key: cord-0334389-xq2wnuks
authors: Lee, K. M.; Junkins, E. J.; Fatima, U. A.; Cox, M. L.; Clancy, K. B.
title: Characterizing menstrual bleeding changes occurring after SARS-CoV-2 vaccination
date: 2021-10-12
journal: nan
DOI: 10.1101/2021.10.11.21264863
sha: a228d12cb391042f86ebc710edecb42fe63c4d0d
doc_id: 334389
cord_uid: xq2wnuks

Many people began sharing that they experienced unexpected menstrual bleeding after SARS-CoV-2 inoculation. This emerging phenomenon was undeniable yet understudied. We investigated menstrual bleeding patterns among currently and formerly menstruating people, with a research design based off our expectations that these bleeding changes related to changes in clotting or inflammation, affecting normal menstrual repair. In this sample, 42% of people with regular menstrual cycles bled more heavily than usual, while 44% reported no change, after being vaccinated. Among people who typically do not menstruate, 71% of people on long-acting reversible contraceptives, 39% of people on gender-affirming hormones, and 66% of post-menopausal people reported breakthrough bleeding. We found increased/breakthrough bleeding was significantly associated with age, other vaccine side effects (fever, fatigue), history of pregnancy or birth, and ethnicity. Changes to menstrual bleeding are not uncommon nor dangerous, yet attention to these experiences is necessary to build trust in medicine.

SARS-CoV-2 inoculation. This emerging phenomenon was undeniable yet understudied. We 32 investigated menstrual bleeding patterns among currently and formerly menstruating people, with 33 a research design based off our expectations that these bleeding changes related to changes in 34 clotting or inflammation, affecting normal menstrual repair. In this sample, 42% of people with 35 regular menstrual cycles bled more heavily than usual, while 44% reported no change, after being 36 vaccinated. Among people who typically do not menstruate, 71% of people on long-acting 37 reversible contraceptives, 39% of people on gender-affirming hormones, and 66% of post-38 menopausal people reported breakthrough bleeding. We found increased/breakthrough bleeding 39 was significantly associated with age, other vaccine side effects (fever, fatigue), history of 40 pregnancy or birth, and ethnicity. Changes to menstrual bleeding are not uncommon nor 41 dangerous, yet attention to these experiences is necessary to build trust in medicine. 49 Menstruating and formerly menstruating people began sharing that they experienced 50 unexpected bleeding after being administered a SARS-CoV-2 vaccine in early 2021. Vaccine trial 51 protocols do not typically monitor for major adverse events for more than seven days, and 52 additional follow up communications do not inquire about menstrual cycles or bleeding. 53 Therefore, manufacturers had no way of addressing the extent to which this observation was a 54 coincidence or a potential side effect of the vaccines. In mainstream media coverage, medical 55 doctors and public health experts hastened to say that there was "no biological mechanism" or "no 56 data" to support a relationship between vaccine administration and menstrual changes. In other 57 cases experts declared that these changes were more likely a result of "stress" (1) (2) (3) (4) . 58 Unfortunately, dismissal by medical experts fueled greater concerns, as both vaccine 59 hesitant and anti-vaccine individuals and organizations began to conflate the possibility of short-60 term menstrual changes with long-term harms to fertility. Pundits, politicians, religious leaders, 61 and wellness influencers worked the oft-used framing of protecting women to advise against the 62 vaccine (5-9). As the SARS-CoV-2 vaccine became available to adolescents, calls to understand 63 the menstrual changes associated with the vaccine increased as parents felt they were weighing 64 their child's pubertal development and future fertility against their risk of getting COVID-19(10, 65 11). 66 There are in fact multiple plausible biological mechanisms to explain a relationship 67 between an acute immune challenge like a vaccine (12), its corresponding and well-known 68 systemic effects on hemostasis and inflammation (13), and menstrual repair mechanisms of the 69 uterus (14-17), The uterine reproductive system is flexible and adaptable in the face of stressors, 70 in order to weather short-term challenges in a way that leaves long-term fertility intact (18, 19) . 71 We know that running a marathon may influence hormone concentrations in the short term while 72 not rendering that person infertile (20); that short-term calorie restriction that results in a loss of 73 menstrual cycling can be overcome by resuming normal feeding (21); that inflammation 74 influences ovarian hormones (22-24); and that psychosocial stressors can correspond to cycle 75 irregularity and yet resilience can buffer one from these harms (25-27). Less severe, short-term 76 stressors can and do influence menstrual cycling and menstruation, and this has been established 77 over forty years of cycle research (19, 20, (28) (29) (30) ; However this work has also established that 78 while sustained early stressors can influence adult hormone concentrations, short-term stressors 79 resolve and do not produce long-term effects (31); This is quite different from the sustained 80 immune assault of COVID-19 itself: studies and anecdotal reports are already demonstrating that 81 menstrual function may be disrupted long-term, particularly in those with long COVID (32). 82 Vaccines function by mobilizing the immune system to protect from disease if exposure 83 occurs. This immune activation is important, although it may also produce a cascade of other 84 localized (e.g., soreness at injection site) or systemic (e.g., fatigue, fever) inflammatory responses. 85 Studies that assess the direct effect of vaccination on the menstrual cycle are few and far between. 86 A study from 1913 identified that the typhoid vaccine was associated with menstrual 87 irregularities, which included missed, late, and early menstruation, discomfort, and heavy 88 bleeding in more than half of their female sample (33). Hepatitis B studies have also indicated 89 that menstruation could be altered (34), and a HPV post-market safety study found that over a Here we share results from our first round of analyses of this instrument (N=39,129), as 117 well as the ways that this early exploration has made it possible to establish the parameters of the 118 phenomenon of post-vaccine menstrual change. We focus specifically on data related to menstrual 119 bleeding (in people who menstruate regularly) or breakthrough bleeding (in people who do not 120 currently menstruate) from the first three months of data collection in order to provide a timely 121 description of trends to clinicians and the public alike. Specifically, we sought to answer the 

Demographics and summary statistics 142 After data cleaning and aggregation of the first three months of data collection (Fig. 1 Summary statistics of menstrual changes and breakthrough bleeding occurrences 164 We examined menstrual experiences overall and by conservatively defined subgroups, 165 based upon self-reported typical pre-vaccine menstrual cycle status. We identified two major 166 groupsthose who regularly menstruate, and those who do not currently menstruate but have in 167 the past. The subgroups of regularly menstruating people are 1) premenopausal people (ages 18-168 45) with spontaneous menstrual cycles and 2) premenopausal people (ages 18-45) with 169 hormonally contracepting cycles who still bleed regularly. The subgroups of non-menstruating 170 people are 1) premenopausal people (ages 18-45) on hormonal treatments that suppress 171 menstruation (e.g., hormonal contraceptives, long-acting reversible contraception (LARC), 172 testosterone), and 2) postmenopausal people (ages 55-80, no period for at least 12 months). Our 173 analyses seeking to characterize the range of typical experiences focus on regularly cycling 174 respondents without diagnosed reproductive conditions. We include respondents with 175 reproductive conditions in our additional analyses of trends of which groups appear to be more at 176 risk for changed bleeding patterns. Our analyses of breakthrough bleeding include people with 177 and without diagnosed reproductive conditions after initial inspection did not reveal those groups 178 to be significantly different. See Fig. 1 and Methods for data cleaning details and detailed 179 subgroup descriptions; subgroup demographics are in Table S3 . 180 We first report descriptive statistics, including the proportions of our sample who 181 experienced changes in bleeding heaviness and duration (for people who regularly menstruate) or 182 breakthrough bleeding (for those who typically do not menstruate), overall and by subgroups. We 183 then analyzed associations between menstrual changes and age, race, and ethnicity, vaccine type 184 (restricted to the most common 2-dose vaccines, Pfizer and Moderna), vaccine symptoms, typical 185 period experience, reproductive history, and diagnosed reproductive conditions focusing on the 186 same sub-groups. As these sample sizes were large, we used alpha threshold p<.001 when testing 187 for associations within subgroups greater than 800, which meant respondents on gender-affirming 188 treatments and postmenopausal respondents we used the typical alpha threshold p<.05. 195 . CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Due to the proportion of people experiencing heavy flow after at least one of the vaccines, 196 we grouped regularly cycling individuals with any heavier flow into one condition ("heavier"; 197 N=7,429 of 17,642; 42.11%), people who experienced no change in flow after either dose into the 198 second condition ("no change"; N=7,684; 43.56%), and the remainder of people who experienced 199 a combination of lighter and no change after their doses into a smaller third condition ("not   200 heavier", N=2,529; 14.34%). In total, 727 were missing dose 1 period flow information and 3,031 201 were missing dose 2 period flow information (Table S5) . We similarly grouped people reporting 202 any longer bleeding into one condition ("longer"; N=5,9780 of 17,366; 34.42%), people who 203 experienced no change in period length after either dose into a second group ("no change", 204 N=8,914; 51.33%), and the remainder of people who experienced a combination of shorter and no 205 change to period length across doses into a smaller group ("not longer", N=2,474; 14.25%). 206 Period length information was missing for 776 people following dose 1 and 3,342 people 207 following dose 2. If information was missing at either dose, we treated it as pairwise missingness, 208 so the respective menstrual change variable was missing. Thus, period flow change is described 209 as heavier, no change, or not heavier and period length change is described as longer, no change, 210 or not longer. Since lighter flow or shorter duration bleeding is meaningful but less common in 211 this sample, we include this condition in our analyses but focus mainly on factors associated with 212 heavier bleeding as compared to no change in bleeding. (Table S4) . Differences between the regularly menstruating groups in post-222 vaccine menstrual flow and duration were statistically significant, with hormonally contracepting 223 people being less likely to experience no change in flow (χ 2 (2, N=12,816)=17.338, p=0.0002) and 224 more likely to experience a longer menstrual duration (χ 2 (2, N=12,608)=49. 689, p=1.62E-11) . 225 When comparing spontaneously cycling and hormonally contracepting regularly 226 menstruating respondents, we found differences between these groups in past pregnancy, parity, 227 usual period flow, usual period length, race, ethnicity, and age (Nsp=11,700 and Nhc=3,855; Table   228 S6). Vaccine type and local/systemic symptoms did not differ between subgroups. . CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Age, gender, race, and ethnicity 246 We tested for significant differences between age and change to one's period flow after 247 vaccination within the regularly cycling subgroups. Age was significantly different only in the 248 spontaneously cycling subgroup (F(2, 9,576)=34.1, p=1.70E-15). Those who reported heavier flow 249 change were older (M=32.9 years) than those with no change (M=31.7; p=8.18E-15) or not 250 heavier change (M=31.7; p=8.06E-08). 251 No significant associations were found in either of the regularly cycling subgroups when 252 we tested for any association between racial group and changes to period flow after vaccination. 253 When we examined for an association with ethnicity, we found a significant relationship in the was small for this relationship (φc=0.060). 258 We examined for significant differences between age and presence of breakthrough Typical menstrual experiences 293 We tested for associations between a respondent's usual menstrual blood flow and 294 whether they experienced a change to that flow after the vaccine in regularly cycling subgroups.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint There was a significant relationship with typical menstrual blood flow and change to menstrual 296 blood flow in the regularly cycling groups (spontaneously cycling: χ 2 (4, N=9,537) =61 Because age, pregnancy, and parity were all associated with a heavier flow in our sample, 331 and these variables have a tendency to be associated with each other, we tested for average age 332 differences in those with past pregnancy/no pregnancy and parous/not parous. Respondents with a 333 history of pregnancy or who were parous were significantly older (pregnancy: t(10,738)=-77.73, 334 p<2.2E -16 ; parity: t(9,440)=-80.678, p < 2.2E -16 ). People who had been pregnant were older 335 (Mage=37.47) than people with no history of pregnancy (Mage=29.08), and parous respondents 336 were also older (Mage=38.13) than non-parous (Mage=29.54). Finally, we looked only at 337 respondents with no history of pregnancy and tested for age differences to try to separate the 338 effect of age and reproductive history. We found those who reported a heavier flow post-vaccine 339 were more likely to be older (F(2, 6,093)=9.091, p=0.0001). Again, the heavier condition was is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Reproductive conditions 344 We then made comparisons for those diagnosed with a specific reproductive condition 345 against the appropriate (i.e., spontaneously cycling or hormonally contracepting) larger sample 346 without any diagnosed conditions (Table 6 ). In spontaneously cycling subgroups, a higher without diagnosed reproductive conditions. We cannot estimate prevalence or incidence based on 358 our methodological approach of this emergent phenomenon. However, our results highlight that 359 these changes affect a large number of people and some of the trends we observe support 360 hypothesis development for additional prospective study in hemostatic and inflammatory changes 361 to the endometrium after an acute immune response (Fig. 4) . 362 In this first analysis, we focus on the heavier bleeding of currently menstruating and 363 breakthrough bleeding of formerly menstruating people, which we define as an increased bleeding 364 phenotype. The increased bleeding phenotype appeared to be the most common post-vaccination 365 change within our sample. Initial forays into our qualitative data suggest a widely variable 366 experience of the increased bleeding phenotype, confounding a straightforward case definition. At 367 this time, we suggest that rather than a threshold quantity to define the increased bleeding CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint bleeding runs the risk of psychological distress for those who experience gender dysphoria with 393 menstruation, and physical harm for people for whom managing menstruation in public is 394 dangerous (53, 54). 395 In addition to our finding of a significant proportion of respondents experiencing some 396 form of increased bleeding, we noticed some trends in who was more likely to have this 397 phenotype. Among premenopausal respondents, those who were older and/or Hispanic (using 398 U.S. census demographic approaches) were more likely to report heavier bleeding post-vaccine. 399 Prior pregnancy and prior birth also were associated with a greater risk of heavier bleeding. 400 Finally, premenopausal, spontaneously cycling respondents who were diagnosed with 401 endometriosis, menorrhagia, and/or fibroids were more likely to report experiencing heavier 402 bleeding post-vaccine compared to those without any diagnosed reproductive condition. We also 403 find that many respondents who had post-vaccine changes did not have them until fourteen days CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint reproduce inequitable power structures of the world. Research which notices and attends to the 492 experiences of people is a necessary first step to restore trust and create transparency in science. 493 We have documented a phenotype of increased menstrual bleeding across a diverse set of 494 currently and formerly menstruating people as a post-vaccine response. In doing so, we help 495 provide evidence and context for clinicians regarding the validity of these experiences and we 496 note future avenues of inquiry for researchers. Recognizing and attending to this emerging 497 phenomenon of bleeding changes can help bolster trust between people who menstruate and 498 medical providers, which is an area that has a long history of medical misogyny and gaslighting 499 (78-81). Current and historic focus on fertility and reproduction in research and clinical trials is 500 insufficient for addressing the changes in bleeding patterns that cause concern in many people. 501 We urge other researchers and funding bodies to increase investment in understanding queer, 502 trans, and nonbinary menstrual experiences, because there is a dearth of existing literature to 503 understand the biosocial context of menstrual bleeding in these groups. Furthermore, we note that 504 postmenopausal bleeding is still massively understudied. Mixed-methods and community based 505 participatory research to address questions that matter to those historically excluded from 506 reproductive and menstruation science is needed in order to provide adequate and culturally and 507 physically relevant care to these populations. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint second dose variables for the single-dose Johnson & Johnson vaccine. As these effects we 592 observed were an emerging phenomenon we focus primarily on descriptive statistics and trends. 593 In each subgroup we focused on menstrual changes following vaccination. In regularly 594 cycling subgroups, changes to menstrual flow and menstrual length were investigated. These were 595 defined as 'heavier', 'not heavier', or 'no change'. In non-menstruating subgroups, breakthrough 596 bleeding was investigated. We describe the approximate frequency of breakthrough bleeding as 597 'after both', 'only dose 2', 'only dose 1', or 'none'. We examined for differences based on 598 whether respondents reported any breakthrough (see measures in online Supplemental for more 599 details). 600 Chi-square test of independence were used in two ways-to examine whether a study 601 variable was associated with bleeding outcomes and to determine whether there was an 602 association between study variables and the subgroup defined as spontaneous versus hormonally 603 contracepting or without versus with diagnosed reproductive conditions. Study variables were not 604 compared either within or across subgroups when the sample sizes per cell were less than 20. For 605 regularly cycling subgroups, we omitted respondents with missing data and respondents who 606 selected 'other' usual menstrual flow or length, most of whom gave detailed unique descriptions 607 via text entry. Association with vaccine type was restricted to Pfizer and Moderna due to a largely 608 US based sample and, therefore, large sample numbers for these vaccines for the early April -609 early July time period. 610 We investigated other factors likely to be associated with reported bleeding experiences 611 by analyzing each subgroup. We calculated the proportions within each independent variable that 612 were associated with each bleeding outcome (flow change or breakthrough depending on the 613 subsample). Chi-square tests of independence were run within each sample subgroup to test for 614 significant associations between bleeding condition and study variables. Pairwise missingness 615 was used to handle missing data. All study variables were nominal, except age. When the 616 contingency table was 2x2, Yate's continuity correction was applied. Alongside the chi-square 617 tests for associations with bleeding outcomes, we calculated Cramér's V (φc) to estimate effect p<.05 which means we have 28% power to detect a similar effect. As a result of large sample 627 sizes in select subgroups, we did expect to find significant associations that related to small 628 differences in proportions. In most subgroups, we had the power to detect an effect of 7% 629 proportion differences with 90% power at p<.001, minimum sample size per proportion was 993, 630 and 3% proportion differences required a minimum of 4,724 individuals per proportion. To detect 631 a 7% proportion difference, we required 993 individuals per sample proportion. With similar 632 proportion differences at 90% power with alpha threshold at p<.05, we required samples of 499 633 for 7% differences and 2,374 for 3% differences. Therefore, the smallest subgroups where p<.05 634 was the alpha threshold, namely people using gender-affirming hormones and post-menopause 635 people, we had 90% power to detect 25% proportion differences and less than 20% power to 636 detect 7% proportion differences. As the main goal of this paper was to describe the experiences 637 of a wide range of people, we acknowledged the limitation of these significance tests and focus on 638 effect estimates and the odds ratio for significant associations. Following the results of the chi-639 square tests, the comparative odds of bleeding conditions, namely heavier flow change and 640 breakthrough bleeding, were calculated for significant associations using odds ratio.

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint All analyses were run in R Studio (87 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint individuals reporting breakthrough bleeding (y-axis) after both doses, only following dose 2, only following dose 1, 914 or no breakthrough bleeding during vaccination time. 915 916 917

. CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Ages are binned based on approximate life stages and the later subgroup analysis. 927

Other vaccines includes Novavax, AstraZeneca, and self-reported other. These percentages do not add up to 100% 928 because while all responded appropriately to whether they had two doses, some did not report the vaccine-type. 929 . CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint CC-BY-NC-ND 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Note: These proportions are calculated without inclusion of missing or omitted categories in the total sample size in 940 each flow change condition displayed at the bottom (displayed percentages should equal 100%). Independent 941 variables are displayed in the rows, and the dependent variable is displayed in the columns. The column N reported is 942 the total N per level of the independent variable used in analysis. The chi-square results refer to the left hand-side (no 943 diagnosed condition) on top then directly below the same comparison in the diagnosed conditions subgroups 944 (shaded). 945

HC is hormonal contraception, PPH is post-partum hemorrhage, VB is vaginal bleeding 946 947 . CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint . CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint Table 5 . Breakthrough bleeding in post-menopausal people. Vaccine and medical history related to breakthrough Note. The reproductive conditions are compared to the first listing (spontaneous or hormonally contracepting). 962

Diagnoses were not mutually exclusive, so some individuals might be represented in the endometriosis as well as the 963 fibroids group, for example, or have other listed diagnoses with too few respondents to make comparisons. The Odds 964

Ratio is calculated for risk of heavier bleeding in the diagnosed group. 965 966 967

. CC-BY-NC-ND 4.0 International license It 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 October 12, 2021. ; https://doi.org/10.1101/2021.10.11.21264863 doi: medRxiv preprint

Is there a link between COVID vaccine and "funky" menstrual periods? Experts say it's too soon to 648 know

Can COVID-19 vaccine impact your menstrual cycle? Doctors address side-effects concerns

Some people are reporting abnormal periods after a COVID-19 vaccine. U. of I. professor is looking 653 for answers

No data' linking Covid vaccines to menstrual changes, US experts say. The Guardian (2021)

Why false claims about Covid-19 vaccines and infertility are so powerful

SPECIAL REPORT The ex-Pfizer scientist who became an anti-vax hero

The Life Cycle Of A COVID-19 Vaccine Lie. npr (2021)

Why Some Orthodox Jewish Women Won't Get Vaccinated

Newsom calls GOP rivals 'anti-vax

Why this mom got the COVID-19 shot but hesitates for her daughter

They're Not Anti-Vaccine, but These Parents Are Hesitant About the Covid Shot

Rapid changes 675 in serum cytokines and chemokines in response to inactivated influenza vaccination. Influenza Other Respi 676 Viruses

Cytokines and hemostasis

Regulation of human endometrial function: 679 mechanisms relevant to uterine bleeding

Inflammation, leukocytes and menstruation

Immunology of Normal and Abnormal Menstruation. Womens Health (Lond Engl)

Endometrial inflammasome activation accompanies 685 menstruation and may have implications for systemic inflammatory events of the menstrual cycle. Human 686 Reproduction

Endometrial thickness is not independent of luteal 688 phase day in a rural Polish population

The ecological context of human ovarian function. 690 Human Reproduction

Reversible luteal phase changes and infertility 692 associated with marathon training. The Lancet

A case report of 694 recovery of menstrual function following a nutritional intervention in two exercising women with amenorrhea 695 of varying duration

Relationships between 699 biomarkers of inflammation, ovarian steroids, and age a menarche in a rural Polish sample

Energetics and the immune 702 system: Trade-offs associated with non-acute levels of CRP in adolescent Gambian girls

Ovarian Function in Late Reproductive Years in Relation to 705 Lifetime Experiences of Abuse

The Influence of Stress on the

Menstrual Cycle among Newly Incarcerated Women. Women's Health Issues

Dispositional resilience as a moderator of the relationship between chronic stress 709 and irregular menstrual cycle

Ovulation Prevalence in Women with Spontaneous Normal-712 Length Menstrual Cycles -A Population-Based Cohort from HUNT3

Luteal Phase Deficiency in Recreational Runners: 715 Evidence for a Hypometabolic State

Epidemiological transitions, reproductive health, and the Flexible Response 718 Model

On fertile ground

How does long COVID affect periods? (2021)

EXPERIENCES WITH PROPHYLACTIC TYPHOID VACCINATION: ITS EFFECT ON 724 MENSTRUATION

Menstrual 726 Abnormalities after Hepatitis B Vaccine

No association between HPV vaccine and reported post-vaccination symptoms in 728 Japanese young women: Results of the Nagoya study

Efficacy and Safety of the mRNA-1273 SARS-CoV-2 Vaccine

A preliminary report of a 740 randomized controlled phase 2 trial of the safety and immunogenicity of mRNA-1273 SARS-CoV-2 vaccine. 741 Vaccine

743 Spontaneous Abortion Following COVID-19 Vaccination During Pregnancy

SARS-CoV-2 spike protein seropositivity from vaccination or infection does not cause sterility

Does mRNA SARS-CoV-2 vaccine 748 influence patients' performance during IVF-ET cycle?

Preliminary 752 Findings of mRNA Covid-19 Vaccine Safety in Pregnant Persons

Receipt of mRNA Covid-19 755 Vaccines and Risk of Spontaneous Abortion

An Archeology of My Desk

Screening for endometrial cancer

Endometrial Cancer Risk With Postmenopausal Bleeding in Women: A Systematic Review and Meta-analysis. 763

Abnormal Uterine Bleeding in Premenopausal Women

Abnormal Uterine Bleeding: Is It an Under-766

Reported Side Effect after Epidural Steroid Injection for the Management of Low Back Pain

Comparing the efficacy of intrauterine lidocaine 771 and paracervical block in decreasing the pain associated with endometrial biopsy: a randomised trial

Pain Experienced Using Two Different 774 Methods of Endometrial Biopsy: A Randomized Controlled Trial

Just because I don't bleed, doesn't mean I don't go through it": Expanding knowledge on trans and 777 non-binary menstruators

779 Queer periods: attitudes toward and experiences with menstruation in the masculine of centre and transgender 780 community

The Routledge Handbook of Anthropology and Reproduction

Committee on Adolescent Health Care, Menstruation in Girls and Adolescents: Using the 785

Menstrual Cycle as a Vital Sign

Cycles as a Fifth Vital Sign

Characterizing physiological 789 and symptomatic variation in menstrual cycles using self-tracked mobile-health data. npj Digit

Understanding menstruation: Influence of gender and ideological factors. A 792 study of young people's social representations

There will be blood: Women's positive and negative experiences with menstruation. Women's 794 Reproductive Health

Maternal vaccines-safety in pregnancy

Vaccination during pregnancy

The effects of active immunization of the mother upon the offspring

Vaccines in Pregnancy. Infectious Disease Clinics of North America

A prospective study of influenza vaccination and time to pregnancy

No association between 806 HPV vaccination and infertility in U.S. females 18-33 years old

Little, Quadrivalent Human Papillomavirus Vaccine and the Young Ovary: Review of Safety Research 808

Following Two Case Series of Premature Ovarian Insufficiency

Human papillomavirus vaccine-associated premature ovarian 811 insufficiency and related adverse events: data mining of Vaccine Adverse Event Reporting System

Menstrual changes after covid-19 vaccination

The role of stem cells in uterine involution. 815 Reproduction

Sexual Violence on Campus: No Evidence that Studies Are Biased Due to Self-817 Selection. Dignity: A Journal of Analysis of Exploitation and Violence

Absence of malice: A meta-analysis of nonresponse bias in 820 counterproductive work behavior research

Disparities in COVID-19 Vaccination Rates across Racial and Ethnic Minority Groups in the United 822 States" (Office of the Assistant Secretary for Planning and Evaluation, U.S. Department of Health and Human 823 Services

Social Media Use in 2021. Pew Research Center: Internet, Science & Tech (2021)

The Making and Unmaking of Ignorance

The Mushroom at the End of the World: On the Possibility of Life in Capitalist Ruins

Against Purity: Living Ethically in Compromised Times

The Medicalization of Birth and Midwifery as Resistance. Health Care for Women International

Catalyzing a Reproductive Health and Social Justice 835 Movement. Matern Child Health

CCinBC Steering Committee, Patient-led decision making: Measuring autonomy and respect in 838 Canadian maternity care

I fought my entire way": Experiences of declining 840 maternity care services in British Columbia

Research electronic data capture 843 (REDCap)-A metadata-driven methodology and workflow process for providing translational research 844 informatics support

The REDCap consortium: Building an international community of software platform 847 partners

Statistical Power Analysis for the Behavioral Sciences

R: A language and environment for statistical computing" (manual

DescTools: Tools for descriptive statistics

rcompanion: Functions to support extension education program evaluation

questionr: Functions to make surveys processing easier

Ggplot2: elegant graphics for data analysis

 870 We first and foremost would like to thank the tens of thousands of people who responded 871 to the survey to tell us about their experiences. We additionally would like to thank (in 872 alphabetical order) Chongliang Luo, Bryana Rivera, Fatima Soumare, Emma Verstraete, 873 and Florence Yung for their contributions to the project.