Measurement properties of sexual function assessment questionnaires in women with endometriosis: A systematic review following COSMIN guidelines

Anne Oppenheimer and Arnaud Fauconnier designed the study. Anne Oppenheimer and Florence Boitrelle reviewed all articles included in the systematic review. Anne Oppenheimer and Arnaud Fauconnier wrote the manuscript. All authors reviewed and modified the manuscript, and approved the final version for submission.

Figure  1 reports the flow chart of the study selection process. A total of 1787 articles were identified using the search equation and a further two were found in the references of the selected articles. Analysis of the titles and abstracts led to the selection of 839 articles, among which 765 were excluded after reading the full text. Of these, 264 (34.5%) were excluded because they were limited to assessment of dyspareunia alone. We finally included 74 articles in the study (Figure  1 ). Chart flow of the studies selected. Table  1 summarizes the characteristics of the included studies. Among the 74 articles included, 25 different PROMs evaluating sexual function were identified (Table  1 ). The Female Sexual Function Index (FSFI) was the most frequently used PROM (34/74 [45.9%] items), followed by the Female Sexual Distress Scale (FSDS) (9/74 [12.2%] items) and the Sexual Activity Questionnaire (SAQ) (8/74 [10.8%] items). However, the FSDS was systematically associated with the use of another PROM evaluating sexual function. Characteristics of the studies included in the systematic review. All studies including women with confirmed endometriosis assessing sexual quality of life or sexual function or sexual distress were retrieved. Structure validity Construct validity Internal consistency Responsiveness and MCID Abbreviations: AD, adenomyosis; B‐PFSF, Brief Profile of Female Sexual Function; BISF‐W, Brief Index of Sexual Functioning for Women; DIE, deep invasive endometriosis; DSFI, Derogatis Sexual Function Inventory; EHP30 SR, EHP30 Sexual relationship; FSDS, Female Sexual Distress Scale; FSFI, Female Sexual Function Index; FSQ, Female Sexual Quotient; GRISS, Golombok Rust Inventory of Sexual Satisfaction; GSSI, Global Sexual Satisfaction Index; ISS, Index of Sexual Satisfaction; KFSP, German version of the Massachusetts General Hospital Sexual Functioning Questionnaire; MFSQ, McCoy Female Sexuality Questionnaire; N, Number of patients in the study; Na, Not applicable; NETA, norethisterone acetate; PISQ 12, Pelvic Organ Prolapse/Urinary Incontinence Sexual Questionnaire (short form); QCSASC, Questionnaire of Cognitive Schema Activation in Sexual Context; QCSASC, Questionnaire of Cognitive Schema Activation in Sexual Context; SAQ, Sexual Activity Questionnaire; SDBQ, Sexual Dysfunctional Belief Questionnaire; SHOW‐Q, Sexual Health Outcomes in Women Questionnaire; SIDI, Subjective Impact of Dyspareunia Inventory; SMQ, Sexual Modes Questionnaire; SQOL‐F, Sexual Quality of life‐Female; SSCS, Sexual Self‐Consciousness Scale; SSFS, short sexual functioning scale; SSRS, revised Sabbatsberg Sexual Self‐Rating Scale; SSS‐W, Sexual Satisfaction Scale for Women; SVQ, Vaginal Changes Questionnaire. Reason for consultation: mixed: infertility, pain, bleeding. Diagnosis: mixed: histology or surgery or imaging or clinically. Both: Hormone and/or surgery. The risks of bias of the measurement property studies classified by the PROMs are presented in Figure  2 . Among all the questionnaires, the SAQ, the FSDS, and the FSFI were those with the highest number of studies classified as “good” and “very good.” Risk of bias of each single study according to the COSMIN Checklist classified by patient‐reported outcome measure (PROM). Note that an article can use several measurement properties per PROM and that a “study” corresponds to one measurement property (there can, therefore, be several measurement property studies for an article). The methodological quality of each single “study” is assessed by using the COSMIN risk of bias checklist (as “very good,” “adequate,” “doubtful,” “inadequate quality”). The assessment of content validity of each PROM is presented in Table  S1 . The two PROMs with inadequate content validity were excluded from further evaluation. Table  2 summarizes the measurement properties and risk of bias of each questionnaire according to the COSMIN criteria properties. The most commonly used measurement properties in the articles were “hypothesis testing” and “responsiveness.” Seven questionnaires had high levels of evidence for these two measurement properties: the FSFI, the SAQ, the Short Sexual Functioning Scale (SSFS), the Sexual Satisfaction Scale for women (SSS‐W), Sexual Quality of Life‐Female (SQOL‐F), the Brief Profile of Female Sexual Function (B‐PFSF), and the Sexual Health Outcomes in Women Questionnaire (SHOW‐Q). These questionnaires also presented no risk of bias when used for these types of measurement (“hypothesis testing” and “responsiveness”). Only one questionnaire, the SAQ, represented almost all measurement properties in the context of endometriosis. The Subjective Impact of Dyspareunia Inventory (SIDI) appears in only one article and is the only specific questionnaire. However, the responsiveness of the SIDI is not available. Measurement properties and risk of bias of sexual function assessment questionnaires according to the COSMIN criteria properties. Endometriosis of the rectovaginal septum vs control: ES = −0.3, p  = 0.029 Endometriosis vs control: Significant difference for total score and pain domain (ES = 0.6) and according to the studies: desire, arousal, lubrication, satisfaction, and orgasm domains 73 : significant difference in pain domain 81 : endometriosis vs. control significant difference in FSFI and all domains 85 : significant difference in sexual arousal domain, satisfaction and sexual pain domain Between two progestins (Dienogest and NETA): No significant difference Medical treatment (Dienogest) vs control: Significant difference ( p  = 0.001) Continuous vs discontinuous hormone treatment: Difference in favor of continuous, ES = 3.57, p  = 0.001 Surgical vs medical treatment: Surgery significantly more deleterious ES = −0.47 At 3 months after surgery 75 significant difference ES = 0.82 p  ≤ 0.001 At 12 months Progression not always significant according to the studies 76 NS 77 Significant improvement total score, and desire score and pain score 87 Significant improvement of CO 2 laser ablation of vaginal endometriosis nodule ES = 1.13 Deep endometriosis and nerve‐sparing surgical technique: Significant improvement of the total score and of the domains: desire, satisfaction, pain, arousal and orgasm no improvement in the lubrication domain At 6 months or 12 months Progestin treatment or pregnancy: total score improvement: ES ranging from 0.4 to 5.4 Improvement in domains: Desire (ES = 0.82 to 1.23), satisfaction (ES = 0.46 to 1), arousal (ES = 0.45 to 1.36), lubrication (ES = 0.83), orgasm (ES = 0.71 to 1.64), dyspareunia (ES = 0.9 to 2.46) GnRH antagonist vs. placebo: Not significant Dihydrogestérone 80 Desire, satisfaction, orgasm scores increased significantly FSFI score increased significantly: ES = 0.84 Correlation with quality of life scales Surgery with immediate or delayed resection of endometriosis lesions High doses progestins Good correlation with quality of life score: EHP5 and EQ5D Not significant when comparing immediate or delayed management SAQ score significant lower when High dose progestins were used ES = −0.44 Surgical or medical treatment High dose progestins Total score ES = 0.3, pleasure domain (ES = 0.9 to 1.2), habit domain (ES = 0.4 [NS] to 1.1, [S]), discomfort domain (ES = −0.79 to −1.06 [S]) No amelioration between T0 and T1 when high dose progestins were used Endometriosis vs. control AD vs. DIE vs. control Endometriosis Endometriosis vs control Dyspareunia not associated with more sexual distress in endometriosis. Significant difference in AD and DIE vs. control ES (AD vs. control) = −1.35 No difference AD vs. DIE without AD Multivariate logistic regression model: sexual distress associated with greater avoidance of sex and negative impact of endometriosis on sex lives Significantly sexual distress (FSDS >15) when DIE compared with control group Continuous vs. sequential treatment Significant improvement in score with continuous treatment ES = −2.4 Dienogest vs. control: ES = ‐ 4.9, p  = 0.001 Deep endometriosis: Significant improvement post op Peritoneal + vaginal endometriosis: No change Hypothesis 23 Comparison of US vs. control (without endometriosis) involvement: significant in all of the following domains: Relaxation after sex (ES = 0.67), satisfying orgasm (ES = 1.10), good communication with partner about sex (ES = 0.82), pain intensity (ES = −0.79), decreased orgasm intensity due to pain (ES = −1), frequency of sex in the last 3 months (ES = 0.57), sexual intercourse interruption due to pain (ES = ‐ 0.58) Comparison of endometriosis surgery with US involvement vs no US involvement: In the absence of US involvement, no benefit of surgery for sexual variety, communication with the partner, and interruption of sexual intercourse for pain Endometriosis with US involvement: Significant for the following items: no satisfaction with sexual frequency (ES = −1.2), relaxation after sex (ES = 1.4 to 1.6), not enough sexual variety (ES = −0.6), not enough sexual duration (ES = −0.8 to −1.2), orgasm (ES = 1.7 to 2.1), communication with partner (ES = 0.45), sexual frequency (ES = 1.3 to 1.4), sexual pain (ES = −1.1 to −1.7), relaxation during sex (ES = −0.8 to 0.9); Not significant for: satisfaction with partner and libido Not significant Desire (D1), Excitement (D2), Frequency of activity (D3), Receptivity (D4), Orgasm (D5), Relationship satisfaction (D6), Medical problems affecting sexuality (D7) Significant with ES = 0.57 to 0.72 for arousal (D2), frequency of activity (D3), orgasm (D5) Not significant: desire (D1), receptivity (D4), relationship satisfaction (D6), medical problems affecting sexuality (D7) Composite score ES 1.73 Hormonal treatment 82 Surgery (digestive segmental resection vs. shaving rectal) Medical Treatment (Elagolix 200 mg) Medical Treatment (Elagolix 200 mg) ES = −0.71, p  = 0.0001 Significant worse score for segmental resection vs. shaving rectal Dyspareunia responders had significant improvement in EHP30 sexual intercourse Significant improvement in EHP30 Comparison of chronic pain (with or without endometriosis) vs. control: Significant for domains relationship frequency, sexual satisfaction, sexual aversion, lack of expression of sensuality, vaginismus Not significant: sexual communication, anorgasmia Total GRISS Comparison of two treatments: Non‐significant difference Responsiveness 43 At 12 months: Significant for improved sexual satisfaction (ES = 0.2 to 0.56) Not significant for sexual problems and satisfaction with partner Not significant between two surgical techniques ES = −0.1 Significant for Sexual interest, sexual activity, sexual satisfaction, sexual pleasure, and ability to reach orgasm. Not significant for Importance of sex Surgical treatment vs healthy women Quality of life scale comparison Segmental surgery vs nodular surgery Operated patients are comparable with healthy women for satisfaction, desire and interference with pelvic problems but not for orgasm Correlation 0.32 with SF36 quality of life scale Not significant Significant for pain and orgasm problems Not significant for desire and arousal Significant difference ES = 0.38 Significant improvement after surgery ES = 0.7 Other (Sex education program) DIE+ adenomyosis vs. DIE AD vs. DIE vs. control DIE vs. control Significant difference Significant higher score for group with no adenomyosis ES = −0.43 Significant difference in AD and DIE vs. control ES (AD vs. control) = 1.89 No difference AD vs. DIE without AD Significant difference Other (Sex education program) Hormonal treatment DIE+ adenomyosis vs. DIE Surgery treatment and DIE Significant difference EF = 0.73 Significant difference: Slightly greater when adenomyosis ES = 2.16 Significant improvement after surgery and DIR DIE+ adenomyosis vs. DIE AD vs. DIE vs. control DIE vs. control Significant difference: Higher with no adenomyosis score ES = −0.33 Significant difference in AD and DIE vs. control ES (AD vs. control) = 0.66 No difference AD vs. DIE without AD Significant lower score when DIE vs. control Hormonal treatment: DIE+ adenomyosis vs. DIE Surgery treatment and DIE Significant difference Slightly greater when adenomyosis No difference when no DIE no adenomyosis ES = 0.73 Significant improvement in B‐PFSF score Sexual dysfunction significantly associated with greater Sexual and relationship concerns, poorer partner support and higher Endurance than participants without sexual dysfunction Internal consistency 72 Cronbach's α  = Factor 1: 0.88 Factor 2: 0.86 Factor 3: 0.75 Factor 4: 0.87 Note : (1) The column “Criteria for good measurement properties (pooled data) (Terwee)” corresponds to the Table 4 ( updated criteria for good measurement properties ) of the COSMIN manual for systematic reviews of PROMs ( https://cosmin.nl/wp‐content/uploads/COSMIN‐syst‐review‐for‐PROMs‐manual_version‐1_feb‐2018.pdf ). (2) The column “Risk of bias (Pooled data) (According to COSMIN Checklist)” corresponds to the Table 7 ( Instructions on downgrading for risk of bias ) of the COSMIN manual for systematic reviews of PROMs ( https://cosmin.nl/wp‐content/uploads/COSMIN‐syst‐review‐for‐PROMs‐manual_version‐1_feb‐2018.pdf ). (3) The column “Level of evidence (according to COSMIN grades)” corresponds to the Table 6 ( modified grade approach for grading the quality of evidence ) of the COSMIN manual for systematic reviews of PROMs ( https://cosmin.nl/wp‐content/uploads/COSMIN‐syst‐review‐for‐PROMs‐manual_version‐1_feb‐2018.pdf ). Abbreviations: AD, adenomyosis; DIE, deep invasive endometriosis; ES, Effect size d Cohen; HADS, Hospital Anxiety and Depression Scale; MCID, Minimal Clinically Important Difference; NETA, norethisterone acetate; NS, non‐significant; RSES, Rosenberg Self‐Esteem Scale; S, significant; SR, sexual relations; US, Utero‐sacral ligaments; vs, vs; Definition responsiveness, pre‐treatment vs post‐treatment (pre‐operative vs. post‐operative). The FSFI had a high level of evidence for discriminating between known groups of patients suffering from various forms of endometriosis and “control” groups without endometriosis (eg endometriosis of the rectovaginal septum vs control group [without endometriosis]: ES = −0.3, p  = 0.029) (Table  2 ). The FSFI also appears to be effective for evaluating medical treatments for the two measurement properties mentioned above. However, in the context of surgical treatments, the “responsiveness” of the FSFI was insufficient (no significant change was noted with FSFI). On the other hand, the SAQ was more relevant for evaluating surgical treatments (good responsiveness) for the total score (ES = 0.3), pleasure (ES = 0.9 to ES = 1.2), and discomfort (ES = −0.79 to ES = −1.06) subdomains. The SSFS and the SHOW‐Q both provide a good evaluation of surgical treatment but not in all domains. However, the SSFS has insufficiently rated measurement properties when referring to the criteria for good measurement properties. The FSDS questionnaire is often used in association with the FSFI. Similar to the FSFI, the FSDS appears to be effective in the evaluation of medical treatments but only for the “responsiveness” measurement property (ES = −2.25 to ES = −5.4, p  = 0.01, for the comparison of continuous vs sequential administration of Dienogest). The other questionnaires are reported in Table  2 with their level of evidence. We also provide the evaluation of each article in Table  S2 .

This systematic review identified 25 PROMs used to assess sexual function in patients with endometriosis. Our analysis will help future research investigators as well as clinicians to choose the most appropriate PROM to evaluate sexual function. Most PROMs lack complete evaluation of measurement properties according to the COSMIN criteria. The two main properties evaluated were “hypothesis testing” and “responsiveness.” This is mainly due to the fact that most studies were not designed to study the psychometric properties of the questionnaire used, but rather to assess sexual function between different groups (defined by the characteristics of the disease), or the responsiveness. The FSFI and the SAQ questionnaires emerged as having the best measurement properties according to the COSMIN criteria. The FSFI questionnaire appeared to be more relevant for evaluating medical treatment, and the SAQ for surgical treatment. The SIDI is the only specific questionnaire. From our initial selection of articles assessing sexual outcomes, 34.5% were excluded because only dyspareunia was assessed. This underlines the difficulty of assessing sexual function in its entirety. Dyspareunia, although an important parameter for evaluating pain in patients with endometriosis, does not fit into the definition of a PROM evaluating sexual function. Indeed, a PROM measuring sexual function must take sexual function into account as a whole and should not be limited to the isolated evaluation of pain during penetration. For example, drug‐induced pain relief might have a positive impact on sexual functioning but a negative impact on libido and desire. 93 Furthermore, changes in dyspareunia are not always correlated with sexual function. 36 This highlights the importance of our work assessing the quality and ability of a PROM to capture the multidimensional aspect in this context. Sexual function may be altered by many factors—endometriosis lesions with pelvic adhesions, fibrosis of the contact organs, and chronic inflammation (related to monthly bleeding, release of prostaglandins and inflammatory mediators, and chemical peritonitis)—resulting in dyspareunia such as genito‐pelvic pain in vaginal penetration. 94 Furthermore, surgery with uterosacral ligament and segmental bowel resection as well as extensive dissection of the pararectal spaces may damage the inferior hypogastric nerve plexus, causing impairment in vaginal lubrication. 95 , 96 We found that the FSFI was the most used questionnaire in endometriosis. There are two main reasons for its success: it is easy to use (19 items, response time of about 15–20 minutes) and the resulting score makes it possible to categorize women into those who do or do not suffer from sexual dysfunction (sexual dysfunction: yes/no) based on a defined cut‐off of 26.5. 97 However, the FSFI was originally developed for patients who have a partner and are sexually active. Hence, a low FSFI score may be related either to the absence of a sexual partner or to the absence of sexual intercourse due to the dyspareunia associated with endometriosis. Even if this questionnaire is easy to use, it takes a relatively long time to answer, and is multidimensional, making it more complicated to interpret. The SAQ, although less used, has been validated in an endometriosis population with a psychometric validation study (with complete validation properties 20 ) and also has the advantage of having a well‐defined minimum clinically important difference, which may be very useful for research. The minimum clinically important difference is described as the smallest difference in score in the domain of interest that patients perceive as beneficial and that would mandate, in the absence of troublesome side effects and excessive cost, a change in patient management. 98 Compared with the FSFI, the SAQ is simpler to use (10 items, response time of about 5–10 minutes), and sexual inactivity is captured by questions asked in Part 2 of the questionnaire (unlike the FSFI). Hence, the score from this questionnaire is only used for patients who are sexually active. One interesting property is that it is unidimensional, 20 which facilitates the calculation and use of the score. Unidimensionality is an interesting property for clinician use in everyday practice to monitor the effect of the treatment of endometriosis on sexual function. 88 A third questionnaire, the SHOW‐Q distinguished itself by good measurement properties for “hypothesis” and “responsiveness,” although it is less frequently used. Various PROMs—such as the SHOW‐Q, FSFI, Derogatis Sexual Function Inventory, Golombok Rust Inventory of Sexual Satisfaction, SSFS, 3 , 4 , 41 , 99 or SAQ 20 —have been used in recent years to assess sexual dysfunction in the general population (other than the endometriosis population). Measurement properties vary according to the context and to the disease. 100 All the aforementioned PROMs have been validated in the general population but only the SAQ has specifically been validated in endometriosis. 20 Some of these questionnaires have shortcomings for assessing sexual function in women with endometriosis: some questions may address topics that are not relevant to these women while neglecting other important aspects of the disease. Interestingly, the SIDI is the only PROM that has been developed and validated specifically for women with endometriosis. This questionnaire assesses the subjective impact of dyspareunia. The validation study has been recently published (June 21, 2022) and did not provide information on the responsiveness of this questionnaire. This information should soon be available if the SIDI is used to evaluate therapeutic strategies. Our study is the first systematic review analyzing PROMs of sexual function in the field of endometriosis. The use of the COSMIN checklist to evaluate their measurement properties allowed us to rigorously evaluate the qualities and shortcomings of the different PROMs reported in the literature through a precise analysis of their measurement properties. Some limitations deserve to be mentioned. Articles of interest were identified using the PubMed database alone. Nevertheless, we also searched for articles in the reference lists of the included articles and only identified two additional articles. Furthermore, one should note that PubMed lists the vast majority of accepted and published articles in the field of endometriosis. Another limitation, as previously explained, is that most of the studies were not designed to study the psychometric properties of the questionnaire but rather to assess sexual function between different groups. Future clinical research should focus on developing a well‐designed questionnaire for use in women with endometriosis. This questionnaire should be specific to endometriosis, similar to the SIDI, and should possess good measurement properties for evaluating both medical treatment, like the FSFI, and surgical treatment, like the SAQ.

In this systematic literature review of sexual function assessment questionnaires in endometriosis, the FSFI and the SAQ questionnaires emerged as having the best measurement properties according to the COSMIN criteria. The FSFI questionnaire would appear to be more relevant for evaluating medical treatment, and the SAQ for surgical treatment. The SIDI is the only specific questionnaire, but its responsiveness remains to be defined. Our study will help the development of future tools by providing an overview of all the measurement properties investigated by PROMs assessing sexual function in a population of patients with endometriosis.

Endometriosis is a chronic benign gynecological pathology characterized by the presence of endometrial cells outside the uterus that impacts the reproductive and general health of women during the course of their life. 1 Endometriosis can cause dyspareunia that may affect sexual function. 2 , 3 , 4 However, pain during intercourse is not the only determinant of sexual health in women: sexuality is a complex and multifactorial phenomenon involving emotional and physiological processes. Endometriosis lesions and their treatment may negatively affect different domains of sexual function. 5 , 6 , 7 Due to the subjective and intimate nature of sexual function, the most appropriate instruments for assessing sexual dysfunction are patient‐reported outcome measures (PROMs), which assess several areas of sexuality with high reliability and validity. 8 PROMs are standardized and validated questionnaires used as support in daily clinical practice to ascertain perception of patients' health status, perceived symptoms, level of impairment, disability, and health‐related quality of life. 9 Evidence has shown that the systematic use of PROMs in daily clinical practice contributes to better communication and decision‐making between clinicians and patients. 10 , 11 , 12 The use of PROMs to assess sexual function in endometriosis provides an opportunity for clinicians to discuss women's expectations of a given treatment and potentially identify incompatibilities. For this reason, it is important to define the most appropriate PROMs for sexual function in endometriosis patients. The COSMIN (COnsensus‐based Standards for the selection of health Measurement Instruments) initiative provides tools to improve the selection of PROMs for research and clinical practice by assessing the risk of bias and the level of evidence for each specific measurement property. 13 We conducted a systematic literature review of the existing PROMs used to measure sexual function of patients with endometriosis. The objective was to identify the different PROMs used to assess sexual function in women with endometriosis and to evaluate their psychometric properties using the checklist developed by the COSMIN initiative. 13

The authors declare no conflicts of interest.

This systematic review was conducted according to the PRISMA (Preferred Reporting Items for Guidelines on Systematic Reporting and Meta‐analysis) guidelines. This literature review was registered on Prospero as 2018 CRD42018102278. We followed COSMIN recommendations for the methodology of the systematic review and for the presentation of PROM assessment qualities. We conducted a Medline search (via PubMed) covering the last 23 years (January 2000 to September 2023). We did not use a search filter. The search strategy was based on two search equations, one dedicated to studies indexed with MeSH terms (Medical Subject Headings), and the other using exclusively free text (such as endometrioma, sexual disorder, sexual distress, quality of sexual life). The MeSH terms used were: quality of life, sexual behavior, sexual dysfunction, dyspareunia, questionnaires, endometriosis, and humans. The search equations were: (((((((((("quality of life"[MeSH Terms]) OR sexual behavior[MeSH Terms]) OR "sexual dysfunction"[MeSH Terms]) OR dyspareunia[MeSH Terms]) OR questionnaires[MeSH Terms]) AND ("2000/01/01"[PDat]: "3000/12/31"[PDat]))) AND ((Clinical Study[ptyp] OR Clinical Trial[ptyp] OR Clinical Trial, Phase I[ptyp] OR Clinical Trial, Phase II[ptyp] OR Clinical Trial, Phase III[ptyp] OR Clinical Trial, Phase IV[ptyp] OR Comparative Study[ptyp] OR Observational Study[ptyp] OR Randomized Controlled Trial[ptyp] OR Review[ptyp] OR systematic[sb] OR Meta‐Analysis[ptyp] OR Journal Article[ptyp] OR Controlled Clinical Trial[ptyp]) AND ("2000/01/01"[PDat]: "3000/12/31"[PDat]) AND Humans[Mesh] AND (English[lang] OR French[lang])))) AND (endometriosis[MeSH Terms] AND ("2000/01/01"[PDat]: "3000/12/31"[PDat]))). An additional manual search was performed using the reference lists of the included articles. Eligible studies were clinical trials or observational studies published in English and/or French, including women over 18 years of age with confirmed endometriosis (diagnosis of endometriosis by histology, surgery or a diagnosis of endometriosis based on expert ultrasound or pelvic magnetic resonance imaging) and assessing sexual quality of life or sexual function or sexual distress by means of a specific PROM or a PROM including sexual function as one of its dimensions. Studies were excluded based on the following criteria: fewer than 10 patients, case reports, studies without PROMs of sexual function per se or without a calculated sexual subscore, qualitative studies without exploitable scores, and studies measuring only dyspareunia, but no other dimension related to sexual function. Two reviewers—the first (AO) and second (FB) authors of the manuscript—independently examined the titles and abstracts of the articles identified in the literature to determine whether the eligibility criteria were met. The final selection of articles was made only after a full‐text review had confirmed that the inclusion criteria were met. The reference lists of full‐text articles were systematically reviewed to identify additional studies for inclusion. Any disagreement between the two reviewers (AO and FB) about the inclusion of an article was resolved by consensus. A third reviewer—the last author of the manuscript (AF)—was also involved if the two main reviewers failed to reach consensus. A standardized extraction form was established by using the COSMIN guidelines. 14 , 15 The COSMIN checklist 13 is a dedicated tool that combines the assessment of the methodological quality of articles on measurement properties with the quality of the PROM itself. 14 In each article using a PROM measuring sexual dysfunction, several properties of each PROM can be assessed simultaneously (structural validity, internal consistency, cross‐cultural validity, reliability, measurement error, criterion validity, construct validity, and responsiveness). It is important to note that one article can use several measurement properties per PROM and that a “study” corresponds to one measurement property (there can, therefore, be several measurement property studies for an article). The final objective of the COSMIN checklist is to assess the risk of bias and the level of evidence for each specific measurement property. We achieved this objective in four steps. First, we assessed the methodological quality of each single “study” by using the COSMIN risk of bias checklist (as “very good,” “adequate,” “doubtful,” “inadequate quality”). 16 Second, we evaluated the content validity of each PROM. If there was high quality evidence that the content validity of a PROM was insufficient, the PROM was not further considered. Third, we evaluated the measurement property of each PROM in each “study” by referring to the COSMIN table Updated criteria for good measurement property (sufficient [+], insufficient [−], inconsistent [±], or indeterminate [?]). 17 Fourth, the overall quality of evidence of a PROM was summarized according to the result of the first two steps to assign overall recommendation level (grade) according to the COSMIN table Grade approach for grading the quality of evidence 17 (high, moderate, low, or very low). The entire methodology of this evaluation is described in Appendix  S1 . The extraction form was pre‐tested by two authors (AO and FB), and disagreements were discussed with a third evaluator (AF). Once all the contentious issues were resolved, two reviewers (AO and FB) independently extracted the following data from the selected articles (using the full text and supporting information): general data (geographical origin, study subject), methods (type of intervention evaluated, study design, type of questionnaire used, measurement properties evaluated, duration of follow up, number of patients), and main results. Differences between treatments were expressed by effect size (ES) according to Cohen's method, which standardizes differences for comparison. An ES of around 0.2 was considered a small effect, around 0.5 moderate, and 0.8 or higher large. 18 , 19 The calculation of the ES was made with the Effect Size Calculator in Excel using n and mean difference.

Appendix S1.