Psychometric Validation of the Sexual Distress Scale in Male and Female Portuguese Samples

ABSTRACT

Background

The Female Sexual Distress Scale and the Female Sexual Distress Scale-Revised—herein called the Sexual Distress Scale (SDS and SDS-R)—are among the most widely used self-report instruments to assess sexual distress, but no version for use in the Portuguese population is available to date.

Aim

The current study aimed to validate the Portuguese version of the SDS/SDS-R in samples of women and men with and without distressing sexual problems.

Methods

A sample of 1,109 participants without distressing sexual problems (761 women) and 382 participants with distressing sexual problems (283 women), ages ranging from 18 to 72 years, were used to examine the psychometric properties of the Portuguese SDS and SDS-R.

Outcomes

Participants completed a survey that included a sociodemographic and health questionnaire, the Portuguese version of the SDS and SDS-R, and measures of sexual satisfaction, sexual quality of life, sexual function, dyadic adjustment, and psychological distress.

Results

Results indicated that the Portuguese SDS and the SDS-R assess 1 general domain of sexual distress and showed good evidences of validity based on content and on relations with other variables. Sexual distress was associated with poorer sexual function, satisfaction, and quality of life, with higher psychological distress, and lower dyadic adjustment. Internal consistency and test-retest (1 month) reliabilities were excellent. Tests of differential functioning of items indicated that the SDS and SDS-R scores can be used to compare women and men on sexual distress, but the SDS/SDS-R scores flagged differential functioning of items and test (DFIT) between participants with and without distressing sexual problems.

Clinical translation

Clinicians and researchers can now make use of the SDS and of the SDS-R in the Portuguese population, facilitating the assessment of sexual distress in clinical settings.

Strengths & Limitations

The Portuguese SDS/SDS-R scores can be compared between women and men, providing information on sexual distress independently of sexual function. With the current evidence, comparisons between individuals with and without distressing sexual problems should be made with caution, as the scores may be biased against the former.

Conclusion

This study provides a validation of the Portuguese version of the SDS/SDS-R that can be used to assess sexual distress in Portuguese women and men and can be used to compare between these 2 groups.

INTRODUCTION

Sexual health is a major dimension of global health and is linked to several aspects of individuals’ overall well-being and quality of life, mental health, as well as the quality and stability of their sexual relationships.^1–3 Sexual dysfunction hampers sexual health and is influenced by biomedical, social, psychological, and interpersonal factors.⁴ Although heterogeneous, sexual dysfunctions are all characterized by an impairment in the person’s ability to respond sexually or to experience sexual pleasure and, importantly, by the concomitant experience of sexual distress. Sexual distress refers to the experience of negative emotional responses, such as frustration, worry, guilt, anxiety, or bother, associated with one’s own sex life. These emotional negative responses correspond to the subjective evaluation of one’s own sexual difficulty and the assessment of this construct has been increasingly recognized by clinicians and researchers in recent years.

The experience of sexual distress is currently considered a necessary condition for the diagnosis of sexual dysfunction according to diagnostic manuals such as the DSM-5⁵ and the ICD-11.⁶ Although both criteria – that is, the presence of a clinically significant sexual difficulty as well as the presence of sexual distress – are necessary conditions for a diagnosis of sexual dysfunction, the assessment of sexual dysfunctions has extensively relied on the assessment of sexual functioning alone. Widely used instruments such as the Female Sexual Function Index, FSFI,⁷ and the International Index of Erectile Function, IIEF,⁸ lack the assessment of sexual distress, a central although separate component of sexual dysfunction and a determinant of sexual health. This is an important omission because sexual distress exerts a large influence in individuals’ sexual and overall well-being.^9,10 The experience of sexual distress is associated with lower positive mood and increased rates of depression,⁹ lower sexual satisfaction,^11,12 poorer sexual function,^13–15 diminished relationship satisfaction and quality,^11,12 and greater motivation to discuss sexually related problems with professionals and to seek treatment.¹⁶ Therefore, it is crucial that valid, reliable, and easy-to-use instruments for the assessment of sexual distress are available to both clinicians and researchers.

There are several advantages of including sexual distress in the assessment of sexual health. It is recognized that prevalence estimates of sexual problems vary widely according to the criteria used to assess them and, when sexual distressed is assessed, estimates of sexual dysfunction are lower.¹⁷ Taking sexual distress into account is hence critical to prevent an overestimation in sexual dysfunction rates and to permit an adequate evaluation and comparability of population-based prevalence’s. Furthermore, impairments with sexual function per se are not informative about how individuals feel about their sexual problems, since not all impairments in sexual functioning are associated with sexual distress.^14,18,19 An adequate assessment of sexual dysfunction should not only include the evaluation of clinically relevant impairments in sexual function but should also consider the extent to which these impairments are considered distressful by the individual. Furthermore, sexual distress can be experienced even in the absence of problems with sexual functioning. Individuals can be distressed by many aspects of their sexual lives other than impaired sexual function, including barriers to attaining pleasure through sex, anticipating negative consequences of sex, or even the experience of relationship problems.^18,20 As such, the availability of an instrument which provides an assessment of sexual distress as a distinct construct from impaired sexual function is valuable.

Several measures of sexual distress are currently available (for a review, ²¹). The most widely used tool for assessing sexuality-related distress are the Female Sexual Distress Scale, FSDS,²² and its revised version, the Female Sexual Distress Scale–Revised, FSDS-R.²³ The FSDS was originally developed as a 12-item, unifactorial scale to evaluate global sexual distress. The revised version, the FSDS-R, includes an extra item that assesses sexual distress specifically related to low sexual desire (ie, “Are you bothered by low sexual desire?”). Based on several evidences of validity, the authors showed that the FSDS and FSDS-R can be used to assess sexual distress and to discriminate between women with and without sexual dysfunction.²² Both the internal consistency and the test-retest reliability of the FSDS and FSDS-R were excellent.^22,23

The FSDS/FSDS-R offers important advantages compared to other measures of sexual distress, such as the Sexual Satisfaction Scale for Women, SSS-W,²⁴ the Sexual Quality of Life-Male, SQOL-M,²⁵ or the Multidimensional Sexuality Questionnaire, MSQ.²⁶ First, some of these previous scales can only be applied to women or men and have known validity and reliability issues.²¹ Second, the FSDS/FSDS-R assesses sexual distress independently from sexual function domains. The respondent reports the frequency of sexual distress irrespective of specific sexual function domains (eg, desire, erectile function), which increases the sensitivity of the scale to other potential sources of sexual distress (eg, relationship problems). Third, although originally developed to assess female sexual distress, the items of the FSDS/FSDS-R are gender and/or sex neutral, as they do not refer to any gender and/or sex specific characteristic. In fact, the scale has been validated in male samples. Santos-Iglesias and colleagues²⁰ demonstrated that the FSDS can also be used to assess sexual distress in male samples (herein, the Sexual Distress Scale, SDS; and Sexual Distress Scale–Revised, SDS-R). The authors found that the SDS and SDS-R are representative of the sexual distress experiences of men. Furthermore, the SDS and SDS-R can be used to assess sexual distress in men, make unbiased comparisons between men and women, and discriminate men with and without distressing sexual problems. Similarly, the internal consistency and test-retest reliabilities were excellent. Finally, the SDS/SDS-R has been translated into several languages (eg, Korean, Polish, Iranian, Turkish) and has been widely used across different populations and cultures^13,27–29 consistently showing strong psychometric properties. Across prior studies, scores on both scales with female and male samples have found moderate to strong positive associations with indices of psychological distress,^13,20,30 moderate to strong negative associations with sexual satisfaction,^31,32 weak to moderate negative associations with sexual function,^13,20,27 weak negative associations with relationship adjustment,³¹ and weak to non-significant associations with sexual attitudes.²⁰

CURRENT STUDY

To date, a Portuguese version of the SDS/SDS-R is not available. The purpose of the present study was to translate, adapt, and validate a Portuguese version of the SDS/SDS-R using male and female samples. More specifically, the goal of the study was to examine whether the SDS/SDS-R can be used to assess sexual distress in Portuguese-speaking samples and whether it can be used to compare men and women and individuals with and without distressing sexual problems. Following the Standards for Educational and Psychological Testing³³ and the argument-approach to validation,^34,35 we proposed that the SDS/SDS-R scores indicate levels of sexual distress in Portuguese men and women. Based on this interpretation, the SDS/SDS-R can be used to: (i) assess levels of sexual distress in Portuguese-speaking men and women; (ii) compare men’s and women’s levels of sexual distress, and (iii) compare the sexual distress of individuals with and without distressing sexual problems. Based on these proposed interpretations and uses, the following assumptions and evidences of validity and reliability were examined

1. Items of the Portuguese SDS/SDS-R would be relevant and representative of sexual distress. The items would be clear and easy to understand for respondents taking the SDS/SDS-R.

2. Consistent with the original version of the SDS/SDS-R, the items would cluster into 1 single factor of sexual distress.

3. The items’ parameters follow the functional form of a graded response model. That is, higher levels of sexual distress would be associated with higher responses options (ie, the thresholds are ordered increasingly). The items would also show different levels of difficulty and discrimination.

4. The SDS/SDS-R scores would be strongly negatively correlated with sexual quality of life and sexual satisfaction scores. The SDS/SDS-R scores would be moderately negatively correlated with sexual function and moderately positively correlated with psychological distress. Finally, the SDS/SDS-R scores would be weakly or not correlated with dyadic adjustment.

5. The items would not show differential item or test functioning across gender (men and women) and between individuals with and without distressing sexual problems.

6. The SDS/SDS-R scores would discriminate between individuals with and without distressing sexual problems.

7. Internal consistency and test-retest reliability values would be excellent.

METHODS

Participants

For this study, we aimed to recruit a sample of men and women with and without distressing sexual problems. The inclusion criteria were being at least 18 years old and being able to proficiently read Portuguese. Eligibility criteria were determined using a brief screening questionnaire before the beginning of the survey. The initial sample comprised 1,507 individuals. Three (0.2%) individuals responded to the female version of the questionnaire but reported their sex was male and 3 (0.2%; 1 women) were non-attentive responders (ie, failed to correctly respond to at least 2 of 3 attention checks embedded in the survey). These participants were not included in the analyses. An additional 10 participants were excluded because they had either out-of-range or missing values on key variables (eg, age, gender). The final sample was comprised of 1,491 participants.

Participants were classified as having distressing sexual problems if they: (i) self-reported sexual difficulties and concomitant sexual distress over the last 3 months; and/or (ii) were currently receiving and/or seeking treatment for their sexual difficulties. Seeking help and/or treatment for sexual difficulties was used as an indicator of sexual distress because past research has found that seeking help and/or treatment is a significant marker of higher levels of sexual distress.¹⁶ Following these criteria, a total of 382 (25.62%) individuals were classified as having distressing sexual problems (283 women and 99 men).

Procedures

Ethics approval for the study was obtained from the institutional ethics board. Different procedures were used to disseminate the study and recruit participants, including: (i) study advertisement through social media networks (eg, academic social media official pages, online newsletters of community and sexual health organizations); (ii) flyers posted in different community and/or sexual health-related sites (eg, University, sexual-health clinics); (iii) referral of participants with sexual difficulties from sexual health professionals working within clinical contexts. All participants were invited to complete the survey online.

Participants provided informed consent online upon following the survey URL link and before beginning the survey. Data collection was performed using an online platform that required no personal identification (eg, IPs were not registered). Upon completing the baseline survey, participants were asked whether they would be interested and available to complete the retest survey 1 month later. In case they responded affirmatively, participants were directed to a web form where they were asked to provide a valid email address and to generate a unique code. This information was stored in a secure database, separate from their survey responses, and was used to conduct the retest assessment. An invitation to complete the retest survey was sent 1 month after the baseline survey. One week after sending out the invitation, an email reminder was sent to those participants who had not yet completed the retest survey. The 1 month follow-up survey was completed by 192 women (18.39%) and 98 men (21.92%). No monetary compensation was provided to participants.

Measures

Sociodemographic and Health Questionnaire

Information on participants’ age, gender, sexual orientation, and relationship status were collected. Participants reported on their current overall health status using a visual analog scale ranging from 0 (worst imaginable health) to 100 (best imaginable health). Participants were also asked to report whether they were sexually active during the past 3 months, and whether they had experienced sexual difficulties in the past 3 months. In case of an affirmative response to the latter, participants were asked to report on whether they were currently receiving and/or seeking treatment for their sexual difficulties.

Sexual Distress Scale-Revised (SDS-R)

The Portuguese version of the 13 item SDS-R was administered (see scale adaptation and translation procedures below). Participants rate their answers on a 5 points scale that ranges from 0 (never) to 4 (always), with higher scores indicating higher sexual distress. A full description of the SDS-R has been provided in the Introduction.

Sexual Quality of Life Questionnaire (SQoL)

The impact of problems with sexual function to one’s sexual quality of life was assessed using the valid and reliable SQoL.^25,36 Women completed the female version (SQoL-F; 18 items) and men completed the male version of the instrument (SQoL-M; 11 items). Using a 6 point Like-type rating scale, participants reported on the emotional well-being related to their sexual life (eg, “When I think about my sexual life, I feel frustrated”). Total scores range from 18 to 108 in the SQoL-F and from 11 to 66 in the SQoL-M, with higher scores indicating higher sexual quality of life. The scales demonstrated good internal consistency in the current study (ω = .77 for women and ω = .92 for men).

Global Measure of Sexual Satisfaction (GMSEX)

The GMSEX is a 5 item scale that assesses satisfaction with the sexual relationship with a partner.^37,38 Participants rate their overall satisfaction with their sexual relationship on 5 seven-point bipolar scales: good–bad, pleasant–unpleasant, positive–negative, satisfying–unsatisfying, valuable–worthless. Total score ranges from 5 to 35, with higher scores indicating greater sexual satisfaction. Reliability in the current study was excellent (ω = .96).

Female Sexual Function Index (FSFI)

The FSFI is a 19 item scale^7,39 that assesses women’s sexual functioning across 6 domains (desire, arousal, lubrication, orgasm, satisfaction, and pain). Total scores range from 2 to 36 with higher scores indicating better sexual function. The FSFI demonstrated excellent internal consistency for all subscales (ω = .84 - .96) in the present study.

International Index of Erectile Function (IIEF)

Men completed the 15 item IIEF,^8,40 a measure of men’s sexual function across 5 domains (sexual desire, erectile function, orgasmic function, intercourse satisfaction, and overall satisfaction). Total scores range from 5 to 75 with higher scores indicating better sexual function. All IIEF subscales demonstrated good internal consistency (ω = .78 - .95) in this sample.

Brief Symptom Inventory 18 (BSI-18)

Psychological distress was assessed using the 18 item BSI.^41,42 Using a 5 point Likert-type scale, individuals report on the intensity of psychopathological symptoms experienced in the last 7 days. The instrument includes 3 subscales: anxiety (eg, “Feeling tense or keyed up”), depression (eg, “Feeling lonely”), and somatization (eg, “Faintness or dizziness”). Total scores range from 0 to 72 (0 – 24 for each subscale), with higher scores indicating greater psychological distress. Internal consistency was good for all subscales (ω = .76 - .92) in the current sample.

Dyadic Adjustment Scale-Revised (DAS-R)

Participants responded to the 14 item DAS-R, a questionnaire to measure relationship quality.^43,44 Respondents are asked to rate aspects of their relationship on a 5- or 6-point Likert-type scale (eg, “How often do you and your partner quarrel?”). The DAS-R is comprised of 3 subscales: (i) consensus, (ii) satisfaction, and (iii) cohesion. Total scores range from 0 to 69, with higher scores indicating greater dyadic adjustment. The DAS-R showed good internal consistency in the current study (ω = .72 - .86).

Scale Translation and Adaptation

After obtaining permission for translation and validation from the original authors, the Portuguese version of the SDS/SDS-R was developed. First, all items were translated to Portuguese by 2 bilingual proficient speakers. Second, a back-translation to English was performed by an independent native English speaker. At this stage, the English translation was compared with the original questionnaire to identify any relevant language differences. The 3 translators discussed the backward translation, and no relevant discrepancies were identified.

Data Analysis

In order to examine whether the SDS-R items were clear and representative of sexual distress (assumption 1), evidences of validity based on content from the perspective of respondents were obtained. We used a sample of 22 participants (12 women) with similar characteristic as the study sample. Participants were provided with a definition of sexual distress (ie, “the experience of negative feelings regarding one’s own sexual life”) and were asked to indicate: (i) whether SDS/SDS-R items were clear and easy to understand (if No, they were asked to provide alternative suggestions to improve clarity and understanding of the items); (ii) whether the items were sufficient to assess sexual distress (if No, they were asked to provide additional items); and (iii) whether any item was not necessary to assess sexual distress and could be deleted (if Yes, they were asked to indicate which ones). Finally, the resulting version was reviewed by a panel of professionals (2 psychologists, 2 sex therapists, 1 gynecologist, 1 pelvic floor physiotherapist, and 1 general and family medicine doctor) to further ensure clarity of the items, clinical and conceptual relevance (eg, whether the items comprehensively reflected representations of sexual distress according to their clinical experience).

All the analyses were conducted separately on the 12- (SDS) and 13-item (SDS-R) versions. Evidences of validity based on internal structure were examined using confirmatory factor analysis (CFA), internal consistency, item analysis via item response theory (IRT), and differential item and test functioning analyses.⁴⁵ Four separate CFAs (Assumption 2) were used to examine the unidimensional structure of the SDS and SDS-R in both women and men. A diagonally weighted least squares estimator was used because of the ordinal nature of the scale and the data were not multivariate normal.⁴⁶ Comparative Fit Index (CFI) and Tucker–Lewis Index (TLI) ≥ .95, and Standardized Root Mean Square Residual (SRMR) ≤ .08 indicated a good fit.^47,48 Internal consistency (Assumption 7) was examined using McDonald’s omega⁴⁹ because the assumption of tau-equivalence was violated. Test-retest reliability (Assumption 7) was tested using Pearson correlations.

Item analyses using IRT for polytomous items (Assumption 3) were calibrated using a Graded Response Model, GRM.⁵⁰ The GRM was compared to a restricted GRM (in which all discrimination parameters were equal) using -2*log-likelihood test. Item level fit was tested using the polytomous extension of the S-χ² statistic⁵¹ and overall model fit was tested using the M² and RMSEA statistic.⁵² All the analyses were conducted using the R package “mirt.”⁵³

The ability to compare across gender and/or sex (men and women) and between individuals with and without distressing sexual problems was tested using both differential item functioning (DIF) and differential test functioning (DTF), Assumption 5. DIF means that the items measure the construct differently across groups. DTF occurs when these between-group differences can be observed at the test level.⁵⁴ Tests of DTF are usually more meaningful than tests of DIF because most between-group comparisons are conducted at the test level.⁵⁵ For polytomous items, the likelihood ratio test (LRT), as implemented in the R package “lordif,”⁵⁶ was used. The chi-square method and alpha of 0.01 were used to detect DIF. The next step was to inspect DTF between gender and/or sex and individuals with and without distressing sexual problems. Once items with DIF were identified, 2 randomly selected items with no DIF were used as anchors. If all items flagged DIF, then 2 items were randomly used as anchors. DTF was examined using the signed and unsigned DTF statistics⁵⁴ using the “mirt” package in R.⁵³

Assumptions 4 and 6 were examined using evidence of validity based on relations with other variables. Pearson correlations were used to examine the association between the SDS/SDS-R scores and other related and unrelated constructs (Assumption 4). Receiver operating characteristic (ROC) curves and t-tests for independent means were used to examine the ability of the SDS/SDS-R scores to discriminate between people with and without distressing sexual problems (Assumption 6).

RESULTS

Descriptive Statistics

Participants’ ages ranged from 18 to 72 (M = 29.12, SD = 9.65). Of the total sample, 1,044 (70%) were women (age M = 28.04, SD = 8.81, range = 18–66) and 447 were men (age M = 31.62, SD = 11.00, range 18–72). The majority of women and men were exclusively to predominantly heterosexual (87.2 and 81%, respectively), were in a romantic relationship (71.3 and 66.2%, respectively), and were sexually active at the time they completed the survey (88.4 and 86.6%, respectively). At the time of completion of the survey, the majority of women and men had an average self-rated health status of 80.14 for women (SD = 16.64) and men (SD = 17.17) on a scale from 0 to 100.

Evidences of Validity Based on Content

Overall, participants described the items as clear and representative. The majority (77%) offered no suggestions to improve the clarity of the measure. Three participants suggested rephrasing 1 of the items to enhance clarity (item 9); suggested reformulations included stating “feelings of regret regarding your sexuality” instead of “regrets about your sexuality”; we implemented this suggestion. One participant suggested that the meaning of “sexual problems” should be better formulated but offered no alternative suggestions. No missing contents or items were identified by any participant; 1 participant suggested that an open-ended question could be provided to “describe what is causing sexual distress to each person.” Based on these results, no other modifications to the measure were deemed necessary. Finally, a multidisciplinary clinical panel composed of 2 psychologists, 2 sex therapists, 1 gynecologist, 1 pelvic floor physiotherapist, and 1 general and family medicine doctor reviewed the resulting version. No additional revisions were suggested, resulting in the final version of the instrument.

Evidences of Validity Based on Internal Structure

Factor Analysis and Internal Consistency

Results of the CFAs for the SDS and SDS-R for women and men showed good fit (Table 1). For women, all the factor loadings were on the very good to excellent range. For men, all factor loadings were excellent, except for items 9 and 13 whose factor loadings were good.⁵⁷ McDonald’s omega values were excellent (> .924) for both women and men (Table 2).

Item Analysis and Differential Item and Test Functioning

The GRM model fit the data better than the restricted GRM (ie, same discrimination parameters for all items), indicating that the items had different discrimination parameters. Results of the calibration and the final item parameters for women and men are presented in Tables 3 and 4, respectively. For women, the discrimination parameters (a) ranged from 1.317 (item 9, SDS-R) to 3.636 (item 4, 12-item version), which indicates a considerable amount of discrimination and moderate to strong association between the items and the latent construct. The location parameters (b) ranged from -0.578 (item 2, 12-item version) to 5.133 (item 9, 13-item version). Overall, most items showed moderate to high levels of difficulty, indicating that items were most efficient at average to moderately high levels of sexual distress. Item 9 (“Regrets about your sexuality”) showed the highest location (ie, difficulty) parameters, indicating that women need very high levels of sexual distress to endorse this item (Table 3), or that this item is most useful to discriminate women with very high levels of sexual distress.

Similar results were found for men. Discrimination parameters ranged from 1.129 (item 13) to 3.093 (item 6, 13-item version), indicating a considerable amount of discrimination. The location parameters ranged from -1.041 (item 2, 12-item version) to 4.822 (item 9, 12-item version). Overall, most items showed moderate to high levels of difficulty, indicating that items were most efficient at average to moderately high levels of sexual distress. Item 9 also showed the highest location (ie, difficulty) parameters, indicating that men need higher levels of sexual distress to endorse this item (Table 4).

For test of differential item and test functioning (DIF and DTF, respectively) (Table 5), the signed DTF (sDTF) statistic represents the bias in the total raw score between groups, whereas the unsigned DTF (uDTF) statistic is an indicator of the average area between the test curves for each group. The sDTF ranges between -TS and TS (TS being the highest possible test score). Positive values indicate that the reference group (men and individuals without distressing sexual problems) scored higher than the focal group (women and individuals with distressing sexual problems). Negative values indicate that the focal group scored higher than the reference group. The uDTF ranges between 0 and TS. Small values indicate that the area between the test curve for each group is small.

Tests of differential item and test functioning (DIF and DTF, respectively) did not reveal any significant problems across gender/sex (Table 5). For gender and/or sex, several items flagged DIF, although they did not result in DTF. From Table 5, we can see that the sDTF statistic was not statistically significant and was very small in size. The positive sign for the SDS suggests that, after being matched on their levels of sexual distress, men (reference group) scored slightly higher than women. The uDTF statistic was also very small (uDTF = 0.10) indicating a small area between the 2 curves. Taken together, these results show that although 3 items flagged DIF, they did not contribute to DTF. Very similar results were found for the SDS-R (except that the sDTF sign was negative). Therefore, the SDS/SDS-R can be used to compare women’s and men’s scores.

Regarding distressing sexual problems, for both the SDS and SDS-R several items flagged DIF and overall test scores were also affected. The significant negative sDTF statistic indicates that, after being matched on their level of sexual distress, the group with distressing sexual difficulties (focal) scored significantly higher than the group without distressing sexual difficulties (reference group). The overall raw score difference of 1 point (1.13 for the SDS-R), or around 2% of the total score (2.17% for the SDS-R), indicates that the group without distressing sexual difficulties scored more favorably (lower distress), and that the groups with and without distressing sexual difficulties cannot be compared on these 2 scales. Because some items and overall scores were biased, we did not examine the ability of the SDS/SDS-R to discriminate between participants with and without distressing sexual problems (Assumption 6), as measurement bias also has the potential to influence diagnostic tests.⁵⁸

Evidences of Validity Based on Relations with Other Variables

The predictions about the correlations with other variables were supported. The SDS/SDS-R were strongly and negatively correlated with sexual quality of life and sexual satisfaction, with the strength of the correlation being higher for sexual quality of life than for sexual satisfaction. Psychological distress was moderately positively associated with sexual distress and sexual function. Finally, as expected, dyadic adjustment total scale and subscales were weakly to moderately negatively associated with sexual distress.

Test-retest Reliability

The test-retest reliability values after 1 month were .794 (P < .001) and .799 (P < .001) for the female 12 and 13-item versions, respectively. For men, the reliabilities were .842 (P < .001) and .845 (P < .001), respectively.

DISCUSSION

The FSDS/FSDS-R and its recently validated version for use in male samples, the SDS/SDS-R, are the most widely used measures of sexual distress. These measures allow the assessment of sexual distress in a wide range of countries and populations (eg, sex and/or gender, health status). This study is the first to provide and validate a Portuguese version of the SDS/SDS-R. The results offer an overall positive picture of the SDS/SDS-R in that it can be used to examine levels of sexual distress in Portuguese-speaking women and men and to compare their scores. Difficulties arose in its usefulness to compare women and men with and without distressing sexual problems. Future studies need to provide a more thorough assessment of this issue and offer alternatives and revisions to address this limitation.

We proposed 1 interpretation and 3 different uses of the SDS/SDS-R in Portuguese-speaking samples. We found evidence that support both the interpretation that the SDS/SDS-R scores indicate levels of distress, as well as support for 2 of the uses. Overall, the assumptions based on inferences of scoring, scaling, generalization, and extrapolation,⁵⁹ for the most part, received support. Evidence of validity based on content show that participants thought that the items were clear, understandable, and all necessary to comprehensively assess the intended construct (ie, sexual distress).

All participants considered that the items that currently comprise the SDS were sufficient to assess sexual distress, with no other items or issues being suggested as necessary additions to the scale. Furthermore, evidence based on internal structure via confirmatory factor analyses corroborated the unidimensional structure of the both the SDS and SDS-R scales, resembling their original factor structure for both women and men.^21,23,30 As such, results indicate that the SDS/SDS-R assesses a unique and general domain—sexual distress—which reflects negative feelings of distress related to one’s sexual function and sexual life. In terms of reliability, the measure demonstrated excellent indices of internal consistency and temporal stability, in line with its original validation studies.^21,23,30 Internal consistency and test–retest reliabilities were excellent for both versions among women and men. We also conducted item analysis to further assess the efficiency and difficulty of each item in tapping into sexual distress. For both women and men, most items showed moderate to high levels of difficulty, with item 9 (ie, “Regrets about your sexuality”) demonstrating the highest difficulty levels such that, in both versions of the SDS/SDS-R, individuals needed higher levels of distress to endorse this item. Overall, SDS/SDS-R items performed well at identifying individuals with different levels of sexual distress as they were able to clearly discriminate between individuals with different levels of sexual distress.

As expected, the SDS/SDS-R also presented good indices of validity in terms of relations with other variables. In line with our predictions, scores on the SDS/SDS-R showed strong negative associations with sexual quality of life and sexual satisfaction scores. This was expected since these constructs encompass positive feelings regarding one’s own sexual life, which contrast with the experience of sexual distress, as supported by prior research.^31,32 Results also indicated that higher scores on the SDS/SDS-R were moderately linked to higher psychological distress and to poorer sexual function, which is consistent with prior evidence.^9,13,15 This reinforces that the experience of sexual distress is an important marker of poorer sexual and psychological well-being. Indeed, women and men who scored higher on sexual distress also scored higher on depression, anxiety, and somatization symptoms, with all these associations being of moderate size. For both women and men, sexual distress was linked to poorer scores in all sexual function dimensions. The strength of these associations mirrors those indicated by prior research—while for women sexual arousal and orgasm showed the highest correlations with sexual distress, the strongest association for men was found for erectile function.^12,18 This aligns with prior research indicating that, across populations, erectile difficulties are 1 of the most prevalent sexual difficulties for men and the complain that shows the strongest associations with levels of sexual distress.^19,60,61

Finally, sexual distress also showed moderate negative associations with dyadic adjustment, for those individuals who were in a relationship. This mild correlation reinforces that the experience of sexual distress can be importantly connected with dyadic dimensions of the sexual relationship, and that it encompasses more than just distress about sexual function. Among the dyadic adjustment dimensions, the strongest association was found for the dyadic consensus dimension—and of moderate size for both women and men—which comprises areas of agreement or disagreement between partners, including sexual relations. These results further support the advantage of assessing sexual distress in a broader manner that goes beyond the mere assessment of distress related to sexual function impairments, as is done in the SDS/SDS-R.

As for test functioning, the SDS/SDS-R did not show differential functioning between women and men, demonstrating that the scale can be used to compare women and men. However, comparisons between individuals with and without distressing sexual problems are not recommended based on the results of this study. The results show that several items were biased against individuals with distressing sexual problems. That is, after being matched on their levels of sexual distress, individuals with distressing sexual problems score higher on these items. For example, item 4 (“Frustrated by your sexual problems”) was biased against individuals with distressing sexual problems compared to those without them. Similar results have been found in men with prostate cancer.³² It might be that the feeling of frustration is very specific to the experience of sexual function difficulties,^62,63 and unlikely to be used by those not experiencing sexual difficulties. Not only did participants with distressing sexual difficulties score higher on several items, but this also affected the overall test score by disproportionately inflating their total score. If we were to wrongly assume that the SDS/SDS-R scores are not biased and compared these 2 groups, we would arrive at wrong conclusions about their true level of sexual distress. Future studies need to examine the sources of bias, as the translation procedure could have introduced differences in the way the items are interpreted between groups. In this regard, cognitive interviewing has proved to be a useful method to identify sources of bias at the item and construct level.⁶⁴ Another alternative is to develop a shorter version that minimizes the influence of biased items on the total score and that facilitates the comparison among individuals with and without distressing sexual problems. It is important to note that our sample was comprised of individuals who self-reported sexual difficulties and distress. Future studies also need to use samples men and women with a clinical diagnosis to see if the scale shows the same limitation. Despite this limitation, it is worth noting that, to date, the SDS and SDS-R are the only measures of sexual distress for Portuguese-speaking samples of women and men in which differential functioning of item and test have been tested, making them the only available option when the purpose is to compare sexual distress across gender.

The present study offers a significant contribution to the literature by providing the first validation of the SDS to the Portuguese context. Still, these results need to be interpreted considering a few limitations. First, results may have limited generalizability since the sample was collected using non-random sampling procedures. Second, although the study was advertised as a “study on psychological and sexual well-being,” the voluntary nature of the participation on such a study may have led to a selection bias.⁶⁵ Finally, women and men in our group with distressing sexual problems had a variety of difficulties. Future research should further explore the psychometric properties of the SDS/SDS-R among more clinically homogenous samples, as to further understand how the scale performs for groups of women and men with different sexual complains.

CONCLUSIONS

Considering current and prior evidence of validity and reliability of the SDS/SDS-R, these instruments are highly valuable tools for researchers and clinical practitioners to assess distress about one’s sexual life. Current results indicate that the Portuguese version of the SDS/SDS-R assesses levels of sexual distress and that it can be used to compare between samples of women and men, although comparisons between individuals with and without distressing sexual problems are limited with the current evidence. The existence of a Portuguese version of the SDS/SDS-R will assist cross-national research contributing to a better understanding of the determinants and processes involved in sexual difficulties, sexual well-being, and quality of life of women and men.

Funding

This research was supported by the Portuguese Foundation for Science and Technology (FCT) and the COMPETE 2020 program (CPUP UIDB/00050/2020, POCI-01-0145-FEDER-0072). IT is supported by a grant from FCT (SFRH/BD/131808/2017).

STATEMENT OF AUTHORSHIP

Conceptualization, I.M.T., P.J.N.; Methodology, I.M.T., P.S.I., P.J.N.; Investigation, I.M.T., P.S.I., P.J.N.; Writing − Original Draft, I.M.T., P.S.I.; Writing − Review & Editing, I.M.T., P.S.I., P.J.N.; Funding Acquisition, I.M.T., P.J.N.; Supervision, P.S.I., P.J.N.

REFERENCES

Author notes