{"paper_id":"1f170011-4fe0-4e0f-bd9c-ff73095ef186","body_text":"725\nCHAPTER 53\nPractice Note: ‘If Only All Women \nMenstruated Exactly Two Weeks Ago’: \nInterdisciplinary Challenges and Experiences \nof Capturing Hormonal Variation Across \nthe Menstrual Cycle\nLauren C. Houghton and Noémie Elhadad\nMenstrual health is important in its own right, but menstrual  characteristics, \nincluding the signs and symptoms that accompany the cycle and their \n underlying hormonal patterns, are also risk factors for chronic diseases such \nas breast cancer and endometriosis, two diseases to which we have dedicated \nour own research careers. One such risk factor for both diseases is early age \nat menarche (Bodicoat et al. 2014; Parazzini et al. 2017). When one of us \ngot her first period at age 12.11 years (LCH knows her exact age because \nit was on Christmas Eve), she was ashamed. Now 24 years and 288 men-\nstrual cycles later, she has personally interviewed over 1000 girls and women \nabout their menstrual cycles and the shame has subsided (Houghton et al. \n2014, 2018, 2019). When the other got her first period at age 13.9 years, she \nsoon realized that periods were going to play a painful part of her life. She \nwas later diagnosed with endometriosis and has since designed technology \nto give a voice to women living with endometriosis (McKillop, Mamykina, \nand Elhadad 2018; Urteaga et al. 2018). Since their first periods, secular \nchanges in societal views have also been dramatic: Teenage girls are coding \ntheir own video games “to rid the world of the menstrual taboo” (“Tampon \nRun,” n.d.), governments are passing “Menstrual Equity” legislation (Zraick \n2018), and women downloading period tracking apps have made them the \nsecond most used health app (PRIORI DATA). These societal and techno-\nlogical changes have also changed the way we and other anthropologists, \n© The Author(s) 2020 \nC. Bobel et al. (eds.), The Palgrave Handbook of Critical Menstruation  \nStudies, https://doi.org/10.1007/978-981-15-0614-7_53\n\n\n726  L. C. HOUGHTON AND N. ELHADAD\nepidemiologists, and data scientists interested in characterizing hormonal var -\niation in menstrual patterns conduct menstruation research.\nAnthropologists have been instrumental in demonstrating that there is \nglobal variation in menstruation, its timing across the life course, the cul-\ntural practices surrounding it, and the underlying hormones that regulate \nit. Starting in the 1980s, biological anthropologists went to the field in var -\nious populations and compared menstrual cycles among women living in \nvastly different ecological contexts with starkly different norms surrounding \nreproductive timing. They demonstrated that the Dogon women of West \nAfrica menstruate three times less than American and European women in \ntheir lifetime (Strassmann 1999). This dramatic difference in lifetime num-\nber of menstrual cycles is explained by the later age at menarche, earlier age \nat first pregnancy, higher parity, and earlier age at menopause in the Dogon  \nwomen.\nThe variation in number of lifetime menstrual cycles is also reflected in \nunderlying variation in reproductive hormones. For example, the overall pat-\ntern of progesterone across the menstrual cycle is similar among American, \nPolish, Lese, and Nepali women, with levels low and flat during the follicu-\nlar phase (days 1–14), rising during ovulation (day 14), peaking during the \nmid-luteal phase (day 21), and slowly declining in the late luteal phase until \nday 28. However, the absolute levels of hormone at each of these phases is \nthree-fold higher in American and Polish women compared to the Lese and \nNepalese (Ellison 1994). The variation in hormone profiles demonstrates nat-\nural variation in ovarian function across populations, and this variation should \nbe considered normal rather than pathological (Ellison et al. 1993).\nTechnological advances in hormone measurement made this important \nresearch possible. Peter Ellison and his colleagues leveraged the advancement \nin assays (Riad-Fahmy et al. 1982) to measure hormones in saliva for field set-\ntings (Ellison 1993). The advantage of using saliva was that anthropologists \ncould collect daily samples from women and, using preservatives, store the \nsamples at room temperature in isolated field locations where refrigeration \nwas not possible (Lipson and Ellison 1989). It was important to collect daily \nsamples to capture the hormonal profile of the entire cycle and to detect day \nof ovulation in order to align cycles.\nEpidemiologists have also been interested in the variation of menstrual \npatterns (timing and frequency) and reproductive hormones because of their \nassociations with breast cancer (Mumford et al. 2012; Whelan et al. 1994; \nTerry et al. 2005). As a result, hundreds of studies have compared hormones \nbetween women with and without breast cancer; between women with dif-\nferent risk factors, such as reproductive events, diets, and physical activity \nlevels; and between women living in populations with different breast can-\ncer incidence rates. Epidemiologists, who typically work with much larger \nstudy populations than anthropologists, conducted these comparisons using \nserum or urine samples collected on specific days of the menstrual cycle. In \norder to correctly time collection, epidemiologists asked women to recall the \n\n53 PRACTICE NOTE: ‘IF ONLY ALL WOMEN MENSTRUATED …   727\ndate of their last menstrual period (LMP) and then count forward to sched-\nule a specimen collection on the targeted day. In most cases the target day \nwould be menstrual cycle day 21, when both progesterone and estradiol are  \nrelatively high.\nRestricting comparison of hormones to specific days is cost-effective for \nlarge-scale studies, but it comes at a scientific cost in studies of premenopausal \nwomen. Characterizing a woman’s “dose” of estrogen based on only one or two \ntimed samples may not capture the full variation in hormones over the men-\nstrual cycle. For example, we know elevated estrogen levels are associated with \nincreased risk for breast cancer among postmenopausal women (from whom \na single sample can be collected since there is no menstrual cycle variation to \naccount for), but we are less confident this association holds in premenopausal \nwomen (Key 2011). Additionally, follicular phase estradiol is associated with \nincreased breast cancer risk, whereas luteal phase estradiol is not (Eliassen et al. \n2006). These gaps in understanding regarding premenopausal hormones and \ntheir relationship with breast cancer are particularly problematic given that breast \ncancer incidence is on the rise, specifically in US women under 40 years old \n(Johnson, Chien, and Bleyer 2013). So while epidemiologists have also contrib-\nuted to knowledge through large cohort studies, their narrow focus on meas-\nuring hormone levels on specific days has limited our understanding of the full \nvariability of hormones in menstruation.\nFurther, calibrating the hormone samples to the cycle day relies on wom-\nen’s ability to recall the date of their LMP, but cycle lengths vary within and \nbetween women and women’s recollections may not be accurate. For exam-\nple, some epidemiologists have followed up with women to confirm that the \nsample collection in fact occurred on or near the targeted day. They pro-\nvided participants, who in this case were highly motivated nurses, with a \n pre-addressed postcard to return with the date when their next period began. \nThey found that specimens were collected as many as four days before or after \nthe target day in the best-case scenarios (Eliassen et al. 2006).\nIn our experience, some women record their menstrual period in their own \npersonal calendar, but many women do not. Over the course of interviewing \nwomen and asking them to recall their LMP, we have observed several esti-\nmation techniques. If her period just finished, she would count back five days \nto the day it began. Or, she would think of where she was the last time it \noccurred and then work out the date based on where she was that month. \nIf her period had not been within the last week or if there was no significant \nevent to help jog her memory, the most common response was to default to \n“oh, about two weeks ago.” This suggests that self-reported LMP is a rather \nrough estimate.\nAdvancements in statistical methods can overcome some of these chal-\nlenges when collecting daily specimens is not feasible. In one study, one of \nus took an alternative approach to targeting a specific cycle day. We collected \nspecimens from all of the women in our study over the same two calen-\ndar days and later asked them to confirm when their next menstrual period \n\n728  L. C. HOUGHTON AND N. ELHADAD\n(NMP) started (Houghton et al. 2016; Troisi et al. 2014). Thus, by chance, \nwomen were at different days of their menstrual cycle and this yielded a col-\nlection of samples randomly distributed across a menstrual cycle. We then \ntook this random distribution and modeled (using cubic splines) the hor -\nmone profile at the population level using individual data on specific days. \nWithout giving the model a specified shape, the curves mimicked biological \ncurves seen in individual women. We were then able to compare hormone \nconcentrations between populations by calculating the area under each curve. \nIn this specific case, we were able to demonstrate that estrogens were actu-\nally higher in Mongolian women than in British women, a finding that was \ncounter to what most people would predict, given that Mongolia has some \nof the lowest breast cancer rates in the world (Troisi et al. 2014). Confirming \nthe date of the NMP was of the utmost importance, to make sure we knew \nthe exact menstrual day on which the sample had been collected. Data sci-\nentists use even more sophisticated modeling and prediction techniques to \npredict the hormone profiles underlying women’s menstrual cycle character -\nistics. For example, the state-of-the-art model developed by Clark and col-\nleagues accurately captures the same hormonal patterns as a dataset of daily \nhormonal measurements from healthy women (Harris Clark, Schlosser,  \nand Selgrade 2003).\nOur own experience in collecting and analyzing menstrual cycle and \nhormonal data mirrors the methodologies in the literature, ranging from \nsmall-scale studies with “in-depth” hormone measurements (daily readings \nthroughout the menstrual cycle) to large-scale cohort analysis with “shallow” \nhormone measurements (under-sampling in time, or sampling only at known \nphases of the cycle). The “deepest” examples can be found in reproductive \nendocrinology, where researchers have established basic knowledge about \nthe cycle through small-scale studies of 12 women and 15-minute interval \nblood serum measurements (Murdoch et al. 1985). But such a deep dive \nis not the route we have chosen to follow. Rather, we are both now excited \nto incorporate mobile apps into our research toolbox to mitigate the ten-\nsion between the need for multiple measurements and the invasive nature of \ntaking such measurements. Mobile health is a game changer for investigat-\ning the menstrual cycle, making data collection possible at an unprecedented \nscale. Women can now use menstrual tracking apps to track their own cycles, \nalong with a range of signs and symptoms—data which previously did not \nget recorded. It is not only possible, women are doing it. Cycle tracking apps \nare some of the fastest-growing health apps and have a loyal base of users \n(Fox and Duggan 2012; Wartella et al. 2016). In 2018, 48% of US females \nages 18- to 22-years-old and 25% of teen girls reported having used a period \ntracking app (Fox 2018). Some apps have as many as 10 million users world-\nwide (“Clue Wants Your Research Proposals on Reproductive Health,” n.d.). \nThere is a great opportunity to leverage self-tracking technology to enable \ncharacterizing the menstrual cycle at scale and for individuals.\n\n53 PRACTICE NOTE: ‘IF ONLY ALL WOMEN MENSTRUATED …   729\nOne promising aspect of apps used for period tracking is their flexibility \nin capturing a range of experiences related to menstruation. Popular track-\ners enable users to self-track their cycle and period lengths, as well as spe-\ncific symptoms (such as, headaches, sleep, bowel movements, pain levels); \npsychosocial elements (such as, moods or socializing patterns); and sexual \nactivity. Many apps further enable users to customize their tracking options, \nthus getting closer to capturing the broader context and the narrative \naround menstruation. While cycle characteristics on their own might not be \nenough for algorithms to model cycle variations, the additional data elements \ncan significantly boost these models’ ability to infer underlying hormonal  \nvariations.\nBeyond self-tracked data, emerging analytics can help us determine the \nideal hormone-sampling rate for future studies or improved diagnostics. \nFor example, Urteaga and colleagues (2017, 2019) showed that machine \n learning-based approaches that combine the aforementioned state-of-the art \nmodel for daily average hormone concentrations with cycle data collected \nusing a mobile phone application can accurately identify the phases within \nthe menstrual cycle, even without daily measurements. Furthermore, the \nmodel allows for generation of hormone patterns with different character -\nistics, such as varying cycle length. Once these methods are validated with \nin-depth hormonal datasets, they have the potential to revolutionize how we \nstudy the variation in menstrual cycle and hormonal patterns in healthy and  \nunhealthy women.\nAs app developers try to perfect their algorithms to predict each user’s next \nperiod more accurately, and researchers try to use app tracking data to pre-\ndict underlying hormonal profiles, this momentum must be accompanied by \nsome caution and critical reflections. First, there is the risk that only using \ndata collected by menstruators who use apps may further marginalize those \nmenstruators who don’t use apps for biological (irregular periods) or socially \npatterned reasons. For example, while smart phone and app use is global, \nwithin emerging markets people with less education are still less likely to \nuse apps than those with more education (Taylor and Silver 2019). Second, \nnew mHealth data collection methods may lead to large menstrual cycles \nstudies at unprecedented scales, but the cultural norms with which the tech-\nnology was developed may obscure the cultural context of the populations \nunder study (Fox and Epstein [Chapter 54] in this volume). Principles of cit-\nizen science might be helpful in engaging on-the-ground perspectives from \ndiverse groups of menstruators. From the collective work of anthropologists, \nepidemiologists, and data scientists, we know that our culture and biology \nboth influence our menstrual health. We should ensure that our methods are \nequally attuned to biological plausibility and cultural context.\n\n730  L. C. HOUGHTON AND N. ELHADAD\nRefeRences\nBodicoat, Danielle H., Minouk J. Schoemaker, Michael E. Jones, Emily McFadden, \nJames Griffin, Alan Ashworth, and Anthony J. Swerdlow. 2014. “Timing of \nPubertal Stages and Breast Cancer Risk: The Breakthrough Generations Study.” \nBreast Cancer Research 16 (1): R18. https://doi.org/10.1186/bcr3613.\n“Clue Wants Your Research Proposals on Reproductive Health.” n.d. Accessed March \n19, 2019. https://helloclue.com/articles/about-clue/clue-research-innovation- \nprogram-call-for-proposals.\nEliassen, A. Heather, Stacey A. Missmer, Shelley S. Tworoger, Donna Spiegelman, \nRobert L. Barbieri, Mitch Dowsett, and Susan E. Hankinson. 2006. “Endogenous \nSteroid Hormone Concentrations and Risk of Breast Cancer among Premenopausal \nWomen.” Journal of the National Cancer Institute 98 (19): 1406–15. https://doi.\norg/10.1093/jnci/djj376.\nEllison, P. T. 1993. “Measurements of Salivary Progesterone.” Annals of the New York \nAcademy of Sciences 694 (September): 161–76. http://www.ncbi.nlm.nih.gov/\npubmed/8215052.\n———. 1994. “Salivary Steroids and Natural Variation in Human Ovarian Function.” \nAnnals of the New York Academy of Sciences 709 (February): 287–98. http://www.\nncbi.nlm.nih.gov/pubmed/8154721.\nEllison, P. T., C. Panter-Brick, S. F. Lipson, and M. T. O’Rourke. 1993. “The \nEcological Context of Human Ovarian Function.” Human Reproduction 8 (12): \n2248–58. http://www.ncbi.nlm.nih.gov/pubmed/8150934.\nFox, Susannah. 2018. “Digital Health Practices among Teens and Young Adults in \nthe U.S.” 2018. https://susannahfox.com/2018/07/31/digital-health-practices- \namong-teens-and-young-adults-in-the-u-s/.\nFox, S., and M. Duggan. 2012. “Mobile Health 2012 | Pew Research Center.” \nn.d. Accessed March 19, 2019. http://www.pewinternet.org/2012/11/08/\nmobile-health-2012/.\nHarris Clark, L., Paul M. Schlosser, and James F. Selgrade. 2003. “Multiple Stable \nPeriodic Solutions in a Model for Hormonal Control of the Menstrual Cycle.” \nBulletin of Mathematical Biology 65 (1): 157–73. https://doi.org/10.1006/\nbulm.2002.0326.\nHoughton, L. C., G. D. Cooper, G. R. Bentley, M. Booth, O. A. Chowdhury,  \nR. Troisi, R. G. Ziegler, R. N. Hoover, and H. A. Katki. 2014. “A Migrant Study \nof Pubertal Timing and Tempo in British-Bangladeshi Girls at Varying Risk \nfor Breast Cancer.” Breast Cancer Research 16 (1). https://doi.org/10.1186/\ns13058-014-0469-8.\nHoughton, Lauren, Davaasambuu Ganmaa, Philip Rosenberg, Dambadarjaa \nDavaalkham, Frank Stanczyk, Robert Hoover, and Rebecca Troisi. 2016. \n“Associations of Breast Cancer Risk Factors with Premenopausal Sex Hormones \nin Women with Very Low Breast Cancer Risk.” International Journal of \nEnvironmental Research and Public Health 13 (12): 1066. https://doi.\norg/10.3390/ijerph13111066.\nHoughton, L. C., J. A. Knight, M. J. De Souza, M. Goldberg, M. L. White, K. \nO’Toole, W. K. Chung, et al. 2018. “Comparison of Methods to Assess Onset of \nBreast Development in the LEGACY Girls Study: Methodological Considerations \nfor Studies of Breast Cancer.” Breast Cancer Research 20 (1). https://doi.\norg/10.1186/s13058-018-0943-9.\n\n53 PRACTICE NOTE: ‘IF ONLY ALL WOMEN MENSTRUATED …   731\nHoughton, Lauren C., Julia A. Knight, Ying Wei, Russell D. Romeo, Mandy \nGoldberg, Irene L. Andrulis, Angela R. Bradbury, et al. 2019. “Association of \nPrepubertal and Adolescent Androgen Concentrations with Timing of Breast \nDevelopment and Family History of Breast Cancer.” JAMA Network Open 2 (2): \ne190083. https://doi.org/10.1001/jamanetworkopen.2019.0083.Johnson.\nJohnson, Rebecca H., Franklin L. Chien, and Archie Bleyer. 2013. “Incidence of \nBreast Cancer with Distant Involvement among Women in the United States, 1976 \nto 2009.” JAMA 309 (8): 800–805. https://doi.org/10.1001/jama.2013.776.\nKey, Timothy J. 2011. “Endogenous Oestrogens and Breast Cancer Risk in \nPremenopausal and Postmenopausal Women.” Steroids 76 (8): 812–15. https://\ndoi.org/10.1016/J.STEROIDS.2011.02.029.\nLipson, S. F., and P. T. Ellison. 1989. “Development of Protocols for the Application \nof Salivary Steroid Analysis to Field Conditions.” American Journal of Human \nBiology 1 (3): 249–55. https://doi.org/10.1002/ajhb.1310010304.\nMcKillop, Mollie, Lena Mamykina, and Noémie Elhadad. 2018. “Designing in the \nDark.” In Proceedings of the 2018 CHI Conference on Human Factors in Computing \nSystems—CHI ’18, 1–15. ACM Press, New York, New York, USA. https://doi.\norg/10.1145/3173574.3174139.\nMumford, Sunni L., Anne Z. Steiner, Anna Z. Pollack, Neil J. Perkins, Amanda C. \nFiliberto, Paul S. Albert, Donald R. Mattison, Jean Wactawski-Wende, and Enrique \nF. Schisterman. 2012. “The Utility of Menstrual Cycle Length as an Indicator \nof Cumulative Hormonal Exposure.” The Journal of Clinical Endocrinology & \nMetabolism 97 (10): E1871–79. https://doi.org/10.1210/jc.2012-1350.\nMurdoch, A. P., P. J. Diggle, W. Dunlop, and P. Kendall-Taylor. 1985. \n“Determination of the Frequency of Pulsatile Luteinizing Hormone Secretion by \nTime Series Analysis.” Clinical Endocrinology 22 (3): 341–46. http://www.ncbi.\nnlm.nih.gov/pubmed/3978835.\nParazzini, F., G. Esposito, L. Tozzi, S. Noli, and S. Bianchi. 2017. “Epidemiology of \nEndometriosis and Its Comorbidities.” European Journal of Obstetrics & Gynecology \nand Reproductive Biology 209 (February): 3–7. https://doi.org/10.1016/j.\nejogrb.2016.04.021.\n“PRIORI DATA—Mobile App Data Intelligence, ASO Tool & Market Research.” \nn.d.\nRiad-Fahmy, D., G. F. Read, R. F. Walker, and K. Griffiths. 1982. “Steroids in Saliva \nfor Assessing Endocrine Function.” Endocrine Reviews 3 (4): 367–95. https://doi.\norg/10.1210/edrv-3-4-367.\nStrassmann, B. I. 1999. “Menstrual Cycling and Breast Cancer: An Evolutionary \nPerspective.” Journal of Women’s Health 8 (2): 193–202. http://www.ncbi.nlm.\nnih.gov/pubmed/10100133.\n“Tampon Run.” n.d. Accessed March 19, 2019. http://tamponrun.com/.\nTaylor, Ken, and Laura Silver. 2019. “Smartphone Ownership Is Growing Rapidly \nAround the World, but Not Always Equally | Pew Research Center.” Pew Research \nCenter, 2019. http://www.pewglobal.org/2019/02/05/smartphone-ownership-\nis-growing-rapidly-around-the-world-but-not-always-equally/.\nTerry, Kathryn L., Walter C. Willett, Janet W. Rich-Edwards, David J. Hunter, and \nKarin B. Michels. 2005. “Menstrual Cycle Characteristics and Incidence of \nPremenopausal Breast Cancer.” Cancer Epidemiology Biomarkers & Prevention 14 \n(6): 1509–13. https://doi.org/10.1158/1055-9965.EPI-05-0051.\n\n732  L. C. HOUGHTON AND N. ELHADAD\nTroisi, Rebecca, Daavasambuu Ganmaa, Isabel dos Santos Silva, Dambadarjaa \nDavaalkham, Philip S. Rosenberg, Janet Rich-Edwards, Lindsay Frasier, et al. \n2014. “The Role of Hormones in the Differences in the Incidence of Breast \nCancer between Mongolia and the United Kingdom.” PLoS One 9 (12): e114455. \nhttps://doi.org/10.1371/journal.pone.0114455.\nUrteaga, Iñigo, David J. Albers, Marija Vlajic Wheeler, Anna Druet, Hans Raffauf, \nand Noémie Elhadad. 2017. “Towards Personalized Modeling of the Female \nHormonal Cycle: Experiments with Mechanistic Models and Gaussian Processes,” \nMachine Learning for Health Workshop at NIPS (ML4H’17).\nUrteaga, Iñigo, Mollie McKillop, Sharon Lipsky-Gorman, and Noémie Elhadad. \n2018. “Phenotyping Endometriosis through Mixed Membership Models of \n Self-Tracking Data,” November. Machine Learning for Healthcare (MLHC’18).\nUrteaga, Iñigo, Tristan Bertin, Theresa Hardy, David Albers, and Noémie Elhadad. \n2019. “Multi-Task Gaussian Processes and Dilated Convolutional Networks for \nReconstruction of Reproductive Hormonal Dynamics,” Machine Learning for \nHealthcare (MLHC’19).\nWartella, Ellen, Vicky Rideout, Heather Montague, Leanne Beaudoin-Ryan, and \nAlexis Lauricella. 2016. “Teens, Health and Technology: A National Survey.” \nMedia and Communication 4 (3): 13. https://doi.org/10.17645/mac.v4i3.515.\nWhelan, E. A., D. P. Sandler, J. L. Root, K. R. Smith, and C. R. Weinberg. \n1994. “Menstrual Cycle Patterns and Risk of Breast Cancer.” American \nJournal of Epidemiology 140 (12): 1081–90. http://www.ncbi.nlm.nih.gov/\npubmed/7998590.\nZraick, Karen. 2018. “It’s Not Just the Tampon Tax: Why Periods Are Political.” The \nNew York Times, July 22, 2018. https://www.nytimes.com/2018/07/22/health/\ntampon-tax-periods-menstruation-nyt.html.\nOpen Access This chapter is licensed under the terms of the Creative Commons \nAttribution 4.0 International License (http://creativecommons.org/licenses/\nby/4.0/), which permits use, sharing, adaptation, distribution and reproduction in \nany medium or format, as long as you give appropriate credit to the original author(s) \nand the source, provide a link to the Creative Commons license and indicate if \nchanges were made.\nThe images or other third party material in this chapter are included in the \nchapter’s Creative Commons license, unless indicated otherwise in a credit line to the \nmaterial. If material is not included in the chapter’s Creative Commons license and \nyour intended use is not permitted by statutory regulation or exceeds the permitted \nuse, you will need to obtain permission directly from the copyright holder.","source_license":"CC0","license_restricted":false}