Physical and psychosocial factors are crucial for maintaining physical and mental health in endometriosis: a longitudinal analysis

Objective: To test the associations of physical and psychosocial fac- tors with physical and mental health in individuals living with endo- metriosis (EM) by means of cross-sectional and longitudinal analyses.

and Measures: Data were gathered via an online survey between February and August 2021. At survey date t1, sociodemo - graphic, EM-related and psychosocial factors as well as physical and mental health of people with EM were assessed. At survey date t2 three months later, physical and mental health was reas - sessed. The sample consisted of n_t1 = 723 (30.60 ± 6.31 years) and n_t2 = 216 (30.56 ± 6.47 years) cis women with EM. Statistical analy - ses included bivariate and partial correlation analyses and hierar - chical regression analyses.

The participants’ physical health was within the average range and their mental health was below-average at t1 and t2. Cross-sectional analyses revealed that worse health was associated with longer diagnostic delay, more surgeries, greater pelvic pain and lower sense of coherence, self-efficacy, sexual satisfaction and satisfaction with the gynecological treatment. In longitudinal anal - yses, pelvic pain and participants’ satisfaction with the gynecologi - cal treatment remained significantly associated with health.

Treatment should address both pelvic pain and psy - chosocial factors to improve long-term physical and mental health in EM. © 2024 t he a uthor(s). Published by Informa UK limited, trading as taylor & Francis group CONTACT sylvia Mechsner [email protected] endometriosis c entre charité, Department of gynaecology with c enter for o ncological surgery, charité –Universitätsmedizin Berlin, c orporate Member of Freie Universität Berlin and humboldt-Universität zu Berlin, 13353 Berlin, g ermany t his article was originally published with errors, which have now been corrected in the online version. Please see c orrection https://doi.org/10.1080/08870446.2024.2376801 https://doi.org/10.1080/08870446.2024.2302486 t his is an o pen a ccess article distributed under the terms of the c reative c ommons a ttribution license ( http://creativecommons. org/licenses/by/4.0/ ), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. t he terms on which this article has been published allow the posting of the a ccepted Manuscript in a repository by the author(s) or with their consent. ARTICLE HISTORY Received 6 April 2023 Accepted 2 January 2024

endometriosis; longitudinal; mental health; physical health; psychosocial; PROMIS PSyChOl OGy & hEAl Th 1157 1.  Introduction Endometriosis (EM) is a chronic inflammatory gynecological disease defined as the growth of tissue similar to the uterine lining outside of the uterus (Giudice, 2010). EM affects about 10% of people born with female-assigned reproductive organs and is associated with infertility and pelvic pain, in some cases chronic pelvic pain (Bulun, 2018). Research shows that EM is associated with impaired mental health (l aganà et  al., 2015; Pope et  al., 2015; van Barneveld et  al., 2021), a reduced quality of life (de Graaff et  al., 2016; Siedentopf et  al., 2008) and debilitated sexual functioning (Aerts et  al., 2018; Pérez-l ópez et  al., 2020). It takes 4–11 years from the first onset of symptoms (Taylor et  al., 2021) and the consultation of several doctors (De Graaff et  al., 2013; Greene et  al., 2009) to receive the correct diagnosis. This long diagnostic delay has repeatedly been described as particularly distressing by individuals affected by the disease (Ballard et  al., 2006; Denny, 2004; Jones et  al., 2004) as it can lead to an exacerbation of pain symptoms, hopelessness, difficulties in patient– doctor-relationships and a general lack of trust in the medical system (Facchin et  al., 2017b). Research highlights the relevance of the relationship between patient and practitioner in this context (Kundu et  al., 2015; Roomaney et  al., 2019) as communi - cation about pain and the recognition of EM symptoms by the attending doctor have been shown to be of particular importance on the path to diagnosis (Bullo & Weckesser, 2021; Drinkell et  al., 2023; Fernley, 2021; Manderson et  al., 2008). According to l ukas et  al. ( 2018) these factors are associated with the satisfaction with medical support in individuals affected by EM. Satisfaction with medical support, in turn, is an import - ant factor in the individual’s adherence in general (Anhang Price et  al., 2014; l ewin et  al., 2001) and thus can be assumed to be critical for their overall health. The association between (chronic) pelvic pain symptoms and health outcomes (physical health, mental health, quality of life) in EM has been widely demonstrated (Brooks et  al., 2020; Comptour et  al., 2020; Facchin et  al., 2017a; Márki et  al., 2017; Netzl et  al., 2022; van Barneveld et  al., 2021), whereas psychosocial factors relevant to the treatment of EM are still understudied. In previous studies, for example, dis - positions, personality traits, resilience, coping styles or stress associated with EM have been examined. The results pointed toward a reduced optimism (Morán-Sánchez et  al., 2020) and higher trait anxiety (Quiñones et  al., 2015) in people affected by EM and an association between painful EM and higher harm avoidance as well as lower exploratory excitability (Facchin et  al., 2016). Other research suggests that resilience in individuals living with EM may be an important factor in their life sat - isfaction (Romaniuk & Oniszczenko, 2023). In other studies, the importance of coping styles for the emotional well-being in EM has been investigated (Eriksen et  al., 2008; Quiñones et  al., 2015; Rees et  al., 2020) and a bidirectional relationship between EM and chronic stress has been suggested (Petrelluzzi et  al., 2008; Reis et  al., 2020). Generally, the experience of EM is considered a disruption in the individual’s life, posing the challenge to integrate this chronic condition into one’s self-concept (Facchin et  al., 2017b ) and to regain interpretative control over one’s own life (Wischmann, 2008) or a sense of coherence (SOC). The concept of SOC is defined as a global orientation that expresses the extent to which one has a pervasive, enduring though dynamic feeling of confidence in the comprehensibility, 1158 J. NETZl ET Al. manageability and meaningfulness of the internal and external demands in one’s life (l angeland, 2014). A higher SOC is associated with lower levels of anxiety and depression, greater physical well-being and lower perceived stress (Zirke et  al., 2007). The concept of self-efficacy is closely related to SOC and refers to the belief in one’s abilities, skills and competencies across various situations. h igher self-efficacy is associated with higher self-esteem and optimism and lower anxiety (Beierlein et  al., 2012). In people living with EM, lower self-efficacy was associated with poorer mental health (Facchin et  al., 2017a), as were negative feelings toward the medical profes - sions and impaired sexual functioning (Roomaney et  al., 2019). Uncontroversially, EM is known to be a heavy sexual burden (la Rosa et  al., 2020; Sullivan-Myers et  al., 2023) and several researchers pointed out the association between impaired sexual functioning and psychological distress in individuals living with the disease (de Graaff et  al., 2016; Facchin et  al., 2017b; Roomaney et  al., 2019). These studies provided important insights into the associations between psychosocial factors and well-being in EM. Still, it remains necessary to investigate additional psychosocial factors in EM to improve the understanding of this complex disease and its physical, psychological and social consequences and, thus, overcome them. Moreover, there have been repeated calls to include longitudinal data in the research (Facchin et  al., 2017a; Márki et  al., 2017). The goal of this study was to increase the knowledge about physical and psycho - social factors associated with self-reported physical and mental health in individuals living with EM. First, cross-sectional associations between putative physical and psy - chosocial predictor variables and physical and mental health were explored. Sociodemographic (age, income, relationship status, EM-related strain on the relation - ship), EM-related (diagnostic delay, time since receiving the EM diagnosis, number of surgeries, number of pelvic pain types, pelvic pain intensity, number of pelvic pain days per month, hormonal treatment, infertility) and psychosocial variables (SOC, self-efficacy, sexual satisfaction, satisfaction with the gynecological treatment) were included. Based on the referenced literature, the included psychosocial factors were assumed to be of high relevance for the well-being in EM. The goal was to test them in a comprehensive statistical model. The statistical procedure was repeated for phys - ical and mental health assessed at t2 three months later, controlling for physical or mental health at survey date t1. By including data from a second survey date, we aimed to increase the knowledge about longer-term associations between physical and psychosocial factors and self-reported health in people living with EM, and con - sequently derive treatment implications. 2.  Material and methods This online study was conducted at the Endometriosis Centre of Charité— Universitätsmedizin Berlin. Data were collected between February and August 2021 using Research Electronic Data Capture (REDCap) tools hosted at Charité— Universitätsmedizin Berlin (harris et  al., 2009, 2019). REDCap is a secure, web-based software platform designed to support data capture for research studies, providing (1) an intuitive interface for validated data capture, (2) audit trails for tracking data manipulation and export procedures, (3) automated export procedures for seamless PSyChOl OGy & hEAl Th 1159 data downloads to common statistical packages and (4) procedures for data integra - tion and interoperability with external sources. 2.1.  Ethical approval The study was conducted in accordance with the Declaration of helsinki, and approved by the Ethics Committee of Charité—Universitätsmedizin Berlin (Number: EA4/248/19). Informed consent to collect and publish the data has been obtained from all partic - ipants via an opt-in item. If participants did not consent, the survey ended automatically. 2.2.  Participants and procedure Participation in the study was publicly advertised (e.g. on Facebook and Instagram). The participants had to meet the following inclusion criteria: being at least 18 years old and pre-menopausal, speaking fluent German, not suffering from a malignant disease or infection, no current pregnancy and no ovariectomy. If any of these criteria were not met, the survey ended automatically. After obtaining informed consent, participants were asked to generate a pseudonym to match the data from survey dates t1 and t2. At the end of survey date t1, participants were asked to send an e-mail to the study team so that their e-mail-addresses could be saved independently of the study data. After three months, participants were sent the browser link to survey date t2. Initially, 734 participants met the inclusion criteria and completed the survey at t1. All of them had been born with a uterus and ovaries and 723 (98%) reported a female, 5 (1%) a male and 6 (1%) a diverse gender identity. To avoid bias in the analyses, these 11 (2%) people were excluded. EM diagnoses by laparoscopy ( n = 624, 86%), magnetic resonance imaging ( n = 25, 4%) or sonography ( n = 74, 10%) were accepted. The final sample comprised n_t1 = 723 cis women with EM at t1 (cis or cisgender describes individuals who identify with the gender they were assigned at birth: cis women were assigned the female gender at birth and identify with the female gender), of whom n_t2 = 216 cis women with EM also completed the ques - tionnaire at the second survey date t2. All of these n_t2 = 216 participants had received the EM diagnosis by laparoscopic surgery. To rule out attrition bias, participants who completed t2 ( n_t2 = 216) were tested against participants who only completed t1 (n = 507) in sociodemographic and EM-related data. In these analyses, no significant differences were found ( Table 1 ). 2.3.  Measures and instruments Participants responded to a comprehensive item battery collecting sociodemographic, clinical and EM-/pelvic-pain-related information. Participants were asked to rate the strain EM puts on their current relationship (0 = not at all to 100 = extremely strained by EM). household net income was assessed with one item (3000€). Participants were 1160 J. NETZl ET Al. asked whether they suffered from dysmenorrhoea, non-menstrual cyclical pelvic pain (e.g. during ovulation), non-cyclical pelvic pain, chronic pelvic pain, pain related to sexual intercourse, dyschezia and dysuria, and to rate the typical intensity of each of the reported types of pain (0 = no pain to 100 = worst pain imaginable). An average pelvic pain intensity item was calculated. Table 1. s ociodemographic, eM-related and psychosocial data at t1 and physical and mental health scores at t1 and t2. Variable t1 sample ( n_t1 = 723) t2 sample ( n_t2 = 216) N M ± sD/Frequency (%) N M ± sD/Frequency (%) pa age 723 30.60 ± 6.31 216 30.56 ± 6.47 .915 Min, Max 18, 51 18, 47 Relationship 723 553 (77%) 216 160 (74%) .829 eM-related strain on the relationship 541 66.98 ± 20.10 157 64.95 ± 21.63 .132 Min, Max [0–100] 0, 100 0, 100 o ther gynecologic diseaseb 723 159 (22%) 216 42 (19%) .280 o ther pain disorder b 723 271 (38%) 216 78 (36%) .619 children b 720 125 (17%) 30 (14%) .107 Infertility 718 174 (24%) 216 45 (21%) .163 hormonal treatment 723 380 (53%) 216 125 (58%) .062 t ime since eM diagnosis (years) 721 3.45 ± 4.18 216 3.49 ± 4.52 .862 Min, Max 0, 30 0, 30 Diagnostic delay (years) 705 9.11 ± 6.45 209 8.91 ± 6.51 .596 Min, Max 0, 32 0, 31 Number of surgeries 721 1.43 ± 1.24 215 1.39 ± 1.20 .209 Min, Max 0, 6 0, 6 Dysmenorrhoea b 723 364 (50%) 216 109 (51%) .967 Non-menstrual cyclical pelvic pain b 723 390 (54%) 216 108 (50%) .165 Non-cyclical pelvic pain b 723 503 (70%) 216 154 (71%) .511 chronic pelvic pain b 723 157 (22%) 216 41 (19%) .245 Pain at sexual intercourseb 723 526 (73%) 216 153 (71%) .449 Dyscheziab 723 408 (56%) 216 119 (55%) .636 Dysuriab 723 237 (33%) 216 59 (27%) .041 Number of pelvic pain types 723 3.58 ± 1.63 216 3.44 ± 1.68 .145 Min, Max 0, 7 0, 7 Pelvic pain days per month 723 12.26 ± 9.44 216 11.69 ± 9.33 .290 Min, Max 0, 31 0, 31 Pelvic pain intensity 723 29.52 ± 16.90 216 27.82 ± 16.76 .077 Min, Max [0–100] 0, 93 0, 89 soc l -9 723 4.17 ± 1.10 216 4.22 ± 1.10 .366 aKsU 714 3.65 ± 0.71 216 3.60 ± 0.72 .188 KFsP-F – sexual satisfaction c 696 13.59 ± 5.35 216 14.00 ± 4.96 .154 FIPs 690 3.32 ± 1.00 216 3.25 ± 0.95 .253 Physical health t1 d 723 45.54 ± 8.34 216 45.90 ± 8.24 .336 Mental health t1 d 723 37.98 ± 7.04 216 38.59 ± 7.27 .068 Physical health t2 d – 216 46.28 ± 8.44 Mental health t2 d – 216 40.45 ± 7.68 aIndependent samples t-tests were used for continuous variables and χ 2-tests were used for categorical variables to compare participants who completed t2 ( n_t2 = 216) with those who only completed t1 ( n = 507, descriptive

not shown). bVariables are reported descriptively only. cl evene’s test showed that the variances for KFsP-F sexual satisfaction were not equal, p = .009. dt -scores: M = 50 ± 10. PSyChOl OGy & hEAl Th 1161 Physical and mental health were measured using the Patient Reported Outcomes Measurement Information System 29 v2.1 (PROMIS-29 Profile v2.1, hays et  al., 2018). Using item response theory methods for item selection, the PROMIS initiative devel - oped a variety of item banks to assess different health domains. The complete ques - tionnaire can be found online and is openly accessible (‘PROMIS-29 Profile v2.1’ , 2023). The PROMIS Profile 29 assesses the seven health domains physical function (e.g. ‘Are you able to do chores such as vacuuming or yard work?’), anxiety (e.g. ‘I felt fearful’), depression (e.g. ‘I felt worthless’), fatigue (e.g. ‘I feel fatigued’), sleep disturbance (e.g. ‘I had a problem with my sleep’), ability to participate in social roles and activities (e.g. ‘I have trouble doing all of my regular leisure activities with others’) as well as pain interference (e.g. ‘how much did pain interfere with your day to day activities?’) with four questions each. Additionally, it comprises one item on the average general pain intensity in the past seven days (0 = no pain to 10 = worst pain imaginable). Physical functioning and the ability to participate in social roles and activities are assessed without a specific timeframe, whereas all other health domains come with the instruction to refer to the past seven days. Reliability scores for the subscales in the study sample were Cronbachs α = .831 to .946 (t1) and Cronbachs α = .828 to .945 (t2). Raw scores and T-scores for each domain and raw scores for the general pain intensity item of sample n_t1 are presented in Appendix A. Based on the scoring instructions provided by hays et  al. ( 2018) and Spritzer and hays ( 2018), physical and mental health summary T-scores were calculated with M = 50 ± 10. These T-scores are derived from the general U.S. population and are commonly used as default values as there are no country-specific scoring algorithms to date (Fischer et  al., 2018). h igher scores in these summary scores represent better physical or mental health. Additionally, they include a direct comparison with the U.S. general population, with M = 50 ± 10 representing the average range. In the calculation of the PROMIS physical and mental health summary scores all seven health domains, that is physical, psychological and social factors, are used, each with a different weighting. Sense of Coherence SOC was assessed using the Sense of Coherence Scale l -9 SOC l -9 (Schumacher, Wilz, Gunzelmann, & Brähler, 2000; possible scores: 1–7, Cronbachs α = .859 (t1), Cronbachs α = .876 (t2)), self-efficacy was assessed according to the Self-Efficacy Scale—Short Form AKSU (Beierlein et  al., 2012; possible scores: 1–5, Cronbachs α = .847 (t1), Cronbachs α = .852 (t2)). Sexual satisfaction was measured along one scale of a German adaptation of the Brief Index of Sexual Functioning KFSP-F (hartmann et al., 2002; possible scores: 3–21, Cronbachs α = .792 (t1), Cronbachs α = .858 (t2)) and satisfaction with the gynecological treatment according to the Freiburg Index of Patient Satisfaction FIPS (Schoenthaler et  al., 2012) with the specific instruction to think about ‘the gynecological examinations and treatments carried out’ (possible scores: 1–6, Cronbachs α = .778 (t1), Cronbachs α = .774 (t2)). h igher scores on all of these scales represent a larger portion of the measured concept, except for the FIPS scale, where lower scores represent greater treatment satisfaction. 2.4.  Statistical analyses Descriptive statistics were calculated by Mean ± Standard deviation. Parametric Pearson correlation analyses between the outcome variables physical and mental health 1162 J. NETZl ET Al. assessed at t1 and the putative predictor variables collected at t1 were calculated to preselect predictor variables for the subsequent regression analyses. The following putative predictor variables were examined: sociodemographic (age, income, relation - ship status, EM-related strain on the relationship), EM-related (diagnostic delay, time since receiving the EM diagnosis, number of surgeries, number of pelvic pain types, pelvic pain intensity, number of pelvic pain days per month, hormonal treatment, infertility) and psychosocial variables (SOC, AKSU, KFSP-F—sexual satisfaction, FIPS). Putative predictor variables that showed significant bivariate associations with the outcome variable were then entered stepwise (enter method) into multiple regression analyses with step I sociodemographic, step II EM-related and step III psychosocial variables. All three steps were calculated for physical and mental health at t1. Pairwise exclusion of missing values was used to keep the sample size at a maximum and to avoid limiting the data set: Since the item on EM-related strain on the relationship had only been answered by participants who were in a relationship at the time, a listwise exclusion of missing values would have resulted in the exclusion of all par - ticipants who were single and thus in a less representative sample. To evaluate the predictive power of each set of predictor variables, R2 and the change in variance explanation are reported for each step. To test whether the results were different when only those participants who had received the EM diagnosis by laparoscopy (n = 624) were considered cross-sectional correlation and regression analyses were repeated for this subsample (Appendix C). Analyses were repeated for the outcome variables physical and mental health assessed three months later at t2, with physical or mental health at t1 included as a control variable to partialize out its influence (Reinders, 2006). Partial correlation analyses between the health outcome at t2 and the sociodemographic, physical and psychosocial putative predictor variables collected at t1 were performed, controlling for the same health outcome at t1. Next, multiple regression analyses with the steps I physical health at t1 or mental health at t1 and II putative predictor variables that were significantly associated in the partial correlation analyses were performed. Again, pairwise exclusion of missing values was used and R2 is reported. Statistical analyses were performed using IBM SPSS Statistics for Windows, Version 26.0 (IBM Corp, 2019) and p-values of p ≤ 0.05 were considered statistically significant. Statistical assumptions of multiple linear regression analysis (Field, 2009) were tested for all regression analyses (Appendices B and C). 3.  Results The descriptive results of sociodemographic, EM-related and psychosocial data col - lected at t1 are presented in Table 1 . At t1, participants were on average 30 years old and three quarters of them were in a relationship. Median monthly net household income was 2001–2500€ with percentile 25 = 1001–1500€ and percentile 75 = >3000€. Most participants did not have children and about one quarter reported being infer - tile. The average time since EM diagnosis was 3.5 years and half of the sample received hormonal treatment. The average delay in diagnosis from the first onset of EM symp - toms was 9 years. Most participants suffered from pelvic pain symptoms and one-fifth of them reported chronic pelvic pain. PSyChOl OGy & hEAl Th 1163 The PROMIS physical and mental health summary T-scores at both t1 and t2 are shown in Table 1 . Compared to the general U.S. population ( M = 50 ± 10), participants reported about half a standard deviation below average physical health and one standard deviation below average mental health at t1 (physical health: 45.54 ± 8.34, mental health: 37.98 ± 7.04) and t2 (physical health: 46.28 ± 8.44, mental health: 40.45 ± 7.68). 3.1.  Cross-sectional analyses Table 2 presents the results of Pearson correlation analyses between the putative predictor variables and physical and mental health scores at t1. Both higher physical health and higher mental health were associated with higher income, lower EM-related strain on the relationship, shorter diagnostic delay, fewer surgeries, fewer pelvic pain types, lower pelvic pain intensity, fewer pelvic pain days per month, a higher SOC, higher self-efficacy, higher sexual satisfaction and higher satisfaction with the gyne - cological treatment. Putative predictor variables that showed significant correlations with the health out - come were then entered into hierarchical multiple regression analyses in three steps: I sociodemographic, II EM-related and III psychosocial predictor variables. The results are presented in Table 3. All three steps were significant for both physical and mental health at t1 ( p < .001). The variance explained by sociodemographic variables was 7.3% for physical and 14.5% for mental health at t1. The inclusion of EM-related variables (step II) increased the percentage of variance explained by 24.6% for physical and 12.7% for mental health at t1. Step III, which additionally included psychosocial predictor variables, led to an increase in variance explanation of 8.4% for physical and of 25.8% for mental Table 2. c ross-sectional correlation analyses between physical and mental health scores at t1 and putative predictor variables. Putative predictor variable Physical health t1 a Mental health t1 a N r p N r p s ociodemographic variables Income 719 .160 <.001 719 .167 <.001 eM-related strain on the relationship 541 −0.218 <.001 541 −0.344 <.001 eM-related variables Diagnostic delay 705 −0.115 .002 705 −0.077 .041 Number of surgeries 721 −0.187 <.001 721 −0.079 .035 Number of pelvic pain types 723 −0.435 <.001 723 −0.347 <.001 Pelvic pain intensity 723 −0.454 <.001 723 −0.408 <.001 Pelvic pain days per month 723 −0.472 <.001 723 −0.360 <.001 Psychosocial variables soc l -9 723 .359 <.001 723 .626 <.001 aKsU 714 .297 <.001 714 .356 <.001 KFsP-F – sexual satisfaction 696 .099 .009 696 .144 <.001 FIPs 690 −0.343 <.001 690 −0.363 .05). 1164 J. NETZl ET Al. health at t1. The percentage of variance explained by the overall models was 40.3% for physical health at t1 and 53.0% for mental health at t1 (see R2 in the footnotes of Table 3). In cross-sectional regression analyses, physical health was higher if the number of surgeries, the pelvic pain intensity and the number of pelvic pain days per month were lower and if SOC, self-efficacy and satisfaction with the gynecological treatment were higher ( Table 3). Mental health was higher if the EM-related strain on the relationship, the pelvic pain intensity and the number of pelvic pain days per month were lower and if SOC and satisfaction with the gynecological treatment were higher ( Table 3 ). Table 3. Results of the cross-sectional multiple regression analyses for the outcome variables physical health at t1 and mental health at t1. o utcome variable Predictor variable a c oefficientsb B (s td.-error) β pb 95%-cI b Physical health t1 c c onstant 46.06 (2.65) <.001 40.85, 51.27 Income 0.18 (0.14) .05 .209 −0.10, 0.45 eM-related strain on the relationship −0.01 (0.02) −0.01 .774 −0.04, 0.03 Diagnostic delay d −0.09 (0.05) −0.07 .051 −0.18, 0.00 Number of surgeries −0.67 (0.24) −0.10 .004 −1.14, −0.21 Number of pelvic pain types −0.16 (0.33) −0.03 .640 −0.80, 0.50 Pelvic pain intensity −0.10 (0.03) −0.19 .002 −0.15, −0.04 Number of pelvic pain days per month −0.22 (0.04) −0.25 <.001 −0.30, −0.15 soc-l9 1.22 (0.33) .16 <.001 0.58, 1.86 aKsU 1.38 (0.48) .12 .004 0.44, 2.31 KFsP-F - sexual satisfaction 0.06 (0.06) .04 .314 −0.06, 0.18 FIPs −1.23 (0.31) −0.15 <.001 −1.83, −0.62 Mental health t1 e c onstant 33.15 (1.99) <.001 29.25, 37.06 Income 0.02 (0.10) .01 .843 −0.18, 0.23 eM-related strain on the relationship −0.04 (0.01) −0.10 .005 −0.06, −0.01 Diagnostic delay −0.02 (0.03) −0.02 .482 −0.09, 0.04 Number of surgeries −0.03 (0.18) −0.01 .885 −0.37, 0.32 Number of pelvic pain types 0.18 (0.25) .04 .476 −0.31, 0.66 Pelvic pain intensity −0.09 (0.02) −0.21 <.001 −0.13, −0.04 Number of pelvic pain days per month −0.11 (0.03) −0.15 <.001 −0.16, −0.05 soc-l9 3.16 (0.24) .49 <.001 2.68, 3.64 aKsU 0.20 (0.36) .02 .580 −0.50, 0.90 KFsP-F - sexual satisfaction 0.01 (0.05) .01 .854 −0.08, 0.10 FIPs −1.02 (0.23) −0.15 <.001 −1.48, −0.57 Note. linear hierarchical regression analyses (enter method). at he number of participants without missing values in the respective predictor variable and included in the pairwise analysis can be found in table 2 , column N. bB = unstandardised beta coefficient. s td. error = standard error of the unstandardised beta coefficient. B = standardised coefficient. p = p-value, probability. 95%- cI = 95% confidence interval for the unstandardised beta coefficient B. cstep I: sociodemographic variables, R2 = .073, p <.001, F(2, 516) = 20.23, p <.001; step II: sociodemographic and eM-related variables, R2 = .319, p <.001, F(7, 511) = 34.16, p <.001; step III: sociodemographic, eM-related and psychosocial variables, R2 = .403, p <.001, F(11, 507) = 31.05, p <.001. t he degrees of freedom are based on the minimum pairwise n. dDiagnostic delay remained a significant predictor variable for physical health at t1 when only those participants who had received the eM diagnosis by laparoscopy ( n = 624) were considered ( p < .05, a ppendix c ). estep I: sociodemographic variables, R2 = .145, p <.001, F(2, 516) = 43.89, p <.001; step II: sociodemographic and eM-related variables, R2 = .272, p <.001, F(7, 511) = 27.22, p <.001; step III: sociodemographic, eM-related and psychosocial variables, R2 = .530, p <.001, F(11, 507) = 52.06, p <.001. t he degrees of freedom are based on the minimum pairwise n. PSyChOl OGy & hEAl Th 1165 When only participants who had received the EM diagnosis by laparoscopy ( n = 624) were included, results were the same in all correlation analyses and both regression analyses with the exception of diagnostic delay remaining a significant predictor for physical health at t1 (see footnotes of Table 3 and Appendix C). In cross-sectional regression analyses, physical health of those participants who had received the EM diagnosis by laparoscopy was higher if the diagnostic delay was shorter, the number of surgeries, the pelvic pain intensity and the number of pelvic pain days per month were lower and if SOC, self-efficacy and satisfaction with the gynecological treatment were higher. 3.2.  Longitudinal analyses Table 4 presents the results of partial correlation analyses between the putative pre - dictor variables assessed at t1 and physical and mental health at t2. In these analyses, the health outcome assessed at t1 was a control variable. When controlling for physical health at t1, higher physical health at t2 was significantly associated with lower EM-related strain on the relationship, fewer pelvic pain types, lower pelvic pain inten - sity, fewer pelvic pain days per month and higher satisfaction with the gynecological treatment at t1. h igher mental health at t2 was significantly associated with higher satisfaction with the gynecological treatment at t1, when controlling for mental health at t1. Figure 1 shows significant results from both cross-sectional and longitudinal correlation analyses. Those putative predictor variables that showed significant associations with the health outcome in the partial correlation analyses were then entered into hierarchical multiple regression analyses as a second step. In the first step, the health outcome at t1 was entered into the analyses. The results are presented in Table 5 . Both steps were significant for both physical and mental health at t2 ( p < .001). The variance explained by entering the health outcome at t1 was 59.7% for physical and 60.3% for mental health. The inclusion of physical and psychosocial predictor variables led Table 4. Partial correlation analyses between physical and mental health scores at t2 and putative predictor variables at t1, controlling for physical and mental health at t1. c ontrol variable Putative predictor variable Physical health t2 a Physical health t1 N r p eM-related strain on the relationship 157 −0.159 .047 Number of pelvic pain types 216 −0.201 .003 Pelvic pain intensity 216 −0.197 .004 Pelvic pain days per month 216 −0.154 .023 FIPs 216 −0.249 <.001 Mental health t2 b Mental health t1 N r p FIPs 216 −0.214 .002 aNo significant associations were found with the variables age, income, relationship status, diagnostic delay, time since eM diagnosis, number of surgeries, hormonal treatment, infertility, soc-l9, aKsU and KFsP-F - sexual sat - isfaction ( p > .05). bNo significant associations were found with the variables age, income, relationship status, eM-related strain on the relationship, diagnostic delay, time since eM diagnosis, number of surgeries, number of pelvic pain types, pelvic pain intensity, number of pelvic pain days per month, hormonal treatment, infertility, soc-l9, aKsU and KFsP-F - sexual satisfaction ( p > .05). 1166 J. NETZl ET Al. to an increase in variance explanation of 4.5% for physical and 1.8% for mental health at t2. The percentage of variance explained by the overall models was 64.2% for physical and 62.1% for mental health at t2 (see R2 in the footnotes of Table 5 ). In addition to the health outcome at t1, satisfaction with the gynecological treatment at t1 remained a significant predictor in both analyses: Physical and mental health at t2 were higher if satisfaction with the gynecological treatment at t1 had been higher ( Table 5 ). Figure 1. significant associations between physical and mental health scores in cross-sectional cor - relation analyses. Bold variables denote significant associations in longitudinal partial correlation analyses with physical health at t2 (control variable: physical health at t1). Italics denote significant associations in longitudinal partial correlation analyses with mental health at t2 (control variable: mental health at t1). Table 5. Results of the longitudinal multiple regression analyses for the outcome variables physical health at t2 and mental health at t2. o utcome variable Predictor variable a c oefficientsb B (s td.-error) β pb 95%-cI b Physical health t2 c c onstant 23.60 (4.00) <.001 15.70, 31.50 Physical health t1 0.67 (0.06) .66 <.001 0.56, 0.79 eM-related strain on the relationship −0.03 (0.02) −0.09 .096 −0.07, 0.01 Number of pelvic pain types −0.30 (0.49) −0.06 .541 −1.27, 0.67 Pelvic pain intensity −0.02 (0.05) −0.03 .721 −0.11, 0.08 Number of pelvic pain days −0.04 (0.06) −0.04 .555 −0.16, 0.08 FIPs −1.36 (0.43) −0.15 .002 −2.20, −0.51 Mental health t2 d c onstant 14.23 (0.78) <.001 9.42, 19.03 Mental health t1 0.78 (0.05) .74 <.001 0.68, 0.87 FIPs −1.14 (0.36) −0.14 .002 −1.84, −0.44 Note. linear hierarchical regression analyses (enter method). at he number of participants without missing values in the respective predictor variable and included in the pairwise analysis can be found in table 4 , column N. bB = unstandardised beta coefficient. s td. error = standard error of the unstandardised beta coefficient. B = standardised coefficient. p = p-value, probability. 95%- cI = 95% confidence interval for the unstandardised beta coefficient B. cstep I: physical health at t1, R2 = .597, p <.001, F(1, 152) = 225.436, p <.001; step II: physical health at t1 and sociodemographic, eM-related and psychosocial variables which showed a significant association in partial cor - relation analyses, R2 = .642, p = .004, F(6, 147) = 43.896, p <.001. t he degrees of freedom are based on the minimum pairwise n. dstep I: mental health at t1, R2 = .603, p <.001, F(1, 214) = 325.178, p <.001; step II: mental health at t1 and sociodemographic, eM-related and psychosocial variables which showed a significant association in partial cor - relation analyses, R2 = .621, p = .002, F(2, 213) = 174.695, p <.001. t he degrees of freedom are based on the minimum pairwise n. PSyChOl OGy & hEAl Th 1167 4.  Discussion Previous research has shown that EM symptoms such as menstrual irregularities, dysmenorrhoea, chronic pelvic pain, dyspareunia and infertility often negatively affect the individual’s psychological and social functioning (laganà et  al., 2017; Vitale et  al., 2016, 2017). Consequently, EM is considered a disabling condition that may signifi - cantly compromise social relationships, sexuality and mental health. In this study, physical and mental health in people living with EM and its associations with physical and psychosocial characteristics were examined. In line with previous findings of impaired health in people with EM (Brandes, 2007), participants in this study reported their mental health to be one standard deviation below and, thus, significantly poorer than the general population’s average. Conversely, the results for physical health were within the average range. This is in contrast to the high level of pain symptoms reported by participants ( Table 1) and could be explained by the design of the PROMIS Profile 29. Firstly, the PROMIS pain intensity item does not focus on pelvic, menstrual or premenstrual pain or the worst pain experienced in a specific time frame but the average general pain intensity in the past seven days. Secondly, some of the items measuring physical functioning may not detect temporary forms of impairment, for example: ‘Are you able to go for a walk of at least 15 min?’ . Most people living with EM would answer in the affirmative and, assuming no chronic pain symptoms are present, which was the case for the majority of the study sample ( Table 1 ), would be most impaired in their physical functioning during their menstruation but not permanently. This has potentially led to an underestimation of the participants’ pelvic pain intensity and the related physical impairment and highlights the difficulty to adequately measure the burdens associated with EM with established, non-EM-specific questionnaires. Nevertheless, the results of both the correlation and regression anal - yses in this study confirm the notion that pelvic pain is a crucial factor for the well-being in EM (Brooks et  al., 2020; Comptour et  al., 2020; Facchin et  al., 2017a; Márki et  al., 2017; Netzl et  al., 2022; van Barneveld et  al., 2021) and demonstrate the significance of psychosocial factors at the same time. These findings have important implications for EM treatment as they highlight the necessity of a biopsychosocial approach to improve both physical and mental health (Netzl et  al., 2023). The cross-sectional correlation analyses have shown that both physical and mental health in people living with EM suffer if the diagnostic delay is long and pelvic pain is present and intense. In addition to that, they indicate that surgeries should be kept at the minimum number necessary to preserve the individual’s health. Notably, the role of the diagnostic delay was of particular importance for physical health in the cases where the diagnosis had been made by means of surgery. A potential explanation for this finding might be that in most cases, treatment is only started after the diagnosis of EM is confirmed and as certain types of EM can only be reliably diagnosed by means of surgery this may consequently lead to a delay in treatment. hormonal symptom-oriented treatments, however, could already be implemented at the time of a suspected diagnosis based on a thorough anamnesis of pain symptoms (Mechsner, 2016; Sillem, 2015). The importance of an early start of EM treatment is underlined by the fact that earlier stages of the disease show higher metabolic activity and are more reactive to hormonal treatment than advanced stages (Schweppe, 2011). 1168 J. NETZl ET Al. Furthermore, the cross-sectional analyses have revealed positive associations between the health outcomes and the household income of participants. The level of income is one of the main factors in the calculation of socioeconomic status (l ampert et  al., 2013) and a lower socioeconomic status is generally associated with poorer health and poorer access to health care (Binder & Rieder, 2014; Donkin et  al., 2017; lampert et  al., 2019). Thus, special attention must be paid to the treatment of financially and socially disadvantaged individuals with EM. As expected, a higher SOC, higher self-efficacy, higher sexual satisfaction and higher satisfaction with the gynecological treatment were associated with better physical and mental health. In line with the presented result on the importance of participants’ treatment satisfaction, previous research has shown that the patient-doctor-relationship is an important factor in the well-being of people with EM (Roomaney et  al., 2019). Adequate education about the disease and its treatment options is crucial to improving patient-doctor relationships and the individual’s trust in the medical system (l ukas et  al., 2018). Additionally, providing individuals living with a chronic illness with disease-specific knowledge can help protect them from misinformation on the internet (Dinh et  al., 2020) and improve self-efficacy and self-care (Wu et  al., 2016). These results suggest that by attending to psychological and interpersonal factors such as the individual’s SOC, self-efficacy and sexual functioning, their overall health could be improved. Included in joint models, sociodemographic, EM-related and psychosocial pre - dictor variables all significantly contributed to the variance explanation in both physical and mental health. As expected, the increase in variance explanation for physical health was greater when EM-related—primarily physical factors with a focus on pain—predictor variables were included, while variance explanation for mental health increased particularly strongly when psychosocial predictor variables were included. By providing these detailed analyses, this study contributes to a better understanding of the complex associations between physical and psycho - social factors and subjective health in people with EM. These results illustrate the necessity and importance of setting individual priorities within a multiprofessional treatment approach depending on each individual’s current treatment needs and potential for improvement. When examining the associations between predictor variables at t1 and physical and mental health at t2, physical and mental health at t1 were included as covariates to partialize out the variables’ influence (Reinders, 2006). In the partial correlation analyses, EM-related variables continued to show significant associations with physical health at t2. In addition, participants’ satisfaction with the gynecological treatment was significantly associated with both physical and mental health at t2 both in partial correlation and regression analyses. Thus, the individual’s evaluation of the gyneco - logical treatment can be assumed to be a key factor in their long-term well-being (Kundu et  al., 2015). By including physical and psychosocial factors as predictor variables and health as dependent variable, the multivariable models in this study assumed health to be a consequence of these factors. Thus, the multivariable models were consistent with previous albeit cross-sectional studies (Facchin et  al., 2015; Márki et  al., 2017). however, the associations between the analysed constructs in this study are more complex and probably not unidirectional: It is very likely that both physical and mental health also PSyChOl OGy & hEAl Th 1169 have an impact on physical, psychological, social and sexual factors and that these factors themselves are interdependent and influence each other as well. Thus, the possibility of health influencing physical and psychosocial factors needs to be acknowledged. There are some limitations to this study. First, the drop-out rate between the two survey points was quite high with less than a third of the initial participants com - pleting the questionnaire at t2. Sample n_t2, however, was tested for a potential attrition bias against those participants who only completed the questionnaire at t1 ( Table 1 ) and as the two groups did not differ in any of the variables included in the analyses, the results of these tests were acceptable to continue with the planned analyses. To test for attrition bias, t-tests were used, which are reasonably robust with differing sample sizes, especially with large samples and equal variances in the test variables (Zimmerman, 1987), both of which were the case. The main reasons for dropping out between the two survey dates were, on the one hand, that the survey was comparatively long as it comprised several questionnaires, and, on the other hand, that we were not allowed to ask for participants’ contact infor - mation directly for data protection reasons. At the end of survey date t1, participants were asked to send an e-mail to the study team to receive the browser link to survey date t2. Additionally, there was no intervention between the two survey points and no major differences were observed between the health outcomes at t1 and t2. Moreover, this study only included two survey points, whereas a reliable statement about the stability of a construct would benefit from at least three survey points to account for measurement errors and ceiling effects. It should be noted, that the data were collected during the COVID-19 pandemic, whose potential impact on physical and mental health was not considered. In addition, not all statistical assumptions of the multiple linear regression analysis were met (Appendix B). Finally, participants who did not identify as female and responded to the survey were excluded to avoid bias in the data. Their participation, however, deserves attention as transgender and non-binary individuals face health disparities and have an increased risk for physical and mental health problems in general (Pulice-Farrow et  al., 2021). Transmasculine or non-binary persons suffering from a gynecological and thus gendered disease such as EM may experience an increase in their gender dysphoria as a consequence, which can lead to a deterioration in their state of health (Pulice-Farrow et  al., 2021). Despite the high drop-out rate, one advantage of this study is the large sample size. A thorough exploration of the associations between the putative predictor variables and physical and mental health was presented. In this context, the large sample is particularly advantageous, as it can be assumed that the reported effects are quite stable. The selection of the putative predictor variables was both based on previous research (Brooks et  al., 2020 ; Comptour et  al., 2020 ; Facchin et  al., 2017a , 2017b ; Márki et  al., 2017; Netzl et  al., 2022; Roomaney et  al., 2019; van Barneveld et  al., 2021 ) and was consistent with a biopsychosocial approach to health (l arsen, 2022 ) by incorporating physical and psychosocial variables. The psychosocial variables considered in this study were SOC, self-efficacy, sexual sat - isfaction and satisfaction with the gynecological treatment, all of which emerged as important for the participants’ health from the analyses. The longitudinal study 1170 J. NETZl ET Al. design is the main strength of this study. Incorporating the health outcome from t1 in the longitudinal analyses enhances their validity and attests to the stability of the identified associations. With this study, PROMIS Profile 29 data from a big sample of cis women living with EM are provided, which can be used as reference and comparison scores worldwide. 4.1.  Conclusion The findings of this study confirm that pelvic pain is a key factor in the well-being of people living with EM and that shortening the diagnostic delay remains a signif - icant goal in EM care. Additionally, the presented results demonstrate the relevance of psychosocial factors for both physical and mental health in EM and highlight the necessity of mental health care in EM management. They suggest that by attending to psychological and interpersonal factors such as the individual’s SOC, self-efficacy, sexual satisfaction and treatment satisfaction in addition to treating their pelvic pain symptoms, their overall and long-term health could be maintained and improved. Moreover, special attention should be paid to the gynecological care of financially and socially disadvantaged individuals. Future EM research and care should strive for an interdisciplinary, biopsychosocial perspective to adequately meet the challenges posed by this complex disease and its consequences.

J.N. holds a doctoral scholarship by the Studienstiftung des deutschen Volkes (German National Academic Scholarship Foundation). We thank all the people who participated in this study. In addition, we would like to thank the authors of the questionnaires that were used in this study: J. Schumacher, G. Wilz, T. Gunzelmann, E. Brähler (SOC-l9), C. Beierlein, A. Kovaleva, C.J. Kemper, B. Rammstedt (AKSU), U. hartmann, K. heiser, C. Rüffer-hesse, G. Kloth (KFSP-F), M. Schoenthaler, E. Farin, W.K. Karzc, P . Ardelt, U. Wetterauer, A. Miernik (FIPS) and R.D. hays, K.l. Spritzer, B.D. Schalet, D. Cella and the PROMIS initiative (PROMIS Profile 29, physical and mental health sum - mary scores). We thank Andreas hetey from Charité – Universitätsmedizin Berlin, Clinical Trial Office, for the support in the use of the REDCap software platform. We thank hannah Arnu for her support during the publication process. Funding No funding was received for conducting this study. Disclosure statement The authors report there are no competing interests to declare. Data availability statement We do not have the approval of the ethics committee or our participants’ consent to make the data acquired and analysed in this study publicly available. The data are available from the cor - responding author upon reasonable request. PSyChOl OGy & hEAl Th 1171

Aerts, l., Grangier, l., Streuli, I., Dällenbach, P ., Marci, R., Wenger, J.-M., & Pluchino, N. ( 2018). Psychosocial impact of endometriosis: From co-morbidity to intervention. Best Practice & Research. Clinical Obstetrics & Gynaecology , 50, 2–10. https://doi.org/10.1016/j.bpo - bgyn.2018.01.008 [Mismatch] Anhang Price, R., Elliott, M. N., Zaslavsky, A. M., hays, R. D., l ehrman, W. G., Rybowski, l., Edgman-l evitan, S., & Cleary, P . D. ( 2014). Examining the role of patient experience surveys in measuring health care quality. Medical Care Research and Review: MCRR , 71(5), 522–554. https://doi.org/10.1177/1077558714541480 Ballard, K., l owton, K., & Wright, J. ( 2006). What’s the delay? A qualitative study of women’s experiences of reaching a diagnosis of endometriosis. Fertility and Sterility , 86(5), 1296–1301. https://doi.org/10.1016/j.fertnstert.2006.04.054 Beierlein, C., Kovaleva, A., Kemper, C. J., & Rammstedt, B. ( 2012). Ein Messinstrument zur Erfassung subjektiver Kompetenzerwartungen: Allgemeine Selbstwirksamkeit Kurzskala (AKSU). (GESIS-Working Papers, Ed.). GESIS - l eibniz-Institut für Sozialwissenschaften. Binder, C., & Rieder, A. ( 2014). Zur Verflechtung von Geschlecht, sozioökonomischem Status und Ethnizität im Kontext von Gesundheit und Migration. Bundesgesundheitsblatt - Gesundheitsforschung - Gesundheitsschutz , 57(9), 1031–1037. https://doi.org/10.1007/s00103-014-2009-1 Brandes, I. ( 2007). l ebensqualität von endometriose-patientinnen – Quality of life of patients with endometriosis. Geburtshilfe Und Frauenheilkunde , 67(11), 1227–1231. https://doi. org/10.1055/s-2007-965696 Brooks, T., Sharp, R., Evans, S., Baranoff, J., & Esterman, A. ( 2020). Predictors of psychological outcomes and the effectiveness and experience of psychological interventions for adult women with chronic pelvic pain: A scoping review. Journal of Pain Research , 13, 1081–1102. https://doi.org/10.2147/JPR.S245723 Bullo, S., & Weckesser, A. ( 2021). Addressing challenges in endometriosis pain communication between patients and doctors: The role of language. Frontiers in Global Women’s Health , 2, 764693. https://doi.org/10.3389/fgwh.2021.764693 Bulun, S. E. ( 2018 ). Endometriosis. In Yen & Jaffe’s Reproductive Endocrinology: Physiology, Pathophysiology, and Clinical Management (8th ed., pp. 609–642). Elsevier Inc. https://doi. org/10.1016/B978-0-323-47912-7.00025-1 Comptour, A., Pereira, B., lambert, C., Chauvet, P ., Grémeau, A.-S., Pouly, J.-l., Canis, M., & Bourdel, N. ( 2020). Identification of predictive factors in endometriosis for improvement in patient quality of life. Journal of Minimally Invasive Gynecology , 27(3), 712–720. https://doi. org/10.1016/j.jmig.2019.05.013 De Graaff, A. A., D’hooghe, T. M., Dunselman, G. A. J., Dirksen, C. D., hummelshoj, l., Simoens, S., … Wullschleger, M.; WERF EndoCost Consortium. ( 2013). The significant effect of endo - metriosis on physical, mental and social wellbeing: Results from an international cross-sectional survey. Human Reproduction (Oxford, England) , 28(10), 2677–2685. https://doi.org/10.1093/ humrep/det284 De Graaff, A. A., van lankveld, J., Smits, l . J., van Beek, J. J., & Dunselman, G. A. J. ( 2016). Dyspareunia and depressive symptoms are associated with impaired sexual functioning in women with endometriosis, whereas sexual functioning in their male partners is not affect - ed. Human Reproduction (Oxford, England) , 31(11), 2577–2586. https://doi.org/10.1093/humrep/ dew215 Denny, E. ( 2004). Women’s experience of endometriosis. Journal of Advanced Nursing , 46(6), 641–648. https://doi.org/10.1111/j.1365-2648.2004.03055.x Dinh, T., Flaxman, T., Shea, K., & Singh, S. S. ( 2020). Endometriosis on the internet - myths or facts? Journal of Minimally Invasive Gynecology , 27(7), S109. https://doi.org/10.1016/j. jmig.2020.08.165 Donkin, A., Goldblatt, P ., Allen, J., Nathanson, V., & Marmot, M. ( 2017). Global action on the social determinants of health. BMJ Global Health , 3(Suppl. 1), e000603. https://doi.org/10.1136/ bmjgh-2017-000603 1172 J. NETZl ET Al. Drinkell, K., Fajzel, h., & Tordon, K. ( 2023). Patient and practitioner: The impact of social factors on diagnostic delay for endometriosis. Undergraduate Research in Natural and Clinical Science and Technology (URNCST) Journal , 7(3), 1–11. https://doi.org/10.26685/urncst.450 Eriksen, h. l. F., Gunnersen, K. F., Sørensen, J. A., Munk, T., Nielsen, T., & Knudsen, U. B. ( 2008). Psychological aspects of endometriosis: Differences between patients with or without pain on four psychological variables. European Journal of Obstetrics, Gynecology, and Reproductive Biology, 139(1), 100–105. https://doi.org/10.1016/j.ejogrb.2007.10.002 Facchin, F., Barbara, G., Dridi, D., Alberico, D., Buggio, l., Somigliana, E., Saita, E., & Vercellini, P . (2017a). Mental health in women with endometriosis: Searching for predictors of psycholog - ical distress. Human Reproduction (Oxford, England) , 32(9), 1855–1861. https://doi.org/10.1093/ humrep/dex249 Facchin, F., Saita, E., Barbara, G., Dridi, D., & Vercellini, P . ( 2017b). “Free butterflies will come out of these deep wounds”: A grounded theory of how endometriosis affects women’s psycholog - ical health. Journal of Health Psychology, 23(4), 538–549. https://doi.org/10.1177/1359105316688952 Facchin, F., Barbara, G., Saita, E., Erzegovesi, S., Martoni, R. M., & Vercellini, P . ( 2016). Personality in women with endometriosis: Temperament and character dimensions and pelvic pain. Human Reproduction (Oxford, England) , 31(7), 1515–1521. https://doi.org/10.1093/humrep/dew108 Facchin, F., Barbara, G., Saita, E., Mosconi, P ., Roberto, A., Fedele, l., & Vercellini, P . ( 2015). Impact of endometriosis on quality of life and mental health: Pelvic pain makes the difference. Journal of Psychosomatic Obstetrics and Gynaecology , 36(4), 135–141. https://doi.org/10.3109 /0167482X.2015.1074173 Fernley, N. ( 2021). That one doctor… Qualitative thematic analysis of 49 women’s written ac - counts of their endometriosis diagnosis. Journal of Endometriosis and Pelvic Pain Disorders , 13(1), 40–52. https://doi.org/10.1177/2284026520984366 Field, A. ( 2009). Discovering Statistics using SPSS (3rd ed.). Sage Publications. Fischer, F., Gibbons, C., Coste, J., Valderas, J. M., Rose, M., & l eplège, A. ( 2018). Measurement invariance and general population reference values of the PROMIS Profile 29 in the UK, France, and Germany. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation, 27(4), 999–1014. https://doi.org/10.1007/s11136-018-1785-8 Giudice, l. C. ( 2010). Endometriosis. The New England Journal of Medicine , 362(25), 2389–2398. https://doi.org/10.1056/NEJMcp1000274 Greene, R., Stratton, P ., Cleary, S. D., Ballweg, M. l., & Sinaii, N. ( 2009). Diagnostic experience among 4,334 women reporting surgically diagnosed endometriosis. Fertility and Sterility , 91(1), 32–39. https://doi.org/10.1016/j.fertnstert.2007.11.020 harris, P . A., Taylor, R., Thielke, R., Payne, J., Gonzalez, N., & Conde, J. G. ( 2009). Research elec - tronic data capture (REDCap) – A metadata-driven methodology and workflow process for providing translational research informatics support. Journal of Biomedical Informatics , 42(2), 377–381. https://doi.org/10.1016/j.jbi.2008.08.010 harris, P . A., Taylor, R., Minor, B. l., Elliott, V., Fernandez, M., O’Neal, l., Mcl eod, l ., Delacqua, G., Delacqua, F., Kirby, J., & Duda, S. N.; REDCap Consortium. ( 2019). The REDCap consortium: Building an international community of software platform partners. Journal of Biomedical Informatics, 95, 103208. https://doi.org/10.1016/j.jbi.2019.103208 hartmann, U., heiser, K., Rüffer-hesse, C., & Kloth, G. ( 2002). Female sexual desire disorders: Subtypes, classification, personality factors, and new directions for treatment. World Journal of Urology , 20(2), 79–88. https://doi.org/10.1007/s00345-002-0280-5 hays, R. D., Spritzer, K. l., Schalet, B. D., & Cella, D. ( 2018). PROMIS®-29 v2.0 Profile physical and mental health summary scores. Quality of Life Research: An International Journal of Quality of Life Aspects of Treatment, Care and Rehabilitation , 27(7), 1885–1891. https://doi.org/10.1007/ s11136-018-1842-3 IBM Corp. ( 2019). IBM SPSS Statistics for Windows, Version 26.0 . Armonk, N y: IBM Corp. Jones, G., Jenkinson, C., & Kennedy, S. ( 2004). The impact of endometriosis upon quality of life: A qualitative analysis. Journal of Psychosomatic Obstetrics and Gynaecology , 25(2), 123–133. https://doi.org/10.1080/01674820400002279 PSyChOl OGy & hEAl Th 1173 Kundu, S., Wildgrube, J., Schippert, C., h illemanns, P ., & Brandes, I. ( 2015). Supporting and in - hibiting factors when coping with endometriosis from the patients’ perspective. Geburtshilfe Und Frauenheilkunde , 75(5), 462–469. https://doi.org/10.1055/s-0035-1546052 la Rosa, V. l., Barra, F., Chiofalo, B., Platania, A., Di Guardo, F., Conway, F., Di Angelo Antonio, S., & lin, l.-T. ( 2020). An overview on the relationship between endometriosis and infertility: The impact on sexuality and psychological well-being. Journal of Psychosomatic Obstetrics and Gynaecology , 41(2), 93–97. https://doi.org/10.1080/0167482X.2019.1659775 laganà, A. S., Condemi, I., Retto, G., Muscatello, M. R. A., Bruno, A., Zoccali, R. A., Triolo, O., & Cedro, C. ( 2015). Analysis of psychopathological comorbidity behind the common symptoms and signs of endometriosis. European Journal of Obstetrics, Gynecology, and Reproductive Biology, 194, 30–33. https://doi.org/10.1016/j.ejogrb.2015.08.015 laganà, A. S., la Rosa, V. l., Rapisarda, A. M. C., Valenti, G., Sapia, F., Chiofalo, B., Rossetti, D., Ban Frangež, h., Vrtačnik Bokal, E., & Vitale, S. G. ( 2017). Anxiety and depression in patients with endometriosis: Impact and management challenges. International Journal of Women’s Health, 9, 323–330. https://doi.org/10.2147/IJWh.S119729 lampert, T., hoebel, J., Kuntz, B., & Waldhauer, J. ( 2019). Soziale Ungleichheit und Gesundheit. In R. haring (Ed.), Gesundheitswissenschaften. Springer Reference Pflege - Therapie - Gesundheit . Springer. lampert, T., Kroll, l. E., Müters, S., & Stolzenberg, h. ( 2013). Messung des sozioökonomischen Status in der Studie „Gesundheit in Deutschland aktuell “(GEDA). Bundesgesundheitsblatt, Gesundheitsforschung, Gesundheitsschutz, 56(1), 131–143. https://doi.org/10.1007/s00103-012-1583-3 langeland, E. ( 2014). Sense of Coherence. In A. C. Michalos (Ed.), Encyclopedia of Quality of Life and Well-Being Research . Springer. https://doi.org/10.1007/978-94-007-0753-5 larsen, l. T. ( 2022). Not merely the absence of disease: A genealogy of the WhO’s positive health definition. History of the Human Sciences, 35(1), 111–131. https://doi.org/10.1177/0952695121995355 l ukas, I., Kohl-Schwartz, A., Geraedts, K., Rauchfuss, M., Wölfler, M. M., häberlin, F., von Orelli, S., Eberhard, M., Imthurn, B., Imesch, P ., & l eeners, B. ( 2018). Satisfaction with medical sup - port in women with endometriosis. PloS One , 13(11), e0208023. https://doi.org/10.1371/ journal.pone.0208023 Manderson, l ., Warren, N., & Markovic, M. ( 2008). Circuit breaking: Pathways of treatment seek - ing for women with endometriosis in Australia. Qualitative Health Research , 18(4), 522–534. Retrieved from http://qhr.sagepub.com.myaccess.library.utoronto.ca/cgi/reprint/18/4/522 https://doi.org/10.1177/1049732308315432 Márki, G., Bokor, A., Rigó, J., & Rigó, A. ( 2017). Physical pain and emotion regulation as the main predictive factors of health-related quality of life in women living with endometriosis. Human Reproduction (Oxford, England) , 32(7), 1432–1438. https://doi.org/10.1093/humrep/dex091 Mechsner, S. ( 2016). Endometriose. Schmerz (Berlin, Germany) , 30(5), 477–490. https://doi. org/10.1007/s00482-016-0154-1 Morán-Sánchez, I., Adoamnei, E., Sánchez-Ferrer, M. l., Prieto-Sánchez, M. T., Arense-Gonzalo, J. J., Casanova-Mompeán, V., Carmona-Barnosi, A., Mendiola, J., & Torres-Cantero, A. M. ( 2020). Is dispositional optimism associated with endometriomas and deep infiltrating endometrio - sis? Journal of Psychosomatic Obstetrics and Gynaecology , 42(1), 50–56. https://doi.org/10.108 0/0167482X.2020.1729732 Netzl, J., Gusy, B., Voigt, B., Sehouli, J., & Mechsner, S. ( 2022). Chronic pelvic pain in endome - triosis: Cross-sectional associations with mental disorders, sexual dysfunctions and childhood maltreatment. Journal of Clinical Medicine , 11(13), 3714. https://doi.org/10.3390/jcm11133714 Netzl, J., Gusy, B., Voigt, B., Sehouli, J., & Mechsner, S. ( 2023). Pain symptoms, sexual and men - tal health at the time of endometriosis diagnosis. Journal of Endometriosis and Pelvic Pain Disorders, 15(2), 72–81. https://doi.org/10.1177/22840265231179004 Pérez-l ópez, F. R., Ornat, l., Pérez-Roncero, G. R., l ópez-Baena, M. T., Sánchez-Prieto, M., & Chedraui, P . ( 2020). The effect of endometriosis on sexual function as assessed with the Female Sexual Function Index: Systematic review and meta-analysis. Gynecological Endocrinology: The Official Journal of the International Society of Gynecological Endocrinology , 36(11), 1015–1023. https://doi.org/10.1080/09513590.2020.1812570 1174 J. NETZl ET Al. Petrelluzzi, K. F. S., Garcia, M. C., Petta, C. A., Grassi-Kassisse, D. M., & Spadari-Bratfisch, R. C. (2008). Salivary cortisol concentrations, stress and quality of life in women with endometri - osis and chronic pelvic pain. Stress (Amsterdam, Netherlands) , 11(5), 390–397. https://doi. org/10.1080/10253890701840610 Pope, C. J., Sharma, V., Sharma, S., & Mazmanian, D. ( 2015). A systematic review of the associ - ation between psychiatric disturbances and endometriosis. Journal of Obstetrics and Gynaecology Canada: JOGC = Journal D’obstetrique Et Gynecologie Du Canada: JOGC , 37(11), 1006–1015. https://doi.org/10.1016/S1701-2163(16)30050-0 PROMIS-29 Profile v2.1. ( 2023). Retrieved August 14, 2023, from https://www.healthmeasures. net/index.php?option=com_instruments&view=measure&id=849&Itemid=992 Pulice-Farrow, l., Gonzalez, K. A., & lindley, l. ( 2021). ‘None of my providers have the slightest clue what to do with me’: Transmasculine individuals’ experiences with gynecological health - care providers. International Journal of Transgender Health , 22(4), 381–393. https://doi.org/10 .1080/26895269.2020.1861574 Quiñones, M., Urrutia, R., Torres-Reverón, A., Vincent, K., & Flores, I. ( 2015). Anxiety, coping skills and hypothalamus-pituitary-adrenal (hPA) axis in patients with endometriosis. Journal of Reproductive Biology and Health , 3(1), 2. https://doi.org/10.7243/2054-0841-3-2 Rees, M., Kiemle, G., & Slade, P . ( 2020). Psychological variables and quality of life in women with endometriosis. Journal of Psychosomatic Obstetrics and Gynaecology , 43(1), 58–65. https:// doi.org/10.1080/0167482X.2020.1784874 Reinders, h. ( 2006). Kausalanalysen in der längsschnittforschung. Das Crossed-lagged-Panel Design. Diskurs Kindheits- Und Jugendforschung , 1(4), 596–587. https://doi.org/10.25656/01 Reis, F. M., Coutinho, l. M., Vannuccini, S., l uisi, S., & Petraglia, F. ( 2020). Is stress a cause or a consequence of endometriosis? Reproductive Sciences (Thousand Oaks, Calif.) , 27(1), 39–45. https://doi.org/10.1007/s43032-019-00053-0 Romaniuk, A., & Oniszczenko, W. ( 2023). Resilience, anxiety, depression, and life satisfaction in women suffering from endometriosis: A mediation model. Psychology, Health & Medicine , 28(9), 2450–2461. Roomaney, R., Kagee, A., & heylen, S. ( 2019). Biopsychosocial predictors of symptoms of de - pression in a sample of South African women diagnosed with endometriosis. Health Care for Women International , 41(3), 308–329. https://doi.org/10.1080/07399332.2019.1624758 Schoenthaler, M., Farin, E., Karcz, W. K., Ardelt, P ., Wetterauer, U., & Miernik, A. ( 2012). The Freiburg Index of Patient Satisfaction: Introduction and validation of a new questionnaire. Deutsche Medizinische Wochenschrift (1946) , 137(9), 419–424. https://doi.org/10.1055/s-0031-1298976 Schumacher, J., Wilz, G., Gunzelmann, T., & Brähler, E. ( 2000). Die Sense of Coherence Scale von Antonovsky - Teststatistische Überprüfung in einer repräsentativen Bevölkerungsstichprobe und Konstruktion einer Kurzskala. Psychotherapie, Psychosomatik, Medizinische Psychologie , 50(12), 472–482. https://doi.org/10.1055/s-2000-9207 Schweppe, K.-W. ( 2011). Endometriose – Entstehung, Diagnostik, Behandlungsmöglichkeiten und Probleme in Klinik und Praxis. Journal Für Reproduktionsmedizin Und Endokrinologie , 8(3), 180–194. Retrieved from http://www.kup.at/kup/pdf/9658.pdf Siedentopf, F., Tariverdian, N., Rücke, M., Kentenich, h., & Arck, P . C. ( 2008). Immune status, psychosocial distress and reduced quality of life in infertile patients with endometriosis. American Journal of Reproductive Immunology (New York, N.Y .: 1989) , 60(5), 449–461. https:// doi.org/10.1111/j.1600-0897.2008.00644.x Sillem, M. ( 2015). Endometriose: Klinik und Therapie. In M. Sillem, F. Siedentopf, & S. Mechsner (Eds.), Leitsymptom chronischer Unterbauchschmerz der Frau. (pp. 55–60). Springer. https://doi. org/10.1007/978-3-662-43669-1 Spritzer, K. l., & h ays, R. D. ( 2018). Calculating physical and mental health summary scores for PROMIS-29 v2.0 and v2.1. Retrieved June 20, 2022, from http://www.healthmeasures.net/ media/kunena/attachments/257/PROMIS29_Scoring_08082018.pdf Sullivan-Myers, C., Sherman, K. A., Beath, A. P ., Cooper, M. J. W., & Duckworth, T. J. ( 2023). Body image, self-compassion, and sexual distress in individuals living with endometriosis. Journal of Psychosomatic Research , 167, 111197. https://doi.org/10.1016/j.jpsychores.2023.111197 PSyChOl OGy & hEAl Th 1175 Taylor, h. S., Kotlyar, A. M., & Flores, V. A. ( 2021). Endometriosis is a chronic systemic disease: Clinical challenges and novel innovations. Lancet (London, England) , 397(10276), 839–852. https://doi.org/10.1016/S0140-6736(21)00389-5 van Barneveld, E., Manders, J., van Osch, F., van Poll, M., Visser, l., hanegem, l ., … l eue, C. (2021). Depression, anxiety and correlating factors in endometriosis: A systematic review and meta-analysis. Journal of Women’s Health , https://doi.org/10.1089/jwh.2021.0021 Vitale, S. G., la Rosa, V. l., Rapisarda, A. M. C., & laganà, A. S. ( 2017). Impact of endometriosis on quality of life and psychological well-being. Journal of Psychosomatic Obstetrics and Gynaecology, 38(4), 317–319. https://doi.org/10.1080/0167482X.2016.1244185 Vitale, S. G., Petrosino, B., la Rosa, V. l., Rapisarda, A. M., & laganà, A. S. ( 2016). A systematic review of the association between psychiatric disturbances and endometriosis. Journal of Obstetrics and Gynaecology Canada: JOGC = Journal D’obstetrique Et Gynecologie Du Canada: JOGC, 38(12), 1079–1080. https://doi.org/10.1016/j.jogc.2016.09.008 Wischmann, T. ( 2008). Psychologische Aspekte bei Endometriose und Kinderwunsch – einige kritische Anmerkungen. Geburtshilfe Und Frauenheilkunde , 68(3), 231–235. https://doi. org/10.1055/s-2007-989480 Wu, S. V., h sieh, N., lin, l., & Tsai, J. ( 2016). Prediction of self-care behaviour on the basis of knowledge about chronic kidney disease using self-efficacy as a mediator. Journal of Clinical Nursing, 25(17–18), 2609–2618. https://doi.org/10.1111/jocn.13305 Zimmerman, D. W. ( 1987). Comparative power of Student T test and Mann-Whitney U test for unequal sample sizes and variances. The Journal of Experimental Education , 55(3), 171–174. https://doi.org/10.1080/00220973.1987.10806451 Zirke, N., Schmid, G., Mazurek, B., Klapp, B. F., & Rauchfuss, M. ( 2007). Antonovsky’s Sense of Coherence in psychosomatic patients - a contribution to construct validation. Psycho-Social Medicine, 4, Doc03. 1176 J. NETZl ET Al. Appendix A.  Descriptive statistics of the 7 PROMIS profile 29 scales Table A1. Descriptive statistics of the 7 PRoMIs health domains at t1 of the complete sample at t1 ( n = 723). PRoMIs Profile 29 scale Min Max M ± sD Mdn (P25, P75) t -score M ± sD Physical functioning 7 20 17.26 ± 2.76 18.00 (16.00, 20.00) 46.91 ± 7.28 a nxiety 4 20 10.91 ± 3.80 11.00 (8.00, 14.00) 60.95 ± 8.22 Depression 4 20 11.34 ± 3.85 12.00 (8.00, 14.00) 60.92 ± 7.54 Fatigue 4 20 15.14 ± 3.76 16.00 (13.00, 18.00) 63.47 ± 8.31 sleep disturbance 4 20 12.52 ± 3.93 12.00 (9.00, 15.00) 55.17 ± 8.31 a bility to participate in social roles and activities 4 20 11.59 ± 3.49 12.00 (9.00, 14.00) 43.76 ± 6.81 Pain interference 4 20 11.73 ± 4.56 12.00 (8.00, 15.00) 60.14 ± 8.31 general pain intensity a 0 10 5.00 ± 2.60 5.50 (3.00, 7.00) – aNo t -score available for the general pain intensity item. Appendix B.  Statistical assumptions of multiple linear regression analysis The assumptions of no multicollinearity and uncorrelated residuals were met by all four models (VIF .10, Durbin-Watson test = 1–3). The assumption of normally distrib - uted residuals was met as well with an exception of the longitudinal model for physical health: In this model, the residuals showed slight deviations from a normal distribution. Partial plots between the residuals of the outcome variables and each predictor variable and plots of the standardised residuals against the standardised predicted values indicated not perfectly linear relationships and heteroscedasticity. The plots of the standardised residuals against the stan - dardised predicted values of the regression models for mental health at t1 and t2, however, were satisfactory so that homoscedasticity could be assumed. Appendix C.  Results for the subsample of participants who had received the EM diagnosis by laparoscopy ( n = 624) Table A2. c ross-sectional correlation analyses between physical and mental health scores at t1 and putative predictor variables including only those participants who had received the eM diagnosis by laparoscopy ( n = 624). Putative predictor variable Physical health t1 a Mental health t1 a N r p N r p s ociodemographic variables Income 620 .166 <.001 620 .171 <.001 eM-related strain on the relationship 467 −0.215 <.001 467 −0.343 <.001 eM-related variables Diagnostic delay 609 −0.133 <.001 609 −0.087 .033 Number of surgeries 622 −0.174 <.001 622 −0.084 .037 Number of pelvic pain types 624 −0.409 <.001 624 −0.319 <.001 Pelvic pain intensity 624 −0.431 <.001 624 −0.379 <.001 Pelvic pain days per month 624 −0.455 <.001 624 −0.350 <.001 Psychosocial variables soc l -9 624 .338 <.001 624 .633 <.001 aKsU 617 .292 <.001 617 .388 <.001 KFsP-F – sexual satisfaction 604 .081 .045 604 .151 <.001 FIPs 600 −0.335 <.001 600 −0.342 .05). PSyChOl OGy & hEAl Th 1177 Statistical assumptions of multiple linear regression analysis The following descriptions refer to the regression models presented in Table A3. The assumptions of no multicollinearity and uncorrelated residuals were met by both models (VIF .10, Durbin-Watson test = 1–3). The assumption of normally distributed residuals was met by both models as well. Partial plots between the residuals of the outcome variables and each predictor variable and plots of the standardised residuals against the standardised predicted val - ues indicated not perfectly linear relationships and heteroscedasticity. The plots of the stan - dardised residuals against the standardised predicted values of the regression model for mental health at t1, however, were satisfactory so that homoscedasticity could be assumed. Table A3. Results of the cross-sectional multiple regression analyses for the outcome variables physical health at t1 and mental health at t1 including only those participants who had received the eM diagnosis by laparoscopy ( n = 624). c oefficientsb o utcome variable Predictor variable a B (s td.-error) β pb 95%-cI b Physical health t1 c c onstant 46.57 (2.86) <.001 40.95, 52.20 Income 0.24 (0.15) .06 .116 −0.06, 0.54 eM-related strain on the relationship −0.01 (0.02) −0.03 .450 −0.05, 0.02 Diagnostic delay −0.11 (0.05) −0.08 .033 −0.21, −0.01 Number of surgeries −0.68 (0.26) −0.10 .010 −1.19, −0.17 Number of pelvic pain types −0.15 (0.36) −0.03 .687 −0.85, 0.56 Pelvic pain intensity −0.09 (0.03) −0.18 .008 −0.15, −0.02 Number of pelvic pain days per month −0.22 (0.04) −0.25 <.001 −0.30, −0.13 soc-l9 1.04 (0.37) .14 .005 0.32, 1.76 aKsU 1.46 (0.54) .12 .007 0.39, 2.53 KFsP-F - sexual satisfaction 0.06 (0.07) .04 .398 −0.07, 0.18 FIPs −1.22 (0.34) −0.15 <.001 −1.88, −0.56 Mental health t1 d c onstant 31.29 (2.09) <.001 27.19, 35.40 Income 0.06 (0.11) .02 .588 −0.16, 0.28 eM-related strain on the relationship −0.04 (0.01) −0.10 .007 −0.06, −0.01 Diagnostic delay −0.03 (0.04) −0.03 .437 −0.10, 0.04 Number of surgeries −0.12 (0.19) −0.02 .525 −0.50, 0.25 Number of pelvic pain types 0.21 (0.26) .05 .421 −0.30, 0.73 Pelvic pain intensity −0.09 (0.02) −0.21 <.001 −0.13, −0.04 Number of pelvic pain days per month −0.10 (0.03) −0.14 .001 −0.16, −0.04 soc-l9 3.16 (0.27) .49 <.001 2.63, 3.68 aKsU 0.45 (0.40) .05 .260 −0.33, 1.23 KFsP-F - sexual satisfaction 0.03 (0.05) .02 .512 −0.06, 0.13 FIPs −0.91 (0.25) −0.13 <.001 −1.39, −0.42 Note. linear hierarchical regression analyses (enter method). at he number of participants without missing values in the respective predictor variable and included in the pairwise analysis can be found in table a2 , column N. bB = unstandardised beta coefficient. s td. error = standard error of the unstandardised beta coefficient. B = standardised coefficient. p = p-value, probability. 95%- cI = 95% confidence interval for the unstandardised beta coefficient B. cstep I: sociodemographic variables, R2 = .073, p <.001, F(2, 448) = 17.61, p <.001; step II: sociodemographic and eM-related variables, R2 = .302, p <.001, F(7, 443) = 27.38, p <.001; step III: sociodemographic, eM-related and psychosocial variables, R2 = .380, p <.001, F(11, 439) = 24.45, p <.001. t he degrees of freedom are based on the minimum pairwise n. dstep I: sociodemographic variables, R2 = .145, p <.001, F(2, 448) = 38.06, p <.001; step II: sociodemographic and eM-related variables, R2 = .260, p <.001, F(7, 443) = 22.26, p <.001; step III: sociodemographic, eM-related and psychosocial variables, R2 = .532, p <.001, F(11, 439) = 45.29, p <.001. t he degrees of freedom are based on the minimum pairwise n.