Abstract
The pattern of secretion of many hormones, including prolactin, is dependent on the circadian rhythm. Night shift work involves exposure to artificial light at night and sleep deficiency, which in turn can affect prolactin synthesis. The aim of the present study was to evaluate a possible association between night shift work characteristics, sleep quality, lifestyle factors and prolactin concentration, using data from a cross-sectional study of nurses and midwives. A cross-sectional study was conducted among 327 nurses and midwives currently working on rotating night shifts, and 330 nurses and midwives working during the day (aged 40-60 years)(388 premenopausal, and 269 postmenopausal). Information about night shift work characteristics, lifestyle, reproductive factors, sleep pattern and other covariates was collected through face-to-face interview, and from a one-week work and sleep diary completed by the subjects. Weight and height were measured. Prolactin concentration was measured in the morning blood sample using the ECLIA method. Associations were analyzed using linear regression models adjusted for important confounders. Analyses were carried out separately in pre- and postmenopausal women.
None of the night shift work or sleep characteristics was significantly associated with prolactin concentration. Prolactin concentration was significantly (p<0.05) inversely associated with smoking and time of blood sample collection. These result were consistent among both pre- and postmenopausal women. Nulliparity was significantly positively associated with prolactin among premenopausal women, but inversely among postmenopausal. Age was related to prolactin among postmenopausal women only.
Our study indicates that rotating night shift work is not associated with prolactin concentration. Smoking, parity, time of blood collection and age among postmenopausal women were significant determinants of prolactin.
Night shift work has been classified by the International Agency for Research on Cancer (IARC) as probably carcinogenic to humans (Group 2A).(IARC Monographs on the Evaluation of Carcinogenic Risk to Humans, 2010) A suppressed synthesis of melatonin, a pineal gland hormone which exhibits anticarcinogenic potential, is a key element of the hypothesis linking night work with an increased risk of breast cancer.(Bukowska, 2011) Lack of sleep and exposure to artificial light that are characteristic of night shift work may evoke disruption of the circadian rhythm (biological clock), which is stabilized by the suprachiasmatic nucleus of the hypothalamus, and disrupt the secretion of hormones that are produced or secreted according to a circadian pattern, (Morris et al. 2012) including prolactin.
Prolactin is a polypeptide hormone synthesized and secreted mainly by lactotrophs, the cells of the anterior pituitary gland. Prolactin is involved in the regulation of biological processes such as growth and development of the mammary glands as well as in milk synthesis and secretion.(Freeman et al. 2000) High prolactin concentration has been linked to breast cancer in experimental studies on animals and in some epidemiological studies.(Bukowska & Peplonska, 2013; Tworoger & Hankinson, 2006) The plasma level of this hormone is the highest during sleep (≈45 ng/ml), and the lowest during the daytime (≈5 ng/ml).(Sassin et al. 1972) The most important factors affecting prolactin secretion are sleep, pregnancy, breastfeeding, stress, and estrogen level. Dopamine, a neurotransmitter, is the main inhibitor of prolactin secretion.(Freeman et al. 2000; Morris et al. 2012) Some studies demonstrated that endogenous circadian rhythm regulates prolactin synthesis in humans independently of the sleep. (Waldstreicher et al. 1996; Weibel & Brandenberger, 1998) Moreover, there are studies indicating the role of other factors in prolactin secretion. These include lifestyle factors, e.g. cigarette smoking (Kapoor & Jones, 2005), physical activity (Baker et al. 1982; Kraemer et al. 1993; McTiernan et al. 2006), and reproductive
parameters, e.g. parity and the number of children (Tworoger & Hankinson, 2008). However, the results have been inconclusive thus far (Tworoger & Hankinson, 2008). Only a few small-scale studies examined the association between night shift work and prolactin level.(Costa et al. 1997; Korompeli et al. 2009; Langley et al. 2012; Schernhammer et al. 2004; Spiegel et al. 1996; Touitou et al. 1990; Weibel & Brandenberger, 1998)
In view of the limited knowledge on the factors that can influence prolactin secretion, we have undertaken a cross-sectional study on nurses and midwives working on night shifts to further explore these issues. The primary focus of the reported research was a possible relationship between night shift work and prolactin concentration. We also examined associations between the sleep quality, lifestyle and reproductive factors and the concentration of prolactin since night shift work may have influence on lifestyle and thus, indirectly, on the concentration of this hormone. Since hormonal milieu in women depends on menopausal stage we performed separate analyses for pre- and postmenopausal women.
Materials and Methods
The study was described earlier.(Peplonska et al. 2012) In short, the subjects for the study were selected from the database of the local Registry of the Chamber of Nurses and Midwives in Lodz, Poland. Nurses and midwives aged 40-60 years, currently working in this occupation were eligible for the study. A total of 1117 nurses and midwives were randomly selected based on the Registry database (ca. 30% of all registered nurses and midwives in this age group) of which 924 (83%) were contacted. The inclusion criteria were confirmed for 866 women, and 725 (84% of the eligible) agreed to participate in the study.
A structured questionnaire was administered during a face-to-face interview. Data were collected on occupational history, demographics, medical and reproductive history, use of steroid medications or sex hormones, physical activity (according to the International Questionnaire on Physical Activity, IPAQ), smoking, alcohol consumption, coffee
consumption, and sleep quality (using the Pittsburgh Sleep Quality Questionnaire). Women were classified to premenopausal group if reported to have menstrual period. A work and sleep diary was completed by the study participants. The timing of falling asleep and waking up, the start and end time of work, and number of awakenings were recorded for seven days preceding the collection of biological material.
Immediately before blood sampling, a short interview was conducted to elicit information on the use of steroid medications, alcohol consumption and cigarette smoking during the last 24 hours. Women were also asked about the date of their last menstrual period. Number of days, which elapsed between day of blood draw and day of last menstrual period in women who were premenopausal was used to determine their menstrual cycle phase. Weight and height measurements were made, and BMI was calculated.
Morning blood samples were collected from 708 women. Excluded from analysis were 51 women who reported taking (a) antidepressants and sedatives which increase prolactin level; (b) medications decreasing prolactin secretion that are used in hyperprolactinemia; and (c) medications used in hormone replacement therapy. Eventually, the study group comprised 657 nurses and midwives (327 working on rotating night shifts and 330 working only during the day).
The study was approved by the ethical review board at the Nofer Institute of Occupational Medicine. A signed informed consent was obtained from each study participant. The study conformed to international ethical standards accordingly to guidelines provided in Portaluppi et al. (Portaluppi et al. 2010)
Night shift exposure assessment
We analysed questionnaire data on night shift work at current job and lifetime occupational exposure, and work and sleep diary records kept for seven days preceding blood sample collection.
Occupational data included information on current job and all previous jobs held for at least six months; these were described in detail in our previous study.(Peplonska et al. 2012) Based on the data gained, the following variables describing current night shift work were identified: current night shift work (Yes/No); night shift frequency (2-7, 8+ night shifts per month), napping during night duty between 11 p.m. – 3 a.m. (Yes/No), and total duration of napping (0, ≤2, >2 hours).
Lifetime occupational history was analyzed using the following characteristics: the total duration of jobs with night shifts (years), the number of night shift hours, and the total number of night shifts over lifetime work history. Diary records were used to assess the most recent exposure over a period of seven days preceding blood sampling. The number of night shifts, of night shift hours, and of consecutive night shifts during that period were calculated and analysed.
Sleep characteristics
Sleep time and the sleep quality index (Buysse et al. 1989) (PSQI) were determined using questionnaire data. They demonstrated the sleep pattern during four weeks preceding the interview. PSQI was calculated according to the formula provided in the instruction of the Sleep Medicine Institute, University of Pittsburgh. PSQI represents a summary score ranging from 0 to 21. Total PSQI score ≤5 is associated with a good and >5 with a poor quality of sleep.
Sleep duration and the number of awakenings during the night immediately before blood sample collection were determined based on work and sleep diary records.
Blood sample collection and prolactin analysis
Blood samples were collected into S-Monovette® heparinized test tubes (Sarstedt AG & Co, Nümbrecht, Germany) between 6:00 a.m and 10:00 a.m. We analyzed prolactin
concentration in blood sample taken in the morning to examine the effects of night shift work. In the case of night shift workers, blood sampling was synchronized with the night shift, so that the sample would be taken at the end of the shift. Samples were collected by trained nurses and delivered to the laboratory immediately afterwards. Serum samples were frozen at -80°C and stored until analysis. Electrochemiluminesence immunoassay method (ECLIA, Cobas®, Roche, Diagnostics, Elecsys 2010) was used to determine prolactin concentration in blood serum. The measurement range for prolactin was 127-637 µIU/ml. Pearson correlation coefficients for 10% paired measurements of prolactin concentration were calculated as a measure of assay reliability, with r2 = 0.991 (p<0.001).
Statistical analysis
Arithmetic means with standard deviations and frequencies of the basic characteristics were calculated. The ANOVA test was used to compare means of continuous variables, and the chi-square test was used to compare distribution of categorical variables among women working rotating night shifts and daytime workers.
The distribution of prolactin concentration was considerably right skewed; therefore, prolactin concentrations were log transformed for the analysis.
Based on the literature review, a list of factors potentially related to prolactin secretion was formulated.(Benker et al. 1990; Faupel-Badger et al. 2010; Freeman et al. 2000; Kapoor & Jones, 2005; Morris et al. 2012; Sassin et al. 1972; Tworoger & Hankinson, 2008) The list included parity, breastfeeding, age at menarche, age at menopause (in postmenopausal women), age at first full-term birth, use of steroid medications or oral contraceptives, menstrual cycle stage (in premenopausal women), family history of breast cancer, BMI, cigarette smoking, physical activity, alcohol consumption, coffee consumption, quality of sleep, age, season of the year and time of blood sample collection. All these factors were
tested in the multivariate model to assess their associations with prolactin. Three variables describing reproduction, i.e, parity, breastfeeding and age at first birth were highly correlated, of two first we built one variable (with 3 categories: nulliparous, parous who never brestfed and parous who brestfed), and age at first birth was deleted from the model due to collinearity.
For the multivariate analyses we used information about alcohol drinking and smoking during 24 hours preceding blood collection instead of averages.
Univariate and multivariate linear regression models were fitted with prolactin concentrations as dependent variable in order to assess crude and adjusted associations between prolactin and night shift work characteristics. All potential confounders listed above were included in the models and backward stepwise analysis was applied to select ones for the final models. The AIC criterion (Kadane & Lazar, 2004) for selecting variables was used. The following variables were retained in the final model: time of blood sampling (hour), cigarette smoking within 24h before blood sampling (Yes/No), calendar season when blood sample was collected (July-September/October-June), parity with breastfeeding (nulliparous, parous who breastfed, parous who never breastfeed); in addition among postmenopausal women age and age at menopause were included. The results of analyses are presented as adjusted geometric means with 95% confidence intervals and p-values for trends.
We also examined a possible association between sleep characteristics of the night preceding blood sample collection and prolactin concentration. This analysis was carried out only in the subgroup of daytime workers because for night shifts workers, the most recent night before blood collection was on duty (sleepless).
Analyses were performed among pre- and postmenopausal women separately. All statistical analyses were carried out using R version 3.1.1.
The characteristics of the study population by current night shift work is presented in Table 1. Women working on rotating night shifts both premenopausal and postmenopausal were slightly younger compared to women working only during the day (one year difference), with a higher physical activity. Statistically significant differences between groups of nurses and midwives working in different systems of work were also found for alcohol consumption the day before blood sampling and time of blood sample collection. Daily coffee consumption in premenopausal women was higher in women working rotating system than in nurses working during the day. Other characteristics were similarly distributed between compared groups.
Most of the subjects working on rotating night shifts (88%) had 2-7 duties per month, usually between 7:00 p.m. and 7:00 a.m. (12h duties), and a long history of night shift work (25 years on average). For the subjects who were currently working days only, but who had a history of night shift work, the total duration of night shifts was approximately 12 years. In this group, the majority (83%) had changed their work schedule to daytime work more than five years before they were recruited for the study, and 3% stopped working night shifts during the preceding year (data not shown).
The results of multivariate regression analysis for lifestyle and reproductive factors are displayed in Table 2. Prolactin was significantly inversely associated with smoking, and time when blood sample was collected in both pre- and postmenopausal women. Nulliparity was positively associated with prolactin when compared to parous who breastfed among premenopausal women, while the opposite association was observed among postmenopausal. Moreover prolactin concentration was inversely associated with age in postmenopausal women.
Crude geometric means of prolactin were significantly higher among both pre- and postmenopausal women working rotating night shift when compared to day workers (premenopausal: 256.0 µIU/ml vs. 203.2 µIU/ml; p <0.001, postmenopausal: 162.2 µIU/ml vs. 141.8 µIU/ml; p=0.020). Differences in the means attenuated and became insignificant in the adjusted analyses. Adjusted geometric means for prolactin concentration, by current shift work status, total duration of jobs with night shift work and current job characteristics in premenopausal and postmenopausal women are presented in Table 3 and Table 4. No associations between either lifetime occupational history of the night shift work or current night shift work characteristics and prolactin concentration were found.
The results of the analyses between the quality of sleep and prolactin concentration are shown in Table 5. None of the analysed sleep characteristics, i.e. the sleep time (average and the night before blood collection) or the number of awakenings showed significant associations with morning blood prolactin concentration.
Discussion
The present study investigated the associations between night shift work characteristics, lifestyle and reproductive factors, quality of sleep, and prolactin concentration (measured in morning blood sample) among premenopausal and postmenopausal women. The analyses adjusted for important confounders did not show any significant inferences between either current or lifetime history of rotating night shift work and prolactin. Time of blood collection, and cigarette smoking, were inversely significantly related to prolactin, consistently in both pre-and postmenopausal women. Nulliparity was positively associated with prolactin among premenopausal but inversely among postmenopausal women. In addition age was inversely related to prolactin among postmenopausal women.
In most crude analyses for association between night shift work and prolactin, we observed statistically significant results. Nevertheless, the differences between the categories diminished and the trends became insignificant after adjustment, which indicates a substantial effect of the covariates included in the models.
Only a few studies investigated the association between night shift work and prolactin concentration.(Costa et al. 1997; Korompeli et al. 2009; Langley et al. 2012; Schernhammer et al. 2004; Spiegel et al. 1996; Touitou et al. 1990; Weibel & Brandenberger, 1998) The study by Costa et al. performed among 10 young (under 30 years of age) nulliparous nurses working at intensive care unit indicated “acute“ and reversible effects of the night shift work. Prolactin concentration at the end of two consecutive night shifts was significantly lower (13.0 ng/ml) than the concentration measured at the beginning of morning shifts (21.4 ng/ml)(Costa et al. 1997). Lower prolactin concentration among night shift workers was also observed in another study that investigated prolactin secretion pattern. Blood samples were collected every two hours over night shift in four oil refinery operators (aged 25-34 years).(Touitou et al. 1990) Our study did not yield similar results, which could be explained by the different characteristics of the study population, such as age, parity, gender, and possibly work environment and occupational history.
To our knowledge, there have been only two studies examining the association between the number of years of night shift work over lifetime occupational history and prolactin concentration in the blood, and no links between the two have been observed, like in our own research. (Langley et al. 2012; Schernhammer et al. 2004) In the present study, we considered not only the total number of years of night shift work, but also the total number of night shifts and of night shift hours for a given job.
Night shift work may lead to circadian misalignment (Smith & Eastman, 2012), thus affecting the human physiology. It may also have influence on lifestyle, social and family life
or reproduction.(Rutenfranz et al. 1977) For example, it has been reported that night shift workers are more likely to smoke,(Trinkoff & Storr, 1998) consume more coffee, (Jermendy et al. 2012) and become overweight or obese.(Morikawa et al. 2007; van Amelsvoort et al. 1999) To address this potential effect of night shift work, we extended our research to investigate associations between selected lifestyle and reproductive factors and prolactin concentration.
We observed an inverse association between cigarette smoking and prolactin level. The effect of chronic smoking on prolactin concentration has been reported only in a few studies. Most of the studies focused on the effects of smoking on prolactin concentration and the duration of breastfeeding. The results of previous studies are consistent in that smoking decreases prolactin level and shortens duration of breastfeeding. (Kapoor & Jones, 2005) The biological mechanism linking cigarette smoking with a lower prolactin concentration has not been fully explained. Nevertheless, it has been postulated that in chronic smokers, nicotine might suppress prolactin secretion. Nicotine was found to up-regulate the secretion of dopamine, the main inhibitor of prolactin secretion, by activating nicotinic receptors of the tuberoinfundibular dopamine neurons.(Bahadori et al. 2013; Kapoor & Jones, 2005)
We found an inverse association between age and prolactin, which was statistically significant among postmenopausal women. This finding is consistent with previous studies where relationship between age and declining prolactin was observed. Significant decrease of prolactin after menopause was explained by parallel decrease of estrogens. (Vekemans & Robyn, 1975)
Other factors that may play a role in prolactin secretion include those related to reproduction (Tworoger & Hankinson, 2008). Postmenopausal women showed lower levels of sex hormones, including prolactin,(Tanner et al. 2011) which was confirmed in our study. Likewise in previous reports (Faupel-Badger et al. 2010; Tworoger & Hankinson, 2008), we
also observed inverse association between parity and prolactin concentration, but only among premenopausal women. The significantly lower prolactin concentration that we observed among postmenopausal nulliparous women when compared to parous who breastfed was unexpected finding when compared to results of the previous epidemiological studies. (Eliassen et al. 2007; Tworoger & Hankinson, 2008; Wang et al. 1988) One potential explanation of the inconsistency with others is that in our analysis we used as the reference women who were parous, and breastfed, while other studies have analyzed parity irrespective of breastfeeding.
Sleep has been reported to be one of the most important factors inducing the secretion of prolactin (Morris et al. 2012). Thus we carried out in-depth analyses using data from a validated questionnaire on night sleep pattern during four weeks preceding the interview, and on the quality of sleep during the night preceding blood sampling, based on diary records. Since all the night shift workers had night duties and experienced a lack of sleep the night before prolactin assessment, the analysis was performed only among day shift workers. The results of the analysis for association between habitual sleep time and prolactin concentration were statistically insignificant. The lack of significant results may be explained by the fact that the PSQI measures average sleep quality within the period of four weeks, and thus may have not reflected the quality of sleep during the night preceding prolactin measurement. The results for sleep time, although, insignificant were in the expected direction, which is consistent with the findings of previous studies. (Benker et al. 1990; Morris et al. 2012; Sassin et al. 1972; Spiegel et al. 1996)
Our study has several strenghts. It was conducted in a well-characterized population of women working in the profession. The response rate was relatively high; blood samples were taken from 97% of the total population of women who agreed to participate in the study. The data were collected via a face-to-face interview by trained interviewers, and detailed
information on both current work characteristics and lifetime occupational history were elicited. All the major confounding factors were evaluated and we were also able to control for the time of blood sampling, which is important for the analysis of most of the biological parameters.(Morris et al. 2012)
A potential limitation of our study relates to the lack of information about stress, a factor that was reported to increase prolactin concentration.(Benker et al. 1990) We were not able to control for stress directly; nevertheless, we analysed the quality of sleep, which was shown to be strongly related to stress, and turned out not to be important confounder in our study. (Vandekerckhove & Cluydts, 2010) Moreover, the results of some previous studies indicate that in nurses working on rotating night shifts, the level of stress is not higher than among nurses working only during the day.(Barton et al. 1995; Korompeli et al. 2009) We also excluded from the analysis the women taking sedatives.
In addition, we acknowledge that the study could have been limited by recall bias, which is typical for the most questionnaire based studies. Information about some of the covariates such as for example age at first and last menstrual period, or past occupational history may have been imprecise. This is less likely, though, that information about current job’s or lifestyle characteristics were affected by poor recall. The potential selection bias, which could be an issue when frequencies are being compared between groups, may have been of less importance in our study, since we analyzed associations between personal characteristics and biological parameter. We also acknowledge that biological material was collected only once from each women, therefore, we could not assess the reproducibility of prolactin measurement in the same women.
Our study had the power of 0.99 for factor explaining 2.6% of variance and 0.80 for factor explaining 1.1% of variance. For example, it had an 80% power to detect 14% of difference in geometric means between night shift workers and daytime workers.
To sum up, in the study population of nurses and midwives, no association was found between rotating night shift work and prolactin concentration in morning blood sample. Such factors as smoking, parity, time of sampling, and age in postmenopausal women were related to the hormone concentration.
Acknowledgements
We appreciate the assistance of the Registry of Nurses and Midwives in Lodz, especially of Mrs. Grażyna Romanowska, in organizing the study. We thank the interviewers for help in collecting data for analysis. We also want to thank all the nurses and midwives for their participation in our project.
Declaration of Interest statement
No potential conflicts of interest were disclosed.
This project was supported by the Norway Grants, under the Polish-Norwegian Research Programme (Grant no. PNRF – 243 – AI – 1/07), and project financed by the statute activity of Nofer Institute of Occupational Medicine in Lodz No. IMP 10.2.
Reference
1 IARC Monographs on the Evaluation of Carcinogenic Risk to Humans. (2010). Painting, Firefighting and Shiftwork.
2 Bukowska A. (2011). Anticarcinogenic role of melatonin--potential mechanisms. Med Pr.62:425-434.
3 Morris CJ, Aeschbach D, Scheer FA. (2012). Circadian system, sleep and endocrinology. Mol Cell Endocrinol.349:91-104.
4 Freeman ME, Kanyicska B, Lerant A, Nagy G. (2000). Prolactin: structure, function, and regulation of secretion. Physiol Rev.80:1523-1631.
5 Tworoger SS, Hankinson SE. (2006). Prolactin and breast cancer risk. Cancer Lett.243:160-169.
6 Bukowska A, Peplonska B. (2013). Night shift work and prolactin as a breast cancer risk factor. Med Pr.64:245-257.
7 Sassin JF, Frantz AG, Weitzman ED, Kapen S. (1972). Human prolactin: 24-hour pattern with increased release during sleep. Science.177:1205-1207.
8 Weibel L, Brandenberger G. (1998). Disturbances in hormonal profiles of night workers during their usual sleep and work times. J Biol Rhythms.13:202-208.
9 Waldstreicher J, Duffy JF, Brown EN, Rogacz S et al . (1996). Gender differences in the temporal organization of proclactin (PRL) secretion: evidence for a
sleep-independent circadian rhythm of circulating PRL levels- a clinical research center study. J Clin Endocrinol Metab.81:1483-1487.
10 Kapoor D, Jones TH. (2005). Smoking and hormones in health and endocrine disorders. Eur J Endocrinol.152:491-499.
11 McTiernan A, Wu L, Chen C, Chlebowski R et al . (2006). Relation of BMI and physical activity to sex hormones in postmenopausal women. Obesity (Silver
Spring).14:1662-1677.
12 Kraemer RR, Blair MS, McCaferty R, Castracane VD. (1993). Running-induced alterations in growth hormone, prolactin, triiodothyronine, and thyroxine
concentrations in trained and untrained men and women. Res Q Exerc Sport.64:69-74.
13 Baker ER, Mathur RS, Kirk RF, Landgrebe SC et al . (1982). Plasma gonadotropins, prolactin, and steroid hormone concentrations in female runners immediately after a long-distance run. Fertil Steril.38:38-41.
14 Tworoger SS, Hankinson SE. (2008). Prolactin and breast cancer etiology: an epidemiologic perspective. J Mammary Gland Biol Neoplasia.13:41-53.
15 Touitou Y, Motohashi Y, Reinberg A, Touitou C et al . (1990). Effect of shift work on the night-time secretory patterns of melatonin, prolactin, cortisol and testosterone. Eur J Appl Physiol Occup Physiol.60:288-292.
16 Costa G, Bertoldi A, Kovacic M, Ghirlanda G et al . (1997). Hormonal Secretion of Nurses Engaged in Fast-rotating Shift Systems. Int J Occup Environ Health.3:35-39.
17 Korompeli A, Sourtzi P, Tzavara C, Velonakis E. (2009). Rotating shift-related changes in hormone levels in intensive care unit nurses. J Adv Nurs.65:1274-1282.
18 Schernhammer ES, Rosner B, Willett WC, Laden F 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.
19 Langley AR, Graham CH, Grundy AL, Tranmer JE et al . (2012). A cross-sectional study of breast cancer biomarkers among shift working nurses. BMJ Open.2:e000532
20 Spiegel K, Weibel L, Gronfier C, Brandenberger G et al . (1996). Twenty-four-hour prolactin profiles in night workers. Chronobiol Int.13:283-293.
21 Peplonska B, Bukowska A, Gromadzinska J, Sobala W et al . (2012). Night shift work characteristics and 6-sulfatoxymelatonin (MT6s) in rotating night shift nurses and midwives. Occup Environ Med.69:339-346.
22 Portaluppi F, Smolensky MH, Touitou Y. (2010). Ethics and methods for biological rhythm research on animals and human beings. Chronobiol Int.27:1911-1929. 23 Buysse DJ, Reynolds CF, III, Monk TH, Berman SR et al . (1989). The Pittsburgh Sleep
Quality Index: a new instrument for psychiatric practice and research. Psychiatry Res.28:193-213.
24 Faupel-Badger JM, Sherman ME, Garcia-Closas M, Gaudet MM et al . (2010). Prolactin serum levels and breast cancer: relationships with risk factors and tumour
characteristics among pre- and postmenopausal women in a population-based case-control study from Poland. Br J Cancer.103:1097-1102.
25 Benker G, Jaspers C, Hausler G, Reinwein D. (1990). Control of prolactin secretion. Klin Wochenschr.68:1157-1167.
26 Kadane JB, Lazar NA. (2004). Methods and criteria for model selection. Journal of the American Statistical Association.99:279-290.
27 Smith MR, Eastman CI. (2012). Shift work: health, performance and safety problems, traditional countermeasures, and innovative management strategies to reduce circadian misalignment. Nat Sci Sleep.4:111-132.
28 Rutenfranz J, Colquhoun WP, Knauth P, Ghata JN. (1977). Biomedical and psychosocial aspects of shift work. A review. Scand J Work Environ Health.3:165-182.
29 Trinkoff AM, Storr CL. (1998). Work schedule characteristics and substance use in nurses. Am J Ind Med.34:266-271.
30 Jermendy G, Nadas J, Hegyi I, Vasas I et al . (2012). Assessment of cardiometabolic risk among shift workers in Hungary. Health Qual Life Outcomes.10:18.
31 Morikawa Y, Nakagawa H, Miura K, Soyama Y et al . (2007). Effect of shift work on body mass index and metabolic parameters. Scand J Work Environ Health.33:45-50.
32 van Amelsvoort LG, Schouten EG, Kok FJ. (1999). Duration of shiftwork related to body mass index and waist to hip ratio. Int J Obes Relat Metab Disord.23:973-978. 33 Bahadori B, Riediger ND, Farrell SM, Uitz E et al . (2013). Hypothesis: smoking
decreases breast feeding duration by suppressing prolactin secretion. Med Hypotheses.81:582-586.
34 Vekemans M, Robyn C. (1975). Influence of age on serum prolactin levels in women and men. Br Med J.4:738-739.
35 Tanner MJ, Hadlow NC, Wardrop R. (2011). Variation of female prolactin levels with menopausal status and phase of menstrual cycle. Aust N Z J Obstet
Gynaecol.51:321-324.
36 Eliassen AH, Tworoger SS, Hankinson SE. (2007). Reproductive factors and family history of breast cancer in relation to plasma prolactin levels in premenopausal and postmenopausal women. International Journal of Cancer.120:1536-1541.
37 Wang DY, De Stavola BL, Bulbrook RD, Allen DS et al . (1988). The permanent effect of reproductive events on blood prolactin levels and its relation to breast cancer risk: a population study of postmenopausal women. Eur J Cancer Clin Oncol.24:1225-1231.
38 Vandekerckhove M, Cluydts R. (2010). The emotional brain and sleep: an intimate relationship. Sleep Med Rev.14:219-226.
39 Barton J, Spelten E, Totterdell P, Smith L et al . (1995). Is there an optimum number of night shifts? Relationship between sleep, health and well-being. Work
