In this study, we aimed to assess the psychometric properties of the Spanish version of the Female Sexual Function Index (FSFI) in a cohort of women with relapsing multiple sclerosis (MS). The results provided robust evidence of the reliability and validity of the FSFI in this specific population. The prevalence of SD in women was high, and its relationship with disability, engagement in a relationship, depression and fatigue were evidenced.
The exploratory factor analysis (EFA) revealed a five-factor structure, which was different from the initially proposed six-factor model [10, 40]. Notably, the 'arousal' items were found to be encompassed within two distinct factors: 'desire’ and 'lubrication'. This departure from the original structure might be attributed to the complex interplay between cognitive and physiological aspects of sexual response, as suggested by recent revisions [41].
Regarding structural validity, in the present study the EFA identified five-factors that explained 85% of the total variance and including the domains of desire/arousal, arousal/lubrication, orgasm, satisfaction and pain, showing the ‘arousal’ items comprised into two factors: item 3, 4 in ‘desire/arousal’ and item 5, 6 in ‘lubrication/arousal’, disappearing as an independent factor. Although Rosen and colleagues initially proposed a five-factor solution for the original FSFI, they chose to extend the FSFI structure to six domains because they considered the clinical distinction between ‘desire’ and ‘arousal’. A systematic review of measurement properties of the FSFI according to COnsensus-based Standards for the selection of health Measurement INstruments (COSMIN) guidelines shows inconsistent structural validity of the FSFI in distinct countries and populations. Although the authors observed a strong criterion validity that supported the use of the scale for the screening of sexual dysfunction, they reported conflicting evidence for some of the measurement properties of the FSFI advising to perform structural factor analysis for each study cohort before its use The authors concluded that there was more evidence against than in favor of the hypothesized 6-factor structure. Most studies direct and indirectly support a five-factor solution through use of confirmatory factor analysis (CFA) and principal component analysis (PCA)[10, 13, 16, 17, 22, 39, 42–45] in which ‘desire’ and ‘arousal’ are interdependent, as originally proposed by Rosen. This is in line with the revisions made in the DSM-5 to the model of FSD in which ‘desire’ and ‘arousal’ have been merged into one category of ‘female sexual interest/arousal disorder’ [1, 46, 47]. ‘Desire’ could be considered as a cognitive component into the category of sexual ‘arousal’ and ‘desire’ can precede or follow ‘arousal’ causing difficulties in differentiating between them in some women therefore causing cross-loading in the factor analysis[48, 49]. Other validations with PCAs resulted in distinct factor structures in different subgroups or nationalities. Such differences may be cultural, related to differing motivations for sex in women with or without arousal disorder [50], or it may represent the extensive overlap in sexual disorders [46, 51, 52]. This could also be due to differences in sample characteristics or measurement properties of the instrument used between studies. We conclude that the five-factor solution is a valid structure that enables the use of the Spanish FSFI in our cohort of women with MS, although larger studies or validation by external groups are warranted.
But, why using FSFI instead of a specific questionnaire for MS like the MSISQ-19[53]? Compared to MSISQ-19, FSFI provides a comprehensive evaluation of multiple domains of female sexual function that can be useful to identify specific affected areas and guide the selection of appropriate interventions [54] and allows for the comparison between cohorts of women with MS versus other neurologic patients, with other pathologies, or even versus health women. FSFI has been widely validated in several populations and languages and, conversely, pinpoints which specific sexual domains are affected and susceptible of a multidisciplinary therapeutic intervention, which was our initial purpose in the present study. It also reflects SD in the previous four weeks, being more sensitive to recent changes. Finally, and importantly, it allows to compare SD between people with or without MS and therefore, the prevalence among distinct populations. We consider that both scales are complementary, and efforts should be made to perform a psychometric validation of the Spanish version of MSISQ-19 in female population with MS.
The prevalence of SD in our cohort was of 42.6% of SD in MS. The prevalence observed was similar to that described in the general female population by others[55] and in the lower limit in women with MS, with studies showing SD ranging from 40 to 80% [56–59]. This may occur because our cohort includes young premenopausal women with an average low degree of disability and relapsing remitting forms of MS. Inclusion of progressive MS women, may increase the prevalence of SD, and further investigations are planned, although, from our clinical experience, the frequency of sexual intercourse in the most affected population was low due to disability. Regarding the affected domains, in our cohort SD was more related to ‘desire’ and ‘arousal’, and the least to ‘pain’. In other studies, the most frequent symptoms varied including ‘dyspareunia’ and ‘anorgasmia’ [60], anorgasmia and decreased vaginal lubrication[61], ‘alteration of the sensory function in the genital area’ and ‘decreased libido’ in women with advanced MS[62], ‘fatigue’, ‘decreased sensations’ and ‘decreased libido’[63]. Symptoms may be influenced by the degree of disability in the study cohort.
We found a relationship between SD and the degree of neurological disability, fatigue, and sphincter dysfunction. To have a stable partner showed like a protective factor. In general, fatigue, depression and anxiety are common symptoms among individuals with MS that can significantly impact on their sexual function. Their relationship with SD appears to be common and independent of other factors studied [64–69]. Subsequently, psychoeducational interventions and medical approaches of anxiety and depressive symptoms have shown a positive impact on SD [70, 71]. Age may also be a predictive factor for SD among women with MS, with older individuals potentially experiencing more difficulties [65, 69, 72–74]. Neurological disability has also been related to SD. As in our sample, the EDSS score, using different cut-off values, has been consistently related to SD in previous reports [72, 75–78]. Specific sphincter symptoms, such as urinary incontinence or bowel dysfunction, could also negatively impact sexual function in individuals with MS [79–81] although we could not confirm this association. These symptoms can lead to discomfort or embarrassment during sexual activity, resulting in a decrease in sexual desire and satisfaction. One study reported specifically overactive bladder as an independent factor of SD in women with MS [82]. In those women who suffer from sphincter symptoms, pelvic floor exercise, transcutaneous tibial nerve stimulation, clean intermittent self-catheterizations and botulinic toxic injection have shown an improvement of SD after non-blinded, single-arm trials [83–87]. The presence of spinal cord lesions did not associate to SD in our cohort. The use of radiological parameters, such as quantitative measures or brain and spinal lesion topography, has shown contradictory results [88] and, in most cases, no relationship was found between radiological variables and SD [66, 76, 89, 90]. CNS depressants were a negative factor affecting SD in our population, although after correcting for multiple comparisons, statistical significance was not confirmed.
In a study by Altman and colleagues, in which MS doctors were surveyed regarding their approach to taking a sexual history of their patients, only 15% reported discussing sexuality with every patient. The lack of evaluation of SD during consultation leads to the remaining as an underdiagnosed and undertreated condition [91]. The most common reason for not doing so was a fear of crossing personal boundaries, lack of spontaneous communication by patients [92] or lack of knowledge [93]. However, some studies show that women found the interview regarding sexuality a positive experience[62]. To provide a more conductive setting in this field, it is recommended to provide the women a safe and confidential environment, using open-ended questions and assess the SD as part of a holistic approach to managing MS [93].
One limitation of the study is the disproportionate number of patients with EDSS < 3 and EDSS > 3, which may have influenced the results of the discriminant validity analysis and could have been affected the prevalence and the most frequent symptoms. On other hand, the present study did not evaluate the responsiveness to intervention, an important psychometric property than allow to assess the capacity of the questionnaire to measure response to treatment on SD. The strengths of the study include being, to our knowledge, the first validation of the Spanish FSFI in women with MS, a relatively large multi-center sample, and completion of the questionnaire by all participants.