercellular communication between oocytes and granulosa cells, interference with thyroid hormone synthesis, and increased of oxidative stress [2].

This study had some limitations. First, at the start of the two-year assessed in this study, women were aged 43–56 years. Although we adjusted for menopausal stage, prior research indicates that PFAS exposures are associated with an earlier age at menopause [25]. Therefore, we cannot rule out the possibility that our findings may, at least in part, reflect menstrual cycle changes associated with loss of ovarian reserve and the transition to menopause, rather than direct effects of PFAS on MCL. It should be noted our findings remain consistent in stratified analyses between premenopausal and early perimenopausal women, suggesting that the impact of menstrual cycle changes among women approaching menopause is minimal. Prospective studies involving younger cohorts of women will provide critical data to confirm whether PFAS exposures directly affect MCL. Findings for each individual chemical require cautious interpretation, as multiple comparisons were not taken into account. Additionally, although we controlled for a range of potential confounding covariates, it is possible that residual or unmeasured confounding might have biased our effect estimates.

This study also had several strengths. Our prospective design with PFAS serum concentrations assessed at baseline and menstrual data measured longitudinally helps address concerns related to reverse causation. Menstrual data were collected prospectively on monthly menstrual calendars, with clear definitions provided for coding the amount of menstrual flow. Consequently, menstrual cycle measures in this study are not subject to recall bias, and thus assessment of MCL and duration of bleeding are likely to be more accurate. Furthermore, this large community-based study included multiple racial and ethnic groups, increasing the generalizability of our results.

In conclusion, this longitudinal study with prospectively recorded menstrual data found that multiple PFAS serum concentrations were associated with increased variance in MCL and reduced menstrual bleeding. These findings support evidence from previous studies that PFAS exposure may alter ovarian function and menstrual cyclicity. Alterations in MCL can serve as early markers of adverse impacts of environmental exposures, with potential implications for a range of reproductive outcomes including fertility and timing of menopause. It also provides confirmatory evidence that menstruation is an important route of PFAS elimination, suggesting the potential health benefits of normal menstrual function in reducing body burden of these persistent environmental contaminants.

Supplementary MaterialAcknowledgements

The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIA, NINR, ORWH, or the NIH.

Clinical Centers: University of Michigan, Ann Arbor – Carrie Karvonen-Gutierrez, PI 2021 – present, Siobán Harlow, PI 2011 – 2021, MaryFran Sowers, PI 1994–2011; Massachusetts General Hospital, Boston, MA – Sherri-Ann Burnett-Bowie, PI 2020 – Present; Joel Finkelstein, PI 1999 – 2020; Robert Neer, PI 1994 – 1999; Rush University, Rush University Medical Center, Chicago, IL – Imke Janssen, PI 2020 – Present; Howard Kravitz, PI 2009 – 2020; Lynda Powell, PI 1994 – 2009; University of California, Davis/Kaiser – Elaine Waetjen and Monique Hedderson, PIs 2020 – Present; Ellen Gold, PI 1994–2020; University of California, Los Angeles – Arun Karlamangla, PI 2020 – Present; Gail Greendale, PI 1994–2020; Albert Einstein College of Medicine, Bronx, NY – Carol Derby, PI 2011 – present, Rachel Wildman, PI 2010 – 2011; Nanette Santoro, PI 2004 – 2010; University of Medicine and Dentistry – New Jersey Medical School, Newark – Gerson Weiss, PI 1994 – 2004; and the University of Pittsburgh, Pittsburgh, PA – Rebecca Thurston, PI 2020 – Present; Karen Matthews, PI 1994–2020.

NIH Program Office: National Institute on Aging, Bethesda, MD – Rosaly Correa-de-Araujo 2020 - present; Chhanda Dutta 2016- present; Winifred Rossi 2012–2016; Sherry Sherman 1994 – 2012; Marcia Ory 1994 – 2001; National Institute of Nursing Research, Bethesda, MD – Program Officers.

Central Laboratory: University of Michigan, Ann Arbor – Daniel McConnell (Central Ligand Assay Satellite Services).

NIA Biorepository - Rosaly Correa-de-Araujo 2019 - Present; SWAN Repository: University of Michigan, Ann Arbor – Siobán Harlow 2013–2018; Dan McConnell 2011–2013; MaryFran Sowers 2000 – 2011.

Coordinating Center: University of Pittsburgh, Pittsburgh, PA – Maria Mori Brooks, PI 2012 - present; Kim Sutton-Tyrrell, PI 2001 – 2012; New England Research Institutes, Watertown, MA - Sonja McKinlay, PI 1995 – 2001.

Steering Committee:Susan Johnson, Current Chair, Chris Gallagher, Former Chair

We thank the study staff at each site and all the women who participated in SWAN.

We also thank the late Ms. Xiaoyun Ye at CDC for her support in PFAS exposure assessment.

Funding statement

The Study of Women’s Health Across the Nation (SWAN) has grant support from the National Institutes of Health (NIH), Department of Health and Human Services, through the National Institute on Aging (NIA), the National Institute of Nursing Research (NINR), and the NIH Office of Research on Women’s Health (ORWH) (Grants U01NR004061, U01AG012505, U01AG012535, U01AG012531, U01AG12539, U01AG012546, U01AG012553, U01AG012554, U01AG012495, and U19AG063720). The study was also supported by the SWAN Repository (U01AG017719). This study was also supported by grants from the National Institute of Environmental Health Sciences (NIEHS) R01-ES026578, R01-ES026964, R01 ES035087, and P30-ES017885, and by the Centers for Disease Control and Prevention (CDC)/National Institute for Occupational Safety and Health (NIOSH) grant T42-OH008455. The content of this article is solely the responsibility of the authors and does not necessarily represent the official views of the NIA, NINR, ORWH, or the NIH.

CRediT authorship contribution statement

Sung Kyun Park: Writing – review & editing, Supervision, Methodology, Investigation, Funding acquisition, Conceptualization. Ellen Gold: Writing – review & editing, Methodology, Investigation. Randolph John: Writing – review & editing, Methodology, Investigation. Antonia Calafat: Writing – review & editing, Methodology, Investigation. Rachael Brooks: Writing – review & editing, Methodology, Investigation. Michelle Hood: Writing – review & editing, Methodology, Formal analysis, Data curation. Siobán Harlow: Writing – original draft, Supervision, Investigation, Conceptualization.

Disclaimer

The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC. Use of trade names is for identification only and does not imply endorsement by the CDC, the Public Health Service, or the US Department of Health and Human Services.

Attestation statements

Data regarding any of the subjects in the study has not been previously published unless specified.

The appropriate checklist for this study design was followed.

Declaration of Competing Interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Data Availability

SWAN provides access to public use datasets that include data from SWAN screening, the baseline visit and follow-up visits (https://agingresearchbiobank.nia.nih.gov/). To preserve participant confidentiality, some, but not all, of the data used for this manuscript are contained in the public use datasets. A link to the public use datasets is also located on the SWAN web site: http://www.swanstudy.org/swan-research/data-access/. Investigators who require assistance accessing the public use dataset may contact the SWAN Coordinating Center at the following email address: swanaccess@edc.pitt.edu.

Supporting information

Supplementary data associated with this article can be found in the online version at doi:10.1016/j.reprotox.2026.109207.

ReferencesHammerKC, VeigaA, MahalingaiahS, Environmental toxicant exposure and menstrual cycle length, Curr. Opin. Endocrinol. Diabetes Obes [Internet] 27 (6) (2020). Available from: 〈https://journals.lww.com/co-endocrinology/fulltext/2020/12000/environmental_toxicant_exposure_and_menstrual.2.aspx〉.DingN, HarlowSD, RandolphJF, Loch-CarusoR, ParkSK, Perfluoroalkyl and polyfluoroalkyl substances (PFAS) and their effects on the ovary, Hum. Reprod. Update 26 (5) (2020) 724–752.32476019 ShulhaiA-M, BiancoV, DoniniV, EspositoS, StreetME, Which is the current knowledge on man-made endocrine- disrupting chemicals in follicular fluid? an overview of effects on ovarian function and reproductive health, Front. Endocrinol [Internet] 15 (2024) 2024. Available from: 〈https://www.frontiersin.org/journals/endocrinology/articles/10.3389/fendo.2024.1435121〉.MendolaP, MesserLC, RappazzoK, Science linking environmental contaminant exposures with fertility and reproductive health impacts in the adult female, Fertil. Steril 89 (2, ement) (2008) e81–e94.18308071 FarrSL, CooperGS, CaiJ, SavitzDA, SandlerDP, Pesticide use and menstrual cycle characteristics among premenopausal women in the agricultural health study, Am. J. Epidemiol 160 (12) (2004) 1194–1204.15583372 WindhamGC, LeeD, MitchellP, AndersonM, PetreasM, LasleyB, Exposure to organochlorine compounds and effects on ovarian function, Epidemiology 16 (2) (2005) 182–190.15703532 OuyangF, PerryMJ, VennersSA, ChenC, WangB, YangF, , Serum DDT, age at menarche, and abnormal menstrual cycle length, Occup. Environ. Med 62 (12) (2005) 878.16299097 PerryMJ, OuyangF, KorrickSA, VennersSA, ChenC, XuX, , A prospective study of serum ddt and progesterone and estrogen levels across the menstrual cycle in nulliparous women of reproductive age, Am. J. Epidemiol 164 (11) (2006) 1056–1064.17023545 ToftG, AxmonA, LindhCH, GiwercmanA, BondeJP, Menstrual cycle characteristics in European and Inuit women exposed to persistent organochlorine pollutants, Hum. Reprod 23 (1) (2008) 193–200.17989067 ZhangY, JiL, HuY, TseLA, WangY, QinK, , Exposure to organophosphate pesticides and menstrual cycle characteristics in chinese preconceptional women, Am. J. Epidemiol 189 (5) (2020) 375–383.31845721 AxmonA, RylanderL, StrombergU, HagmarL, Altered menstrual cycles in women with a high dietary intake of persistent organochlorine compounds, Chemosphere 56 (8) (2004) 813–819.15251296 MendolaP, BuckGM, SeverLE, ZieleznyM, VenaJE, Consumption of PCB-contaminated freshwater fish and shortened menstrual cycle length, Am. J. Epidemiol 146 (11) (1997) 955–960.9400337 ShiX, WangX, PengL, ChenY, LiuC, YangQ, , Associations between polybrominated diphenyl ethers (PBDEs) levels in adipose tissues and female menstrual cycle and menstrual bleeding duration in Shantou, China, Environ. Pollut 301 (2022) 119025.35183664 EskenaziB, WarnerM, MocarelliP, SamuelsS, NeedhamLL, PattersonDGJr, , Serum dioxin concentrations and menstrual cycle characteristics, Am. J. Epidemiol 156 (4) (2002) 383–392.12181109 JukicAM, CalafatAM, McConnaugheyDR, LongneckerMP, HoppinJA, WeinbergCR, , Urinary Concentrations of phthalate metabolites and bisphenol a and associations with follicular-phase length, luteal-phase length, fecundability, and early pregnancy loss, Environ. Health Perspect 124 (3) (2016) 321–328.26161573 ATSDR. Toxicological profile for Perfluoroalkyls. Atlanta, GA: Agency for Toxic Substances and Disease Registry (ATSDR); 2021.ITRC. History and Use of Per-and Polyfluoroalkyl Substances (PFAS). Interstate Technology Regulatory Council (ITRC); 2020.SigmundG, VenierM, ÅgerstrandM, CousinsIT, DeWittJ, DiamondML, , Scientists’ statement on the chemical definition of PFASs (Available from:), Environ. Sci. Technol. Lett [Internet] (2025), 10.1021/acs.estlett.5c00478.ButenhoffJL, OlsenGW, Pfahles-HutchensA, The applicability of biomonitoring data for perfluorooctanesulfonate to the environmental public health continuum, Environ. Health Perspect 114 (11) (2006) 1776–1782.17107867 LinP-ID, CardenasA, HauserR, GoldDR, KleinmanKP, HivertM-F, , Temporal trends of concentrations of per- and polyfluoroalkyl substances among adults with overweight and obesity in the united states: results from the diabetes prevention program and NHANES, Environ. Int 157 (2021) 106789.34333293 KudzinGP, DoddsJN, Kirkwood-DonelsonKI, SchiffenbauerA, SarkarK, Noroozi FarhadiP, , Evaluating legacy and emerging pfas in human blood collected from 2003 to 2021, Environ. Sci. Technol 60 (4) (2026) 3068–3079.41562486 KnoxSS, JacksonT, JavinsB, FrisbeeSJ, ShankarA, DucatmanAM, Implications of early menopause in women exposed to perfluorocarbons, J. Clin. Endocrinol. Metab 96 (6) (2011) 1747–1753.21411548 TaylorKW, HoffmanK, ThayerKA, DanieJL, ls, polyfluoroalkyl chemicals and menopause among women 0–65 years 2 of age (NHANES), Environ. Health Perspect 122 (2) (2014) 145–150.24280566 WuJ, HarlowSD, RandolphJF, GoldEB, ParkSK, Endocrine-disrupting chemicals and female reproductive aging, Semin Reprod. Med 42 (04) (2024) 330–360.39879998 DingN, HarlowSD, RandolphJF, CalafatAM, MukherjeeB, BattermanS, , Associations of perfluoroalkyl substances with incident natural menopause: the study of women’s health across the nation, J. Clin. Endocrinol. Metab 105 (9) (2020) e3169–e3182.32491182 HarlowSD, HoodMM, DingN, MukherjeeB, CalafatAM, RandolphJF, , Per- and Polyfluoroalkyl substances and hormone levels during the menopausal transition, J. Clin. Endocrinol. & Metab 106 (11) (2021) e4427–e4437.34181018 FeiC, McLaughlinJK, LipworthL, OlsenJ, Maternal levels of perfluorinated chemicals and subfecundity, Hum. Reprod 24 (5) (2009) 1200–1205.19176540 LyngsøJ, Ramlau-HansenCH, HøyerBB, StøvringH, BondeJP, JönssonBAG, , Menstrual cycle characteristics in fertile women from Greenland, Poland and Ukraine exposed to perfluorinated chemicals: a cross-sectional study, Hum. Reprod 29 (2) (2014) 359–367.24163265 LumKJ, SundaramR, BarrDB, LouisTA, Buck LouisGM, Perfluoroalkyl chemicals, menstrual cycle length, and fecundity: findings from a prospective pregnancy study, Epidemiol [Internet] 28 (1) (2017). Available from: 〈https://journals.lww.com/epidem/fulltext/2017/01000/perfluoroalkyl_chemicals,_menstrual_cycle_length,.15.aspx〉.WeiZhou, LuluZhang, ChuanliangTong, FangFang, ShashaZhao, YingTian, Plasma Perfluoroalkyl and Polyfluoroalkyl Substances Concentration and Menstrual Cycle Characteristics in Preconception Women. Environmental Health Perspectives 125(6):067012.28657892 ChenY, ZhouL, XuJ, ZhangL, LiM, XieX, , Maternal exposure to perfluorooctanoic acid inhibits luteal function via oxidative stress and apoptosis in pregnant mice, Reprod. Toxicol 69 (2017) 159–166.28219760 RodriguezG, Faundes-LathamA, AtkinsonLE, An approach to the analysis of menstrual patterns in the critical evaluation of contraceptives, Stud. Fam. Plan 7 (2) (1976) 42–51.HarlowSD, CainK, CrawfordS, DennersteinL, LittleR, MitchellES, , Evaluation of four proposed bleeding criteria for the onset of late menopausal transition, J. Clin. Endocrinol. & Metab 91 (9) (2006) 3432–3438.16772350 LebduskaE, BeshearD, SpataroBM, Abnormal uterine bleeding, Med Clin. North Am 107 (2) (2023) 235–246.36759094 MunroMG, CritchleyHOD, FraserIS, the FIGO Menstrual disorders committee. the two figo systems for normal and abnormal uterine bleeding symptoms and classification of causes of abnormal uterine bleeding in the reproductive years: 2018 revisions, Int. J. Gynecol. & Obstet 143 (3) (2018) 393–408.OlsenGW, ChurchTR, MillerJP, BurrisJM, HansenKJ, LundbergJK, , Perfluorooctanesulfonate and other fluorochemicals in the serum of American Red Cross adult blood donors, Environ. Health Perspect 111 (16) (2003) 1892–1901.14644663 KatoK, BasdenBJ, NeedhamLL, CalafatAM, Improved selectivity for the analysis of maternal serum and cord serum for polyfluoroalkyl chemicals, J. Chromatogr. A 1218 (15) (2011) 2133–2137.21084089 SternfeldB, AinsworthBE, QuesenberryCP, Physical activity patterns in a diverse population of women, Prev. Med 28 (3) (1999) 313–323.10072751 BaeckeJA, BuremaJ, FrijtersJE, A short questionnaire for the measurement of habitual physical activity in epidemiological studies, Am. J. Clin. Nutr 36 (5) (1982) 936–942.7137077 UpsonK, ShearstonJA, KioumourtzoglouM-A, An epidemiologic review of menstrual blood loss as an excretion route for per- and polyfluoroalkyl substances, Curr. Environ. Health Rep 9 (1) (2022) 29–37.35267175 WongF, MacLeodM, MuellerJF, CousinsIT, Enhanced elimination of perfluorooctane sulfonic acid by menstruating women: evidence from population-based pharmacokinetic modeling, Environ. Sci. Technol 48 (15) (2014) 8807–8814.24943117 RickardBP, RizviI, FentonSE, Per- and poly-fluoroalkyl substances (PFAS) and female reproductive outcomes: PFAS elimination, endocrine-mediated effects, and disease, Toxicology 465 (2022) 153031.34774661 GeraciM, Linear Quantile Mixed Models: The lqmm Package for Laplace Quantile Regression, J. Stat. Soft 57 (13) (2014) 1–29.GeraciM, BottaiM, Linear quantile mixed models, Stat. Comput 24 (3) (2014) 461–479.HeffernanAL, CunninghamTK, DrageDS, AylwardLL, ThompsonK, VijayasarathyS, , Perfluorinated alkyl acids in the serum and follicular fluid of UK women with and without polycystic ovarian syndrome undergoing fertility treatment and associations with hormonal and metabolic parameters, Int J. Hyg. Environ. Health 221 (7) (2018) 1068–1075.30037723 Di NisioA, RoccaMS, SabovicI, De Rocco PonceM, CorsiniC, GuidolinD, , Perfluorooctanoic acid alters progesterone activity in human endometrial cells and induces reproductive alterations in young women, Chemosphere 242 (2020) 125208.31896193 SingerAB, WhitworthKW, HaugLS, SabaredzovicA, ImpinenA, PapadopoulouE, , Menstrual cycle characteristics as determinants of plasma concentrations of perfluoroalkyl substances (PFASs) in the Norwegian Mother and Child Cohort (MoBa study), Environ. Res 166 (2018) 78–85.29879567 KristensenSL, Ramlau-HansenCH, ErnstE, OlsenSF, BondeJP, VestedA, , Long-term effects of prenatal exposure to perfluoroalkyl substances on female reproduction, Hum. Reprod 28 (12) (2013) 3337–3348.24129614 ParkSK, PengQ, DingN, MukherjeeB, HarlowSD, Determinants of per- and polyfluoroalkyl substances (PFAS) in midlife women: evidence of racial/ethnic and geographic differences in PFAS exposure, Environ. Res 175 (2019) 186–199.31129528 DingN, HarlowSD, BattermanS, MukherjeeB, ParkSK, Longitudinal trends in perfluoroalkyl and polyfluoroalkyl substances among multiethnic midlife women from 1999 to 2011: The Study of Women′s Health Across the Nation, Environ. Int 135 (2020) 105381.31841808 WiseLA, WesselinkAK, SchildrothS, CalafatAM, BetheaTN, GellerRJ, , Correlates of plasma concentrations of per- and poly-fluoroalkyl substances among reproductive-aged Black women, Environ. Res 203 (2022) 111860.34403666 ClarkKL, ShuklaM, GeorgeJW, GustinS, RowleyMJ, DavisJS, An environmentally relevant mixture of per- and polyfluoroalkyl substances (PFAS) impacts proliferation, steroid hormone synthesis, and gene transcription in primary human granulosa cells, Toxicol. Sci 200 (1) (2024) 57–69.38603627 YangM, LeeY, GaoL, ChiuK, MelingDD, FlawsJA, , Perfluorooctanoic Acid Disrupts Ovarian Steroidogenesis and Folliculogenesis in Adult Mice, Toxicol. Sci 186 (2) (2022) 260–268.35104888

Characteristics of study sample in the Study of Women’s Health Across the Nation (N = 951).

At follow-up visit 2 (1998–1999)
Mean (SD)
Age, years48.0 (2.4)
Physical activity score (possible range 3–15)7.8 (1.7)
N (%)
Race/ethnicity
 Black171 (18.0)
 Chinese137 (14.4)
 Japanese165 (17.4)
 White478 (50.3)
Education
 High school or less162 (17.1)
 Some college or more785 (82.9)
Body mass index
 < 25 kg/m2448 (47.1)
 25–29.9 kg/m2244 (25.7)
 ≥ 30 kg/m2259 (27.2)
Current smoker
 No863 (90.7)
 Yes88 (9.3)
Menopause status
 Pre-menopause210 (22.1)
 Early peri-menopause685 (72.0)
 Late peri-menopause37 (3.9)
 Unknown due to HT use19 (2.0)
Fibroids
 No872 (91.7)
 Yes79 (8.3)
Oral contraceptive use ever
 No244 (25.7)
 Yes707 (74.3)
Site
 Southeast Michigan148 (15.6)
 Boston, Massachusetts145 (15.2)
 Oakland, California221 (23.2)
 Los Angeles, California284 (29.9)
 Pittsburgh, Pennsylvania153 (16.1)
At baseline (1996–1997)
N (%)
Length of typical menstrual flow during the last year
 1–2 days14 (1.5)
 3–7 days900 (94.6)
 More than 7 days37 (3.9)
Average cycle length during typical menstrual cycle during the last year
 Less than 26 days219 (23.4)
 26–32 days660 (70.6)
 33–35 days17 (1.8)
 More than 35 days18 (1.9)
 Too variable to say21 (2.3)
Longitudinal measures during 2-year window
Median (IQR)
Individual 25th percentile cycle length, days25 (23–27)
Individual 50th percentile cycle length, days28 (26–31)
Individual 75th percentile cycle length, days31 (28–45)
Individual 90th percentile cycle length, days41 (31–71)
Individual 25th percentile bleed length, days5 (4–6)
Individual 50th percentile bleed length, days6 (5–7)
Individual 75th percentile bleed length, days7 (5–8)
Individual 90th percentile bleed length, days7 (6–9)
N (%)
At least 1 cycle with bleed length > 8 days468 (49.2)
At least 1 cycle with heavy bleeding on 3 or more days264 (27.8)

Medians and interquartile ranges (IQRs) of PFAS concentrations (ng/mL) in serum collected in 1999–2000 for the Study of Women’s Health Across the Nation (N = 951).

PFASMedian (IQR)
n-PFOA4.0 (2.7–5.7)
PFNA0.6 (0.4–0.8)
PFHxS1.4 (0.9–2.2)
n-PFOS16.8 (12.0–23.8)
Sm-PFOS7.0 (4.5–10.4)
PFOS23.9 (17.1–34.7)
MeFOSAA1.4 (0.9–2.3)
EtFOSAA1.2 (0.7–2.2)

n-PFOA (linear perfluorooctanoic acid), PFNA (perfluorononanoic acid), PFHxS (perfluorohexane sulfonic acid), n-PFOS (linear perfluorooctane sulfonic acid), Sm-PFOS (branched perfluorooctane sulfonic acid isomers), MeFOSAA (2-(N-methyl-perfluorooctane sulfonamido) acetic acid), and EtFOSAA (2-(N-ethylperfluorooctane sulfonamido) acetic acid).

PFOS is the sum of n-PFOS and Sm-PFOS.

Differences in cycle length per doubling of PFAS serum concentrations at the 25th, 50th, 75th, and 90th percentiles of cycle length in the Study of Women’s Health Across the Nation (N = 931).

25th Percentile50th Percentile75th Percentile90th Percentile
PFASβ (95% CI)p-valueβ (95% CI)p-valueβ (95% CI)p-valueβ (95% CI)p-value
n-PFOA0.05 (−0.47, 0.58)0.8360.11 (−0.20, 0.42)0.487−0.05 (−0.83, 0.74)0.9050.53 (−1.33, 2.40)0.568
PFNA0.56 (0.06, 1.05)0.0280.10 (−0.32, 0.53)0.632−0.14 (−0.68, 0.40)0.610−0.01 (−1.45, 1.43)0.992
PFHxS−0.19 (−0.68, 0.31)0.4580.08 (−0.16, 0.31)0.5080.05 (−0.25, 0.35)0.7570.23 (−0.68, 1.14)0.617
n-PFOS−0.50 (−1.20, 0.20)0.161−0.28 (−0.86, 0.29)0.3250.65 (−0.21, 1.51)0.1361.73 (0.57, 2.90)0.004
Sm-PFOS−0.32 (−0.72, 0.08)0.113−0.03 (−0.31, 0.25)0.8340.38 (−0.25, 1.01)0.2321.54 (0.05, 3.03)0.043
PFOS−0.52 (−1.09, 0.05)0.074−0.31 (−0.85, 0.22)0.2470.85 (−0.11, 1.80)0.0832.77 (1.96, 3.58)< 0.001
MeFOSAA0.35 (−0.18, 0.87)0.189−0.04 (−0.33, 0.24)0.771−0.21 (−0.56, 0.14)0.234−0.32 (−1.53, 0.88)0.591
EtFOSAA0.19 (−0.12, 0.49)0.222−0.12 (−0.33, 0.09)0.261−0.24 (−0.54, 0.06)0.108−0.20 (−1.05, 0.66)0.649

β coefficients and 95% confidence intervals (CIs) were calculated using linear quantile mixed models (LQMMs). All models adjusted for race/ethnicity, menopause status, body mass index, physical activity, education, and baseline cycle length.

n-PFOA (linear perfluorooctanoic acid), PFNA (perfluorononanoic acid), PFHxS (perfluorohexane sulfonic acid), n-PFOS (linear perfluorooctane sulfonic acid), Sm-PFOS (branched perfluorooctane sulfonic acid isomers), MeFOSAA (2-(N-methyl-perfluorooctane sulfonamido) acetic acid), and EtFOSAA (2-(N-ethyl-perfluorooctane sulfonamido) acetic acid). PFOS is the sum of n-PFOS and Sm-PFOS.

Differences in bleed length and odds of long or heavy bleeds per doubling of PFAS serum concentration in the Study of Women’s Health Across the Nation (N = 951).

Bleed LengthLong BleedHeavy Bleed
PFASβ (95% CI)p-valueOR (95% CI)p-valueOR (95% CI)p-value
n-PFOA−0.29 (−0.43, −0.14)< 0.0010.75 (0.62, 0.89)0.0010.64 (0.47, 0.88)0.006
PFNA−0.21 (−0.34, −0.09)0.0010.75 (0.63, 0.88)< 0.0010.64 (0.48, 0.85)0.002
PFHxS−0.28 (−0.37, −0.18)< 0.0010.72 (0.63, 0.82)< 0.0010.61 (0.48, 0.78)< 0.001
n-PFOS−0.28 (−0.42, −0.15)< 0.0010.68 (0.57, 0.82)< 0.0010.60 (0.44, 0.81)0.001
Sm-PFOS−0.20 (−0.32, −0.09)< 0.0010.79 (0.68, 0.92)0.0020.65 (0.50, 0.85)0.002
PFOS−0.29 (−0.42, −0.16)< 0.0010.69 (0.57, 0.82)< 0.0010.58 (0.43, 0.79)< 0.001
MeFOSAA−0.02 (−0.13, 0.08)0.6590.97 (0.83, 1.12)0.6660.91 (0.70, 1.18)0.480
EtFOSAA0.00 (−0.08, 0.09)0.9750.97 (0.87, 1.09)0.5900.95 (0.78, 1.15)0.590

β coefficients or odds ratios (ORs) along with 95% confidence intervals (CIs) were calculated using linear mixed models for continuous bleed length and generalized linear mixed models with a logit link function for binary long bleed and heavy bleed. All models adjusted for race/ethnicity, menopause status, body mass index, smoking status, and typical flow length at baseline.

n-PFOA (linear perfluorooctanoic acid), PFNA (perfluorononanoic acid), PFHxS (perfluorohexane sulfonic acid), n-PFOS (linear perfluorooctane sulfonic acid), Sm-PFOS (branched perfluorooctane sulfonic acid isomers), MeFOSAA (2-(N-methyl-perfluorooctane sulfonamido) acetic acid), and EtFOSAA (2-(N-ethyl-perfluorooctane sulfonamido) acetic acid). PFOS is the sum of n-PFOS and Sm-PFOS.