Arabic Translation of Female Sexual Function Index and Validation in an Egyptian Populatio

Validation of the Arabic Female Sexual Function Index
of 9
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  ORIGINAL RESEARCH—OUTCOMESASSESSMENT Arabic Translation of Female Sexual Function Index andValidation in an Egyptian Population jsm_2471 3370..3378 Tarek H. Anis, MD,* Samah Aboul Gheit, MD, † Hanan S. Saied, MD, ‡ and Samar A. Al_kherbash, MD § *Kasr Al-Eini School of Medicine, Cairo University, Department of Andrology, Cairo, Egypt;  † Kasr Al-Eini School ofMedicine, Cairo University, Department of Obstetrics and Gynecology, Cairo, Egypt;  ‡ Sahel Teaching Hospital,Department of Dermatology, Cairo, Egypt;  § Yemen International Hospital, Department of Dermatology, Taiz, YemenDOI: 10.1111/j.1743-6109.2011.02471.x A B S T R A C T  Introduction.  Female sexual dysfunction (FSD) is a prevalent health problem that has been inadequately investigatedin the Arab world. An Arabic assessment instrument for FSD is urgently needed.  Aim.  To validate the Arabic version of the Female Sexual Function Index (ArFSFI).  Methods.  This is a cross-sectional study conducted between January and April 2010. Eight hundred and fifty-five women (16–60 years old) participated in the study. Validation was carried out on aspects of face, content, discrimi-nant, and criterion (concurrent) validity. Construct validity was evaluated using principal component analysis.Reliability studies on test–retest and on internal consistency were conducted with Pearson correlation and Cron-bach’s alpha, respectively. The best cutoff point for the ArFSFI to differentiate cases and noncases was determinedusing a receiver operating characteristic (ROC) curve.  Main Outcome Measure.  Parameters of validity and reliability of the ArFSFI and its domains.  Results.  ArFSFI total score and scores of various domains showed high test–retest reliability (r from 0.92 to 0.98). ArFSFIdomainsshowedhighinternalconsistency( a from0.85to0.94).Sixhundredandforty-fourwomen(75.32%)mettheDiagnosticandStatisticalManualofMentalDisorders,4thedition,criteriafordiagnosisofsexualdysfunction while211women(24.68%)showednormalfunction.TheArFSFIwasfoundtohavegooddiscriminantvalidity.There were significant differences between the mean scores of women with sexual dysfunction and those of women without sexual dysfunction. A total score of 28.1 was taken as the cutoff point for the ArFSFI to distinguish between women with FSD and women with normal function (sensitivity 96.7%, specificity 93.2%). The ArFSFI showed an excellent overall performance (area under the curve [AUC]  =  0.985, 95% confidence interval 0.978–0.992). Conclusion.  The ArFSFI is a validated, reliable, and locally accepted tool for use in the assessment of FSD in theEgyptian population.  Anis TH, Aboul Gheit S, Saied HS, and Al_kherbash SA. Arabic translation of femalesexual function index and validation in an Egyptian population. J Sex Med 2011;8:3370–3378.  Key Words.  Female Sexual Function Index; Arabic Version; Validation Studies; Psychometry  Introduction  T he Arab world refers to Arabic-speakingcountries stretching from the Atlantic Oceanin the West to the Arabian Sea in the East, andfrom the Mediterranean Sea in the North to theHorn of Africa and the Indian Ocean in the South-east. It consists of 21 countries and territories with a combined population of 360 million peoplestraddling North Africa and Western Asia. The last decade has witnessed an exponentialincrease of medical research in the area of malesexual dysfunction, coming from Arabic-speakingcountries. On the other hand, female sexual func-tion research in that part of the world is still in itsinfancy. Unlike the case of female sexual function,many validated Arabic tools to assess the malesexual function have been developed [1–3]. Thedelay in the advancement of female sexual functionresearch is attributed largely to the conservative3370  J Sex Med 2011;8:3370–3378 © 2011 International Society for Sexual Medicine  nature of the Arabic culture toward sex, in general,and female sexuality, in particular. The lack of an Arabic self-report instrument that demonstratesboth reliability and validity is another obstacle forthe advancement of female sexual medicineresearch.Withoutsuchatool,outcomeresearchinfemale sexuality will continue to lag behind that of male sexuality. Many self-report assessment tools have beendeveloped to help in identifying problems with the various components of female sexual response.Some of these tools are rather general scales focus-ing on all or several aspects of female sexualresponse like the Golombok–Rust Inventory of Sexual Satisfaction (GRISS) [4], the Brief Index of Sexual Functioning for Women [5], the DerogatisInterview for Sexual Functioning [6], the FemaleSexual Function Index (FSFI) [7], and SexualFunction Questionnaire [8]. Other scales focus ona particular population or a particular aspect of female sexual function, such as the HypoactiveSexual Desire Disorder (HSDD) Screener [9] forthe assessment of hypoactive sexual desire disor-ders in postmenopausal women, the DecreasedSexual Desire Screener [10] for identifyingHSDD, or The Female Genital Self-Image Scale[11] for the assessment of genital self-imagesatisfaction.FSFI, which was developed by Rosen et al. in2000, is a 19-item multidimensional self-reportingmeasure that quantifies six domains of femalesexual dysfunction (FSD), including desire,arousal, lubrication, orgasm, satisfaction, and pain[7]. It has been shown in several validation studiesthat FSFI is highly reliable and valid [12–15]. Thecontents are not too sexually explicit, which may make it suitable for use in the more conservative Arabic society. So far, FSFI has been translatedinto more than 20 languages, and it has becomethe de facto “gold standard” in the assessment of female sexual function and an indispensable tool inclinical research of FSD [16]. Objective  The objective of this study is to validate the Arabic version of the Female Sexual Function Index(ArFSFI). Methods Study Design and Setting   This is a cross-sectional study conducted between January and April 2010 at the outpatient clinic of Cairo University Hospital, a 4,500-bed hospitalreceiving referrals from all over Egypt. Study Population  A nonprobability (accidental sampling) method was used. Females willing to join the study andfulfilling the inclusion criteria were included in thestudy.Theinclusioncriteriawereasfollows:(i)agebetween 16 and 60 years; (ii) married and sexually active during the past 6 months; (iii) able to giveconsent; and (iv) able to read and understand Arabic. Instruments  The following instruments were used in the study:1 A sociodemographic profile form to obtaininformation about study participants, such asage, educational level, living environment, as well as medical and menstrual history. History of female genital cutting (FGC), and the typeand age at FGC was also obtained.2 The ArFSFI is the FSFI translated into the Arabic language.3 TheDiagnosticandStatisticalManualofMentalDisorders, 4th edition (DSM-IV), publishedby the American Psychiatric Association, is a widely used instrument for diagnosing psy-chiatric illnesses. DSM-IV was used as the“gold standard” against which the ArFSFI was validated [17]. Study Procedures  The study procedures are shown in Figure 1. Phase 1: Translation, and Face andContentValidity  The Arabic version of the FSFI was prepared usingthe back-translated technique [18]. During thetranslation process, simple formal Arabic wasadapted to make the questionnaire clear andunderstandable. The ArFSFI was then evaluatedfor face validity through a pilot study, in which it  was administered to 20 female medical students. It  was observed whether ArFSFI was “on its face” agood translation of the srcinal instrument. The ArFSFI was then presented to a panel of sexualmedicine experts for content validity assessment. The panel consisted of four experts from theDepartment of Andrology, Cairo University, withat least 20 years clinical experience in sexual medi-cine. Upon reviewing the panel’s comments, theauthors modified wordings that were not clear,confusing, or would not be accepted by Arabic women. Validation of Arabic Female Sexual Function Index   3371  J Sex Med 2011;8:3370–3378  Phase 2: Reliability andValidity Assessment  The final version of the ArFSFI was then adminis-tered to the sample group of 855 female attendeesof the Cairo University outpatient clinic on twooccasions. At the first visit, all participants whofulfilled the inclusion criteria were given an expla-nation about the study, and a written consent wasobtained.Participantswereassuredoftheanonym-ity and confidentiality of the data obtained. ThesociodemographicformandtheArFSFIwerefilledout in a room with some privacy. Upon completionof the ArFSFI, participants were interviewed by oneofthethreefemaleauthorsofthestudy,andtheDSM-IV was used to diagnose patients with sexualdysfunction. Reliability  Participants who were willing to come back wereasked to do so in 2–4 weeks after the first interview to fill out the ArFSFI for test–retest reliability. The ArFSFI scores (total score and scores of  various domains) from the two visits were com-pared using the paired  t  -test. The correlationbetween the two visits’ mean scores (total scoreand scores of each domain) was also measuredusing the Pearson correlation coefficient. The internal consistency of the ArFSFI wasexamined by calculating the Cronbach’s alpha values for assessing the relatedness of each item inevery domain. A Cronbach’s alpha  0.8 indicateshigh internal consistency [19]. Validity   The concurrent (criterion) validity was measuredusing the sensitivity and specificity of the ArFSFIagainst the DSM-IV as the “gold standard” instru-ment. Multiple cutoff scores from the ArFSFI were compared against DSM-IV diagnosis todetermine the most sensitive and specific cutoff score for the questionnaire to detect FSD. Thesensitivity and specificity data were used to pro-duce the receiver operating characteristic (ROC)curves. The ROC curve is a graphical plot of thesensitivity, or true positive rate, vs. false positiverate (1-specificity or 1-true negative rate), for abinary classifier system as its discrimination thre-shold is varied. The discriminant validity of the ArFSFI wascalculated with an independent   t  -test that com-pared the mean scores (ArFSFI total score andscores of various domains) between cases (women with sexual dysfunction) and noncases (women without sexual dysfunction). To assess the ArFSFI domain structure and toevaluate the construct validity, principal compo-nentanalysisusingvarimaxrotationwithextractingsix factors was used. Statistical Analysis  The data were analyzed with IBM ® SPSS ®  version16.0 for Apple OSX (SPSS Inc., Chicago, IL,USA). For test–retest reliability, the correlationbetween scores of the first and second visits wasdetermined using Pearson product–moment cor-relations. The scores of the two visits were alsocompared using paired  t  -test. Internal consistency reliability was estimated with the Cronbach’s alphacoefficient. Discriminant validity was assessed by comparing mean scores of cases and noncasesusing  t  -test for independent samples. For concur-rent validity, the ROC curve was used to deter-mine the best cutoff value of the ArFSFI todifferentiate cases and noncases, and the areaunder the curve (AUC) was calculated. We calcu-lated the sample size of case and noncase groupsfor AUC of 0.95 and a standard error of 0.01, andfound it to be equal to 250 in each group. Main Outcome Measures ParametersofvalidityandreliabilityoftheArFSFI,and its domains. The main outcome measures are the parametersof validity and reliability of the ArFSFI, and itsdomains. Translation of the FSFIPilot study - administered on 20 female medical studentsContent validity and experts review855 females recruited from women attending Cairo University outpatient clinicFirst visit Given information Signing written consent Completing sociodemographic data form  ArFSFI administered on 855 females Clinical interview with DSM-IV case N=644, noncase N=211Second visit  ArFSFI administered on 50 females for test–retest reliability analysis Figure 1  Study flow chart. 3372  Anis et al.  J Sex Med 2011;8:3370–3378  Results  The study included 855 women fulfilling theinclusion criteria. Two hundred and eighty-one of the participants (32.86%) presented for annualhealth checkup, 196 of the participants (22.92%) were accompanying family members, 294 of theparticipants (34.38%) were patients complainingof mild gynecological or dermatological illness,and 84 of the participants (9.82%) were volunteerfemale doctors and nurses. A summary of sociode-mographic characteristics of the study participantsis presented in Table 1. Reliability  Test–Retest Fifty patients (5.85%) were able to come back toretake the ArFSFI. Agreement between ratings at  visits 1 and 2 was found to be high. The Pearsonproduct–moment correlation coefficient (r) for thesix domains ranged from 0.922 to 0.983, as shownin Table 2. When the mean scores of each domain in thetest and retest were analyzed with a paired  t  -test,no significant differences were found between thetwo visits in the total ArFSFI score or any of the sixdomains (Table 3). Internal Consistency  Cronbach’s alpha statistic was used to determinethe internal consistency reliability of ArFSFIdomains. The analyses were conducted with thedata obtained from the first administration(N  =  855). Table 4 presents the Cronbach’s alpha values of ArFSFI domains. The Cronbach’s alpha valuesoftheArFSFIdomainsrangedfrom0.854to0.945, indicating that ArFSFI has good internalconsistency reliability. The range of Cronbach’salpha statistic is from  - 1 to  + 1. Generally, a Cron-bach’s alpha value less than 0.6 is considered pooror weak, 0.6–0.8 is considered moderate but satis-factory, and a value  0.8 indicates high internalconsistency [20]. FaceValidity  Nineteen of the 20 participants of pilot study (95%) reported that the ArFSFI was clearly  Table 2  Test–retest reliability of the Arabic version of theFemale Sexual Function Index DomainPearson(r) N  =  50Significant(two-tailed)Desire 0.922 0.000Arousal 0.980 0.000Lubrication 0.971 0.000Orgasm 0.983 0.000Satisfaction 0.961 0.000Pain 0.971 0.000Total score 0.983 0.000 Table 1  Sociodemographic characteristics of theparticipants Characteristics Count PercentageAge distribution < 20 years 116 13.620–30 years 440 51.5 > 30 years 299 35.0Education levelPrimary 115 13.5Preparatory 130 15.2Secondary 176 20.6Technical 28 3.3University 406 47.5Living environmentRural 245 28.7Urban 610 71.3Female genital cuttingYes 532 62.2No 323 37.8MenopauseYes 119 13.92No 736 86.08Duration of marriage (years) < 10 167 19.5310–20 512 59.88 > 20 176 20.58 Table 3  Test–retest mean scores and paired sample  t  -test DomainTest mean(SD) N  =  50Retest mean(SD) N  =  50Mean differenceSignificance2 tailsMeandifferences(SD)95% confidenceinterval of thedifferencet dfLower UpperDesire 4.164 (0.560) 4.212 (0.612)  - 0.048 (0.237)  - 0.116 0.020  - 1.429 49 0.159Arousal 4.290 (0.518) 4.314 (0.506)  - 0.024 (0.102)  - 0.053 0.005  - 1.661 49 0.103Lubrication 4.884 (0.437) 4.896 (0.421)  - 0.012 (0.104)  - 0.042 0.018  - 0.814 49 0.420Orgasm 4.736 (0.538) 4.744 (0.542)  - 0.008 (0.098)  - 0.036 0.020  - 0.573 49 0.569Satisfaction 5.488 (0.333) 5.464 (0.348) 0.024 (0.095)  - 0.003 0.051 1.769 49 0.083Pain 2.024 (0.474) 2.040 (0.458)  - 0.016 (0.113)  - 0.048 0.016  - 1.000 49 0.322Total score 25.586 (1.616) 25.650 (1.623)  - 0.064 (0.302)  - 0.150 0.022  - 1.497 49 0.141 Validation of Arabic Female Sexual Function Index   3373  J Sex Med 2011;8:3370–3378
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks