Dyspareunia and Functioning Among Somali Women Who Have Experienced Female Genital Cutting.

Table 2 provides background information about the individuals in our sample. Overall, our sample reported homogenous demographic information, which was expected given the inclusion criteria (i.e., Somali ethnicity, experienced FGC, had sex in the last year). Thirteen participants held an ethnic identity in addition to Somali, either Oromo ( n = 10) or Arab/Bantu ( n = 3) (not shown in table). Everyone was born outside of the United States and many lived in a refugee camp for some period. Everyone was female and Muslim (not shown in table), and most were married. Most participants had not attended college. Nearly two-thirds reported a household income less than the U.S. 2021 poverty threshold of $30,000 for a family of five ( Office of the Assistant Secretary for Planning and Evaluation, n.d. ). Our research study used a community-based participatory research (CBPR) approach ( Wallerstein & Duran, 2010 ), which engaged community members as partners during study development, data collection, analysis, results interpretation, and dissemination. Faculty specializing in human sexuality collaborated with a community organization that specializes in CBPR approaches, SoLaHmo. An 11-member community advisory board (CAB) composed of Somali health professionals, school administrators, and community leaders met 13 times to provide culturally informed guidance on research design and methods – including input on survey content, data analysis and interpretation, and dissemination of study results. CAB members and community researchers on the team advised on acceptability of commonly used terms in sexual health research; for example, we were advised to not use the phrase sexual pleasure. Additionally, “sex” was defined for participants as vaginal intercourse, as indicated by previous research in this Somali population ( Connor, Abdi, et al., 2025 ) and recommended by community researchers on the study team. All verbiage used in the surveys, recruitment, and consent forms were reviewed by community researchers and advisory board members. The University of Minnesota Institutional Review Board approved the protocol. At the beginning of each data collection session, interviewers tried to create a safe and comfortable environment for participants. As part of the informed consent process, a written consent form in English or Somali was reviewed with the participant. Interviewers gave an overview of the study purpose and reviewed interview procedures, making sure to allow time for questions and participants’ demonstration of their understanding of the study before signing the consent form. Women were able to take breaks and pause if needed at any time during data collection. Interviewers remained nearby to answer questions or assist with any technological challenges but were not able to view participant screens or answers. We estimated, via G*power, that we would need to recruit 300 individuals to be able to test our hypotheses, based on 80% power to detect a difference of approximately 18% to compare the difference in proportions to the normal distribution with a significance level of 0.05. Because this is the first quantitative analysis to our knowledge to examine differences in variables of interest with this population, we specified a two-sample test in the power analysis such that either a larger sample or larger effect would be needed to detect a difference. We used convenience and snowball sampling methods to recruit and collect survey and interview data from our sample of Somali women in the Minneapolis-Saint Paul 7-county metropolitan area from September 2021 – April 2023. Women were approached by bilingual community researchers, fluent in Somali and English, in two University-based clinics in the Minneapolis-St. Paul metro area, and in community settings throughout the metro area (e.g., apartment buildings with large numbers of Somali residents, Somali shopping malls, community centers, resettlement agencies, personal community connections). Participants were encouraged to refer others who met the inclusion criteria to the study. To avoid recruitment of people from the same social network, we limited participant referrals to two referrals per participant. Care was taken to ensure that we had participants from across the metro area by tracking how our recruitment numbers matched the population numbers in each of the seven metro counties (e.g., 64% of Somalis lived in Hennepin County and 63% of our sample came from Hennepin County). Because the focus of this study was on dyspareunia and sexual functioning, we intentionally recruited women who had been sexually active in the past year so that they could answer all questions related to sexuality. Inclusion criteria included Somali identity, experience of FGC, being sexually active in the previous year, speaking Somali or English, and being aged 18–45 years. Exclusion criteria included being unable to consent. Face-to-face recruitment in difficult to access populations with low literacy rates has been shown to yield a higher recruitment rate than written methods ( Paskett et al., 2008 ). Recruiters approached women prior to or after their medical appointments at the two participating medical clinics and conducted outreach recruitment by approaching women at community venues. Of the eligible women approached, 78% participated, resulting in a total of 300 participants. Almost half of the participants were recruited by participant referral ( n = 140, 47%), followed by health clinic recruitment ( n = 87, 29%), community recruitment ( n = 43, 15%), and participants from previous research ( n = 28, 9%). Participants were reimbursed $75 for participating and an additional $30 for child-care, as needed. During scheduling, the recruiters informed women about the data collection process (e.g., consent form, personal sexual and relationship questions, average length of survey) and arranged a meeting location. Interviews were conducted in-person at the participant’s preferred location – most ( n = 286, 95%) chose their or a friend’s home. We used audio computer-assisted self-interviewing (ACASI) to administer our questionnaire/survey (described below). This technology was used for the entire survey with the exception of FGC and deinfibulation history. We selected this technology primarily to accommodate the specific needs of our population regarding language, literacy, and sensitivity discussing FGC and sexuality; this technology has been used successfully in similar studies ( Beauclair et al., 2013 ; Bhatnagar et al., 2013 ; Caldwell & Jan, 2012 ; Robinson et al., 2013 ). ACASI allowed participants to use headphones connected to a laptop computer to privately listen to instructions, questions, and responses that were digitally recorded onto an ACASI program while corresponding text was displayed on the computer screen. The computer also had a privacy screen so that the interviewer was unable to see the answers. This enabled literate participants to simultaneously read and listen to the questions while allowing those who had low levels of literacy to rely on the audio component of the survey. Visual analogues were added to scales as appropriate, with colorful graphics that clarified numerical and scale ratings and constructs (e.g., faces for emotions, jars of beans for frequency) ( Blum et al., 2014 ). Based on feedback from the CAB and community researchers, we developed Somali-appropriate images. ACASI programming automatically provided options of “unsure,” “don’t know,” and “not applicable” for each question. We created audio files in both English and Somali versions of the survey. We asked respondents to select their preferred language; 67.7% ( n = 203) of the interviews were conducted in Somali and 32.3% ( n = 97) in English. We used a cross-cultural equivalence process informed by Flaherty and colleagues ( Flaherty et al., 1988 ) to review and adapt our measures ( Connor, Sheik, et al., 2025 ). This process included a panel retreat with eight community and academic researchers reviewing identified measures for semantic and content equivalence. We also solicited feedback from our CAB. All survey items were translated using a team approach ( Behr & Braun, 2023 ; Hollifield et al., 2013 ; Johnson-Agbakwu et al., 2016 ). Sociodemographic variables included age, ethnicity, religion, birth country, years in United States, years in refugee camp, current relationship status, education, employment, and household income. History about FGC was also gathered, such as who performed the procedure and where it was done. The National Survey of Sexual Attitudes and Lifestyles Sexual Function measure (the Natsal-SF) was developed and validated using a general population and clinical sample in Britain ( Mitchell et al., 2012 ). Six sexual health problems from the Natsal-Sexual Function (SF) measure were administered with the stem, In the last year , have you experienced any of the following for 3 months or longer ? Participants answered “Yes” or “No” to having no interest in sex, having no enjoyment in sex, feeling anxious during sex, feeling no excitement or arousal during sex, having difficulty reaching orgasm, and feeling genital pain as a result of sex. The latter item was used to classify respondents as having or not having dyspareunia in analyses. The Changes in Sexual Functioning Questionnaire, Short version (CSFQ-14) is a 14-item clinical and research instrument identifying five scales of sexual functioning, on either of two 5-point Likert-type frequency scales from 0 (not at all) to 4 (every day) or 0 (not at all) to 4 (extreme); higher scores indicate higher sexual functioning. For example, a sample item is “ In the last year , how often did you want to have sex?” The internal consistency and factor structure of the CSFQ-14 as a global measure of sexual dysfunction has been previously established ( Keller et al., 2006 ) by utilizing 12 of the items to create an overall composite of sexual functioning. Due to feedback from community researchers about content equivalence, we modified this version of the CSFQ by removing three items and added one item about the importance of orgasm (i.e., “ In the last year, how important was having an orgasm to you? ”), resulting in 10 items that were summed and then averaged. In our sample, the scale demonstrated good internal consistency, with an alpha of 0.88. Scores for the modified measure could range from 0 to 4. A brief structured interview, designed by our team, was conducted to gather information about FGC types, deinfibulation history, and reinfibulation history. We used an iterative process with eight watercolor-based visual images to determine type of FGC. Uncut vulva, Type 1a, Type 1b, Type 2a, Type 2b, Type 2c, Type 3a, and Type 3b, Type 4 were described to participants in words. This method is described further in Chaisson et al. (2023) and Johnson-Agbakwu et al. (2023) . Interviewers described each watercolor and participants reviewed the watercolors and described modifications that happened at the time of their FGC. Through dialogue, the interviewer and participant determined what type of FGC occurred when she was a child. We previously found that through dialoguing about each illustration, participants had a greater understanding of what each type represented and therefore was a more valid representation of FGC history than self-report alone. For women who were infibulated, the interviewer asked whether and when she was deinfibulated with four options: never, deinfibulated before ever having sex, deinfibulated after sex and before labor and delivery, deinfibulated during labor and delivery. Any participant who was deinfibulated was asked if she made a request to be sewn open, allowed to heal naturally (i.e., no sewing, which typically results in some skin fusing back to the original state of closure), be sewn partially shut, or be sewn shut to the size of the original infibulation. To create a variable that captured FGC type with deinfibulation history, we created a four-category variable. This variable – infibulation/deinfibulation history – captures modifications to the genital area (Category 1 = never infibulated, i.e., experienced type 1, 2, or 4; Category 2 = deinfibulated before sex; Category 3 = deinfibulated after sex, i.e., before or during labor and delivery; Category 4 = type 3, never deinfibulated). Because of the small number who answered “deinfibulated after sex and before labor and delivery,” we combined this response with deinfibulated during labor and delivery to create Category 3. The Bicultural Involvement Questionnaire ( Szapocznik et al., 1980 ) was modified for use with Somali women ( Johnson-Agbakwu et al., 2016 ) and was utilized in the present study. The M-BIQ explores five domains/subscales of acculturation – speak American, speak Somali, enjoy American culture, enjoy Somali culture, and preference for Somali or American culture ( Johnson-Agbakwu et al., 2016 ). Items were answered using a 3-point Likert-type scale ranging from 1 (never), 2 (sometimes), and 3 (always). Sample items included “ How often do you enjoy music from your cultural background?” and “ How often do you enjoy American music ?” We utilized the American scales as a covariate to determine acculturation to American culture by summing across 20 items, resulting in a range of 20 to 60. The internal consistency of this subscale in our sample was good (α = .82). This relationship satisfaction instrument was composed of three questions about the participant’s relationship in the past year (e.g., In the last year , how satisfied were you with your marriage or relationship ?) using a 5-point scale ranging from 1 to 5 (strongly dissatisfied, dissatisfied, neither, satisfied, and strongly satisfied) ( Schumm et al., 1983 ). Scores were created by summing the three items, with higher scores indicating stronger marital satisfaction. The scale has been shown to be a reliable and valid measure of marital satisfaction ( Schumm et al., 1983 ), and demonstrated excellent internal consistency in our sample (α = 0.95). An item from the Refugee Health Screener ( Hollifield et al., 2013 ) was used to assess global emotional distress. The RHS is a screener assessing depression and trauma symptoms in refugees. It is available in Somali, as well as 15 other languages. For this analysis, we used the Distress Thermometer alone to determine emotional distress. This item includes a drawing of a thermometer, with a “0” (“ no distress – things are good ”) at the bottom and a “10” ( extreme distress – I feel as bad as I ever have ”) at the top. Respondents were asked to “pick a number on the line (0–10) that best describes how much distress you have been experiencing in the past week, including today .” To control for potential impact of birth control on sexual function ( Ghorbani et al., 2021 ), participants were asked if they ever used hormonal birth control for at least 6 months (i.e., pills, patch, ring, hormonal implant, hormonal IUD) and classified into two groups: no or yes. To control for impact of vaginal birth on sexual function, participants were asked how many vaginal births they had, including stillbirths, and then classified into two groups: no vaginal births or had vaginal births. Based on interventions for pain with sexual intercourse in existing research and clinical guidelines, including but not limited to pain associated with FGC, we developed a list of potential interventions besides deinfibulation ( Bergeron et al., 2016 ; Connor, Abdi, et al., 2025 ; Hartmann, 2025 ; King et al., 2014 ; Morin et al., 2017 ; Santangelo et al., 2023 ). Participants were asked whether they saw people (e.g., health care provider), used medications (e.g., painkillers), or used home remedies (e.g., Vaseline, coconut oil) to reduce dyspareunia symptoms. Participants were asked if they had any pain in their genitals other than when having sex in the past year for at least 3 months. Response options were yes or no. The McGill Pain Questionnaire short form (MPQ) is an assessment of pain intensity, and its reliability and validity have been extensively documented ( Melzack, 1987 ). It was translated into Somali by Perovic and colleagues ( Perovic et al., 2021 ) and used to assess sexual pain in women who experienced FGC. In pilot testing, our participants reported that this measure was too long and repetitive. We reduced the scale to 6 items, using our previous qualitative research to identify words women commonly used to describe dyspareunia related to FGC ( Connor, Abdi, et al., 2025 ). We kept 5 of the original 15 items (i.e., throbbing, sharp, hot-burning, splitting, itching) and added “swelling” as a response item because women often used this term to describe their pain in the qualitative interviews. Respondents rated their level of pain on a 0–10 scale with 0 = none/no pain to 10 = worst possible pain. In our sample, this scale maintained good internal consistency (α = .81). All participants who began data collection completed the survey. There was a small amount of missing data; 1.65% of values were missing. Eight individuals with unknown deinfibulation status were not included in regression analyses. For the remaining participants, analyses were carried out using multiple imputation by chained equations (MICE) using SPSS ( Rubin, 2004 ; van Buuren, 2018 ). SPSS created five completed data sets with values imputed using information from all study variables, in addition to income, to impute values for each variable. Analyses were done on each of the five datasets, and then pooled ( Li et al., 1991 ). We report the pooled analyses. To accomplish the first study goal, descriptive statistics were used to determine frequency of demographic and FGC history variables, dyspareunia and other types of sexual dysfunction, and interventions for dyspareunia. Correlations between study variables were also examined. To accomplish the second study goal, a logistic regression analysis examined associations between infibulation and deinfibulation history and timing with dyspareunia presence (yes/no). This allowed us to identify if group membership (i.e., infibulation and deinfibulation history and timing) changes the odds of experiencing dyspareunia. Further, a linear regression analysis examined associations between infibulation and deinfibulation history and timing with sexual functioning. In the latter analysis, sexual functioning was examined through the CSFQ, computing a composite score of sexual function across domains. As described above, information about type of infibulation, deinfibulation history, and timing of deinfibulation (if applicable) was combined to create a categorical infibulation/deinfibulation history variable with four levels: never experienced infibulation, never deinfibulated, deinfibulated before sex, and deinfibulated after sex. In regression analyses, “never experienced infibulation” was examined as the comparison group. Both regression analyses adjusted for age, education, ever used hormonal birth control, history of vaginal birth, American acculturation (American alignment), marital satisfaction, and emotional distress. In addition, linear regression of sexual functioning on study variables included sexual pain presence as a covariate. To verify that the model’s assumptions were reasonably met, Hosmer-Lemeshow goodness-of-fit tests were run on the models. The test detected no problems with modeling assumptions.

Most participants underwent FGC outside of a medical setting and by a traditional practitioner. Notably, the women in our sample, on average, reported high marital satisfaction and low emotional distress (see covariates section of Table 2 ). About 66% of the sample reported infibulation; of these, about 24% were never deinfibulated. Deinfibulation most commonly occurred during labor and delivery. Because so few participants reported they had been deinfibulated after having sex but before labor and delivery, for analytic purposes, this group was added to the deinfibulated during labor and delivery and hereafter referred to as the deinfibulated after sex group. The eight individuals with unknown deinfibulation status were not included in the regression analyses. To determine if a woman was reinfibulated (a descriptive variable not included in subsequent analyses), we asked women who were deinfibulated if they asked to be sewn open, left to heal naturally (not sewn open or shut), or sewn closed (to original size or partially), followed by a question about whether or not their request was honored. For example, if a woman said she asked to be sewn closed and her request was honored, she was coded as sewn closed. However, this contributed to missing data, such that if a participant answered “I wanted to ask but I did not” ( n = 6), “I did not ask” ( n = 28), or that their preference was not honored ( n = 9); we unfortunately did not ask a follow-up question to learn if they were actually reinfibulated. Additionally, another 19 participants did not answer the first part of the question. Thus, 41% of participants had missing data for this variable. Table 3 presents the percentage of participants who reported sexual problems in the previous year on the Natsal survey. Of the six sexual concerns on the Natsal, current dyspareunia was the most common issue; 22% of the respondents reported ongoing pain with sexual intercourse, closely followed by difficulties with not enjoying sex (18%), no interest in sex (17%), difficulty having orgasms (15%), sex anxiety (12%), and no arousal (9%). Among those who reported ongoing pain, the average pain intensity was 3.18 (out of 10; SD = 2.32) (not shown in table). Many of the women sought some form of relief from their pain although only about a third sought out a medical provider ( Table 4 ). Of the 65 participants who reported dyspareunia, the majority did not seek medical attention (e.g., going to a health care provider; Table 4 ) but rather used pharmaceutical or other pain reduction strategies (e.g., painkillers, Vaseline, soaking in bath, etc.). Six percent of women reported they had genital pain that was not related to having sex (not shown in table). Table 5 includes a correlation matrix of dichotomous and continuous variables used in regression models. As no correlation was higher than 0.40 in magnitude, multicollinearity was not a concern across variables. Additionally, we descriptively explored the percentage of participants who had pain in each infibulation/deinfibulation group. Among those who were never infibulated, 14.7% reported dyspareunia (hypothesized to have the lowest pain rate and therefore the reference group). The group that was deinfibulated after sex had a similar rate of dyspareunia (14.6%), whereas rates of dyspareunia were higher for those never deinfibulated (33%) and highest for those deinfibulated before sex (45.0%). Given descriptive differences in rates of reporting dyspareunia, we tested differences in the odds of reporting dyspareunia for each group through logistic regression. This allowed us to examine if deinfibulation changed the odds of dyspareunia compared to those never infibulated. The adjusted R squared equaled 0.18. The Hosmer-Lemeshow test demonstrated adequate fit (χ 2 = 6.89, p = .55). Table 6 displays the odds ratios of dyspareunia by infibulation and deinfibulation history group while adjusting for covariates. The reference group was the group hypothesized to have the lowest probability of dyspareunia, i.e., those who were never infibulated. The comparison groups included one of three groups: never deinfibulated, deinfibulated before sex, and deinfibulated after sex (including during labor and delivery). Based on these findings, those who were never deinfibulated and those deinfibulated before sex had a higher probability of dyspareunia than those who were cut but never infibulated ( OR = 2.89 and 4.48). There was no statistically significant difference between those who were deinfibulated after sex and those who were never infibulated. Among the covariates, only emotional distress was associated with dyspareunia, such that higher emotional distress was associated with higher odds of dyspareunia. Table 7 shows results from the linear regression of sexual function, assessed by the modified CSFQ, on infibulation/deinfibulation history and other study variables. The overall adjusted R squared equaled .10 and the overall model was significant ( F = 3.88, p < .001). Infibulation/deinfibulation history items were dummy coded into 0 or 1. Infibulation/deinfibulation history was not associated with sexual function, indicating one’s infibulation or deinfibulation history was not associated with level of sexual function. Notably, sexual function was related to two covariates. Those who had higher marital satisfaction reported better sexual function. Those who reported dyspareunia reported worse sexual function. These associations were independent of one another.

There is a stereotypical portrait of women who have experienced FGC, namely that they experience dyspareunia and a lack of sexual enjoyment and orgasm. This stereotype is not fully supported by research. In our sample, it was clear that Somali women who experienced FGC had considerable diversity in their sexual experiences. Many women were negatively impacted by FGC but did not seek medical assistance for these negative sexual outcomes, only over-the-counter painkillers, lubricants, or other non-medical aides. Our findings suggest that psychosocial variables played an important role in understanding women’s sexual experiences – even in the context of a history of infibulation and deinfibulation. It is important that health care providers understand and assess biopsychosocial factors and refer women to care that is individualized to their personal experiences.

Our first goal in this study was to document (1) frequency of dyspareunia, (2) frequency of other types of sexual dysfunction, and (3) interventions for dyspareunia in a sample of 300 Somali women living in the United States who had experienced FGC. Of the six sexual concerns assessed, current sexual pain was the most common issue; 22% of our respondents reported ongoing dyspareunia. A majority of the women sought some form of relief from their pain via pharmaceutical or other pain reduction strategies; only about a third sought out a medical provider. Previous studies of the impact of FGC on sexual functioning have concluded that FGC does impact sexual functioning, although the nature and extent of this impact have varied across studies ( Nzinga et al., 2021 ; Pérez-López et al., 2020 ). Heterogeneity of results across studies may in part be due to different amounts of cutting by traditional practitioners, differences between clinical and community samples, and the intersection of cultural norms and sexuality in both the country of origin and the host country ( Johansen, 2002 ; Johnsdotter, 2018 ). Even amongst women who have been infibulated, there may be differences in how their anatomy was modified. For example, surgeons have reported that as many as 40–50% of infibulated women have an intact clitoris under the infibulation scar ( Krause et al., 2011 ; Nour et al., 2006 ). Thus, similar types of FGC may have a more significant impact on sexual functioning for some women and less for others. Based on our previous research with similar, but smaller samples ( Connor et al., 2016 ), we originally believed the rate of dyspareunia for Somali women who experienced FGC could be as high as 40%. However, based on our current findings, as well as a recent chart review study of 506 patients who experienced FGC where 25% reported dyspareunia ( Bazzoun et al., 2021 ), we estimate that dyspareunia among Somali migrants who have experienced FGC is likely around 20–25%—still a considerable percentage. Dyspareunia has been linked to FGC in previous meta-analyses examining differences between women with and without FGC in groups of women from similar backgrounds ( Nzinga et al., 2021 ; Pérez-López et al., 2020 ). Our results support these meta-analytic findings. For example, Mitchell et al. (2013) found that 5–10% of their sample of UK women with no known FGC reported dyspareunia using the Natsal-SF; in contrast, our sample of Somali women with FGC reported 2–4 times that pain frequency. Only 32% of women in our sample who experienced pain sought medical care; they instead used over-the-counter painkillers, lubricants, or other non-medical methods. These results indicate that Somali women with FGC and their medical providers need more information about effective interventions for dyspareunia. The number of participants who sought medical care was too low for us to analyze health-seeking correlates, though this is another area worth exploration. We cannot say with certainty why most participants with dyspareunia did not seek medical care. It is possible that Somali women who have experienced FGC expect to experience pain as a result of FGC. Therefore, the meaning attached to pain may be different from those who have unexplained dyspareunia ( Jacobson et al., 2018 ). Other researchers have described a stoicism in Somali women when it comes to reporting and discussing pain ( Finnström & Söderhamn, 2006 ; Jacobson et al., 2018 ; Johansen, 2002 ). It is possible that a combination of cultural norms and expectations lead to an underreporting of pain and/or an inclination toward acceptance of pain that reduces health-seeking behavior among Somali women who have experienced FGC. Although current dyspareunia was the most common sexual problem in our sample, other important sexual functioning difficulties were not far behind: not enjoying sex (18%), no interest in sex (17%), not having orgasms (15%), sex anxiety (12%), and no arousal (9%). There is some evidence that our sample may not be much different in sexual interest, sexual arousal, and orgasm than other samples of women who have not experienced FGC ( Leiblum et al., 2006 ; Mitchell et al., 2013 ). Contrary to how women who have experienced FGC are often characterized ( Johnson-Agbakwu & Manin, 2021 ), Somali women in our sample and others continue to have sexual interest at similar or higher rates than other samples of women with other ethnicities, reported a desire to be sexually active, and described sex as important to them and their partners ( Catania et al., 2007 , Connor, Abdi, et al., 2025 ; Jacobson et al., 2018 ). In our sample, the percentage of women who reported no sexual interest using the Natsal-SF was lower than the British general population sample ( Mitchell et al., 2013 ). However, with regard to not enjoying sex and feeling anxiety about sex, our survey sample reported a higher percentage with these problems than in the Natsal sample ( Mitchell et al., 2013 ). The differences between the samples were negligible with respect to arousal problems and orgasm problems. Because we did not test differences between studies statistically, and data collection years and methods were not the same, it is important not to overstate the differences between the samples. However, one can surmise that there are differences in how problems with sexual interest compare to other sexual functioning domains among Somali women who have experienced FGC and other women who have not experienced FGC. In the British sample, sexual interest was the most prevalent problem, which suggests that even when women are aroused, report no pain, and enjoy sex, there is a group of women who continue to have concerns about sexual interest ( Mitchell et al., 2013 ). In our sample, sexual interest persisted for some women even when there was dyspareunia, little enjoyment, and anxiety. We highlight the complexity of sexual functioning in women with FGC, which is relevant for clinicians who are working cross-culturally and assessing sexual health. Deinfibulation is a recommended intervention for women who have been infibulated to improve sexual health, as well as to prevent obstetric and urinary complications ( World Health Organization, 2016 ). Pre- and post-operative studies of deinfibulation suggest that deinfibulation reduces dyspareunia and improves sexual satisfaction while having minimal complications ( Berg et al., 2018 ; Krause et al., 2011 ; Nour et al., 2006 ). Our second goal was to examine rates of infibulation and deinfibulation and test associations of these procedures with dyspareunia and function. Our hypotheses were that (1) participants who experienced FGC types 1, 2, and 4 (i.e., not infibulated) would report a lower rate of dyspareunia and better sexual function as compared to those infibulated (FGC Type 3) and (2) participants who were deinfibulated before ever having sex would have a lower probability of dyspareunia and higher sexual function than those never deinfibulated or deinfibulated after they had sex for the first time. Our hypotheses on deinfibulation were partially supported. As expected, in women with Type 3 FGC (infibulation), never being deinfibulated increased the odds for experiencing dyspareunia. The likelihood of increased pain with infibulation and no subsequent deinfibulation was also suggested by previous studies that found increased pain in women who were infibulated (Type 3) compared to non-infibulated forms of FGC (Types 1, 2, and 4; Bazzoun et al., 2021 ; Connor et al., 2016 ; Rouzi et al., 2017 ). When is the best time for deinfibulation to improve dyspareunia outcomes? We hypothesized that deinfibulation before ever having sex would improve women’s pain outcomes since they would be less likely to ever experience painful sex. Contrary to our hypothesis, being deinfibulated prior to ever having sex increased the odds for experiencing sexual pain, while being deinfibulated after sex did not increase that risk. In fact, there was no difference in dyspareunia risk for women deinfibulated after sex (most of whom were deinfibulated during labor and delivery) and those never infibulated. Possible explanations are that deinfibulation prior to sex could be a risk factor for later dyspareunia due to increased scar tissue and/or the possibility of more extensive infibulation necessitating earlier deinfibulation ( Taraldsen et al., 2022 ). It is also possible that some women were deinfibulated outside of a clinic setting before attempting sex. We do not know if women in the community samples waited an adequate amount of time before having sex after deinfibulation. It is possible that those deinfibulated prior to childbirth were more willing to admit a concern and that the same group would be more likely to admit to experiencing pain. When examining deinfibulation history in relation to dyspareunia and sexual function, future studies would benefit from learning more about the reasons and outcomes of the timing of deinfibulation when assessing the impact of this intervention. In addition, future studies should investigate in more depth the circumstances surrounding a woman’s deinfibulation history, and if possible, verify self-report with medical charts. Also, contrary to our hypothesis, deinfibulation was not associated with overall sexual functioning as assessed by a composite score. There are few studies that systematically assessed the impact of deinfibulation on different sexual function domains as we did. We speculate that different varieties of FGC within types vary in the amount of clitoral tissue removed or damaged and the presence of more undamaged clitoral tissue in all types of FGC assists in maintaining sexual desire, arousal, and orgasm, regardless of infibulation and deinfibulation history. Cultural and psychosocial factors may also impact sexual function. This is something future studies will need to explore. Generally, participants in our sample were not emotionally distressed and were happy in their marital relationships. Despite these variables lacking much statistical variance, emotional distress was associated with dyspareunia and lower marital satisfaction was associated with worse sexual function. These findings point to the importance of psychosocial factors in participants’ sexual lives. In our previous work, Somali women discussed their belief that having a supportive and compassionate spouse at the beginning of their sexual relationship protected them from developing chronic pain with sexual intercourse and led to increased happiness with their sexual relationship ( Connor, Abdi, et al., 2025 ). Supportive spouses helped seek out necessary medical care, provided pharmaceutical interventions (e.g., lubrication or painkillers), and stopped sexual activity that was too painful. We believe these behaviors can lead to reduced risk of dyspareunia, better sexual functioning, and increased marital satisfaction. However, longitudinal research is needed to understand these associations. Major strengths of our study were that we recruited a large sample of 300 Somali women with FGC. We used a CBPR approach to reach participants across our large metropolitan area with the largest Somali population in the country. Throughout the design and implementation of the study, we incorporated community researchers and CAB member feedback to develop a protocol that was centered around community needs. This CBPR method was responsible for the excellent cooperation we obtained from the community and the fact we met our target recruitment goals without much trouble. We spent significant time and effort to use a cross-cultural equivalency process to develop the content equivalence of our study measures, and used a data collection method (ACASI) specifically developed for sensitive topics and hard-to -reach, marginalized populations that was particularly appropriate for our topic and population. The appropriateness of these methods led to a high response rate (78% of all eligible women we approached agreed to participate), a 100% survey completion rate, and a low percentage of missing data (1.65%). While our sampling strategy relied on the use of convenience and snowball sampling methods, we supplemented those strategies with the use of census data to closely match percentages of participants from each metropolitan county in our sample with the percentages of Somali individuals in each of the seven counties of the metropolitan area. Still, our sample was not a random sample and may not be representative of the community of Somali women with FGC in Minneapolis-St. Paul. We also controlled for demographic variables in our analyses. Weaknesses included limits on the types of analyses we could do based on a lower sample size of women who experienced dyspareunia than expected and the lack of variance in some variables. Two issues emerged: (1) the percent of the population with dyspareunia was less than our pilot studies indicated, and (2) we had low variability on several important variables (e.g., infibulation/deinfibulation history), which led to smaller than expected cell sizes. To account for these concerns, we combined two groups into one (after sex/before labor and delivery). This allowed us to have power needed for the analyses presented in this manuscript. We were also unable to explore questions related to how the perineum was managed (e.g., sewn open post deinfibulation) due to a lack of confidence in participants’ ability to accurately report or recall this information. We intentionally recruited sexually active women so that we had a large enough sample on specific questions about sexual experiences, yet this also excluded women who chose not to be sexually active due to pain or other reasons. This limits our ability to generalize our findings to individuals who do not have vaginal sex. Though we took many steps to ensure privacy, it was important that the interviewer be present to help with questions, and we cannot say how the presence of an interviewer may influence responses, although previous studies using this technology showed that respondents were more likely to report higher numbers of stigmatized behaviors (e.g., abortions and miscarriages ( Caldwell & Jan, 2012 ; Dolezal et al., 2012 ; Estes et al., 2010 ; Lindberg & Scott, 2018 ). Lastly, we relied on the self-report of our female respondents to recall and report the type of FGC they had. While we interviewed respondents carefully and methodically and showed them anatomically correct drawings of the various types of circumcision created specifically by us for our population, we were unable to do a clinical exam of their vulva to verify the type and extent of their circumcision, infibulation, and deinfibulation. A biopsychosocial approach to sexuality provides a comprehensive understanding of a person’s sexual health ( Berry & Berry, 2013 ). Multi-site studies with larger sample sizes are needed to examine the role of potential biological factors (e.g., extent of cutting, amount of remaining clitoral tissue, obstetric trauma, deinfibulation history), psychological factors (e.g., depression, emotional distress, trauma), social factors (e.g., relationship satisfaction), and cultural factors (e.g., beliefs about FGC) in determining dyspareunia and sexual function. Further research on the most appropriate measures to assess sexual health domains in women with FGC is warranted. Some scholars suggest that measures of sexual function besides the FSFI should be used ( Pérez-López et al., 2020 ). For example, Catania et al. (2007) reported that almost 38% of their sample of women with infibulation living in Italy had difficulty completing the FSFI. We used both the Natsal-SF and the CSFQ ( Table 3 ). Both the Natsal-SF and the CSFQ scales were preferred over the FSFI by our community researchers; participants did not express confusion about the Natsal-SF and CSFQ questions. Inclusion of male partners in future research would generate further understanding of how men contribute to both positive and negative outcomes. Longitudinal studies can assist in determining how these factors may intersect and predict sexual health. More information about how the perineum was managed during and after a deinfibulation procedure, in addition to extent of scar tissue, may illuminate some of the unexpected findings about deinfibulation in the present study. We recommend that future studies include more questions about perineum management, and, if possible, include chart review for verification. Finally, there is a need for intervention studies that examine the impact of sexual counseling in teaching women about their remaining clitoral tissue, how to cope with dyspareunia, and how to access sexual desire and arousal. The intersection of culture and sexuality is complex. Though research has pointed to the negative impact of FGC, it has also highlighted that dyspareunia and impaired sexual functioning are not a uniform experience. It is important that medical providers put aside their assumptions about FGC and conduct a full assessment to understand how FGC impacts their patient. For example, though studies of the deinfibulation procedure demonstrate that it may reduce pain, we found that some women still experience dyspareunia after deinfibulation. Most studies have not examined other potential causes of dyspareunia, such as endometriosis; thus, the provider should not assume that FGC is the only potential cause of pain. Additionally, 85% of women in our sample experienced orgasm. This highlights that a large majority of Somali women with FGC still experience sexual desire and arousal. Many women who have experienced infibulation express a positive view toward deinfibulation, particularly when it aligns with cultural norms ( Connor et al., 2024 ; Ziyada et al., 2020 ). Yet, as noted above, there continues to be a lack of clarity about when best to deinfibulate. We do not believe research findings are conclusive enough to refrain from deinfibulation prior to pregnancy. Additionally, clinicians should explore treatment options beyond deinfibulation, such as physical therapy and sex therapy. Health care providers should weigh the patient’s life circumstances to make recommendations.

Female genital cutting (FGC), also known as female genital mutilation or female circumcision, is defined as modification of the female genitals, often including full or partial excision of the prepuce, external clitoral glans, labia majora, and/or labia minora, with no medical benefit ( World Health Organization, 2008 ). As shown in Table 1 , The World Health Organization (WHO) classified FGC into four types based on the amount and nature of the incision, and whether there was a partial covering of the vaginal introitus (infibulation). FGC commonly occurs in 28 countries throughout Africa, across the Middle East and Southeast Asia – with many migrants who have experienced FGC settling in the United States and Europe ( UNICEF, 2024 ). Due to global migration patterns, it is estimated that since 1990, the number of girls and women affected by FGC and living in the United States has risen by 224%, resulting in a large population of female migrants who are living with the aftereffects of FGC and being treated in the American health care system ( Goldberg et al., 2016 ). Minnesota has one of the largest Somali populations, many of whom have experienced FGC ( Mather & Feldman-Jacobs, 2016 ). Due to ongoing civil unrest and climate-based crises, Somali refugees continue to seek refuge in the United States at high rates and represent one of several ethnic groups who need health care that takes into account their experience with FGC ( International Organization for Migration (IOM) & United Nations Refugee Agency [IOM], n.d. ). Nearly all women in Somalia (98%) have undergone FGC ( UNICEF, 2024 ; World Health Organization, 2008 ), with the large majority experiencing the most extensive form – Type 3 FGC (infibulation). Long-term complications of FGC may include obstetric complications (such as perineal tears, episiotomy, prolonged labor), infertility, dysuria, pelvic floor distress, and sexual health problems ( Binkova et al., 2021 ; Lurie et al., 2020 ; Nzinga et al., 2021 ). The impact of female genital cutting (FGC) on sexual health is a primary concern for health care providers who treat women affected by FGC ( Berg & Denison, 2012 ; Gareau et al., 2025 ; Nzinga et al., 2021 ). Multiple studies have reported that infibulation increases one’s risk of sexual dysfunction, including pain and sexual arousal problems ( Bazzoun et al., 2021 ; Connor et al., 2016 ; Rouzi et al., 2017 ). Studies have consistently identified an increased risk of dyspareunia among women with FGC. For example, in their meta-analysis, Berg and Denison (2012) found FGC resulted in a 50% increased risk of dyspareunia. Lurie et al. (2020) completed a more recent meta-analysis and found that FGC resulted in a 2–3-fold risk of dyspareunia. Experiencing infibulation (Type 3 FGC) may increase one’s risk of dyspareunia and other forms of sexual dysfunction to a greater degree than other types because the vaginal orifice is obstructed ( Bazzoun et al., 2021 ; Connor et al., 2016 ; Rouzi et al., 2017 ). Meta-analyses of studies comparing sexual function and satisfaction for women with FGC to those without FGC have concluded that FGC can have a negative impact on sexual health ( Nzinga et al., 2021 ; Pérez-López et al., 2020 ). In each meta-analysis, desire, arousal, lubrication, orgasm, and sexual satisfaction were lower in women who were impacted by FGC than those who were not. However, these authors noted that effect sizes of findings vary across studies and hypothesized that publication bias, the multi-dimensionality of sexuality (i.e., biopsychosociocultural factors), and other unidentified factors may influence the level of impact FGC appears to have on individuals ( Bazzoun et al., 2021 ; Connor et al., 2016 ; Rouzi et al., 2017 ); this was true across each domain of sexuality that was analyzed.

An intervention for women who have been infibulated and are concerned about negative effects is planned deinfibulation ( Nour et al., 2006 ). In this medical procedure, the circumcision/infibulation scar tissue and edges are sutured open. This exposes the underlying structures, including the remaining clitoral tissue, the urethra, and the vaginal opening ( Nour et al., 2006 ). Deinfibulation is necessary for optimizing a vaginal delivery and may also be necessary for the flow of urine and menstrual blood, and vaginal penetration when the narrowed introitus is too small ( Nour et al., 2006 ). Though qualitative studies describe couples attempting to deinfibulate without medical assistance by using a razor or repeated attempts to enter with a penis ( Connor et al., 2024 ; Johansen, 2002 ), it is unknown how the sexual health outcome of medical intervention differs between deinfibulation by a medical provider versus by a husband. Unless noted otherwise, when we refer to deinfibulation in this paper, we refer to this medical procedure. Deinfibulation can decrease dyspareunia and is considered a low-risk procedure ( Bazzoun et al., 2021 ; Nour et al., 2006 ). However, there is contradictory data on whether deinfibulation itself and the timing of deinfibulation improves sexual arousal and sexual satisfaction ( Jordal et al., 2022 ). Medical professional guidance about when to deinfibulate is inconsistent and reflects varied outcomes from research ( Jordal et al., 2022 ; Warren & Tower, 2021 ). Much of this guidance is related to pregnancy and childbirth and is less focused on overall sexual health. For example, the World Health Organization (2018) recommends “either antepartum or intrapartum deinfibulation … to facilitate childbirth.” This latter recommendation is particularly important during the first pregnancy, when the infibulation is most likely to be complete. Deinfibulation prior to sex could lead to fewer negative and traumatic memories of vaginal intercourse ( Battle et al., 2017 ; Connor, Abdi, et al., 2025 ), and possibly a lower risk of perineal tears ( Paliwal et al., 2014 ). However, in practice, due to cultural and familial beliefs and/or pressure, many patients do not consent to deinfibulation prior to labor when it is offered ( Connor et al., 2024 ; Johansen, 2019 ; Paliwal et al., 2014 ; Safari, 2013 ), and they are even less likely to consent prior to marriage and first intercourse. Complications of deinfibulation during labor (e.g., increased tearing, postpartum hemorrhage, episiotomies) have been noted when compared to deinfibulation prior to labor ( Albert et al., 2015 ; Paliwal et al., 2014 ). However, these studies have lacked adequate statistical power ( Esu et al., 2017 ) and it is unknown if complications during labor lead to long-term problems in sexual pain. Berg et al. (2018) also noted a lack of clarity in most studies exploring the impact of deinfibulation due to failure to control for potential confounding variables. In one Norwegian retrospective chart review study of 624 nulliparous Somali women, deinfibulation during labor lowered the risk of obstetric anal sphincter injury as compared to deinfibulation prior to labor [OR = .48, 95% CI = .27–.86 ( Taraldsen et al., 2022 )], possibly due to scar tissue or other confounding factors not measured. This study was adequately powered and controlled for demographic variables (e.g., age, education) and birth conditions. Their outcome variables did not include sexual function. Common interventions for dyspareunia and/or genital pain due to other diagnoses, such as pelvic floor therapy for vulvodynia ( Hartmann, 2025 ), have not been studied in women who have experienced FGC. To our knowledge, no study has looked at whether Somali women experiencing pain in the FGC-context engage in traditionally based interventions (e.g., vaginal steaming), over the counter interventions (e.g., readily available pain medicine, lubrication), or other medical and behavioral interventions (e.g., numbing creams like lidocaine, pelvic floor therapy). Instead, studies have focused on surgical interventions ( Berg et al., 2018 ; Johnson-Agbakwu & Warren, 2017 ). There is a need to understand if non-surgical interventions are acceptable and utilized by this population ( Johnson-Agbakwu & Warren, 2017 ). To our knowledge, there has not been a quantitative study of dyspareunia and function in Somali women in the United States who have experienced FGC. Our first goal was to document (1) rates of dyspareunia, (2) other types of sexual dysfunction, and (3) interventions for dyspareunia in a sample of Somali women living in Minnesota who had experienced FGC. Our second goal was to examine rates of infibulation and deinfibulation and test associations with dyspareunia and sexual function. We hypothesized that participants who experienced FGC types 1, or 2, and those who were not infibulated (Type 4) would report a lower rate of dyspareunia and better sexual function in comparison to those infibulated (FGC Type 3). We also hypothesized that participants who were deinfibulated before first sexual intercourse would have a lower probability of pain and higher sexual function than those never deinfibulated or deinfibulated after having been sexually active.