During the study period (6 months, with 90% of the answers received during the first month), 778 women filed the online survey. Five hundred sixty-eight women (73%) had regular menstruations (around every month), and 210 (27%) did not. As it was stated in the diffusion material and the short introduction at the top of the survey that the questionnaire was for women with menstruation, it suggests that the 27% of “no” answer correspond to women with irregular menstruation, rather than no menstruation at all. But as it was not further assessed, we focused our analysis on the regularly menstruated women (Fig. 1). Menstrual blood donation for research was generally well accepted with 78% of women ready to donate. However, the menstrual cup is not a very popular item of menstrual hygiene as only 18% of the participants used it (either regularly or occasionally). Regarding to endometriosis, more than half of the women who answered our survey (58%) were affected by the disease.

In the regularly menstruating population, 42 women (7.4%) did not answer at the question for menstrual blood donation, so they were removed from the downstream analyses as it was central to our study. Among the 526 regularly menstruating women who expressed their opinion on menstrual blood donation, 93 women (17.7%) did not have a defined answer concerning endometriosis diagnosis (either no answer, suspected endometriosis, or did not know), and they also were excluded from downstream analyses. A total of 433 women had a defined status for endometriosis (affected with a confirmed diagnosis or not) and gave opinion about menstrual blood donation (Fig. 1, Table 1).

When comparing women who self-declared with and without endometriosis (Table 1), we could see that there is a significant difference in the age distribution with an overrepresentation of women in the 33–37 age range and fewer younger women below 28 years old in the endometriosis group (p < 0.001). There is also a significant association between the endometriosis status and willingness to donate menstrual blood: women with endometriosis are more likely to donate their menstrual blood (83% of women with endometriosis versus 68% of women with endometriosis, p < 0.001). Endometriosis patients also exhibited more often heavy menstrual bleeding (62% of endometriosis women versus 40% in unaffected women, p < 0.001). Unsurprisingly, women with endometriosis are more likely to experience dysmenorrhea, and in particular, they suffer from pelvic pain at every menstruation (82%) compared to women without endometriosis (23%, p < 0.001). Of note, dysmenorrhea was common in unaffected women with half of them (51%) occasionally experiencing pelvic pain during menstruation, in addition to the 23% experiencing it every month. While there was no difference with endometriosis affected and unaffected women concerning hormonal contraception use, endometriosis patients were more often trying to conceive (37% vs 8%, p < 0.001). Endometriosis affected and unaffected women were as likely to use a menstrual cup (Fig. 2). However, the majority of women in both populations are not using menstrual cups and are not willing to use it.

Distribution of women according to their use of a menstrual cup. Left panel: Menstrual cup use and endometriosis status. Right panel: Menstrual cup use and willingness to donate menstrual blood

In addition to the endometriosis status as described above, we were able to highlight associations between other variables and menstrual blood donation (Table 2). Indeed, women experiencing dysmenorrhea every month were more likely to be willing to donate menstrual blood (68 vs 51%) than women never experiencing it or only occasionally (p < 0.01).

A strong limiting factor seems to be related to the use of a menstrual cup. Indeed, women who do not want to try a menstrual cup are less willing to donate their menstrual blood (47 vs 79%) compared to women ready to try it or already using it (p < 0.001).

There were also tendencies with fewer women who use hormonal contraception likely to donate menstrual blood (25% vs 36%, p = 0.053) and more women that are trying to conceive likely to donate menstrual blood (30% vs 19%, p = 0.057).

Age and abundance of the menstrual blood flow did not have an impact on menstrual blood donation. Within women with endometriosis, the subtype of endometriosis did not affect the willingness to donate menstrual blood.

After transforming variables to numerical value, a correlation matrix was generated, and correlation between all variables is further detailed in Fig. 3. None of the variables, except endometriosis status and endometriosis subtype, displayed a correlation coefficient greater than 0.6, the threshold used to eliminate a variable for the subsequent generalized linear model. In addition to the associations already described above between endometriosis and willingness to donate menstrual blood with the other variables, there is significant correlation between a few other variables (Fig. 3A). Some were expected, such as the negative correlation between hormonal contraception and trial for conception (r =  − 0.26, p < 0.001), as well as hormonal contraception and heavy menstrual bleeding (r =  − 0.17, p < 0.01). There was also a positive correlation between age and trial for conception (r = 0.13, p < 0.05). Heavy menstrual flow and dysmenorrhea are also positively correlated (r = 0.29, p < 0.001).

Graphical representation of the correlation matrix. A Graphical representation of the correlation matrix for all the studied variables in all women (n = 392 to 433, as some data were missing). B Graphical representation of the correlation matrix for the indicated variables in women with self-declared endometriosis (n = 279 to 299). C Graphical representation of the correlation matrix for the indicated variables in self-declared unaffected women (n = 129 to 134). The size and color of the dots represent the correlation coefficient value (see color scale). Pearson statistical test and adjusted p-value using the FDR (false discovery rate) method. *p < 0.05, **p < 0.01, and ***p < 0.001. Here, the name of the variables was shortened to allow an easier readability: “cup” stands for menstrual cup use, “donation” for willingness to donate menstrual blood, “bleeding” for heavy menstrual flow, “dysmenorrhea” for pelvic pain during menstruations, “contraception” for taking hormonal contraception, “repro” for trying to conceive, and “typeendo” for the type of endometriosis

Among women with self-declared endometriosis (Fig. 3B), no new correlation could be observed. Of note, the negative correlation between hormonal contraception and heavy menstrual bleeding that was found in the whole population was not observed in the endometriosis subgroup, neither was the positive correlation between the trial for conception and dysmenorrhea. When taking into account the subtype of endometriosis in affected women, only age was still positively correlated (r = 0.18, p < 0.01), indicating that the other correlations with menstrual blood donation, heavy menstrual bleeding, and dysmenorrhea are associated with the presence of endometriosis but not its subtype.

Among women who self-declared being unaffected by endometriosis (Fig. 3C), there is a specific correlation that could be observed: a significant negative correlation between age and dysmenorrhea (r =  − 0.38, p < 0.001), with young people more affected by dysmenorrhea.

We used generalized linear model with a stepwise selection procedure to identify potential predictive factors of menstrual blood donation (Fig. 4). The only factor that is significantly predictive in our study is the use of menstrual cup (p < 0.001): women using a menstrual cup are indeed nine times more likely to be willing to donate menstrual blood (95% confidence interval [4.065:24.155], Fig. 4). Experiencing dysmenorrhea and an endometriosis diagnosis have a tendency to predict the willingness to donate menstrual blood (95% confidence interval [0.890:4.774] and [0.891:3.094], respectively).

Menstrual cup use predicts willingness to donate menstrual blood. A generalized linear model was built to assess predictive variables for menstrual blood donation from 392 complete set of answers, with a stepwise model selection procedure that iteratively removed the less predictive variable until the model could not be improved anymore. Odds ratio (OR) for each retained covariates and confidence intervals (2.5–97.5%) are plotted. Here, the name of the variables was shortened to allow an easier readability: “endo” stands for endometriosis diagnosis, “dysmenorrhea” for pelvic pain during menstruations, “cup” stands for menstrual cup use, and “contraception” for the use of a hormonal contraception