Sexual function in women with a history of sexual assault: A cross-sectional study

Sigrid Bergsmyr: data curation, formal analysis, methodology, investigation, data curation and analyses, visualization, interpretation of results, original drafting, reviewing and editing manuscript. Signe Nilssen Stafne: conceptualization, design/methodology, investigation including data collection, visualization, interpretation of results, original drafting, reviewing and editing manuscript, supervision, funding acquisition. Cecilie Therese Hagemann: conceptualization, design/methodology, data curation, formal analysis, investigation including data collection, visualization, validation, interpretation of results, original drafting, reviewing and editing manuscript, supervision, project administration, funding acquisition.

Approval of this study by the Regional Committee for Medical and Health Research Ethics (REK‐Midt, ref. #245815) was granted on March 26, 2024. Data Protection Impact Assessment (DPIA) was performed in spring 2022, and the reuse of data was regarded as not impacting the participants' consent to participate. The institutional review board (R&D) at the Department of Obstetrics and Gynecology at St. Olavs hospital, Trondheim approved the study on April 30, 2024 (Elements ref. #2021/14758). We have permission from the MAPI trust to use a specific translated Norwegian version of the FSFI in this study. All participants received written information about the original study, including that participation in the study is voluntary and that their data could be withdrawn at any time. Updated information about the current reuse of data and further analyses were available on request and posted on the research group's public website .

Part of this study, the publishing process has been funded by The Joint Research Committee (FFU), a collaboration between St. Olavs hospital HF and the Faculty of Medicine and Health Sciences, NTNU, which manages research funds from Helse Midt‐Norge RHF (Application number 38957 and reference number 2024/75315).

A total of 397 women (47% response rate) completed the survey. 20 Three of the 397 women (1%) did not answer the question on sexual assault and were therefore excluded from all analyses (Figure  1 ). Distribution of 397 participants completing an online survey on symptoms of complex pelvic pain in the spring of 2022 according to sexual assault status. 20 Three participants were excluded from analyses because of missing values on sexual assault status, hence the number analyzed was 394. Descriptive characteristics of the study population ( N  = 394) are presented in Table  1 (left column). Table  1 also shows the distribution of total FSFI mean scores among the 261 participating women with complete FSFI scores by the same characteristics. The total FSFI mean score among sexually active women was 26.4 (SD 5.5; 95% CI 25.8 to 27.1). Postmenopausal women and those reporting previous pelvic injury had borderline significantly lower total FSFI mean scores than premenopausal women and those not reporting injury, respectively. Women with fibromyalgia had significantly lower total FSFI mean scores than those without. Descriptive characteristics of 394 survey participants (left column), and the distribution of mean total Female Sexual Function Index (FSFI) scores by the same characteristics among the 261 participants with complete FSFI scores in an online survey on symptoms of complex pelvic pain. Higher Female Sexual Function Index (FSFI) scores indicating better sexual function. ANOVA (analysis of variance) test. Postmenopausal women excluded. Mental Health Inventory‐5, score 0‐100. Here, a score of ≤ 76 is regarded as below cut‐off, that is, worse mental health. Fibromyalgia Survey Diagnostic Criteria – 2016: WPI ≥ 7 and SSS ≥ 5, alternatively WPI 4‐6 and SSS ≥9. In addition, pain must have lasted for ≥ 3 months and be localized in at least four of five body regions: axial, left/right upper limbs and left/right lower limbs. Of the 394 women included, 120 (30.4%) reported a history of sexual assault: 38 (9.6%) reported child sexual assault, 56 (14.2%) adult sexual assault, and 26 (6.6%) both child and adult sexual assault. Those reporting both child and adult sexual assault were younger, had a higher mean BMI, and more often had lower education and were unemployed than the other participants (Table  S1 ). Having a present partner was equally distributed among participant groups ( p  = 0.2), and there was no significant difference in sexual activity past four weeks between women with different assault histories ( p  = 0.7). Those exposed to sexual assault as a child and those exposed both as a child and as an adult more often had vulvodynia and moderate/severe pelvic pain in the past four weeks than those exposed as an adult and those not exposed. Those with a history of both child and adult sexual assault had lower total MHI‐5 scores ( p  < 0.001), and a higher proportion of those with a history of sexual assault as a child or both as a child and as an adult had fibromyalgia compared to the other two participant groups (adult assault and no history of sexual assault) ( p  = 0.02) (Table  S1 ). Participants exposed to sexual assault as an adult and those exposed both as a child and as an adult, reported lower total FSFI mean scores compared to those exposed as a child and those not exposed ( p  = 0.005) (Table  2 ). However, a significant mean difference was only observed between the group exposed to child sexual assault vs. those exposed to both child and adult sexual assault, with a negative mean difference in total FSFI score of 4.9 ( p  = 0.02) (data not shown). Distribution of total and subdomain scores of the questionnaire Female Sexual Function Index (FSFI) by sexual assault type among 394 study participants completing an online survey on symptoms of complex pelvic pain. Note : ANOVA (analysis of variance) test used for all analyses, df =3. Higher scores indicating better sexual function. Zero answers scored as missing, n =261. Regarding the FSFI subdomains, participants exposed to sexual assault as an adult and those exposed both as a child and as an adult demonstrated lower scores in the arousal ( p  = 0.008) and orgasm ( p  = 0.003) subdomains compared to those exposed as a child and those not exposed (Table  2 ). However, significant differences in mean arousal scores were found only between participants exposed as a child vs. those exposed both as a child and as an adult and between those not exposed vs. those exposed both as a child and as an adult, with negative mean differences of 1.0 ( p  = 0.02) and 0.8 ( p  = 0.02), respectively (data not shown). For the orgasm scores, a significant difference was observed only between participants not exposed vs. those exposed as an adult, with a negative mean difference of 0.7 ( p  = 0.02). In the pain subdomain, both participants exposed to sexual assault as a child and those exposed both as a child and as an adult reported lower scores than the other groups ( p  = 0.005) (Table  2 ), but a significant difference in mean pain scores was found only between those not exposed vs. those exposed both as a child and as an adult, with a negative mean difference of 1.0 ( p  = 0.04). In total, 113 (43.3%) of the 261 participants with a valid total FSFI score, reported low sexual function (total FSFI score ≤ 26.55). A higher proportion of participants reporting adult sexual assault (59.0%) or both child and adult sexual assault (55.6%) had total FSFI scores below the cut‐off value compared with participants reporting no history of sexual assault (39.4%) or child sexual assault (37.5%), although this difference did not reach statistical significance ( p  = 0.09) (Table  3 , left columns). In the logistic regression analyses, only adult sexual assault status gave statistically significant higher odds of 2.2 to 2.5 of reporting low sexual function in all models, the association being strongest in the fully adjusted model (Table  3 , right column). We performed subgroup analyses and found that the effect sizes between those exposed to adult sexual assault and those without a history of sexual assault on overall sexual function remained essentially unchanged among healthy volunteers (OR 2.5 (95% CI 0.8 to 7.9)) and among participants without persistent pelvic pain (OR 2.4 (95% CI 1.0 to 6.0)) (Tables  S2 and S3 ). Logistic regression analyses showing sexual assault status by low sexual function among 261 study participants completing an online survey on symptoms of complex pelvic pain. No history of sexual assault is used as reference in all comparisons. Odds ratios of unadjusted and three adjusted models (see details in footnotes) are shown. Low sexual function defined as a total FSFI score ≤ 26.55. Unadjusted logistic regression model. Adjusting in a logistic regression model for age only (< 30, 30‐39, 40‐49, ≥ 50 years). Adjusting in a logistic regression model for postmenopausal status only (yes or no). Adjusting in a multivariable logistic regression model for postmenopausal status (yes or no); BMI (< 18.5, 18.5‐24.9, 25‐29.9, ≥ 30); and employment status (employed/student or unemployed). Percentage within the row (history of sexual assault) category. Chi square analysis, df =3.

In the present study, approximately 10% of participants reported child sexual assault, 14% reported adult sexual assault, and 7% reported experiencing both child and adult sexual assault. Sexual activity did not differ between participants with and without a history of sexual assault. However, the total FSFI mean score among the sexually active participants was only 26.4, which is lower than the cut‐off level defined as low sexual function. Being exposed to adult sexual assault and to both child and adult sexual assault, but not to child sexual assault, was associated with lower total FSFI mean scores compared to no exposure, but after adjustments, only adult sexual assault participants had more than twice the odds of low sexual function compared to those with no history of sexual assault. The same pattern was found with arousal and orgasm with those reporting both child and adult sexual assault and those reporting adult sexual assault, respectively, having the lowest scores. However, in the pain subdomain, child sexual assault and both child and adult sexual assault participants had the lowest mean scores. We found no difference in sexual activity between participant groups. Previous research shows that survivors of sexual assault may either increase or decrease sexual activity post‐assault. 11 , 14 Increased sexual activity can serve as a coping mechanism or a way to regain control and bodily autonomy. 32 Conversely, some adult sexual assault survivors may engage in risky sexual behaviors or use sex as self‐harm due to psychological distress and substance use. 33 , 34 Others may adopt avoidance strategies, experiencing sexual aversion rooted in fear or anxiety. Female rape survivors have reported sexual inactivity due to fear of touch and anxiety around men. 15 These varied responses highlight the importance of clinicians recognizing and validating diverse coping mechanisms in sexual behavior following sexual assault. Doing so is essential in supporting survivors in the process of sexual healing and normalization. As many as 80% of the women had a present partner, and there were no differences between the participant groups, which has been described before. 14 We did not examine the role of the partner in the assaults, nor did we investigate their involvement in post‐assault rehabilitation. Fava et al. have pointed out that supportive partners, effective communication, level of commitment, and relationship duration buffer the negative (i.a. sexual) outcomes associated with adult sexual assault. 11 We observed that participants exposed to both child and adult sexual assault had the lowest MHI‐5 scores, followed by those exposed to adult sexual assault and those exposed to child sexual assault. Psychological issues like depression, anxiety, and suicidality are closely linked to reduced sexual functioning in survivors. 18 , 35 A study from Stockholm found that 60% of women reported worsened sexual functioning six months post‐assault, strongly associated with post‐traumatic stress disorder (PTSD) and depression. 16 These findings emphasize the complex relationship between mental health and sexual well‐being after assault. Integrated treatment approaches that address both PTSD and sexual dysfunction are crucial, as highlighted in recent post‐assault intervention studies. 36 , 37 Adult sexual assault exposure more than doubled the odds of low sexual functioning compared to those with no history of sexual assault or those with child sexual assault, indicating a moderate effect. 38 Participant reporting both child and adult sexual assault had a nearly 5 points lower total mean FSFI score than those reporting child sexual assault, statistically significant, and probably also clinically relevant. Similarly, survivors of adult sexual assault had a total mean FSFI score that was 3.5 points lower than that of survivors of child sexual assault ( p  = 0.1). Although minimal clinically important differences for FSFI are still being explored, these findings align with previous research and underscore the relevance of sexual functioning as a key outcome in post‐assault studies. 11 , 14 , 15 Our cross‐sectional design does not allow causal conclusions, but our findings may indicate a possible dose–response tendency between assault exposure and low sexual function, with women reporting both child and adult sexual assault seemingly showing the lowest scores across sexual function subdomains. Similarly, Lutfey et al. showed that exposure to multiple abuse types (childhood emotional, adult sexual, and adult emotional) doubled the odds of reduced sexual function, and that the association was stronger for women with two or more abuse types. 14 Kinzl et al. also found that female university students with multiple child sexual assault incidents reported more sexual desire and orgasm disorders than those with no or single child sexual assault experiences, suggesting that repeated trauma may have distinct negative effects. 39 However, other studies indicate that trauma type 18 and severity may not predict sexual outcomes, 11 implying that other factors could be more relevant to post‐assault sexual healing. These findings highlight the complexity of sexual recovery and the need for individualized, trauma‐informed care. Surprisingly, child sexual assault participants reported significantly higher total FSFI mean scores, indicating better sexual functioning than the adult sexual assault group, and even higher scores than those with no history of sexual assault. However, for the latter comparison, the CIs overlapped, and those with no history of sexual assault in our study demonstrated relatively low mean scores. Furthermore, except for the pain subdomain, women exposed to child sexual assault reported the highest scores across the FSFI subdomains. A recent systematic review noted that child sexual assault is not consistently associated with adult sexual difficulties, especially in community samples. 40 The authors suggested that women in non‐clinical populations without, for example, psychological disorders, may demonstrate greater sexual resilience despite having a child sexual assault history. 40 Although our sample comprised healthy volunteers, most were gynecological patients or members of vulvodynia and endometriosis associations. Thus, as expected, subgroup analyses limited to healthy volunteers yielded the same pattern: No significant difference was found between those exposed to child sexual assault and those with no history of sexual assault in overall sexual function. This may indicate that the isolated effect of child sexual assault on adult sexual function in women may be less substantial than commonly presumed. Participants in the adult sexual assault and both child and adult sexual assault groups scored 0.7 to 1 points lower in the arousal and orgasm subdomains than participants in the no history of sexual assault and child sexual assault groups. Similarly, the Swedish study found arousal as the most affected subdomain post‐adult sexual assault, followed by lubrication. 16 While other studies have highlighted sexual satisfaction as particularly affected by adult sexual assault, 17 , 18 we observed only a borderline significant association. In a Dutch RCT, the combined desire and satisfaction scores post‐assault ranged from 5.5 to 5.9 (possible range: 2–12), 37 which were notably lower than the corresponding scores observed in our adult sexual assault (7.2) and both child and adult sexual assault (6.7) groups. Whatsoever, both randomized groups in the Dutch study improved significantly over the 12‐week treatment period using eye movement desensitization and reprocessing (EMDR) therapy for PTSD. 37 Sexual contact may trigger PTSD symptoms, and research shows PTSD severity negatively affects sexual satisfaction. 18 Addressing this link is vital for developing trauma‐informed interventions that support sexual recovery. These findings underscore the importance of integrating mental health treatment with sexual rehabilitation to maximize outcomes for adult sexual assault survivors. Both participants exposed to child sexual assault and those exposed to both child and adult sexual assault had lower FSFI pain scores than those with no history of sexual assault, consistent with findings that dyspareunia is common among sexual assault survivors. 14 , 15 While Lutfey et al. reported scores around 5.0 for assaulted women, our sample showed lower scores (3.6–4.3), suggesting more pronounced pain‐related sexual problems in our sample. Despite using a modified 10‐item version of FSFI, their results should be comparable to ours. 14 Similarly, Postma et al. found worsened pain scores among young rape survivors (4.7) versus non‐traumatized controls (5.4). 15 These findings highlight the need to address dyspareunia as a key component of post‐assault care, emphasizing its relevance in both clinical assessment and therapeutic intervention strategies. Our participants reporting child sexual assault and both child and adult sexual assault more frequently reported vulvodynia and moderate/severe pelvic pain than our participants with adult or no history of sexual assault, consistent with findings from Nordic gynecology outpatient clinics. 25 , 41 A meta‐analysis also linked sexual abuse to persistent pelvic and chronic pain conditions. 42 These results suggest a strong association between gynecological pain and sexual assault history. However, it is unclear whether reduced sexual functioning is mediated by pelvic or widespread pain, or if persistent pain and central sensitization act as independent mechanisms affecting post‐assault sexual health. One of our study's strengths is the use of a comprehensive and detailed survey, incorporating a variety of validated questionnaires. Additionally, no difference was found in the frequency of attempted intercourse over the past month between participants with different sexual assault histories, which could have biased FSFI scores when zero answers were scored as missing. Furthermore, FSFI scores among study participants demonstrated high internal reliability (Cronbach's alpha was ≥0.85 for all subdomains, and 0.92 for all FSFI items). Aside from zero answers on the FSFI questionnaire, there were few missing responses in the survey, allowing for robust statistical analyses and comparisons. Another strength is the adequate sample size, which allowed comparisons between child sexual assault, adult sexual assault, and both child and adult sexual assault participants versus those with no history of sexual assault. However, in some cases, Fisher's exact test was needed for valid statistical analyses. There are several potential limitations of this study. As a result of excluding sexually inactive participants from several comparisons, the number of participants included in the analyses on sexual function was reduced, except for the desire subdomain, hence introducing type 2 statistical error. This was especially apparent in the logistic regression analyses with small subgroup sizes of child sexual assault survivors ( n  = 24) and both child and adult sexual assault survivors ( n  = 18) resulting in reduced statistical power. Although our logistic regression models adjusted for several variables, the possibility of unmeasured confounding remains. Additionally, a limitation of the present study was the absence of a priori power calculation, it being a secondary analysis of a validation study. 19 , 43 As previously stated, child sexual assault participants reported the highest total mean FSFI scores. However, it is important to note that even though we excluded those without sexual activity past four weeks, 44 all participant groups in the present study had markedly lower mean total FSFI scores compared to the healthy control group in Rosen's study. 22 Consequently, the mean level of sexual functioning in the present study sample is probably lower than that of a hypothetic community sample, thus potentially obscuring possible associations. Also, the prevalence of pain conditions known to affect sexual functioning, such as vulvodynia and persistent pelvic pain, is likely higher in our study sample than in a community‐based sample. 14 Nevertheless, the observed associations appear robust. Sensitivity analyses showed that the effect sizes for differences in overall sexual function between the adult sexual assault and no sexual assault groups were consistent among healthy volunteers and among participants without persistent pelvic pain. Although our participants were recruited widely, our analyses were limited to sexually active women living in Norway and may not be generalizable to sexually inactive women and women in other countries. A concern has been addressed by not including males in such studies. 11 Self‐reported data may involve recall bias and inaccurate reporting, especially for child sexual assault occurring long ago. Recruitment for validating a pelvic floor instrument may have introduced selection bias, leading to an overrepresentation of women with pelvic pain and sexual problems. A higher proportion of participants reported sexual assault compared to Norwegian population data and a Nordic gynecological sample. 7 , 41 This could reflect selection bias or result from how the question was phrased. However, trauma type 18 and severity may have limited impact on sexual functioning. 11 Further research is recommended using a larger, non‐clinical sample more representative of the general female population.

Our findings confirm the previously demonstrated association between a history of adult sexual assault and low overall sexual function. However, the different sexual assault exposure types seem to relate differently to the various dimensions of sexual functioning, with adult sexual assault being associated with lower arousal and orgasm scores, child sexual assault with worse sexual pain scores, and both child and adult sexual assault with less favorable scores on all these three subdomains. It is essential that healthcare professionals treating individuals with pelvic pain or diminished sexual function remain vigilant to the potential impact of sexual assault history and are encouraged to address this issue proactively and sensitively in clinical practice.

Sexual functioning is defined as the absence of difficulty moving through the stages of sexual desire, arousal, and orgasm, as well as subjective satisfaction with the frequency and outcome of individual and partnered sexual behavior. 1 According to WHO, sexual health is fundamental to the overall health and well‐being of individuals, couples, and families and to the social and economic development of communities and countries. 2 However, more women than men report low sexual function and the extent of female sexual problems represents a public health problem. 3 Low sexual function in women has a potentially negative impact on quality of life and interpersonal relationships. 4 Sexual functioning requires an intact connection between the sex organs, hormone‐producing glands, the brain, and the rest of the body. However, both biological, psychological, interpersonal, and social factors can influence women's sexual health. Thus, low sexual function can have a multifactorial etiology. Sexual violence is a widespread problem which largely affects the female population. In a recent analysis of sexual violence against girls (<18 years), age‐standardized global and Western Europe prevalence was 18.9% and 20.7%, respectively. 5 WHO estimates from 2013, found lifetime prevalence of non‐partner sexual violence to be 12.6% in high‐income countries, increasing to 32.7% if also intimate partner physical and/or sexual violence was included. 6 In line with this, a 2023 survey conducted by the Norwegian Centre for Violence and Traumatic Stress Studies (NKVTS) found that 23% of women in the general Norwegian population reported having been raped, and half of these experienced the assault before the age of 18 years. 7 Being subjected to child, adolescent, and/or adult sexual violence is found to impact physical and mental health negatively. 8 , 9 , 10 Today's knowledge on how sexual functioning is affected by a history of sexual violence, especially in adulthood, is inconclusive. 11 A recent meta‐analysis reported reduced sexual functioning among adult women with a history of child sexual abuse, 12 and another review highlighted that sexual desire and arousal were the most frequently affected subdomains. 13 In contrast, the effects of adult sexual assault on sexual activity and functioning have received comparatively less attention. A recent review comprising studies conducted almost exclusively in the United States, revealed inconsistent findings, with only four out of the 14 included studies indicating that being exposed to adult sexual assault was associated with a decline in general sexual functioning. 11 Some studies have found no differences in reported levels of sexual activity following sexual assault. 14 , 15 However, when examining overall sexual function, even after adjusting for factors such as depression, individuals reporting sexual abuse during adolescence or adulthood (defined in this study as abuse occurring after the age of 14) exhibited nearly twice the odds of experiencing low sexual functioning compared to non‐abused counterparts. 14 Like survivors of child sexual assault, individuals exposed to adult sexual assault may be disproportionately affected in specific subdomains of sexual functioning. Sexual arousal, 16 satisfaction, 17 , 18 lubrication, 15 , 16 and pain during sexual activity 14 , 15 appear to be particularly impacted. Nevertheless, the review by Fava et al. identified several methodological limitations in the existing literature, including insufficient differentiation between child and adult sexual assault exposure, frequent use of non‐validated assessment tools, and issues related to selection bias. 11 The objective of this study was to examine the association between a history of sexual assault (child sexual assault; adult sexual assault; or both child and adult sexual assault vs. no history of sexual assault) and various aspects of sexual function, including level of sexual activity, overall sexual functioning, and the specific subdomains sexual desire, arousal, lubrication, orgasm, satisfaction, and pain.

The authors have no conflicts of interest to disclose.

This study was a cross‐sectional exploratory secondary analysis using self‐reported data originally collected to validate the Norwegian version of the Amsterdam Complex Pelvic Pain Symptom Scale. 19 Invited women answered an online questionnaire between March 5th and May 3rd 2022 and were from the following 3 categories: (1) patients referred to St. Olavs hospital general gynecology outpatient clinic ( n  = 173); (2) members of Vulvodynia or Endometriosis Associations ( n  = 70) in Norway; and (3) “healthy volunteers,” that is, female students or university employees ( n  = 154). Recruitment procedure and flow chart have been described elsewhere. 20 Inclusion criteria were female over the age of 18 years; not being pregnant; able to read, write, and understand Norwegian; and capable of providing consent. 20 Participants did not get any form of economical compensation. After securely accessing the online questionnaire, participants consented to participation. Data from the survey were retrieved encrypted through the eFORSK database, a web‐based application developed and maintained by Central Norway Regional Health Authority's IT department, Hemit HF. 20 All analyses have been carried out using deidentified data in a secure St. Olavs hospital research area. Sexual activity was assessed with one single question: “When was the last time you attempted sexual intercourse?” with the following answer options: “less than” or “more than one month ago.” Sexual function was assessed using self‐reported answers to a translated Norwegian version of the Female Sexual Function Index (FSFI) questionnaire provided from the MAPI Trust. Over the past years, the FSFI has been considered the gold standard for the measurement of women's sexual function 21 in clinic and in research and includes the following six subdomains of sexual function: desire, arousal, lubrication, orgasm, satisfaction, and pain. The FSFI has subdomain score scales from zero/one to five, higher scores indicating better sexual function, and possible FSFI full‐scale scores range from 2.0 to 36.0. Rosen et al. demonstrated high internal consistency for the subscales of the FSFI, with Cronbach's alpha values of 0.82 or higher in all subdomains. 22 The FSFI assesses sexual function over the past four weeks. Zero responses, that is, no sexual activity past four weeks, are valid for 15 of 19 items. Hence, except for the desire subdomain, scores were only calculated for participants without zero responses. 21 In this survey, item 15 (satisfaction subdomain) mistakenly included a zero option; 105 such responses were, hence, treated as missing. If half or fewer subdomain items were missing, mean imputation was applied, 23 yielding 261 total FSFI scores. In the present study, Cronbach's alphas were 0.90, 0.88, 0.92, 0.86, 0.85, and 0.93 for desire, arousal, lubrication, orgasm, satisfaction, and pain subdomains, respectively, and 0.92 for total FSFI scores, implying good internal reliability. Furthermore, Wiegel et al. proposed a total FSFI score of 26.55 as the most optimal cut‐off level to differentiate women with sexual dysfunction from women without sexual dysfunction. 24 However, as sexual distress is not measured by the FSFI, the term “low sexual function” is better fit than “sexual dysfunction” to describe a total FSFI score of ≤26.55 and hence, this term is therefore used in this study. Clinically relevant cut‐off scores for the subdomains are not established, neither are clinically significant changes in scores in treatment studies. 21 The main exposure variable explored in this study was sexual assault, mapped with the previously used question 25 : “Have you ever been forced into sexual intercourse/sexual assault?” Response alternatives were: “ No, never ;” “ Yes, as a child (before 18) ;” “ Yes, as an adult (after 18) ;” and “ Yes, as a child and as an adult .” We categorized these into (1) no history of sexual assault (no sexual assault), (2) sexual assault as a child only (child sexual assault), (3) sexual assault as an adult only (adult sexual assault), or ( 4 ) sexual assault both as a child and as an adult (both child and adult sexual assault). Variables included in the descriptive analyses were the women's self‐reported age in years; body mass index (BMI, calculated from self‐reported height in cm and weight in kg); country of origin; educational level ( lower (i.e., completed high school or a lower degree) and higher (i.e., completed a university degree)); employment status ( employed , student , and unemployed ); and marital status ( in a relationship or not ). Self‐reported gynecological variables included: pregnancy and childbirth history; current use of hormonal contraceptives; menopausal status; and history of pelvic injury (pelvic surgery, radiation therapy, and/or injury to genital area). Mental health was measured with the Mental Health Inventory‐5 (MHI‐5) of the Short Form 36 (FS‐36) quality of life questionnaire, which is a 5‐item questionnaire with a total score ranging from 0 to 100, where lower scores indicate worse mental health. 26 A total score of ≤76 was used as cut‐off as it has been found to be associated with the highest sensitivity and specificity for any mood or anxiety disorder. 27 Finally, self‐reported symptoms of fibromyalgia were assessed using the Symptom Severity Scale (SSS) score ranging from 0 to 12 and the Widespread Pain Index (WPI) ranging from 0 to 19. Both are part of the Fibromyalgia Survey Diagnostic Criteria – 2016, which were used to classify participants as meeting the criteria of fibromyalgia (named as having fibromyalgia in this paper) (WPI ≥7 and SSS ≥5 or WPI 4–6 and SSS ≥9, and additionally the pain should be localized in at least 4/5 body regions and has lasted at least 3 months) or not. 28 Approval of this study by the Regional Committee for Medical and Health Research Ethics (REK‐Midt, ref. #245815) was granted on March 26, 2024. Data Protection Impact Assessment (DPIA) was performed in spring 2022, and the reuse of data was regarded as not impacting the participants' consent to participate. The institutional review board (R&D) at the Department of Obstetrics and Gynecology at St. Olavs hospital, Trondheim, approved the study on April 30, 2024 (Elements ref. #2021/14758). We have permission from the MAPI trust to use a specific translated Norwegian version of the FSFI in this study. All participants received written information about the original study, including that participation in the study is voluntary and that their data could be withdrawn at any time. Updated information about the current reuse of data and further analyses were available on request and posted on the research group's public website . Means, standard deviations (SD), and corresponding 95% confidence intervals (CIs), as well as frequencies and proportions were calculated to present descriptive characteristics of the study sample. Comparisons were made by chi‐square analyses and analyses of variance (ANOVA). The normality of data was assessed by visual inspection of frequency histograms and Q–Q plots. 29 We also performed multiple logistic regression analyses without stepwise selection to explore associations between a history of sexual assault (“no history of sexual assault” used as reference in all comparisons) and FSFI score (dichotomous), and due to small sample size we were only able to adjust for age alone (model b); menopausal status alone (model c); and menopausal status, BMI, and employment status (model d). These variables were included in the models because of their previously demonstrated association with and potentially confounding effect on sexual function. 30 , 31 We performed sensitivity analyses to evaluate whether the association was still valid in different subgroups (healthy volunteers and participants without persistent pelvic pain). All data analyses were conducted in IBM SPSS Statistics version 29.0.1.

Table S1. Sociodemographic, gynecological and other clinical characteristics by sexual assault history among 394 women participants completing an online survey on symptoms of complex pelvic pain between March 5th and May 3rd, 2022. Table S2 . Logistic regression analyses showing sexual assault status by low sexual function among 118 study participants recruited from healthy volunteers (subgroup analyses), completing an online survey on symptoms of complex pelvic pain. No history of sexual assault is used as reference in all comparisons. Odds ratios of unadjusted and three adjusted models (see details in footnotes) are shown. Table S3 . Logistic regression analyses showing sexual assault status by low sexual function among 172 study participants without persistent pelvic pain (subgroup analyses) and completing an online survey on symptoms of complex pelvic pain. No history of sexual assault is used as reference in all comparisons. Odds ratios of unadjusted and three adjusted models (see details in footnotes) are shown.