Psychosocial factors at work and fertility and menstrual disorders: a systematic review

biopsychosocial medicine; endocrine; gynecology; occupational health; 83 reproductive health. 84 85 86 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 4 Statement of authorship 87 Acknowledgment 88 The authors wish to thank Dr. Daisuke Shigemi (obstetrician-gynecologist) for a valuable suggestion 89 on the variable selection. 90 91 Author’s contribution 92 NS, KI, KW, EM, HE, AI, KT, YH, YKom, MIi, YO, ASa, YA, MIw, YKob, RI, ASh, A T, and NK 93 substantially contributed to the conception and design of the study, writing of the protocol, and 94 revising it critically for important intellectual content. All authors approved the final version of the 95 manuscript. 96 97 Funding statement 98 This study is supported by The University of Tokyo Occupational Mental Health (TOMH) Research 99 Fund, obtained from the Department of Mental Health, Graduate School of Medicine, The 100 University of Tokyo, Japan. 101 102 Role of the Funder/Sponsor: The sponsors had no role in the design and conduct of the study; 103 collection, management, analysis, and interpretation of the data; in the preparation, review, or 104 approval of the manuscript; and in the decision to submit the manuscript for publication. 105 106 Registry and registration number of the study/trial: The study protocol was registered at the 107 UMIN registry (registration number: UMIN000039488). The registration date is 14th Feb 2020. 108 URL: https://upload.umin.ac.jp/cgi-bin/ctr/ctr_view_reg.cgi?recptno=R000044704 109 110 Conflict of interest: All authors declare no conflict of interest related the submitted work. 111 112 Data availability statement : All data produced in the present study are available upon reasonable 113 request to the authors. 114 115 116 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 5

117

This systematic review aimed to evaluate the association between psychosocial 118 factors at work and menstrual abnormalities or fertility by collecting the literature that had 119 utilized a longitudinal or prospective cohort design. 120 Data sources : MEDLINE, EMBASE, PsycINFO, PsycARTICLES, and Japan Medical 121 Abstracts Society electronic databases were searched for published studies from inception to 122 26th February 2020. 123 Study eligibility criteria: The inclusion criteria for this systematic review were defined as 124 follows: (P) Adult female workers (over 18 years old), (E) Presence of adverse psychosocial 125 factors at work, (C) Absence of adverse psychosocial factors at work, and (O) Any menstrual 126 cycle disorders, menstrual-related symptoms, or fertility. The study design was limited to 127 prospective/longitudinal studies. 128 Study appraisal and synthesis methods : The included studies were descriptively summarized 129 in a narrative format. 130

Database searching yielded 12,868 abstracts. After the screening process, six studies 131 were included. The outcomes were fertility (n=3), early menopause (n=1), endometriosis (n=1), 132 and serum hormones (n=1). Three included studies presented significant findings: women with 133 high job demands were significantly less likely to conceive; working over 40 hours per week 134 and lifting or moving a heavy load >15 times per day significantly increased the duration of 135 pregnancy attempts; women with rotating night shift work had increased risk of earlier 136 menopause. All the study (n=3) that examined the association of night shift/rotating work with 137 fertility outcomes showed no significant difference. No study investigated the association of 138 psychosocial factors at work with menstrual abnormality. 139

This review revealed insufficient high-level evidence on the association of 140 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 6 psychosocial factors at work with fertility and menstrual disorders. Future well-designed studies 141 are needed. 142 Trial registration : The study protocol was registered at the UMIN registry (registration 143 number: UMIN000039488). The registration date is 14th Feb 2020. 144 URL: https://upload.umin.ac.jp/cgi-bin/ctr/ctr_view_reg.cgi?recptno=R000044704 145 Review protocol: Protocol paper is available in preprint format. 146 URL:https://assets.researchsquare.com/files/rs-179301/v1/48845b10-5ec4-4d48-8918-3dcf0e0e147 dded.pdf?c=1631873381 148 149 150 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 7

151 Menstrual abnormalities are a crucial issue in women’s health. Menstrual cycle 152 irregularities increased the risk of premature death, 1 ovarian and breast cancer, 2,3 153 cardiovascular disease,4 and diabetes. 5 Moreover, fertility is essential for reproductive 154 health and rights among preconceptual working-aged women. Menstrual characteristics 155 and fertility are thus important indicators in occupational health settings. 156 Menstrual cycle and ovarian functions are regulated by sex hormones (i.e., the 157 hypothalamic-pituitary-ovarian [HPO] axis); however, they are often disrupted by 158 psychosocial factors6,7 through activation of the hypothalamic-pituitary-adrenal [HPA] 159 axis.8 Psychosocial factors at work provide potential risks of activating the HPA axis 160 and releasing cortisol due to psychological stress reactions, leading to menstrual 161 disorders. For instance, some literature showed significant associations: low job control, 162 co-worker support, and job security with high menstrual pain, 9 low supervisor support 163 with menopausal symptoms, 10 strenuous work with premenstrual symptoms, 11,12 work 164 schedule variability with irregular cycles, 13,14 time pressure and fast work speed with 165 perimenstrual symptoms,15 and job strain (a combination of high job demands and low 166 job control) with cycle length irregularities. 16 The study design of many of them was 167 cross-sectional, leading to limited evidence and suggestions for appropriate practices in 168 the women’s occupational health field. 169 The HPA-HPO axis interaction is affected by circadian sleep control, 17,18 The 170 relation between night/shift work and menstrual disorders has been thus often described, 171 and a meta-analysis by Stocker et al. (2014), 19 which included a cross-sectional design, 172 showed that shift work increased the risk of menstrual disruption and infertility 19. 173 However, more recent literature showed inconsistent results in the association of shift 174 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 8 work with ovarian function.20-22 Lim et al. (2016) reported no association between shift 175 work and ovarian functions or menstrual disorders in a cross-sectional design. 20 In 176 contrast, Minguez-Alarcon et al. (2017) reported that non-day work schedules were 177 inversely related to oocyte production and quality among women receiving fertility 178 treatment in a cross-sectional design.21 Only one longitudinal study by Albert-Sabater et 179 al. (2016) reported that nurses on the rotating shift did not show an increased risk of 180 having menstrual disorders compared with day staff.22 Since most studies were adopted 181 a cross-sectional design (i.e., one cycle was treated as one individual outcome), the 182 topics of shift work with menstruation require further investigations. 183 The association between psychosocial factors at work and menstrual 184 abnormalities or fertility has not been systematically presented yet. Longitudinal studies 185 should be summarized for concrete conclusions. The aim of this systematic review was 186 thus to investigate whether psychosocial factors at work impact menstrual 187 characteristics or fertility, limiting the literature that has utilized a longitudinal or 188 prospective cohort design, assessing the outcome both at baseline and follow-up. 189 190

191 Study design 192 The method was reported according to the Preferred Reporting Items for Systematic 193 Review and Meta-Analysis (PRISMA 2020) guidelines. 23 The study protocol has been 194 registered at the UMIN registry (registration number: UMIN000039488 URL: 195 https://upload.umin.ac.jp/cgi-bin/ctr/ctr_view_reg.cgi?recptno=R000044704). The 196 registration date is 14th Feb 2020. The protocol paper has been available elsewhere.24 197 198 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 9 PECO and Eligibility criteria 199 The eligible participants, exposures, comparisons, and outcomes (PECO) of this 200 systematic review are listed in Table 1. PECO was: P: adult female workers (over 18 201 years old), E: adverse psychosocial factors at work (+), C: adverse psychosocial factors at 202 work (-), and O: menstrual disorders, related symptoms, reproductive outcomes. The 203 study design was restricted only longitudinal or prospective manner. The study was also 204 excluded if the cross-sectional analysis method was applied, even if the study used a 205 longitudinal design. The language was limited to English or Japanese. 206 207 Search and Information Sources 208 Databases including PubMed (MEDLINE), Embase, PsycINFO, PsycARTICLES, and 209 the Japan Medical Abstract Society database were searched for publications from 210 inception to 26th February 2020. The details of the search terms are shown in Appendix 211 1. The search terms for psychosocial factors at work were defined based on the previous 212 meta-analysis.25-28 The search terms for outcome variables were initially selected by an 213 investigator (NS) who reviewed previous systematic reviews handling menstrual-related 214 health.19,29-32 The medical doctor in obstetrics and gynecology (outside the research 215 members) confirmed the search terms before searching. The terms were amended to 216 exclude infectious, inflammatory disease, and pelvic organ prolapse due to the lack of 217 probability that psychosocial work-related factors cause these. 218 219 Study selection 220 Microsoft Excel (Washington, USA) was used to manage all identified studies. An 221 investigator (NS) excluded duplicate records. The remaining articles were shared with 222 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 10 16 investigators (NS, KI, KW, EM, HE, AI, KT, YH, YKom, MIi, YO, ASa, YA, MIw, 223 YKob, and RI); Pairs of them independently assessed the title and abstract of each 224 article according to the eligibility criteria (shifting phase: first screening). Subsequently, 225 two investigators independently reviewed the full texts that were included in the first 226 screening. Any disagreements were discussed until reaching a consensus by all authors. 227 The reasons why studies were excluded were recorded during the full-text review phase. 228 229 Data collection process and data items 230 Data were extracted independently from eligible articles by 16 investigators (NS, KI, 231 KW, EM, HE, AI, KT, YH, YKom, MIi, YO, ASa, YA, MIw, YKob, and RI). An 232 investigator (NS) integrated and formatted the information on publication year, study 233 design, the country of study origin, the number of participants completing the baseline 234 survey and included in the statistical analysis, demographic characteristics of the 235 participants (i.e., age, occupation), the length of follow-up and attrition rate, exposure 236 variables (i.e., adverse psychosocial factors at work), and outcome variables (i.e., 237 menstrual abnormalities, fertility). 238 239 Risk of Bias in individual Studies 240 The risk of bias in observational studies of exposures (ROBINS-I) tool was used to 241 assess the quality of included studies. Investigators (NS, KI, KW, ASa, and YH) 242 independently scored the bias classified as low, high, or unclear. Discrepancies in 243 quality assessment among the investigators were solved by discussion and consensus 244 among all authors. A Summary of Findings (SoF) will be created using the Grading of 245 Recommendations, Assessment, Development and Evaluation (GRADE) approach to 246 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 11 grade the certainty of evidence. 247 248 Synthesis of Results and Meta-Analysis 249 For the main analysis, we synthesized all types of psychosocial factors at work and all 250 types of menstrual-related disorders/symptoms. Meta-analysis was conducted if at least 251 three eligible studies were found to have the same outcome. If a meta-analysis was not 252 appropriate (i.e., only two or fewer studies were eligible and included), the results were 253 presented in a narrative format. 254 255 Changes to the protocol 256 To examine the association between exposure and outcome longitudinally, the study 257 applied a cross-sectional fashioned analysis (i.e., one menstrual cycle was treated as an 258 individual unit of outcome) was excluded. 259 260

261 Database searching 262 Database searching yielded 12,868 abstracts (PubMed n = 2,823, EMBASE n = 9,340, 263 PsycINFO n = 608, Japan Medical Abstracts Society n = 97). After removing 1,435 264 duplicates, 11,433 records were included in the first screening, after which 11,395 265 records were excluded, and 38 records proceeded to full-text screening. Subsequently, 266 32 studies that did not meet the criteria for design (n=13), 11-16,20-22,33-36 participants 267 (n=4),37-40 exposures (n=10), 41-50 outcomes (n=2), 51,52 article type (n=1), 53 and 268 duplicates (n=2) were excluded. Finally, six studies were included in the systematic 269 qualitative review.54-59 The meta-analysis was not conducted due to insufficient studies 270 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 12 based on our protocol. The flowchart of study selection is shown in Figure 1. 271 272 Study description 273 The characteristics of the six included studies are shown in Table 2. Three studies 274 treated fertility outcomes. 55,57,59 Earlier menopause, 56 and incident diagnosis of 275 endometriosis,58 and serum reproductive hormone 54 were each measured by one study. 276 No study was found which outcome was menstrual abnormalities. Three of the six 277 studies retrieved their data from the nurse’s health study (NHS). All studies measured 278 shift work as an adverse psychosocial factor at work. Each study reported job 279 demands59 and long working hours. 57 280 281 Although authors designed the protocol to conduct a meta-analysis if at least three 282 eligible studies were found with the same outcome, fertility outcomes of included 283 studies varied, and it was difficult to synthesize the findings: time to pregnancy, 55 the 284 current duration of participants’ ongoing pregnancy attempt,57 and conception.59 285 286 Regarding the three studies focusing on fertility outcomes, Barzilai-Pesach et al. 287 (2006)59 prospectively evaluated the association between women’s occupational stress 288 factors (i.e., job demands, decisional latitude, shift work, and working hours, measured 289 using a self-reported questionnaire) and the outcome of fertility treatments. They 290 followed 75 women (mean age = 31.3) who had worked at least three months before 291 intake, had a female fertility problem and attended fertility clinics between 1999 and 292 2000. Information regarding the outcome of fertility treatments was based on a 293 follow-up 18 months after the initial intake through fertility clinics. Women with high 294 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 13 job demands were less likely to conceive (relative risk [RR] = 0.60 [95% confidence 295 interval [CI]: 0.42 to 0.96]). There was no significant association of decision latitude, 296 shift work, or working hours with conception. Gaskins et al. (2015) 57 evaluated the 297 association of work schedule (i.e., work hours and total duration of night shift work) 298 and physical factors (i.e., standing job and lifting load) with fecundity (i.e., current 299 duration of participants’ ongoing pregnancy attempts) among female nurses, using the 300 data from an internet-based cohort study of female nurses in the USA and Canada (NHS 301 3). For assessment of fecundity, the authors took a participant’s first report of ongoing 302 pregnancy attempt after the baseline questionnaire as her outcome in three follow-up 303 survey questionnaires conducted every six months. As a result, 1,739 women (median 304 age = 33) were included in this analysis. Long working hours (over 40 hours per week) 305 and lifting or moving a heavy load (>15 times per day) were related to a longer median 306 duration of pregnancy attempts (increasing 20% [95% CI: 7% to 35%] and 49% [95% 307 CI: 20% to 85%], respectively). Additionally, a subgroup analysis found that the 308 association between heavy moving and lifting with reduced fecundity was even more 309 pronounced among overweight or obese women. Night shift work over the past year 310 was not significantly associated with fecundity. Willis et al. (2019) 55 prospectively 311 evaluated the association of shift work with fecundability. They followed North 312 American women (living in Canada or the United States) who are not using 313 contraception and aged 21–45 years attempting pregnancy for up to 12 months or until 314 conception, using web-based surveys. Time to pregnancy was estimated by reports of 315 their last menstrual period, usual menstrual cycle length, and the number of cycles they 316 had attempted pregnancy at study entry. As a result, 6,873 women who had shift/night 317 work ranging from 14.3% to 23.0% were included in the study. There was no 318 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 14 association between shift work and fecundability. 319 320 Stock et al. (2019) 56 prospectively evaluated the association between nurses’ 321 rotating night shift work (i.e., the number of months with at least three nights per month 322 in the past two years) and menopausal age, using the data from an ongoing prospective 323 cohort study (NHS 2). This cohort included 116,429 female registered nurses in the 324 United States, aged between 25 to 42, at the baseline survey in 1989, followed from 325 1991 through 2013. Participants were asked whether their menstrual periods had ceased, 326 at what age, and if this was due to natural menopause, chemotherapy or radiation, or 327 surgery. Of those whose menstrual periods had not stopped before the last follow-up, 328 27,456 had natural menopause. The mean age of natural menopause in these women 329 was 50 (SD ±4.0) years old. Among them, 2,524 had menopause at under 45 years old. 330 Significant associations were found between the number of months of night shift work 331 in the past two years and earlier menopause (HRs = 1.06 to 1.11 among women with 332 <10 months, 10-19 months, and with ≥ 20 months, compared to none of night shift 333 work). Cumulative rotating night work of 10 or more years was also associated with a 334 higher risk of menopause among women reaching menopause under age 45. These 335 analyses did not consider any confounders such as psychological stress, job strain, or 336 chronotype-related tolerance. 337 338 Schernhammer et al. (2011)58 reported an association between rotating night shift 339 work and endometriosis risk within the NHS2. Participants were asked whether they 340 had ever had physician-diagnosed endometriosis. Analyses of incident diagnosis of 341 endometriosis were restricted to those women who reported laparoscopic confirmation 342 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 15 of their diagnosis. This study included 2,062 laparoscopically confirmed cases 343 documented during 16 years of follow-up evaluation from 89,400 women without 344 diagnosed endometriosis at baseline. Overall, there was no association between rotating 345 night shift work and endometriosis. When the cases were categorized by infertility 346 status, the risk was elevated among women with concurrent infertility and ≥ 5 years of 347 rotating night shift work (RR = 1.71 [95% CI: 1.18 to 2.49]; P for trend =0.005), 348 compared with women with no rotating night shift work. 349 350 Michels et al. (2020) 54 examined the influences of night/shift work on the serum 351 reproductive hormones. The participants were recruited in the United States from the 352 BioCycle Study (2005–2007), a prospective cohort of 259 healthy women (mean age = 353 24) not using oral contraceptives. At baseline, participants were asked if they were 354 currently employed and if they worked nights or rotating shifts. Fasting blood samples 355 were collected at eight clinic visits across each menstrual cycle. Fertility monitors were 356 used to assist in the timing of mid-cycle clinic visits. Two cycles were followed. The 357 proportion of participants who had night/shift work was 34.4%. The results showed that 358 night/shift work did not influence estradiol or progesterone. Night/shift work showed 359 lower testosterone (-9.9%) than employees without night/shift work, but the clinical 360 implication of the finding was not mentioned. 361 362 Risk of bias assessment 363 The summary of the risk of bias assessment by ROBINS-I is shown in Table 3 . The 364 overall judgment was all “serious” due to confounding or selection of reported results. 365 The certainty of evidence was judged to be very low for all outcomes, downgraded for 366 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 16 very serious risk of bias and publication bias (Appendix 2). 367 368

369 Principal Findings 370 This review investigated the association between psychosocial factors at work 371 and menstrual abnormalities or fertility by limiting the study design to longitudinal, 372 assessing the outcome at baseline and follow-up. Of the six included studies, three 373 showed non-significant associations of shift work with fertility. 55,57,59 Instead, high job 374 demands, long working hours, and heavy lifting were significantly related. 57,59 Rotating 375 night shift work was not associated with the onset of endometriosis,58 but was with early 376 menopause.56 One study identified slight changes in the sex hormones by shift work. 54 377 All the studies had a serious risk of bias. There were still insufficient well-designed 378 studies available to conduct meta-analysis and draw conclusions. However, the present 379 study clarified the lack of evidence, achieved hypotheses to be examined in the future, 380 and suggested further research directions. 381 382 Comparison with Existing Literature 383 All three studies 55,57,59 that investigated the risk of shift/night work for 384 pregnancy outcomes showed non-significant associations. These studies were not 385 included in the recent meta-analysis published in 2014 19 that examined the influence of 386 shift work on reproductive outcomes, showing that shift work increased the risk of 387 infertility (crude odds ratio [OR]: 1.80 [95%CI: 1.01- 3.20]). This meta-analysis 388 included four cross-sectional fashioned studies (one study was a retrospective cohort) 389 from 1996 to 2003. 60-63 These studies retrieved a pregnant population and obtained the 390 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 17 occupational data retrospectively. Three of them did not show significant 391 associations,60-62 and one did.63 Certainly, some studies seemed to find that women with 392 night/shift work tend to take a long time to become pregnant. 60-62 However, the 393 significance disappeared after adjusting the demographic and job characteristics 394 information. The adjusted results of the meta-analysis also showed non-significant 395 associations (adjusted OR: 1.12 [0.86 – 1.44]). 19 From both our findings and previous 396 reviews, we cannot conclude that shift/night work itself has an influence on fertility. 397 More studies should be conducted using rigorous longitudinal study designs that 398 account for possible confounding factors (psychosocial factors at work or lifestyle 399 factors) to examine the association between shift work and fertility outcomes. 400 401 Two studies examined psychosocial factors at work other than shift/night work. 402 These studies showed that high job demands, long working hours, and heavy lifting 403 were significantly associated with infertility. 57,59 These findings were consistent with 404 previous cross-sectional studies that were not included in the present review, which 405 suggested that long working hours and physically demanding jobs lead to less ovarian 406 function and infertility. 21,61 In contrast, another study reported a non-significant 407 association.60 Therefore, job demands, long time working, and lifting are possibly 408 related to fertility outcomes. Although the previous study suggested that psychological 409 stress reactions (i.e., depression, low positive affect) declined ovarian function, 64 410 whether these psychosocial factors at work affect fertility through such distress or not 411 (direct impact) still needs examination. Psychological stress theoretically disrupts 412 HPA-HPO axis interactions 8 and has been associated with higher levels of oxidative 413 stress, which potentially declines ovarian function. However, there is a lack of evidence 414 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 18 to investigate the associations between psychosocial factors at work and fertility 415 outcomes in a robust design. Further prospective studies will be expected in this area. 416 417 One study in this review showed the increased risk of early menopause from 418 shift/night work.56 Rotating night shift work experience in the prior 2 years increased 419 the risk of earlier menopause. The subgroup analysis with women under 45 showed that 420 over 10 years of cumulative rotating night shift work exposure increased the risk of 421 early menopause. 56 This finding is consistent with the recent systematic review that 422 included cross-sectional studies investigating job stress and menopause; the authors 423 reported that age at menopause and severity of menopausal symptoms were both 424 influenced by job-related factors such as high job strain. The review identified the risk 425 factor for early menopause as smoking, type of work (e.g., service work), and shift work. 426 Early menopause has been regarded as a significant outcome because it increases the 427 risk of cardiovascular disease and all-cause mortality. 65 Further investigations are 428 needed to prevent the deterioration of female workers’ health if any occupational factors 429 cumulatively affect early menopause. In this review, one study showed no significant 430 association of rotating night shift work with the onset of endometriosis diagnosed by 431 laparoscopy.58 Night shift work was found to decrease melatonin levels, 66 reduced the 432 aromatase activity and worked as a potential antiestrogenic agent. 67 Elevated estrogen 433 levels caused by the low melatonin levels from night shift work could promote 434 endometriosis growth. A population-based case-control study reported that any night 435 shift work was associated with a 48% increased risk of endometriosis. 68 The authors of 436 the included study discussed the possible bias that women with clinical pain symptoms 437 or with children might be likely to be out of the workforce in rotating work. 438 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 19 Additionally, laparoscopic confirmation of endometriosis among those who never 439 experienced infertility is restricted to those with clinical symptoms, leading to 440 underestimating the risk in those who never experienced infertility. The associations of 441 shift work with endometriosis still need further investigation. 442 443 No study investigated the association of psychosocial factors at work with 444 menstrual abnormality using longitudinal data. This review was limited to studies that 445 evaluated the outcome (menstrual abnormality) in both baseline and follow-up. 446 Furthermore, studies were excluded that analyzed the menstrual cycle as one unit of 447 outcome. No study that met these criteria was found in this review. Some studies 448 analyzed the cycle data as one unit.14,16,22,34 That method did not allow for a before/after 449 comparison of data from one person and was considered a cross-sectional study. 450 Menstrual abnormality can be caused not only by occupational stress but also by 451 multiple other factors (e.g., body mass index [BMI], non-occupational stress). A 452 longitudinal study evaluating the change in menstrual abnormalities will be needed to 453 estimate the risk of psychosocial factors at work. 454 455 Strengths and Limitations 456 This systematic review was the first review to investigate the evidence of the 457 associations between psychosocial factors at work and menstrual abnormalities or 458 fertility, which limited the longitudinal well-designed studies. However, this systematic 459 review has several limitations. First, we did not conduct a meta-analysis due to an 460 insufficient number of studies. The concrete evidence thus was not provided for the 461 clinical questions. Second, this review is limited by English and Japanese language 462 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 20 restriction; therefore, studies in other languages may have been missed. Third, 463 publication bias may occur, but this review did not assess that bias. 464 465

and Implications 466 This systematic review investigated the association of psychosocial factors at work with 467 menstrual abnormalities and fertility outcomes, and six articles were included. This 468 review presented insufficient high-level evidence of well-designed studies on this topic. 469 Because psychosocial factors at work are modifiable risk factors for women’s 470 reproductive outcomes, further investments are needed to determine the implications for 471 occupational health practice. 472 473

474 We extend great thanks to cooperated researcher Yumi ASAI for her su pport of our procedures of 475 screening the literature. 476 477 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 21

478 1. Wan g Y-X , Ar v izu M, Ri c h- Ed w ards JW, e t al . Me n strual cyc l e r e gulari t y and l e ngth 479 acr oss th e re produc t iv e li f es pan a n d r is k o f p re ma t ure mo r tality : p r os pe c tiv e 480 coh ort st ud y . BMJ ( Cl inical re search ed ). 202 0; 3 71:m34 6 4. 481 2. Kam dar BB, T er gas AI, Mate e n F J , B hayani NH, Oh J. Ni gh t -sh i ft wo rk and r i sk of 482 bre ast c ancer: a s y st ematic r evi e w and met a -a nalysi s . Breas t can c er r e se arch and 483 tre atment . 2 013;13 8 (1 ):291- 3 01 . 484 3. Tav ani A , Ricc i E, L a Vec c hi a C, e t al . Infl u en ce of me n s t ru a l a n d re p rodu c tiv e 485 fac tors on ovari a n ca ncer risk in women wit h a n d withou t fam il y h is t ory of b r eas t 486 or ovaria n c ancer. In ternat i on al journa l of e pi d emiolo g y. 2000; 2 9( 5):799-80 2. 487 4. Sol omon CG, H u F B, D u naif A, et a l. M e nstr ual cyc l e irre g ul arity a nd r isk for f u tur e 488 car diovas cu l ar di se a se. T he Jo u rn al o f C li ni c al En do c rin o lo gy & M etabo l ism . 489 200 2;87(5 ): 2 013-20 1 7. 490 5. Sol omon C G, Hu FB, Du na i f A, et a l . Lo n g or highl y i rr e gular me nstrua l cy c l es a s 491 a m arker fo r risk o f type 2 di abetes me llitus . J ama. 2 0 01 ;286(1 9 ): 2421-2426 . 492 6. Chr ousos G P , To rpy D J , Gold PW. Inter a ct ions b et ween th e 493 hyp othala mi c -pitui t ar y-adre n al axis a nd the f em a le r e pr oducti v e sy st e m: c li n ica l 494 implicati on s . Ann I nt ern Me d . 1998;129( 3):229 -2 4 0. 495 7. Mag iakou M A, Ma st o rak o s G , Web s te r E , Chrous os GP. Th e 496 hypothala mi c -p itui t ar y-adre n al ax i s and th e fema le repro d uctive syst em. An nals 497 of the Ne w Y ork Ac a de my of S ci ences. 19 97;816 :4 2 -56. 498 8. Len nartss on A K, Kus h ni r MM, B e rg quis t J, Bi l lig H, Jo n sdotti r I H. Sex st e roi d 499 lev els te m po ra r ily i nc r ease in re s ponse to ac u te p sychos oc i al str e ss i n he a lth y 500 men an d w om e n. Intern a ti onal j our n al of ps y ch op h ysiolo gy : of f icial jou r na l o f 501 the Inter na t ional O rg anizat i on of Psych ophysi ol o gy. 20 1 2; 84(3): 2 46 -253. 502 9. Lás zló K D, Gyorf f y Z, Ad á m S, Cs oboth C, Kopp MS. W o rk -r e lat e d st r es s fa c tors an d 503 men strual pai n: a na tion-w i de r e pre s en tat i ve s ur vey. J o ur na l of psycho so m atic 504 obs tetric s a nd gyn a ec ology. 20 08;29(2): 133-13 8. 505 10. Bar iola E , Jac k G, P it ts M , R i ach K , S a rr el P. Emplo y me nt c o ndi t io n s and 506 wor k-rela te d st r essors ar e as sociat e d wit h me n op au s al sym pt o m r e portin g a mon g 507 p er imenopau s al and post m en op a usal wom en . Menop au s e ( Ne w Y ork , NY) . 508 201 7;24(3 ): 2 47-251 . 509 11. Smi th D, W ei N , Wa n g R. S ubj e ct iv e hea l th c o mplain ts a m on g female h os p ita l n u rse s 510 in Mainla nd China. J Occup H ea lth Safet y (Aus t N Z). 20 0 5; 21:51- 5 9. 511 12. Hou rani LL, Yu a n HX , Br a y RM. P s yc ho s oc i al a n d Li f est y le C or re l ate s of Pr em e ns t rua l 512 Sym ptoms am o ng Mil i ta ry Wom e n. Journal of Wom en s Healt h . 2004;1 3 (7 ):812-821 . 513 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 22 13. Mes sing K, Saurel-C ubizol l es MJ, Bou rgine M, Kami n sk i M. Me nstrua l -c ycle 514 cha racter is t ics and wo rk co n dition s of w o rke r s i n poultry slaugh te r house s and 515 can neries . Scandi na v ian jo u rn al of work , e nv i ronmen t & h ea lth. 516 199 2;18(5 ): 3 02-309 . 517 14. Su SB, L u CW , Ka o Y Y , Gu o HR . Effec t s of 12 - ho ur r o ta ting s hifts o n me ns t rua l c y cle s 518 of photoe le c tronic wo rkers i n Taiwan. C hronobio l Int. 2 00 8;25(2 ) :2 37-248. 519 15. Tis sot F , Me s si ng K . Pe r ime n st ru a l Sy m pt oms and W or k ing C onditi on s Among H os p ital 520 Wor kers in Que b ec . Ame r ic an jo ur n al o f in du s trial medic i ne . 199 5; 2 7(4):5 11 - 522 . 521 16. Fen ster L, Walle r K , C hen J, et al. Psy cholog ic a l S tress i n the W or kpl ac e an d 522 Men strual F u nction . A m J Ep i de miol. 199 9;149( 2) : 127-13 4 . 523 17. Gam ble K L , R e sue h r D, J oh ns o n C. Shift wor k an d circ ad i an dy s r egulation o f 524 reproduct io n . Fron t ie rs in e nd ocrinolog y. 201 3; 4 :92. 525 18. Sel lix M , Me n ak er M. Circadian clocks in ma m mal i an r epr o du ct i ve p hysio l ogy : 526 eff ects o f t h e “oth e r” biolo gical cl oc k on ferti l it y. Di s co ve r y m e di c ine . 527 201 1;11(5 9) : 273-28 1 . 528 19. Sto cker LJ, Mac k lo n N S, Cheong YC , Bewl ey SJ. Influ e nc e o f s hift work o n e arly 529 reproduct iv e ou t comes: a s y st ematic re vie w an d m et a -analy si s . O bstet G yn e col. 530 201 4;124( 1) : 99-110 . 531 20. Lim AJ, H uan g Z, C h ua SE, K ramer M S , Yong EL. S l eep D ur a tion, Exer c is e, Sh i ft Wor k 532 and Poly c ys tic Ov a ria n Sy n drome- Re late d Ou tc o mes i n a H e althy Po pulation: A 533 Cro ss-Sec ti o nal St u dy . PloS on e. 2016;1 1(11): e0 1 67048. 534 21. Min guez-A la r con L , S o uter I, W illiam s PL, et al . Occ u pational fac to r s a nd ma r ker s 535 of ov a ri an re se r ve an d respo ns e a mo ng wo men a t a fertil i ty c e nt re . O cc u p En v iro n 536 Med . 2017 ;7 4 (6):42 6 -4 31. 537 22. Alb ert-Sa ba t er JA, M a rti n ez JM , B a ste V , M o en BE , R onda-Pere z E. Co m pari so n o f 538 men strual di s or de r s i n h ospita l n u rsing st aff a c cordin g t o s h i ft work p a tt ern . 539 J C lin Nu rs . 2016; 2 5( 21-22) : 32 91-3299. 540 23. Pag e M J, McKenz i e J E, Bo ssuyt PM, et al . T he P RI SMA 2020 sta t e ment: an u pd ate d 541 gui deline for r epo rti n g syst e ma tic rev i ew s. BM J (Clin i ca l re se a rch ed). 542 202 1;372: n7 1 . 543 24. Sas aki N , Imamu r a K , Wat an a be K, et a l. P syc ho s ocial Facto r s a t W or k a n d Mens t rua l 544 Dis orders : a Pr ot o col for a Sys t em ati c Re vi e w and Met a -a na l ysi s . 2 021 ; 545 htt ps://a ss e ts.res e ar chsqua r e. com/files /rs-17 93 0 1/v1/4 8 84 5b10-5 e c4 -4d48-891 8-546 3dc f0e0ed de d .pdf?c = 16 318733 8 1 . Accessed 3 Apr il 2022. 547 25. Egu chi H , Wa t ana b e K, Kawa k am i N, et a l. P s ychoso ci a l fa c tors at wor k an d 548 inf lammat or y mar k er s: prot oc o l fo r a syst ematic review an d met a -a nalysi s . BM J ope n. 549 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 23 201 8;8(8) :e 0 22612. 550 26. Ima mura K, T sut s um i A , A s ai Y, et al. A s so ciatio n b e t ween psyc h os ocial f ac tor s 551 at wo r k an d health ou tcomes a fte r retire ment: a pr ot o col f or a syste ma t ic re view 552 and meta- an a lysis. BM J open . 2 019;9(8): e03077 3. 553 27. Sak uraya A, W at a nab e K , K a wakami N , et al . Wo r k-rela t ed psychos oc i al facto r s and 554 ons et of m e ta bo l ic sy ndrome a mon g worker s: a sy s te ma t ic re view and meta-a n al ysis 555 pro tocol. B M J open . 2 017;7( 6 ): e016716. 556 28. Wat anabe K, S ak u ray a A , K a wakami N , et al . Wo r k-rela t ed psychos oc i al facto r s and 557 met abolic sy n dr om e on set a m ong workers: a sys te m atic r evi e w and m et a -ana ly sis . 558 Obe s Rev. 2 0 18;19( 1 1) :1557- 1 56 8. 559 29. Sol iman A M , Y ang H, D u EX, Kell e y C, Wi n k el C . Th e dire ct a n d indi r ec t c osts 560 ass ociate d wi th e ndomet r io si s : a s yst e ma tic lit er a ture re v ie w. H uman repro du c tion 561 (Ox ford, En g land). 20 16;31( 4 ): 712-722. 562 30. Cer queira R O, Fr e y BN, Le cl e rc E, Bri e tz ke E. V itex a g nus c as t us fo r preme ns t rual 563 syn drome and preme n st rual d ysphor ic dis order: a system a ti c r eview. Archive s o f 564 wom en's m en t al hea l th . 2017 ; 20 (6):713-7 19. 565 31. Chi affari no F, R i cci E, Cipr i an i S, C hi ant e ra V, P arazzi n i F. Ci ga r ett e sm o kin g 566 and r is k of uterin e myoma : s y st emati c review and me t a-anal ys i s. E u r J Obs t e t 567 Gyn ecol R ep r od Bio l . 2016;1 9 7: 63-71. 568 32. Son g JA, Lee MK, Min E, Ki m ME , Fike G, H ur MH . Effects of a r om at h erapy o n 569 dys menorr he a : A s ys temati c rev ie w a nd me ta - analys i s. I nt J N urs S tud . 570 201 8;84:1 -1 1 . 571 33. Kim O , Ahn Y, Lee H Y, e t a l. Th e K or ea Nu rse s ' He a lt h S tu dy: A Pr o specti ve C o hor t 572 Stu dy. Jo ur n al of w om en's h e al th (2002) . 2017 ;2 6 (8):89 2 -8 99. 573 34. Hjo llund NH, Ko ld Je n sen T, Bonde JP, e t a l. Jo b s t rain and tim e to preg na ncy . 574 Sca ndinav ia n journ a l of wor k , environme nt & h ea l th. 19 9 8; 24(5): 3 44 -350. 575 35. Sch ernham me r E S, Ros n er B , W il le t t WC, L ad e n F , Col di t z GA , H ankins o n SE . 576 Epi demiol og y of U ri nary Me la t oni n in Wome n and I ts Relat ion t o O t her H o rm ones and 577 Nig ht Wor k. Cancer Ep idem B i om ar. 2004; 13(6): 93 6 -943. 578 36. Wan g Y , G u F , D eng M, et a l . R ota t ing s hi f t work and m en st r ual chara ct e r is tic s 579 in a coho rt of Chi n es e nurs e s. BMC Wome ns Hea lt h . 2016 ; 16 :24. 580 37. Jey aseela n L, R ao PS . Effect of oc c upa t io n on men s tr ua l cycle l e ng th : causal m o del. 581 Hum Biol. 1 9 95;67( 2 ): 283-29 0 . 582 38. Bur dorf A, Brand T, J a dd oe V W , Hofman A , Mack e n bach J P, Steeg e rs E A. T h e ef f ect s 583 of w or k -re l at ed m aterna l ri sk f actors o n tim e to pre gnancy , p r et e rm b irt h a nd b irt h 584 wei ght: t he Genera t io n R St u dy . Occup E nviron M e d. 201 1 ;6 8(3):1 9 7- 204. 585 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 24 39. Wan g X S, Travi s RC, Re ev e s G , et al . Ch ara c te ristic s of th e M i ll ion Women S tud y 586 par ticipa nt s who ha ve a n d h a ve no t wo r ke d a t n i ght. S c and i na vi a n jou r na l of w ork, 587 env ironme nt & heal t h. 2012; 3 8( 6):590-59 9. 588 40. Law son C C , Johns o n CY, Ch av a rro J E, e t al. W or k s ch ed u le an d physic a ll y de ma n ding 589 wor k in re l at io n to men st r ual func ti on: t h e Nur s es ' Health S tu d y 3. S c an d i na v i a n 590 jou rnal o f w ork, e n vi ronmen t & health. 2015;4 1( 2 ):194- 2 03 . 591 41. Tab ernero -R i co P M, Ga r ci a- V elasco JA. O bs er v ati o na l stu d y of t h e s o cia l 592 det ermina nt s o f h e alth in s u bf er t ile v er sus nons ub f ertile w om en. J ournal of h uman 593 rep roduct iv e scien c es . 2019 ; 12 (3):240. 594 42. Zhu J L , K nudsen LE , A n de rsen A M, Hjollu nd NH, Olsen J. Ti m e t o pr egnanc y a mong 595 Dan ish labora tory techni c ia ns w ho were a pa r t of the Nati o na l B i rth Co h ort . 596 Sca ndinav ia n journ a l of wor k , environme nt & h ea l th. 20 0 5; 31(2): 1 08 -114. 597 43. Pot ter J, B o uye r J, Trus s el l J , M or eau C . P r emenstr u a l s yndrom e pr ev al e n ce a n d 598 flu ctuati on ov e r time: resu l ts f r om a Fren ch po pulation- b as ed s urvey. Jour n al o f 599 wom en's h ea l th (20 0 2) . 2009 ; 18 (1):31-39 . 600 44. Ass adi S N. Is Being a H e al th ‑ car e Worker a R isk F actor f o r Wo m en’s Repro d uc tive 601 Sys tem? n te r nation a l Journa l o f Prevent ive Me di c ine. 2 0 13 ;4(7). 602 45. Gas kins A J, Ch av a rro JE , Ric h -E dwards J W, et al . O ccupat io n al us e of high- l eve l 603 dis infect an t s an d fec u nd it y a mo ng n urses. S candin a vi an j ou r nal of w ork, 604 env ironme nt & heal t h. 2017; 4 3( 2):171-18 0. 605 46. Sun S, Ch en F. Wo m en's empl o ym ent t raje ctorie s d u ri ng early adult hood i n u rb a n 606 Chi na: A co h ort co m pa rison. So c Sci Res . 2017 ;6 8 :43-58 . 607 47. Geu kes M, A nema J R , v an A al s t MP, de Me ne z es R X, Oos t er hof H. I mprove m ent o f 608 men opausa l sy m ptoms an d the impa c t o n w ork a bility : A re t rospec ti v e cohort p ilo t 609 stu dy. Ma tu r itas. 2 01 9;120: 2 3- 28. 610 48. Kuc zmierc zy k AR, La b rum A H , J ohn s on CC . PER CEPTIO N OF F AM ILY A ND WORK EN VIRON M ENT S 611 IN WOME N WIT H PREME N ST RUAL S Y NDR O ME . J ou rnal o f p sy ch o som a ti c re se a rch. 612 199 2;36(8 ): 7 87-795 . 613 49. Dav ydov DM, Shapir o D, Golds t e in IB , C hi cz- D eM et A . M oo ds i n e ve ryday situ at i ons: 614 eff ects o f me n strual c ycle, w or k, an d s tress h o rmones . J ou r nal o f psycho so m atic 615 res earch. 2 0 05;58( 4 ): 343-34 9 . 616 50. Wel ler L, We ll e r A, Ko r esh-Ka m in H , Ben-S hoshan R. Mens t ru al s y nch r on y in a sa mple 617 of workin g w omen. P sy choneu r oe ndocrinol ogy. 1 99 9 ;24(4) : 44 9-459. 618 51. Woo ds NF , M i tchell ES, Perc iv a l D B, S m it h-D i Ju lio K. Is th e m e no pausal tra ns i tion 619 str essful ? Ob s er va t ions o f per ceived str e ss fro m the S e att l e M idl i fe Wo m en' s 620 Hea lth St ud y . Meno p au se (Ne w Y ork, NY). 2009; 16 ( 1):90- 9 7. 621 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 25 52. Dag her RK , McG ov e rn PM, Dowd BE, L un d be rg U . Po s tpartum d e p re s si ve sym p to m s an d 622 the c om b ined lo a d o f pai d an d u np aid wo r k: a long it u dinal a n al ys i s. Int Arch O ccu p 623 Env iron H ea l th. 20 1 1; 84(7): 7 35 -743. 624 53. Lyn ch R, Lum ma a V, P an c hanath a n K , et al. I nte g ra ti o n invol ves a tra d e- off be t wee n 625 fer tility and stat us fo r Wo rl d W a r II eva cuees. Nat Hum Behav . 201 9 ; 3(4):337 - 345 . 626 54. Mic hels K A , Men d ol a P, Sc h liep K C , et al. T he i nf l uences of s le e p du ra t ion, 627 chr onotype, and nig h tw or k on th e ov ar i an cycle. Ch r on obi ol Int. 628 202 0;37(2 ): 2 60-271 . 629 55. Wil lis S K, Hatc h EE , Wes se l ink AK, Rothman KJ , Mikkel s en E M , W is e L A. F emale s lee p 630 pat terns, shif t work, and fe c und a bi li t y in a No r th Amer i c an p r econce pt i on c o hor t 631 stu dy. Fe rt i lity a n d steril i ty . 2019;11 1(6):1 20 1 -1210 e 12 01. 632 56. Sto ck D, Kni g ht J A, R aboud J, et a l . Rotat ing n i gh t sh if t work and m en o pau s al age . 633 Hum an rep ro d uction (O xford, En gland). 2 019;34 (3 ) :539-5 4 8. 634 57. Gas kins AJ, Rich-E dw a rds J W, L awson C C, Sch e r nhammer ES, Miss m er SA , C h av arro JE . 635 Wor k sc he d ule an d phy sical factors i n relati o n to f e cundit y in nurses . O ccu p 636 Env iron M ed . 2015; 7 2( 11):77 7 -7 83. 637 58. Sch ernham me r E S, V itonis AF , R ic h -Ed w a r ds J , M is smer SA . Rot a t ing nigh t sh ift wor k 638 and the ri sk o f en d ome t ri os i s in preme nopaus a l wome n . A m J O bs tet G yn e col . 639 201 1;205( 5) : 476 e4 7 1- 478. 640 59. Bar zilai- Pe s ach V , S h einer E K , Shein e r E, Po t ashnik G, Shoham - Va rdi I. Th e e f fec t 641 of w om en ' s o ccupat i on al ps y cholog ic str ess o n o u tcome of ferti li t y t reat me nts . 642 Jou rnal o f o ccupat i on al and en vironment al med ic i ne. 20 0 6; 48(1): 5 6- 62. 643 60. Spi nelli A, Fi g à- Ta l ama n ca I , O s bo rn J. Ti me t o pr egnanc y an d oc cu p ation in a g rou p 644 of Italia n w omen. I nt ernati o na l journal of ep id e miolog y . 1997;2 6 (3 ):601-609 . 645 61. Tun tisera ne e P, Ol sen J , G e ater A, Kor- ananta ku l O. Ar e l on g w o rking hours a n d 646 shi ftwork r is k f a ctors f o r s ub f ecu n d i ty? A st udy among co u ples fr om s ou t her n 647 Tha iland. O c cupati o na l and E nv ironmenta l Medi ci n e. 199 8 ;5 5(2):9 9 -1 05. 648 62. Zhu JL, H jol l un d N, B oggild H , Ol se n J. Sh i ft w ork and s u bf ecundi t y: a ca u sa l link 649 or an art ef a ct? Oc c up ationa l a nd enviro nmenta l m edicin e . 2003;6 0 (9 ):e12-e12 . 650 63. Bis anti L, Olsen J, B a ss o O, T h onneau P , Karm a u s W . Sh i ft work an d s ub fe c un d ity : 651 a Eu ropean m u lt ic e nte r st ud y . Jo u rn al o f occ u pa tional a nd enviro n me nt a l me d i c ine . 652 199 6;38(4 ): 3 52-358 . 653 64. Ble il M E , Adler NE, P as ch L A, et a l. De pres s i ve symp to m atolog y, psy ch o lo g ica l 654 str ess, a nd ov ar i an res er v e: a role f or psy c hologi ca l fa c to rs in o v ari a n a g ing ? 655 Men opause ( N ew Yor k , NY). 2 0 12 ;19(11):1 176. 656 65. Muk a T , Ol i ver-Wi ll i ams C, Kun u ts or S , et al . As so c iation o f ag e at o ns e t of 657 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint 26 men opause an d time s ince o n se t o f men op a use wi t h cardio v as cu l ar ou t co mes , 658 int ermedi at e va sc u lar trait s , and all-c ause m ort a li ty : a s y stematic r ev i ew a n d 659 met a-anal ys i s. JAM A c ardiol o gy . 2016;1( 7):767 -7 7 6. 660 66. Bur ch J B, Y ost MG, Joh n so n W , Al l en E . M e lat on i n, s lee p, a n d s h if t w o rk a dap ta t ion . 661 Jou rnal o f o ccupat i on al and en vironment al med ic i ne. 20 0 5; 47(9): 8 93 -901. 662 67. Cos S, M a rtí n ez ‐ C ampa C, Me d iav i ll a M D , Sán c he z‐Bar c el ó EJ. Me l at on i n mo d u l ate s 663 aro matase act iv i ty in MCF‐7 h um a n breast c an c er c ell s . Jo u rnal of pin e al r e se a rch . 664 200 5;38(2 ): 1 36-142 . 665 68. Mar ino JL, Holt VL, C h en C , Da v is S. Shif t w or k , hCLOCK T 3 11 1C po lym o rp hi s m , an d 666 end ometri os i s risk . E pidemi o lo gy (Cambr idge, Ma s s). 20 0 8; 19(3): 4 77 . 667 668 . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint Table 1 Eligibility criteria in the present systematic review Inclusion criteria Exclusion criteria Design /circle6 Longitudinal or prospective cohort design /circle6 Outcome was not assessed at least at two time point. /circle6 Menstrual cycle was cross-sectionally analyzed as an independent unit. Participants /circle6 Adult female workers (over 18 years old) /circle6 Pregnant workers /circle6 Clinical students Exposures /circle6 Presence of adverse psychosocial factors at work N/A Comparisons /circle6 Absence of adverse psychosocial factors at work N/A Outcomes /circle6 Menstrual disorders and related symptoms: 1) diverse menstrual dysfunctions (e.g., cycle disorder, hypermenorrhea), 2) gynecological disease/syndrome (e.g., endometriosis, polycystic ovarian syndrome), 3) menstrual-related symptoms (e.g., PMS, menopausal symptoms) /circle6 Reproductive outcomes (e.g., infertility, time to be pregnant) /circle6 Biological outcomes (e.g., serum sex hormone) /circle6 Pregnancy-related outcomes (e.g., premature birth) /circle6 Malignant outcomes (e.g., cancer) Article type /circle6 Original article published by peer-reviewed journal (including advanced online publication) /circle6 Conference abstract /circle6 Letter/correspondence Language /circle6 Written in English or Japanese N/A . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint Table 2. The descriptions of the included studies (N=6). 1st author, year, (country) Design (cohort name, duration) Baseline N (N for analysis) Participants Exposure Comparison Outcome F/U (%) Main results Michels KA., 2020 (U.S.) Prospective cohort (The BioCycle Study, 2005–2007) 259 (259) Healthy, regularly menstruating women aged 18–44 years, who did not used oral contraceptives, other medications or vitamins; had not been pregnant in the last six months; did not have any diagnoses of chronic conditions; did not have gynecologic surgeries. With night/shift work. Without night/shift work. Reproductive hormone Eight times across one to two menstrual cycle (100%) Women reporting night/shift work (n = 77) had lower testosterone (percent difference = -9.9% [95% CI: -18.4 to -0.4]). No significant relationship in estradiol, ovulatory LH, FSH, ovulatory FSH, luteal progesterone, and the risk of sporadic anovulation. Willis SK., 2019 (Canada and U.S.) Prospective cohort (PRESTO) 8,772 (6,873) Women aged 21 to 45 years, attempting pregnancy, who are not using contraception or fertility treatment. Working rotating shifts (hours that shift in a predictable way from day-to-day, week-to-week, or month-to-month) in the past month. No night/shift work Fecundability (per-cycle probability of conception) Every 8 weeks or until reported conception, initiation of fertility treatment, cessation of pregnancy attempts. (86.2% a) No association between shift work and fecundability. Stock D., 2019 (U.S.) Prospective cohort (NHS2 1989-2011) 116,429 (80,840) Female registered nurses between ages 25 and 42, who did not experience natural or medically induced menopause; did not use menopausal hormone therapy (HT), or did not Rotating night shift (recent number of months worked rotating night shifts, which defined as working Never worked on rotating nightshifts Age at natural menopause over 22 years of follow-up/ earlier menopause younger than 45 years Every 2 years follow-up questionnaire s were sent to cohort members. Response rates to the Women who had worked 10–19 months of rotating night shifts had the largest point estimate (HR = 1.11 [95% CI: 1.01 to 1.21]) and women who had worked ≥ 20 months had similar

(HR = 1.09 [95% CI: . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint have cancer diagnosis (excluding non-melanoma skin cancer). sometime between midnight and 2 AM). NHS 2 questionnaire s have been at 90% throughout the follow-up period. 1.02 to 1.16]). Cumulative rotating night work of 10 or more years was also associated with a higher risk of menopause among women reaching menopause under age 45 (HR10–19 years = 1.22 [95% CI: 1.03 to 1.44]; HR ≥ 20 years = 1.73 [95% CI: 0.90 to 3.35]). Gaskins AJ., 2015 (U.S. and Canada) Prospective cohort (NHS3 2014-2017) 38,016 (1,739) Female registered nurses trying to get pregnant. Typical work schedule over past year (evenings only/ nights only/ rotating with nights/ rotating no nights); weekly hours of nursing work over past year (1-20h/ >40h); frequency of night work in past month ( ≤ 1/ 2-3/ >3); duration of rotating night shifts (3y); duration of permanent night shifts (3y); duration of night shifts (4y); duration of Typical work schedule over past year (days only); weekly hours of nursing work over past year (21-40h); frequency of night work in past month (none); duration of rotating night shifts (never); duration of permanent night shifts (never); duration of night shifts (never); duration of standing/walking (5-8h); frequency of moving or lifting a heavy load (none). The current duration of their ongoing pregnancy attempt. Every 6 months (no data available) Working over 40 hours per week was related to increasing 20% (95% CI 7% to 35%) the median duration of pregnancy attempt compared to working 21-40 h/week (p =0.006). After adjustment for demographic and work-related confounders, women lifting or moving a heavy load >15 times per day had a 49% (95% CI 20% to 85%) longer median duration of pregnancy attempt (p =0.002). Typical work schedules over the past year, frequency of night work in the past month and duration of rotating or non-rotating night shifts were not also significantly associated with fecundity. . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint standing/walking (0 or less than 1h/ 1-4h/ >8h); frequency of moving or lifting a heavy load (1-5/ 6-15/ >15 times). Schernha mmer ES., 2011 (US) Prospective cohort (NHS 2, 1989-2005) 89,400 (2,062) Female registered nurses of ages 25-42 years, without the diagnosis of endometriosis or a history of infertility at baseline. Duration of work on rotating nightshift (1-4.9yr / ≧ 5yr). Never worked on rotating nightshifts. Diagnosis of endometriosis by laparoscopic confirmation. Every 2 years, follow-up questionnaire s were sent to cohort members. Response rates to the NHS II questionnaire s have been at 90% throughout the follow-up period. Overall, there was no association between rotating nightshift work and risk of endometriosis. When the cases were categorized by infertility status, risk was elevated among women with concurrent infertility and ≥ 5 years of rotating nightshift work (RR = 1.71 [95% CI: 1.18 to 2.49]; P for trend =0.005), compared with women with no rotating nightshift work. Barzilai-P esach V ., 2006 (Israel) Prospective cohort (1999-2000 ). 160 (75) Female working patients with a fertility problem, who attended the Fertility and In-Vitro Fertilization clinics at the Soroka University Medical Center. Job demands (high), type of job (full-time job), decision latitude (minimal), shift work (yes). Job demands (low), Type of job (part-time), decision latitude (high), shift work (no). Conception in which the data was obtained from the clinical records. 18 months (100%) Women with high job demand were less likely to conceive (RR = 0.60 [95% CI: 0.42 to 0.96]). No significant association of decision latitude, nor shift work with conception. a. The authors calculated from the data described in the paper (5926/6873*100). NHS: The Nurses’ Health Study. PRESTO: Pregnancy Study Online . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint Table 3. Summary of risk of bias assessment for the included studies analyzed by ROBINS-I (N=6). Michels et al (2020) #3722 Willis et al. (2019) #4093 Stock et al. (2019) #4307 Gaskins et al. (2015)#7240 Schernhammer et al. (2011)#9737 Barzilai-Pesach et al. (2006)#11436 1. Confounding S S M S S S 2. Selection of Participants L L L L S S 3. Classification of intervention M M M M S S 4. Deviations from intended interventions L L L L L L 5. Missing data S M S L S S 6. Measurement of outcomes L S S S L L 7. Selection of reported result S M S S S S Overall judgment S S S S S S Note: L: low risk of bias, M: moderate risk of bias, S: serious risk of bias, C: critical risk of bias, N: no information ROBINS-I: The risk of bias in observational studies of exposures . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint IdentificationScreeningIncluded Figure 1 PRISMA 2020 Flow Diagram of systematic review search results. Records identified from database: (n=12,868) • PubMed: n=2,823 • EMBASE: n=9,340 • PsycINFO/Articles: n=608 • Japan Medical Abstracts Society: n=97 Records removed before screening: Duplicate records removed (n=1,435) Reports sought for retrieval (n=38) Records excluded: (n=11,395) Reports excluded: (n=32) • Design: n=13 • Participants: n=4 • Exposures: n=10 • Outcomes: n=2 • Article type: n=1 • Duplicate: n=2 Studies included in review (n=6) Reports of included studies (n=6) Identification of studies via databases Records screened (n=11,433) Reports not retrieved (n=0) Reports assessed for eligibility (n=38) . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint Appendix 1 Search terms for PubMed (employe*[tw] OR manag*[tw] OR colleague*[tw] OR worksit*[tw] OR “work”[tw] OR works*[tw] OR work’*[tw] OR worka*[tw] OR worke*[tw] OR workg*[tw] OR worki*[tw] OR workl*[tw] OR workp*[tw] OR occupant*[tw] OR company*[tw] OR offic*[tw] OR busines*[tw] OR workplace[mh]) AND (("Stress, Mechanical"[Mesh] OR "Lifting"[Mesh] OR "Moving and Lifting Patients"[Mesh] OR "Weight-Bearing"[Mesh] OR "Biomechanics" OR “Physical Exertion"[Mesh] OR "Torsion, Mechanical"[Mesh] OR "Postural Balance"[Mesh] OR "Walking"[Mesh] OR "Recovery of Function"[Mesh] OR "Relaxation"[Mesh] OR (static[Title/Abstract] AND posture) OR (awkward[Title/Abstract] AND posture) OR (dynamic[Title/Abstract] AND posture) OR static work[Title/Abstract] OR dynamic load*[Title/Abstract] OR lift*[Title/Abstract] OR carry*[Title/Abstract] OR hold*[Title/Abstract] OR pull*[Title/Abstract] OR drag*[Title/Abstract] OR push*[Title/Abstract] OR manual handling[Title/Abstract] OR force*[Title/Abstract] OR biomechanic*[Title/Abstract] OR walking*[Title/Abstract] OR postural balance[Title/Abstract] OR flexion*[Title/Abstract] OR extension*[Title/Abstract] OR turning[Title/Abstract] OR sitting[Title/Abstract] OR kneeling[Title/Abstract] OR squatting[Title/Abstract] OR twisting[Title/Abstract] OR bending[Title/Abstract] OR reaching[Title/Abstract] OR standing[Title/Abstract] OR sedentary[Title/Abstract] OR repetitive movement*[Title/Abstract] OR monot onous work[Title/Abstract] OR relaxation[Title/Abstract] OR recovery of function[Title/Abstract] OR physical demand*[Title/Abstract] OR physically demand*[Title/Abstract]) OR ("Stress, Psychological"[Majr] OR "Social Support"[Majr] OR "Job Satisfaction"[Mesh] OR "Work Schedule Tolerance"[Mesh] OR "Employee Performance Appraisal"[Mesh] OR "Employee Grievances"[Mesh] OR "Social Justice/psychology"[Mesh] OR "Personnel Downsizing"[Mesh] OR "Staff Development"[Mesh] OR "Organizational Culture"[Mesh] OR "Bullying"[Mesh] OR "Prejudice"[Mesh] OR "Social Discrimination"[Mesh] OR "Interpersonal Relations"[Mesh] OR "Communication/psychology"[Mesh]) OR (psychosocial[Title/Abstract] OR job strain[Title/Abstract] OR work strain[Title/Abstract] OR work demand*[Title/Abstract] OR job demand*[Title/Abstract] OR high demand*[Title/Abstract] OR low control[Title/Abstract] OR lack of control[Title/Abstract] OR work control[Title/Abstract] OR job control[Title/Abstract] OR decision latitude[Title/Abstract] OR work influence*[Title/Abstract] OR demand resource*[Title/Abstract] OR effort reward*[Title/Abstract] OR time pressure*[Title/Abstract] OR recuperation*[Title/Abstract] OR work overload*[Title/Abstract] OR work over-load*[Title/Abstract] OR recovery[Title/Abstract] OR coping[Title/Abstract] OR work ability[Title/Abstract] OR social support[Title/Abstract] OR support system*[Title/Abstract] OR . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint social network*[Title/Abstract] OR emotional su pport[Title/Abstract] OR interpersonal relation*[Title/Abstract] OR interaction*[Title/Abstract] OR social capital [Title/Abstract] OR justice*[Title/Abstract] OR injustice*[Title/Abstract] OR job satisfaction[Title/Abstract] OR work satisfaction[Title/Abstract] OR boredom[Title/Abstract] OR skill discretion*[Title/Abstract] OR staff development[Title/Abstract] OR discrimination[Title/Abstract] OR harass*[Title/Abstract] OR work-place conflict*[Title/Abstract] OR workplace violen*[Title/Abstract] OR work-place violen*[Title/Abstract] OR bullying[Title/Abstract] OR ageism[Title/Abstract] OR homophobia[Title/Abstract] OR racism[Title/Abstract] OR sexism[Title/Abstract] OR victimization*[Title/Abstract] OR silent workplace*[Title/Abstract] OR role ambiguity[Title/Abstract] OR role-conflict*[Title/Abstract] OR work-role*[Title/Abstract] OR working hour*[Title/Abstract] OR working time[Title/Abstract] OR day-time[Title/Abstract] OR night-time[Title/Abstract] OR shift work*[Title/Abstract] OR work shift*[Title/Abstract] OR temporary work[Title/Abstract] OR full-time[Title/Abstract] OR part-time[Title/Abstract] OR flexible work*[Title/Abstract] OR organizational change[Title/Abstract] OR organisational change[Title/Abstract] OR lean production[Title/Abstract] OR job security[Title/Abstract] OR job insecurity[Title/Abstract])) AND (menstruation OR menstrual OR “menstrual pain” OR “period pain” OR “period pains” OR menorrhagia OR hypermenorrhea OR dysmenorrhea[MeSH] OR premenstrual OR “PMS” OR “premenstrual syndrome” OR “PMDD” OR “premenstrual dysphoric disorder” OR amenorrhea OR menoxenia OR metrorrhagia OR endometrio* OR myoma* OR leiomyoma OR fibroids OR adenomyosis OR ovarian cyst* OR endometrioma OR polycystic ovar* OR menopaus* OR hot flush* OR climacteric OR estrogen* OR estradiol OR progesterone* OR prolactin OR lutenizing OR LH OR “lutenizing hormone” OR follicul* OR FSH OR “follicle stimulating hormone” OR gonadotropin OR GnRH OR “gonadotropin releasing hormone” OR “basal body temperature” OR infertility OR fertility OR fecund* OR subfecundity OR reproductiv*) AND (longitudinal OR prospective OR cohort OR (follow AND up) OR observational OR (case AND (control or cohort)) OR case-control OR case-cohort) . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint Appendix 2: GRADE overview. Quality Assessment Studies Risk of Bias Inconsistency Indirectness Imprecision Publication bias Overall Quality of Evidence Fecundability 3 studiesa Very serious Not serious Not serious Serious e Very serious g Very low quality Diagnosis of endometriosis 1 studyb Very serious NA f NA f NA f Very seriousg Very low quality Age at natural menopause 1 studyc Very serious NA f NAf NA f Very seriousg Very low quality Reproductive hormone 1 studyd Very serious NA f NA f NA f Very seriousg Very low quality a Willis SK., 2019; Gaskins AJ., 2015; Barzilai-Pesach V., 2006. b Schernhammer ES., 2011. c Stock D., 2019. d Michels KA., 2020. e one study has small sample size f not applicable since no pooled estimate was calculated heterogeneity g no protocol available . CC-BY-NC-ND 4.0 International licenseIt is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) The copyright holder for this preprint this version posted June 17, 2022. ; https://doi.org/10.1101/2022.06.17.22276473doi: medRxiv preprint