Sleep-related factors and circulating levels of sex hormones in premenopausal Japanese women

Abstract

Sleep disruption and circadian disruption have been proposed to be risk factors of breast cancer. The present study examined the associations of sleep-related factors, referring to night shift work, sleep habits, and sleep disturbances, with the plasma levels of sex hormones in premenopausal Japanese women. Study participants were 432 women who had regular menstrual cycles less than 40 days long. Information on their history of night shift work and sleep disturbances was obtained using a self-administered questionnaire. Information on their sleep habits, such as usual wake-up times, bedtimes, and ambient light level while sleeping, was obtained in an interview. The participants’ height and weight were measured. Plasma concentrations of estradiol, testosterone, dehydroepiandrosterone sulfate, sex hormone-binding globulin (SHBG), FSH, and LH were also measured. After controlling for the phase of the menstrual cycle and other covariates, more years of night shift work ≥ once a week during the past 10 years was significantly associated with a lower SHBG and a higher free estradiol level. Shorter sleep duration was significantly associated with the higher total, bioavailable, and free testosterone levels. Sleep disturbance by awaking after sleep onset was significantly associated with a high free estradiol level. The data suggest that long-term night shift work, short sleep duration, and arousal during sleep are associated with higher estradiol or testosterone levels in premenopausal women.

ENDOGENOUS SEX HORMONES are known to play a contributory role in the etiology of breast cancer. Indeed, high levels of circulating estrogens and androgens were associated with an increased risk of premenopausal breast cancer in a recent meta-analysis [1]. Sleep disruption and circadian disruption have been proposed as risk factors for breast cancer [2]. Specifically, these factors include sleep disturbance, sleep duration, night shift work, and exposure to light at night. The relationship of these factors to breast cancer risk has not been established. If these factors are associated with the risk of breast cancer, elevated levels of endogenous sex hormones may be the intermediates in the pathway. However, studies on the associations of these factors with sex hormone levels are still scarce, especially among premenopausal women, and none in Japan. We previously assessed the associations of sleep status at midnight, night shift work, and ambient light exposure while sleeping to sex hormone levels among postmenopausal Japanese women [3]. In the present study, we examined whether these sleep-related factors, additionally including sleep duration and sleep disturbances, are associated with sex hormone levels in premenopausal Japanese women.

Subjects and Methods

Study population

This study was part of one designed to assess the relationships among lifestyle, environmental factors, and women’s health, as described elsewhere [4]. Study subjects were participants in a medical health check-up program provided by a general hospital in Gifu, Japan, between October 2003 and March 2006. A total of 1,103 women participated in the study with a response rate of 74.5%.The study was approved by the ethical board of the Gifu University Graduate School of Medicine (No. 25–57). Written informed consent was obtained from all subjects.

Measurements

Each woman responded to a self-administered questionnaire asking for information on their demographic characteristics, smoking and drinking habits, sleep characteristics, history of night shift work, diet, physical activity, and medical and reproductive histories. Women who reported that they had worked night shifts were asked the dates and the frequency of night shifts for each period. Sleep disturbances were assessed by asking participants how often, over the past month, they had experienced trouble falling asleep for >30 min. after going to bed and how often they woke up in the middle of the night or early morning. If the participants reported “≥3 times/week”, they were categorized as having respective sleep disturbance. Sleep medication use was also based on self-reported use ≥3 times/week. These questions were developed by Doi et al. [5] based on the Pittsburgh Sleep Quality Index [6].

Data on usual sleep habits and the ambient light level in the bedroom were collected through face-to-face interviews. At the interview, participants were asked to report their current usual bedtimes and wake-up times on weekdays and weekends and how long these habits continued to be stable. Based on this information, sleep duration and sleep status at or after 1:00 a.m. were identified. The ambient light level in the bedroom while sleeping was asked in six response categs adopted from the study by Davis et al. [7] (from level 1, the subjects wore an eye mask to keep out light, to level 6, she could read comfortably). Height and weight were measured. Fasting blood samples were drawn from participants around at 8:00 a.m. of the same day and plasma were stored at –80°C until the assay.

The present study restricted study subjects to women who were not pregnant or breastfeeding and had regular menstrual cycles of less than 40 days. The first day of the ongoing menstrual cycle was recorded. A total of 501 women who were not pregnant or breastfeeding reported that they had regular menstrual cycles of less than 40 days. Referring to the study by Verkasalo et al. [8], subjects whose date of blood donation differed by more than 40 days from the onset of the last menses were not included in the analysis (n = 20). Furthermore, women were excluded if they were using oral contraceptives, hormone therapy, or steroid (n = 16), if they had cancer, diabetes mellitus, chronic hepatitis, or thyroid disease (n = 22), or if they had not responded to the questions on history of night shift work (n = 4). Blood samples were not obtained from 6 women. Thus, 432 women aged 20–54 years comprised the study population.

Plasma estradiol and testosterone were measured using electro chemiluminescent immunoassay with kits purchased from the Roche Diagnostic Japan, Tokyo, Japan. Plasma dehydroepiandrosterone sulfate (DHEAS) was measured by chemiluminescent enzyme immunoassay using kits purchased from the Beckman Coulter, Tokyo, Japan. Plasma LH and FSH were measured using chemiluminescent immunoassay with kits purchased the Abbot Japan Co. Ltd., Tokyo, Japan. Sex hormone-binding globulin (SHBG) were measured using immunoradiometric assay with kits purchased from Diagnostic Products Corporation, Los Angeles, USA. The interassay coefficients of variation (CV) were ≤3.5% for estradiol, ≤3.6% for testosterone, ≤10.6% for DHEAS, ≤7.9% for SHBG, ≤5.7% for LH, and ≤5.3% for FSH. Plasma SHBG was not measured in one woman and LH and FSH, were not measured in 14 women, because of insufficient volume. Free estradiol and bioavailable and free testosterone were calculated, respectively, using the measured estradiol or testosterone, albumin, and SHBG concentrations [9].

Statistical analyses

Hormone concentrations were logarithmically transformed. The geometric means of hormones for categorized variables were calculated using analysis of covariate models. A linear trend was assessed using continuous values. The day of the menstrual cycle at blood donation was determined by ‘backward dating’ counted backwards from the estimated date of her next menses. Details are described elsewhere [10]. The day of the menstrual cycle at blood donation was categorized into the early follicular (days ≤–21 of the cycle), late follicular (days –20 to –17), peri-ovulatory (days –16 to –13), early luteal (days –12 to –9), midluteal (days –8 to –4), or late luteal (days ≥–3) phases. The distribution of premenopausal women according to the day of cycle was follows: days ≤–21 of the cycle, n = 98; days –20 to –17, n = 60; days –16 to –13, n = 66; days –12 to –9, n = 67; days –8 to –4, n = 84; and days ≥–3, n = 57. Age, the phase of the menstrual cycle at blood donation, body mass index, smoking status, alcohol intake, parity, and history of breast feeding were included as covariates in models to assess the associations between study variables and hormone levels. Significance was defined as two-sided p < 0.05. All the statistical analyses were performed using SAS programs (SAS Institute Inc., Cary, NC, USA).

In sensitivity analyses, we repeated analyses after excluding women who reported partial hysterectomy or oophorectomy. We also repeated the analyses after restricting study subjects to those who donated blood samples in the follicular phase or those who had menstrual cycles of 25–38 days and had FSH levels ≤10 IU/L. For the relations between sleep disturbance by waking after sleep onset and hormone levels, we repeated analyses after excluding women who were taking sleeping pills. Total sleep time calculated as the difference between bedtime and wake-up time, minus the time to fall asleep was reported to correlate with actigraphy-assessed sleep duration (r = 0.57) [11]. Although information on the time to fall asleep was not obtained, an analysis was conducted for the relations between sleep duration and hormone levels after restricting study subjects to those who reported no difficulty in falling asleep.

Results

Table 1 shows the characteristics of the study subjects. Only 23 women (5.3%) had ever worked a night shift, including nine current workers. The means of sleep duration were 6.5 and 7.4 hours on weekdays and weekends, respectively. The prevalence of sleep disturbances were less than 3%. None of the women used sleep medication ≥3 times /week.

Table 1 Characteristics of 432 premenopausal women

Variables
Age (years)	39.8 ± 5.6
Body mass index (kg/m2)	20.9 ± 2.7
Current smokers, n (%)	25 (5.8)
Ex-smokers, n (%)	17 (3.93)
Alcohol intake (g/d)	7.9 ± 17.1
Age at menarche (years)	12.8 ± 1.2
Age at regular menstrual cycle started (years)	17.6 ± 6.7
Nulliparous, n (%)	92 (21.3)
History of breast feeding, n (%)	326 (75.5)
Night shift work
Current/former night shift workers, n (%)	23 (5.3)
Total duration of night shift work ≥ once/week (years) in the past 10 years† (years)	2.9 ± 3.2
Sleep
Bedtime on weekdays	11:44 ± 0:59
Wake-up time on weekdays	6:11 ± 0:43
Bedtime on weekends	11:51 ± 1:01
Wake-up time on weekends	7:15 ± 1:09
Duration of sleep on weekdays (hrs.)	6.5 ± 0.9
Duration of sleep on weekends (hrs.)	7.4 ± 1.2
Difficulty in falling asleep for >30 min. n (%)	10 (2.3)
Awaking after sleep onset, n (%)	12 (2.9)
Blood levels
Total estradiol (pg/mL)	104.2 (21–517)
Free estradiol (pg/mL)	2.08 (0.43–10.1)
Testosterone (ng/dL)	25.7 (8.9–74.6)
Bioavailable testosterone (ng/dL)	10.6 (2.9–38.3)
Free testosterone (ng/dL)	0.38 (0.11–1.37)
Dehydroepiandrosterone sulfate (μg/dL)	123.3 (53.0–286.8)
Sex hormone binding globulin (nmol/L)	61.2 (16.0–233.9)
FSH (IU/L)	4.88 (1.35–17.6)
LH (IU/L)	4.06 (0.87–18.7)

Continuous variables are expressed as means ± standard deviation or geometric means with 95% confidence interval. Categorical variables are expressed as numbers with percentage.

^† Among current/former night shift workers

History of night shift work was not associated with any hormone level after controlling for covariates (Table 2). However, when night shift work was evaluated as the total duration of the work with at least once a week in past 10 years, women who had worked night shifts for four or more years had significantly lower SHBG and higher free estradiol levels than those who had never worked a night shift during the past 10 years (p = 0.003 and 0.03, respectively; p values are not shown in the table). The trends were also significant (p for trend = 0.0001 and 0.03, respectively).

Table 2 Adjusted geometric means of sex hormones by night shift work and sleep characteristics in premenopausal Japanese women

Variables	n	Total E2 (pg/mL)	Free E2 (pg/mL)	Total T (ng/dL)	Bioavailable T (ng/dL)	Free T (ng/dL)	DHEAS (μg/dL)	SHBG (nmol/L)	FSH (IU/L)	LH (IU/L)
Night shift work
History of nightshift work
Never	409	104.6	2.07	25.8	10.6	0.38	123.6	61.9	4.91	4.05
Ever	23	99.8	2.10	24.0	10.6	0.38	119.4	51.0	4.46	4.18
p for difference		0.77	0.93	0.52	0.96	0.96	0.70	0.17	0.46	0.84
Total duration of night shift work ≥ once/week during past 10 years
0 yr	417	103.8	2.06	25.6	10.5	0.38	122.9	62.7	4.90	4.05
1–3 yr	8	72.6	1.39	25.5	10.4	0.37	132.8	70.7	5.05	3.59
4+ yr	5	186.4	4.42	26.6	14.1	0.51	132.5	25.9	3.88	4.34
p for trend		0.12	0.03	0.98	0.27	0.28	0.73	0.0001	0.36	0.87
Sleep habits
Wake-up time on weekdays
3:00–5:59 a.m.	97	103.2	2.08	25.7	11.0	0.39	125.6	59.8	4.79	3.80
6:00–6:59 a.m.	270	107.8	2.15	25.7	10.5	0.38	121.3	61.0	4.87	4.11
7:00–11:30 a.m.	53	89.9	1.73	24.2	9.4	0.34	122.6	67.2	5.48	4.53
p for trend		0.21	0.06	0.03	0.003	0.003	0.52	0.13	0.16	0.32
Bedtime on weekdays
9:00–10:59 p.m.	47	111.7	2.25	24.2	10.2	0.39	112.2	59.5	4.94	4.22
11:00 p.m.–00:59 a.m.	319	103.0	2.05	25.6	10.5	0.38	123.5	61.3	4.94	4.07
1:00–4:00 a.m.	54	105.5	2.08	26.1	10.4	0.38	125.8	64.2	4.87	4.11
p for trend		0.86	0.93	0.35	0.45	0.39	0.10	0.97	0.56	0.57
Wake-up time on weekends
3:00–5:59 a.m.	33	91.1	1.81	22.0	9.1	0.33	125.6	61.8	5.64	3.39
6:00–6:59 a.m.	99	104.7	2.12	28.1	11.8	0.43	121.3	57.3	4.77	4.03
7:00–11:30 a.m.	287	105.6	2.09	25.1	10.2	0.37	122.6	63.0	4.90	4.07
p for trend		0.79	0.93	0.94	0.82	0.81	0.66	0.49	0.89	0.99
Bedtime on weekends
9:00–10:59 p.m.	42	102.1	2.02	24.5	10.0	0.36	114.1	62.7	5.07	3.85
11:00 p.m.–00:59 a.m.	302	104.9	2.08	25.6	10.5	0.38	121.5	61.6	4.94	4.06
1:00–4:00 a.m.	75	102.4	2.04	25.7	10.6	0.38	130.9	60.3	4.81	4.32
p for trend		0.82	0.77	0.97	0.96	0.93	0.14	0.99	0.81	0.26
At or after 1:00 a.m. on weekdays
Asleep	366	104.1	2.07	25.5	10.5	0.38	117.0	61.1	4.94	4.09
Awake	54	105.5	2.08	26.1	10.4	0.38	112.5	64.2	4.87	4.11
p for difference		0.91	0.97	0.76	0.95	0.96	0.62	0.60	0.87	0.95
At or after 1:00 a.m. on weekends
Asleep	344	104.5	2.08	25.5	10.4	0.38	120.6	61.8	4.96	4.04
Awake	75	102.4	2.04	25.7	10.6	0.38	130.9	60.3	4.81	4.32
p for difference		0.84	0.88	0.92	0.86	0.87	0.32	0.78	0.70	0.50
Sleep duration on weekdays
<6.0 hrs.	95	109.0	2.22	26.7	11.3	0.41	127.5	60.9	4.72	4.07
6.0–6.9 hrs.	167	107.6	2.15	27.2	11.3	0.41	124.8	58.3	4.89	4.14
7.0–7.9 hrs.	128	95.2	1.86	23.5	9.4	0.34	121.0	65.5	5.15	3.93
8.0+ hrs.	30	112.1	2.13	22.2	0.87	0.31	101.9	64.9	4.87	4.60
p for trend		0.34	0.16	0.02	0.006	0.005	0.07	0.38	0.51	0.86
Sleep duration on weekends
<6.0 hrs.	30	88.9	1.84	22.8	9.8	0.36	127.5	56.6	5.10	4.64
6.0–6.9 hrs.	78	103.6	2.04	26.0	10.6	0.38	121.8	62.9	4.82	4.03
7.0–7.9 hrs.	152	111.5	2.24	26.5	11.1	0.40	127.7	59.1	4.97	4.08
8.0+ hrs.	160	101.2	1.98	25.0	10.0	0.36	117.0	64.1	4.91	4.03
p for trend		0.97	0.86	0.90	0.83	0.80	0.44	0.52	0.78	0.39
Ambient light level in the bedroom
Dark	76	108.3	2.14	24.6	10.1	0.36	119.2	60.7	4.78	3.97
Middle	327	103.6	2.06	25.5	10.5	0.38	123.3	61.7	4.92	4.06
Light	11	100.1	2.02	32.4	13.4	0.49	124.0	64.6	6.54	6.32
p for trend		0.34	0.42	0.13	0.17	0.15	0.46	0.93	0.24	0.23
Sleep disturbance
Difficulty in falling asleep for >30 min. ≥3 times/week
No	418	104.3	2.08	25.7	10.6	0.38	123.4	61.3	4.91	4.10
Yes	10	109.2	2.17	25.6	10.8	0.38	128.7	50.2	4.10	3.25
p for difference		0.85	0.85	0.98	0.93	0.98	0.74	0.33	0.34	0.33
Waking after sleep onset ≥3 times/week
No	407	102.3	2.05	25.7	10.6	0.38	123.0	60.4	4.94	4.09
Yes	12	156.0	3.18	26.6	11.1	0.40	123.7	58.9	3.92	4.03
p for difference		0.06	0.046	0.82	0.82	0.78	0.96	0.89	0.21	0.95

Data are adjusted for age, body mass index, smoking status, alcohol intake, parity, breastfeeding, and the day of the menstrual cycle using analysis of covariance models.

E2, estradiol; T, testosterone; DHEAS, dehydroepiandrosterone sulfate; SHBG, sex hormone binding globulin.

An earlier wake-up time on weekdays was significantly associated with higher total and free-testosterone levels. A shorter sleep duration on weekdays, but not on weekends, was significantly associated with an increase in total, bioavailable, and free testosterone levels. The trends for bioavailable and free testosterone were also significant when the mean sleep duration throughout weekdays and weekends was calculated; the geometric means for <6.0 vs. ≥8.0 hrs. were 26.4 vs. 22.7 ng/dL for total testosterone (p for trend = 0.08), 11.3 vs. 8.6 ng/dL (p for trend = 0.03), and 0.40 vs. 0.31 ng/dL for free testosterone (p for trend = 0.03).

Women who had sleep disturbance by waking after sleep onset had a significantly higher free estradiol level. This association remained significant after additional adjustment for duration of working night shift ≥ once/week; the geometric means of free estradiol was 3.20 and 2.04 pg/mL in women with and without this disturbance, respectively (p = 0.04). The association between duration of working night shift ≥ once/week and free estradiol level also remained significant after adjustment for sleep disturbance by waking after sleep onset (p for trend = 0.03).

The results from sensitivity analyses are shown in Supplementary Tables 1–3. The findings were not altered substantially except that the associations between sleep duration and testosterone levels were attenuated when study subjects were restricted to only 158 women who donated blood samples in the follicular phase (Supplementary Table 2).

Discussion

In the present study of premenopausal Japanese women, a longer duration of night shift work at least once per week in the past 10 years was associated with higher free estradiol and lower SHBG levels. The shorter sleep duration was associated with higher levels of total, bioavailable, and free testosterone and sleep disturbance by awaking after sleep onset was associated with a high free estradiol level. These sleep-related factors could be potential risk factors for breast cancer, since estrogen and testosterone levels have been linked to the risk of breast cancer. Our findings also suggest that premenopausal women with these sleep-related characteristics may have altered hormonal profiles.

Our previous study of postmenopausal Japanese women, the experience of night shift work was associated with a higher total estradiol level. To our knowledge, eight studies have compared sex hormone levels between night shift and non-night shift workers in normal premenopausal women [12-19]. Some of them observed that night shift work or longer duration of night shift work was associated with higher estradiol [12-14] or testosterone [15] levels, but the others did not [16-19]. There has been no report on night shift work and SHBG level. Although the observed positive association of duration of night shift work with the free estradiol level may support the reported increased breast cancer risk among night shift workers, our observation may reflect the association of duration of night shift work with the SHBG level. SHBG has been associated with metabolic factors [20]. On the other hand, increased risk of diabetes and metabolic syndromes among night shift workers has been suggested [21]. Alteration of the metabolic profile due to night shift work may have led to an increased free estradiol level.

Only one study reported a relation between sleep duration and sex hormone levels (urinary gonadotropin and estrogen and progesterone metabolites) among premenopausal women [22]. In this study, a short-duration sleeper (<8 hrs.) had a lower urinary FSH level than long-duration sleepers (≥8 hrs.), but there was no differences in urinary estrone-3-glucuronide, the primary urinary metabolite of estradiol. Our finding of the higher total and free testosterone associated with the shorter sleep duration may suggest that shorter sleep is a risk factor for premenopausal breast cancer. So far, two prospective studies have evaluated the association between sleep duration and risk of breast cancer among Japanese women. Both studies showed that short sleep duration (≤6 hrs.), as compared with 7 hrs. [23] or ≥8 hrs, [24] of sleep, was associated with an increased risk of breast cancer. However, other studies outside of Japan reported mixed findings, as reviewed by Samuelsson et al. [2].

The relationship between sleep disturbance and sex hormone levels has been well studied among postmenopausal women or women who were in menopausal transition [25]. Only five studies have reported this relationship among premenopausal women [26-30]. Two of them had findings contrary to our results; sleep efficiency: the ratio of the time of sleep between initial sleep onset and morning awakening to the total sleep time, measured by actigraphy, was significantly positively associated with urinary estrone-3-glucronide [26]. The number of arousals after sleep onset was significantly inversely associated with estradiol level [27]. However, in an intervention study, partial sleep deprivation by awakening at 1:30 a.m. increased estradiol level [28]. As speculated by Baumgartner et al. [28], sleep disturbance may induce enhanced noradrenergic and dopaminergic activity, which in turn may cause alternation in the secretion of estrogen. Some prospective studies observed that having difficulty sleeping (i.e., waking frequently and not falling back to sleep) was associated with an increased risk of breast cancer [31, 32], which may not contradict our results. However, studies on breast cancer risk and sleep characteristics other than sleep duration are scarce [33], and our study captured only some measures of sleep disruption.

Limitations of this study include measurements of hormone levels in a single blood sample in time without restriction to the day of the cycle, which may have caused large measurement errors for hormone levels and misclassification of the phase of the cycle for the adjustment. The history of night shift work and sleep characteristics were ascertained by self-report. Although the associations between sleep duration and total and free testosterone levels were not substantially altered among those who have never had trouble falling asleep, there could be measurement errors. However, it is unlikely that such measurement errors are dependent on hormone levels. The present study’s sample size is comparable to those in some other studies that used untimed blood samples with adjustment for the phase of the menstrual cycle [15, 34, 35]. However, the numbers of night shift workers and women having sleep disturbance were still small.

In conclusion, we observed that long-term night shift work, shorter sleep duration, and sleep disturbance due to awaking after sleep onset were associated with circulating estradiol or testosterone levels in premenopausal Japanese women. However, the sample size of the present study was small, and the associations were not consistently observed among the variables related to night shift work and sleep characteristics, thus not permitting any strong conclusions. Larger studies with more accurate measurements of exposure and hormone levels among premenopausal women are needed.

Disclosures

Funding

This work was supported by a grant from the Ministry of Education, Culture, Sports, Science, and Technology, Japan (grant numbers 19390172 and 22150001).

Conflict of interest

None of the authors have any potential conflicts of interest associated with this research.

References

• 1  Endogenous Hormones and Breast Cancer Collaborative Group,  Key  TJ,  Appleby  PN,  Reeves  GK,  Travis  RC, et al. (2013) Sex hormones and risk of breast cancer in premenopausal women: a collaborative reanalysis of individual participant data from seven prospective studies. Lancet Oncol 14: 1009–1019.
• 2   Samuelsson  LB,  Bovbjerg  DH,  Roecklein  KA,  Hall  MH (2018) Sleep and circadian disruption and incident breast cancer risk: an evidence-based and theoretical review. Neurosci Biobehave Rev 84: 35–48.
• 3   Nagata  C,  Nagao  Y,  Yamamoto  S,  Shibuya  C,  Kashiki  Y, et al. (2008) Light exposure at night, urinary 6-sulfatoxymelatonin, and serum estrogens and androgens in postmenopausal Japanese women. Cancer Epidemiol Biomarkers Prev 17: 1418–1423.
• 4   Nagata  C,  Nakamura  K,  Oba  S,  Hayashi  M,  Takeda  N, et al. (2009) Association of intakes of fat, dietary fibre, soya isoflavones and alcohol with uterine fibroids in Japanese women. Br J Nutr 101: 1427–1431.
• 5   Doi  Y,  Minowa  M,  Okawa  M,  Uchiyama  M (1998) Development of the Japanese version of the Pittsburgh sleep quality index. Jpn J Psychiatr Treat 13: 755–763 (In Japanese).
• 6   Buysse  DJ,  Reynolds CF  III,  Monk  TH,  Berman  SR,  Kupfer  DJ (1989) The Pittsburgh sleep quality index: a new instrument for psychiatric practice and research. Psychiatry Res 28: 193–213.
• 7   Davis  S,  Kaune  WT,  Mirick  DK,  Chen  C,  Stevens  RG (2001) Residential magnetic fields, light-at-night, and nocturnal urinary 6-sulfatoxymelatonin concentration in women. Am J Epidemiol 154: 591–600.
• 8   Verkasalo  PK,  Thomas  HV,  Appleby  PN,  Davey  GK,  Key  TJ (2001) Circulating levels of sex hormones and their relation to risk factors for breast cancer: a cross-sectional study in 1092 pre- and postmenopausal women (United Kingdom). Cancer Causes Control 12: 47–59.
• 9   Sodergard  R,  Backstrom  T,  Shanbhag  V,  Carstensen  H (1982) Calculation of free and bound fractions of testosterone and estradiol-17 beta to human plasma proteins at body temperature. J Steroid Biochem 16: 801–810.
• 10   Nagata  C,  Wada  K,  Nakamura  K,  Hayashi  M,  Takeda  N, et al. (2011) Associations of body size and reproductive factors with circulating levels of sex hormones and prolactin in premenopausal Japanese women. Cancer Causes Control 22: 581–588.
• 11   St-Onge  MP,  Campbell  A,  Salazar  I,  Pizinger  T,  Liao  M, et al. (2019) Information on bedtimes and wake times improves the relation between self-reported and objective assessments of sleep in adults. J Clin Sleep Med 15: 1031–1036.
• 12   Schernhammer  ES,  Rosner  B,  Willett  WC,  Laden  F,  Colditz  GA, et al. (2004) Epidemiology of urinary melatonin in women and its relation to other hormones and night work. Cancer Epidemiol Biomarkers Prev 13: 936–943.
• 13   Bracci  M,  Manzella  N,  Copertaro  A,  Staffolani  S,  Strafella  E, et al. (2014) Rotating-shift nurses after a day off: peripheral clock gene expression, urinary melatonin, and serum 17-β-estradiol levels. Scand J Work Environ Health 40: 295–304.
• 14   Gomez-Acebo  I,  Dierssen-Sotos  T,  Papantoniou  K,  García-Unzueta  MT,  Santos-Benito  MF, et al. (2015) Association between exposure to rotating night shift versus day shift using levels of 6-sulfatoxymelatonin and cortisol and other sex hormones in women. Chronibiol Int 32: 128–135.
• 15   Papantoniou  K,  Pozo  OJ,  Espinosa  A,  Marcos  J,  Castaño-Vinyals  G, et al. (2015) Increased and mistimed sex hormone production in night shift workers. Cancer Epidemiol Biomarkers Prev 24: 854–863.
• 16   Graham  C,  Cook  MR,  Gerkovich  MM,  Sastre  A (2001) Examination of the melatonin hypothesis in women exposed at night to EMF or bright light. Environ Health Perspect 109: 501–507.
• 17   Davis  S,  Mirick  DK,  Chen  C,  Stanczyk  FZ (2012) Night shift work and hormone levels in women. Cancer Epidemiol Biomarkers Prev 21: 609–618.
• 18   Langley  AR,  Graham  CH,  Grundy  AL,  Tranmer  JE,  Richardson  H, et al. (2012) A cross-sectional study of breast cancer biomarkers among shift working nurses. BMJ Open 2: e000532.
• 19   Peplonska  B,  Bukowska  A,  Lie  JA,  Gromadzinska  J,  Zienolddiny  S (2016) Night shift work and other determinants of estradiol, testosterone, and dehydroepiandrosterone sulfate among middle-aged nurses and midwives. Scand J Work Environ Health 42: 435–446.
• 20   Wang  Q,  Kangas  AJ,  Soininen  P,  Tiainen  M,  Tynkkynen  T, et al. (2015) Sex hormone-binding globulin associations with circulating lipids and metabolites and the risk for type 2 diabetes: observational and causal effect estimates. Int J Epidemiol 44: 623–637.
• 21   Khosravipour  M,  Khanlari  P,  Khazaie  S,  Khosravipour  H,  Khazaie  H (2021) A systematic review and meta-analysis of the association between shift work and metabolic syndrome: the roles of sleep, gender, and type of shift work. Sleep Med Rev 57: 101427.
• 22   Touzet  S,  Rabilloud  M,  Boehringer  H,  Barranco  E,  Ecochard  R (2002) Relationship between sleep and secretion of gonadotropin and ovarian hormones in women with normal cycles. Fertil Steril 77: 738–744.
• 23   Kakizaki  M,  Kuriyama  S,  Sone  T,  Ohmori-Matsuda  K,  Hozawa  A, et al. (2008) Sleep duration and the risk of breast cancer: the Ohsaki Cohort Study. Br J Cancer 99: 1502–1505.
• 24   Cao  J,  Eshak  ES,  Liu  K,  Muraki  I,  Cui  R, et al. (2019) Sleep duration and risk of breast cancer: the JACC Study. Breast Cancer Res Treat 174: 219–225.
• 25   Morssinkhof  MWL,  van Wylick  DW,  Priester-Vink  S,  van der Werf  YD,  den Heijer  M, et al. (2020) Associations between sex hormones, sleep problems and depression: a systematic review. Neurosci Biobehav Rev 118: 669–680.
• 26   Li  DX,  Romans  S,  De Souza  MJ,  Murray  B,  Einstein  G (2015) Actigraphic and self-reported sleep quality in women: associations with ovarian hormones and mood. Sleep Med 16: 1217–1224.
• 27   De Zambotti  M,  Colrain  IM,  Baker  FC (2015) Interaction between reproductive hormones and physiological sleep in women. J Clin Endcrinol Metab 100: 1426–1433.
• 28   Baumgartner  A,  Dietzel  M,  Saletu  B,  Wolf  R,  Campos-Barros  A, et al. (1993) Influence of partial sleep deprivation on the secretion of thyrotropin, thyroid hormones, growth hormone, prolactin, luteinizing hormone, follicle stimulating hormone, and estradiol in healthy young women. Psychiatry Res 48: 153–178.
• 29   Hollander  LE,  Freeman  EW,  Sammel  MD,  Berlin  JA,  Grisso  JA, et al. (2001) Sleep quality, estradiol levels, and behavioral factors in late reproductive age women. Obstet Gynecol 98: 391–397.
• 30   Kische  H,  Ewert  R,  Fietze  I,  Gross  S,  Wallaschofski  H, et al. (2016) Sex hormones and sleep in men and women from the general population: a cross-sectional observational study. J Clin Endcrinol Metab 101: 3968–3977.
• 31   White  AJ,  Weinberg  CR,  Park  YM,  D’Aloisio  AA,  Vogtmann  E, et al. (2017) Sleep characteristics, light at night and breast cancer risk in a prospective cohort. Int J Cancer 141: 2204–2214.
• 32   Hurley  S,  Goldberg  D,  Behren  JV,  DeHart  JC,  Wang  S, et al. (2020) Sleep deficiency and breast cancer risk among postmenopausal women in the California teachers study (CTS). Cancer Causes Control 31: 1115–1128.
• 33   Lateef  OM,  Akintubosun  MO (2020) Sleep and reproductive health. J Circadian Rhythms 18: 1.
• 34   Zheng  J,  Islam  RM,  Skiba  MA,  Bell  RJ,  Davis  SR (2020) Associations between androgens and sexual function in premenopausal women: a cross-sectional study. Lancet Diabetes Endocrinol 8: 693–702.
• 35   Borugian  MJ,  Spinelli  JJ,  Gordon  PB,  Abanto  Z,  Brooks-Wilson  A, et al. (2014) Fasting insulin and endogenous hormones in relation to premenopausal breast density (Canada). Cancer Causes Control 25: 385–394.