Hormone Fluctuations, Dietary Intake, and Daily Activity in Premenopausal Women: An 8-Week Observational Study.

The menstrual cycle is a series of hormonal changes that follow an infradian rhythm each month ( 1 , 2 ), resulting in physiological alterations ( 3 ) and physical symptoms ( 4 , 5 ). These alterations have been thought to influence numerous physiological systems in exercise science research, introducing intra- and intervariability. Fluctuations in endogenous sex hormones could confound study outcomes and have contributed to a history of both naturally menstruating (NM) women and those using hormonal contraceptives (HC) being excluded from research ( 6 – 8 ). Further, previous data, including women, have often been collected without incorporating standard protocols for menstrual cycle tracking over multiple menstrual cycles ( 9 , 10 ). The menstrual cycle is divided into two distinct phases: follicular, from the first day of bleeding (menstruation) to the end of ovulation ( 11 ), and luteal, from the first day after ovulation to the start of menstruation. Each phase is characterized by a rise or fall of hormones, including estradiol, progesterone, follicle-stimulating hormone (FSH), and luteinizing hormone (LH). During menstruation, estradiol, progesterone, and LH are at their lowest levels in the cycle, and FSH begins to rise to support the maturation of developing follicles. As these follicles develop, they produce increasing levels of estradiol, and when estradiol peaks, it triggers an LH surge leading to ovulation. Following ovulation, progesterone peaks in the luteal phase before decreasing as menstruation approaches. Sex hormone binding globulin (SHBG), a known modulator of androgens and estrogens, affects the bioavailability of estrogens ( 12 ). Greater levels of SHBG increase the amount of estrogens bound to SHBG, thus decreasing the amount of free circulating estrogens ( 12 ). Throughout the menstrual cycle, SHBG stays rather stable, but may fluctuate in response to changes in estrogens, particularly during the luteal phase, following the pre-ovulation peak in estradiol ( 12 ). HCs are typically designed to stabilize hormone levels and suppress ovulation by applying synthetic estradiol and progestin, inhibiting an LH surge ( 13 ). As a result, HC users often experience lower intraindividual hormonal variability across cycles ( 13 ). HCs that contain synthetic estradiol can significantly increase SHBG production, which may inherently reduce time fluctuations, but overall elevate SHBG, especially when compared to NM controls ( 14 ). Importantly, women using HC still often report menstrual cycle–related symptoms, including fatigue, mood changes, and bloating, which may impact their everyday behavior, such as physical activity and dietary choices ( 15 ). Fluctuations in physical activity and sedentary behavior across the menstrual cycle may be attributed to negative premenstrual symptoms, although these effects are highly individual ( 7 , 16 ). However, physical activity may also help alleviate symptoms associated with the menstrual cycle, and higher levels of physical activity have been linked to improvements in premenstrual symptoms and cycle characteristics such as shorter menses duration, resolution of luteal phase deficiencies, and promotion of normative cycle length ( 17 ). Specifically, the luteal phase is often marked by feelings of fatigue, bloating, and discomfort, which may indirectly result in decreased physical activity, increased sedentary behavior, and decreased exercise engagement ( 18 ). Much of this research ( 7 , 17 ) has utilized self-report methods to estimate menstrual cycle phase as opposed to confirming phase with ovulation assessments, thereby limiting the ability to determine specific-phase fluctuations in behavior. Higher levels of physical activity and exercise are positively associated with SHBG levels, and this association is largely mediated by the amount of adipose tissue ( 19 ). In response to potential fluctuations in physical activity and exercise by menstrual cycle phase, SHBG may fluctuate, given an individual’s activity level. Further, because SHBG is regulated in part by energy status and metabolic signaling, dietary intake may further modulate its role in female hormone fluctuations. In parallel, dietary intake may be related to hormonal fluctuations throughout the menstrual cycle due to potential impacts on appetite, food cravings, and overall eating patterns ( 20 ), particularly during the luteal phase ( 21 ). However, data are often conflicting, highlighting the complexity of menstrual cycle research, and the potential contribution of genetic factors ( 22 , 23 ). For example, some data suggest that energy consumption and macronutrient intake were not significantly different by phase ( 22 ), whereas others indicated that protein intake was increased during the luteal phase, but not other macronutrients or total intake ( 23 ). Moreover, a recent meta-analysis found that resting metabolic rate increases modestly during the luteal phase; however, subgroup analysis of studies published since 2000 reported a smaller effect that was no longer significant ( 24 ). Estrogens may play a notable role in female obesity, and increased bioavailable estrogens have been associated with decreased energy consumption ( 25 ). In addition, progestins may be associated with binge eating behaviors primarily in the luteal phase ( 25 ), suggesting potential complex relationships with dietary intake and female reproductive hormones. Leptin, a key satiety hormone that plays a pivotal role in energy balance, may fluctuate during the menstrual cycle, potentially contributing towards changes in dietary intake ( 26 ). Leptin interacts with hypothalamic pituitary signaling and plays an integrative role in the regulation of LH, FSH, and ovarian function ( 27 ). Leptin levels are influenced by estradiol and progesterone, with estradiol enhancing leptin sensitivity and progesterone contributing to luteal-phase increases ( 26 , 27 ). This interplay illustrates how changes in energy status and leptin signaling can influence reproductive hormone patterns across the menstrual cycle. Overall, dietary intake may be related to hormone fluctuations throughout the menstrual cycle, and leptin may be indirectly related to the secretion of LH, FSH, and estradiol. Women experience a greater sleep need compared to men and require more sleep per night for optimal functioning ( 28 ). However, many American women fail to meet their sleep needs, with young-adult women experiencing more severe sleep disturbances and poorer overall sleep quality than their male counterparts ( 29 ). It is unclear how sleep patterns change during the menstrual cycle ( 30 ) or with HC use ( 31 ). Some research suggests that sleep stages may be influenced by hormonal fluctuations associated with the menstrual cycle ( 32 ), while other studies have found no significant variation in sleep patterns or sleep quality across menstrual cycle phases ( 30 , 33 ). Some evidence exists that suggests that progesterone increases during the luteal phase are associated with greater sleep disturbances as measured by polysomnography ( 34 ). However, these studies often lack confirmation of menstrual cycle phase, and further high-quality scientific data is needed. Additionally, the use of HC may alter sleep patterns, though findings are mixed on whether HC improves or worsens sleep compared to NM women ( 31 ). Emerging mechanistic evidence suggests that these hormones may influence behavior and metabolism. Estradiol has been shown to enhance glucose uptake, increase fat oxidation, and modulate leptin within mice models ( 35 ). Progesterone, in contrast, is associated with elevations in core body temperature and shifts in substrate utilization towards greater reliance on lipids within mice models, which may contribute to fatigue, decreased energy efficiency, and changes in sleep quality during the luteal phase ( 36 , 37 ). Together, these mechanisms suggest plausible biological pathways through which menstrual cycle phase or HC use may alter physical activity and dietary intake. Despite growing recognition of potential interplays among female reproductive hormones, dietary intake, physical activity, and sleep quality, a dearth of evidence exists examining how these factors interact throughout the menstrual cycle, particularly among women with a natural menstrual cycle and those using HC. Understanding these interactions is crucial for understanding female health. The present study aimed to investigate the relationships between hormone levels and dietary intake, physical activity, exercise, sedentary behavior, and sleep quality while accounting for menstrual cycle phase and HC usage. Secondarily, we aimed to examine whether dietary intake, physical activity, exercise, sedentary behavior, and sleep differed by menstrual cycle phase.

Healthy, premenopausal women ( n = 16) aged 18–40 yr with a body mass index (BMI) between 18.5 and 40 kg·m −2 were recruited to participate in this 8-wk observational study (Fig. 1 ). Two participants in the NM group were excluded from data analysis due to having an incomplete menstrual cycle and blood measurements. Both NM and HC users were included. NM participants included any women who self-reported having a normal menstrual cycle (≥21 and ≤35 d, and ≥9 cycles·yr −1 ) or those who were using a copper intrauterine device ( 9 ). Participants were excluded if they had discontinued HC use in the last 6 months ( 10 ), had amenorrhea, oligomenorrhea (menstrual cycle >35 d), an irregular menstrual cycle (<8 menstrual cycle·yr −1 ), currently pregnant, diagnosed with polycystic ovarian syndrome, endometriosis, or fibroids, a self-reported untreated eating disorder, or weight change ±3.5 kg in the last 2 months. All participants gave written informed consent before participation, and the study was approved by the University of Idaho Institutional Review Board (#23-177) in accordance with the Declaration of Helsinki. CONSORT Flow Diagram. CONSORT flow diagram illustrating participant recruitment, enrollment, and reasons for data exclusion. After giving informed consent at visit 0, participants completed eight laboratory visits (visits 1–8) over an 8-wk period between the hours of 0800 and 1200. This schedule was designed to capture data across two consecutive hormonal cycles. Time of day (±2 h) and day of the week (7 d ± 1 d) for visits 1–8 remained consistent for all participants (Fig. 2 ). Weekly study procedures. At visit 1, participants underwent a dual-energy X-ray absorptiometry (DXA) scan to assess body composition. From visits 1 through 8, participants completed weekly assessments, including the Automated Self-Administered 24-h Dietary Recall (ASA-24), the Pittsburgh Sleep Quality Index (PSQI), the Simple Physical Activity Questionnaire (SIMPAQ), and provided a blood sample at each visit. NM participants were asked to monitor their cycles for 4 wk before visit 0 using a mobile tracking application (Flo Health Inc.) and at-home ovulation test kits (Clearblue Advanced Digital Ovulation Test). Participants shared the data with the researchers to confirm NM status, a regular menstrual cycle, and ovulation. Following the initial monitoring period, NM participants continued tracking and returned to the laboratory weekly for visits 1–8. Each visit was classified based on ovulation testing and mobile app tracking following phase definitions outlined by Elliot-Sale et al. ( 9 ): menstrual phase — days of active bleeding; follicular phase — post-menstrual visits occurring before ovulation; and luteal phase — post-ovulation visits occurring before the onset of the next bleed. HC users followed the same visit schedule. Visits that occurred during the withdrawal bleed were classified as the menstrual phase, aligning with the NM “bleeding” phase. Subsequent visits were classified based on the typical 28-d pill cycle, with the active pill phase corresponding to the follicular phase and the luteal-equivalent phase defined as the latter part of the active pill cycle before the next withdrawal bleed. If a participant did not experience a withdrawal bleed, the initial visit of a new cycle was classified as the follicular phase equivalent. Anthropometrics and body composition were measured at visit 1. Height was measured to the nearest 0.01 cm (InBody BSM170), and weight was measured to the nearest 0.01 kg (Detecto apex. Body composition was assessed using dual-energy X-ray absorptiometry (DXA; Hologic DXA Scanner, Hologic Horizon). One whole-body scan was performed while lying supine, and testing was completed according to the manufacturer’s directions and specifications, and scans were analyzed with APEX software (v. 4.5.2.1). The test–retest coefficient of variation (CV; %) was 1.1% for lean mass and 0.69% for fat mass. Participants completed eight weekly 24-h dietary recalls using the validated Automated Self-Administered 24-h Dietary Recall (ASA-24, version 2022) developed by the National Cancer Institute. Any 24-h recall that was <500 kcal was deemed incomplete ( 38 ) and excluded from subsequent analyses. Total calorie intake (kcal), carbohydrate (g), total fat (g), protein (g), and added sugars (g) were included for analyses. Participants completed eight weekly Simple Physical Activity Questionnaires (SIMPAQ) to assess time spent engaging in physical activity, sedentary, and exercise behaviors ( 39 ). The SIMPAQ consists of eight questions related to sleep and wake time, time spent in sedentary behaviors, and time spent walking, exercising, and time spent in any other physical activities excluding walking or exercising over the past 7 d ( 39 ). Researchers summed the time spent in each domain for a total of 24 h·d −1 . Participants completed eight weekly Pittsburgh Sleep Quality Index (PSQI) questionnaires to assess week-to-week changes in sleep quality ( 40 ). The PSQI is a validated self-report metric to evaluate sleep quality and fatigue, which asks participants 10 questions related to sleep disturbances they may experience. Although originally validated for a 1-month recall period, prior studies have used the PSQI with a 1-wk recall and reported adequate psychometric performance ( 41 ). Researchers then calculated sleep duration, sleep disturbances, sleep efficiency, and overall sleep quality to determine the total PSQI score for each participant ( 40 ). Internal consistency for all PSQI components was acceptable (Cronbach’s α = 0.74), supporting its reliability for use on a weekly basis in this sample. Participants underwent eight weekly blood draws from an antecubital vein by a trained phlebotomist into a 9.5 mL serum separator vacutainer tube (SST; Becton Dickinson) at visits 1–8. Samples were centrifuged for 15 min at 1300 × g (Hamilton Bell, VanGuard V6500). The supernatant was transferred into 1.5 mL aliquots using disposable pipets and stored at −80°C until analysis. Serum hormone (progesterone, LH, SHBG, estradiol) concentrations were measured in duplicate using individual, commercially available ELISA kits (Monobind, Inc.). Calibrated, electronic pipets were used to transfer samples and add reagents. The intra-assay CVs for each hormone were <6% for each hormone analysis. Serum leptin samples were diluted and analyzed using manufacturer-supplied diluent before analysis using an automated assay and were measured in triplicate (Ella Bio-Techne). The intra-assay CV for leptin was <4%. All hormone testing was completed by the Center for Bioanalytical and Bionutrient Discovery at the University of North Texas. Visits were classified as “bleeding/phase 1,” “follicular/phase 2,” or “luteal/phase 3” for analysis purposes. Statistical analyses were performed using commercially available open-source statistical software (R; Version 4.4.1). Before analysis, outliers that were not physiologically plausible were removed. This resulted in one incomplete dietary recall being removed and two E2 hormone measurements from two participants being removed. Mixed-effects models were used for primary analyses and are relatively robust to violations of normality assumptions; therefore, formal testing of normality was not required ( 42 ). Linear mixed-effects models were first used to determine whether body composition and dietary intake predicted hormone concentrations. Mixed-effects models were also used to determine if hormone concentrations predicted physical activity behaviors and sleep quality. Separate models were run for the NM and HC groups. Repeated measures correlations were used to determine relationships between hormone concentrations and lifestyle and dietary metrics. Additional mixed-effects models were used to evaluate differences in hormone concentrations and lifestyle factors across menstrual cycle phases for both NM and HC. Models were fitted with group (NM vs HC), menstrual phase (bleeding, follicular, luteal), and their interactions as fixed effects. Type 3 sums of squares with Satterthwaite approximation for degrees of freedom were used to evaluate main effects and interactions. When significant effects were observed, post hoc comparisons were conducted using estimated marginal means with Holm correction. Participant ID was included as a random intercept in all models to account for repeated measures within individuals. An alpha level of 0.05 was used to determine statistical significance for all models ( 42 ).

Participant demographics are displayed in Table 1 , and most participants were Caucasian (87.5%). Two participants were excluded from data analysis; one was excluded due to an irregular menstrual cycle, and the other for missing blood samples at all time points. Participant demographics at baseline. Most participants had a BMI in the normal-weight range (71%). There were no significant differences between NM and HC for any body composition variable. In both NM and HC participants, body fat percentage was a significant predictor of leptin concentrations (NM: β = 1.8 μg·L −1 per 1% increase in body fat, P = 0.002; HC: β = 1.6 μg·L −1 ; P = 0.016). Additionally, lean body mass was a significant predictor of SHBG concentrations in NM only (β = 7.8 nmol·L −1 , P = 0.013). Hormone concentrations by group and cycle phase are displayed in Table 2 . There was a significant phase main effect ( F (2, 91.90) = 8.10, P = 0.0006) and interaction effect for progesterone ( F (2, 91.90) = 10.23, P < 0.0001). Post hoc tests revealed progesterone was significantly higher in the luteal phase compared with both the follicular phase (difference = 33.7 nmol·L −1 ; P <0.0001) and during menses (difference = 35.5 nmol·L −1 ; P = 0.0001) (Figure 1, Supplemental Digital Content 1, https://links.lww.com/TJACSM/A338 ). There were no differences in progesterone for the HC group ( P > 0.05). There were no significant main effects or interactions for estradiol, SHBG, LH, or leptin. Hormone concentrations by group over the menstrual cycle. Bold and italics indicates statistical significance. *Denotes significantly lower than all other phases ( P < 0.05) FSH, follicle stimulating hormone; HC, hormonal contraceptive users; IU/L, international units/liter; LH, luteinizing hormone; SHBG, sex hormone binding globulin; mg/L, micrograms/liter; NM, naturally menstruating; nmol/L, nanomoles/liter; pmol/L, picomoles/liter. There were no significant group or time main effects and no significant interactions for any marker of dietary intake (Table 3 ). Dietary intake across the menstrual cycle. Daily calorie intake was a significant positive predictor of estradiol (β = 0.77 pmol·L −1 ·kcal −1 , P = 0.023), and daily carbohydrate intake was a significant negative predictor of LH (β = −0.0051 IU·L −1 ·g −1 of Carbohydrate, P = 0.013) Daily added sugar consumption positively predicted estradiol (β = 69.04 pmol·L −1 ·g −1 of added sugar; P = 0.001), progesterone (β = 1.14 nmol/·L −1 ·g −1 of added sugar, P = 0.006), and SHBG (β = 1.28 nmol·L −1 ·g −1 of added sugar, P = 0.003), and negatively predicted LH (β = −0.042 IU·L −1 ·g −1 of added sugar, P = 0.015). Further, daily protein intake was a significant negative predictor of SHBG (β = −0.304 nmol L −1 ·g −1 of added sugar, P = 0.032) and total fat intake was a significant positive predictor of estradiol concentrations (β = 18.15 pmol L −1 ·g −1 of added sugar, P = 0.003). Correlational analyses in NM participants revealed that the total caloric intake was positively associated with estradiol ( r = 0.351; 95% confidence interval [CI] 0.064–0.584; P = 0.018; Fig. 3 A) and negatively associated with LH ( r = −0.327; 95% CI −0.566 to −0.037; P = 0.029; Fig. 3 B). Additionally, total fat intake was positively associated with estradiol ( r = 0.427; 95% CI 0.152–0.640; P = 0.003; Fig. 3 C). Total carbohydrate intake was negatively associated with LH ( r = −0.406; 95% CI −0.625 to −0.128; P = 0.006; Fig. 3 D). Similarly, added sugar intake was negatively associated with LH ( r = −0.380; 95% CI −0.606 to −0.097; P = 0.010; Fig. 3 E) but positively associated with progesterone ( r = 0.429; 95% CI 0.155–0.642; P = 0.003; Fig. 3 F), estradiol ( r = 0.477; 95% CI 0.213–0.677; P = 0.0009; Fig. 3 G), and SHBG ( r = 0.427; 95% CI 0.153–0.640; P = 0.003; Fig. 3 H). Repeated measures correlations among naturally menstruating participants between (A) daily total caloric intake (kcal·d −1 ) and estradiol concentrations, (B) daily total caloric intake (kcal·d −1 ) and luteinizing hormone (LH), (C) daily total fat intake and estradiol, (D) daily total carbohydrate intake and LH, (E) daily added sugar intake LH, (F) daily added sugar and progesterone, (G) daily added sugar and estradiol, and (H) daily added sugar and sex hormone binding globulin (SHBG).The thick black line represents the overall trend across all participants. The thin gray lines represent the trend for individuals over multiple time points. The gray dots represent data points for individuals at each time point. When only analyzing data during the luteal phase, negative associations were observed between total caloric (Fig. 4 A) and carbohydrate (Fig. 4 B) intake and LH concentrations (kcal: r = −0.567; 95% CI −0.836 to −0.076; P = 0.028; carbohydrate: r = −0.590; 95% CI −0.846 to −0.111; P = 0.021), and between total fat intake and leptin concentrations ( r = −0.644; 95% CI −0.869 to −0.196; P = 0.010; Fig. 4 C). Further, there was a positive relationship between added sugar intake and SHBG concentration ( r = 0.618; 95% CI 0.155–0.858; P = 0.014; Fig. 4 D). In contrast, during the follicular phase, total fat intake was positively correlated with estradiol ( r = 0.589; 95% CI 0.110–0.846; P = 0.021; Fig. 4 E). Repeated measures correlations among naturally menstruating participants during only the luteal phase between (A) daily total caloric intake (kcal) and luteinizing hormone (LH), (B) daily total carbohydrate intake and LH, (C) daily total fat intake and leptin, and (D) daily added sugar intake and sex hormone binding globulin (SHBG). Repeated measures correlations among naturally menstruating participants during only the follicular phase between total fat intake and estradiol (E) and for between total fat intake and leptin in hormonal contraceptive participants (F). The thick black line represents the overall trend across all participants. The thin gray lines represent the trend for individuals over multiple time points. The gray dots represent data points for individuals at each time point. Daily total fat intake was a significant positive predictor of leptin concentrations (β = 0.054 µg·L −1 ·g −1 of fat intake, P = 0.025). Further, there were significant correlations between these two variables ( r = 0.348; 95% CI 0.060–0.582; P = 0.019; Fig. 4 F). No other dietary behavior predicted hormone concentrations, and there were no other significant correlations with dietary behaviors among HC participants. There were no significant differences in time spent engaging in sedentary, walking, or exercise behaviors between NM and HC, and no differences between menstrual cycle phases (Table 4 ). In NM, higher leptin concentrations were associated with slightly greater sedentary time (β = 0.072 h·ng −1 ·mL −1 leptin, P = 0.048). Between-participant variability in sedentary time was larger (~2.83 h) than within-person day-to-day variability (~1.79 h). However, there were no significant correlations between physical activity behaviors and hormone concentrations in either NM or HC. Activity and sleep behaviors by menstrual cycle phase. AU, arbitrary unit. There were no differences in global PSQI scores between NM and HC participants or between menstrual cycle phases (Table 4 ). Global PSQI scores were not associated with any hormone in either group.

The primary aim of the present exploratory study was to investigate interactions among lifestyle behaviors and potential relationships to progesterone, estradiol, LH, SHBG, and leptin among NM and HC premenopausal women. Notably, our study confirmed multiple group-specific associations between hormone concentrations and both body composition and dietary intake; however, we did not observe group-by-phase interactions or main effects. Although prior studies have compared mean dietary intake across menstrual cycle phases and reported higher caloric, fat, and carbohydrate consumption during the follicular and luteal phases, our findings remain consistent with this literature when viewed through a hormonal lens ( 43 , 44 ). Estradiol concentrations naturally peak in the late follicular phase and rise again during the luteal phase, and in our sample, higher estradiol levels were similarly associated with greater total energy, fat, and added sugar intake. Additionally, we identified that total fat intake was positively associated with estradiol in the follicular phase. During this phase, as ovarian follicles mature and estradiol production rises, higher dietary fat intake has been associated with greater circulating estradiol concentrations ( 23 , 43 , 44 ). Our finding that higher progesterone concentrations predicted greater added sugar intake is consistent with a substantial body of literature linking luteal-phase hormones and increased preference for sweet, carbohydrate-rich foods. Several observational and experimental studies report elevated cravings and intake of sweets in the luteal phase, when progesterone is high, and link those behavioral changes to ovarian hormone fluctuations ( 26 , 45 ). In NM women, higher carbohydrate and added sugar intake was associated with lower LH, a hormone that peaks in the late follicular phase in preparation for ovulation. To the authors’ knowledge, while exploratory, this is a novel and unique finding as dietary associations with LH have previously only been investigated in the context of fiber ( 46 ), sodium ( 47 ), or fat intake ( 48 ). The present study did not identify any significant associations between LH and total fat intake, corroborating findings of Mumford et al. ( 48 ), who also examined reproductive hormones across 8 weeks. Importantly, associations between LH and dietary intake may be largely driven by body composition, and in particular, body fat mass. Thus, these findings warrant further investigation among individuals with low and high body fat percentages. In contrast to previous findings by Huang et al. ( 49 ), our study identified a positive association between added sugar intake and SHBG among NM participants. However, those findings were observed in a postmenopausal population, during which SHBG levels are markedly increased due to lower circulating androgens and estrogens, and therefore may not be directly comparable or applicable to the present premenopausal study population ( 49 , 50 ). Among participants using HCs, our study observed that leptin was positively associated with body fat percentage and total fat intake. A secondary analysis suggested a potential relationship between body fat percentage and total fat intake; however, this was nonsignificant ( P = 0.0673). Previous research has identified that when HC users have increased circulating leptin levels compared with control, which may explain why we identified this finding among HC participants and not those who are NM ( 51 ). Further, although leptin was positively associated with total fat intake among HC participants, this relationship was weak and was characterized by a wide CI. This may explain why a similar association was not observed in NM participants. We found that higher protein intake was associated with lower SHBG concentrations in NM participants. This aligns with prior research in men and pubertal adolescents showing that greater animal protein consumption is linked to reduced hepatic SHBG production ( 52 , 53 ). We identified no significant associations in dietary intake by menstrual cycle phase, consistent with findings of Bryant et al. and Gorcyzca et al., who found that while diet intake modestly changed throughout the menstrual cycle, these changes were not significant by menstrual cycle phase ( 22 , 23 ). Our findings were contradictory to those of Brown et al. and Johnson et al. ( 43 , 44 ). Indeed, the findings of Brown et al. ( 43 ) may be limited as they asked participants to self-categorize their intake as “less than usual,” “about the same,” or “more than usual,” rather than reporting the specific foods and quantities consumed. This method provides only a subjective and highly generalized estimate of dietary behavior, which lacks the precision needed to evaluate meaningful changes in macronutrient or energy intake. Further, the findings of Johnson et al. ( 44 ) indicated a 686 kJ difference between the follicular and luteal phases, which converts to approximately 160 kcal, suggesting a relatively small change in total energy intake. Our study identified no significant differences in dietary intake between HC and NM participants. Previous research has indicated that women using HCs may have greater overall carbohydrate intake compared to NM women ( 54 ), while others have indicated that HCs may reduce cycle-related fluctuations in appetite and food intake ( 55 ). Overall, previous findings examining dietary intake between HC and NM participants are inconsistent, and the lack of observed differences in our sample aligns with the broader evidence indicating that any HC-related changes in dietary intake are subtle, highly variable, and not consistent across populations or study designs. The present study identified a significant positive association between leptin concentrations and sedentary time in NM participants only. Leptin, primarily produced by adipose tissue, is often reflected by total body fat mass. Individuals with greater body fat mass tend to accumulate more sedentary time due to less time active ( 56 ). Accordingly, our finding that higher leptin levels were associated with slightly greater sedentary duration likely reflects these underlying physiological and behavioral patterns. Although the effect size was small, the direction of association is consistent with literature linking higher leptin and adiposity to lower habitual physical activity and greater sedentary behavior ( 56 , 57 ). Aside from leptin, the present study did not identify any other significant associations with self-reported walking, exercise, sedentary behavior, or sleep by menstrual cycle phase. In both athletic and nonathletic populations, women often report avoiding exercise or physical activity due to symptoms associated with the menstrual cycle ( 58 , 59 ). Numerous methods of measuring physical activity and exercise have been used in previous research, including self-report methods, accelerometry, and smart devices, potentially explaining our inconsistent findings ( 58 , 59 ). In agreement with objective findings by Alzueta et al. ( 30 ), our study identified no significant differences in sleep quality by menstrual cycle phase. Conversely, research has suggested that sleep quality may improve during the follicular phase and decline during the luteal phase ( 60 ). Given the established relationship between physical activity and exercise and sleep quality, it is perhaps not surprising that perceived sleep quality did not differ by menstrual cycle phase, particularly since physical activity and exercise remained unchanged. Importantly, these findings suggest that perceived self-reported sleep quality does not appear to change in healthy young women across the menstrual cycle, indicating that hormonal fluctuations may not significantly impact perceived sleep experiences among NM or HC. While we did not observe any differences in perceived sleep quality across menstrual cycle phase or between NM and HC users, this finding should be interpreted within the limitations of subjective sleep reporting and high interindividuality. The lack of phase-based differences found in the present study challenges earlier findings that claim reduced physical activity or impaired sleep during the luteal phase, in addition to assumptions that behavioral patterns fluctuate predictably across the menstrual cycle. While previous models have indicated strong hormonal/behavioral relationships based on estradiol or progesterone concentrations, recent literature suggests that phase-based effects are small, inconsistent, and/or highly individual ( 61 , 62 ). Our findings align with this evolving perspective, contributing to a growing body of literature suggesting that menstrual behavioral responses may be more heterogeneous and context-dependent than previously assumed. This study is strengthened by the inclusion of both NM and HC women, populations that have historically been excluded from exercise science research ( 6 – 8 ). The study employed an observational study design and included two menstrual cycles, providing eight time points for all participants. Additionally, the study was comprehensive and measured dietary intake, self-reported physical activity, sleep quality, and hormone concentrations each week, improving our ability to examine potential relationships over time. Despite these strengths, there are a few limitations to address. Importantly, the study was exploratory, and several outcomes in this study were primarily assessed with self-report measures. Inherently, this may limit sensitivity to detect subtle changes across the menstrual cycle and may introduce participant fatigue. Our analyses only included 14 participants, the majority of whom were Caucasian, which may limit the generalizability of the present findings. Due to the weekly testing schedule, it is possible that participants’ hormonal fluctuations were not captured precisely, resulting in considerable variability in hormonal profiles. Current evidence suggests that menstrual cycles have a high intra- and intervariability, and it may be premature to apply our findings to all women of different cultural and ethnic backgrounds. Unintentionally, most of the included participants had a normal-weight BMI, potentially limiting the variability that could occur between participants in different BMI groups, such as underweight, overweight, or obese. Future research should employ objective measures of physical activity and sleep to monitor daily behavior in addition to subjective measures, thus improving the knowledge of within- and between-phase variability. Future research would be strengthened by including a daily and weekly symptom questionnaire to examine how behaviors change in relation to hormonal cycle symptoms. This exploratory study demonstrated that estradiol positively predicted daily calorie intake, added sugar intake, and was positively associated with total fat intake and progesterone positively added sugar intake among NM women, but not women using HCs. Our study identified positive associations between total fat intake and leptin among HC women, but not NM women. Leptin was associated with greater sedentary time in NM women only. Self-reported sleep quality, walking, and exercise behavior did not differ by menstrual cycle phase and are not associated with any hormone fluctuations. These findings are novel and important in providing a holistic view of the menstrual cycle, helping to bridge the gap between physiological changes and behavioral patterns in healthy premenopausal women. Importantly, menstrual cycles vary greatly across individuals in both timing and hormonal patterns, underscoring that behavioral and hormonal relationships are unlikely to be uniform across all women. Additional research examining food cravings and objective measurements of physical activity and sleep is warranted to further examine potential complex interactions between behavior and hormonal fluctuations across multiple menstrual cycles. The authors would like to thank the undergraduate research assistants in the Human Performance Laboratory involved in the data collection of this project. A. A. B. C, A. J. C., and A. B. were supported by an Institutional Development Award from the National Institute of General Medical Sciences of the National Institutes of Health under Grant #P20GM152304 during preparation of this manuscript. The contents are solely the responsibility of the authors and do not necessarily reflect the official views of the NIH. Author contributions: Conceptualization, A. A. B. C., A. F. B.; methodology, A. A. B. C., A. F. B.; formal analysis, A. A. B. C., A. J. C.; investigation, A. A. B. C., G. I., M. S.; data curation, A. A. B. C.; writing—original draft preparation, A. A. B. C., A. J. C.; writing—review and editing, A. A. B. C., A. J. C., G. I., M. S.; supervision, A. F. B. All authors have read and agreed to the published version of the manuscript. The authors report conflicts of interest. The results of the present study do not constitute endorsement by the American College of Sports Medicine.