Design and data collection
This cross-sectional questionnaire-based study is reported according to Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (for checklist, see supplementary information)[11].
This population-based cross-sectional study was conducted with a 2-stage cluster sampling design in Sari, Iran. First, the necessary scientific permissions were attained from the Mazandaran University of Medical Sciences. Subsequently, approval was obtained from the Ethics Committee of Mazandaran University of Medical Sciences (IR.MAZUMS.REC.94.1734).
The Sample
The research was conducted with a 2-stage cluster sampling design. In the first stage, all the healthcare centers of Sari city were included in the study, as maximum heterogeneity was attained. In the second stage, reproductive age women were selected at systematic random sampling method from each center. The sample in each center was determined considering the probability of selection in proportion to population size (or estimated population size). In each center, a list of eligible women in each center was provided and numbered and a fixed sampling interval was determined. The interval was attained by dividing the population size by the desired sample size. Then the individuals were contacted and invited to participate in the study.
Participants
A total of 1000 reproductive age women from 20 health centers of Sari participated in the study, conducted between October and January 2015–2016. The study population was aware of the nature of the survey, and their verbal and written informed consent was achieved. Note that, none of the study participants were under the age of 16, so there was no need for parental or guardian consent.
Participants in this study included reproductive-aged women who have been married and lived with their sexual partner for at least 6 months and were willing to participate in the study. In addition, women who were pregnant or breastfeeding for the first six months, and those who went through premature ovarian failure were excluded from the study.
Outcome Measurements
Women who participated in the study were initially questioned on their socio-demographic characteristics including age, age of first intercourse, number of children, duration of marriage, body mass index, level of satisfaction with income, education level, physical activity, current smoking and alcohol use. Subsequently, women were asked to fill in the Sexual Interest and Desire Inventory-Female and Female Sexual Distress Scale-Revised forms.
Sexual Interest and Desire Inventory-Female (SIDI-F) were designed by Clayton et al in 2006 and consisted of 13 items plus 5 diagnostic items. The probable score range for each individual falls within 0 to 51, and a higher score indicates better sexual desire. The reliability of this tool has been excellent and the internal consistency coefficient of this tool was reported with the Cronbach's Alpha 0.90%. The validity of this tool was also examined by observing and correlating it with other valid instruments of sexual function[12]. Moreover, the validity and reliability of the Persian version of this tool was assessed in Iran. In the Persian version, the internal consistency was obtained with the Cronbach's Alpha 0.90% and the evaluation of the test-retest with a two week interval showed the proper reliability of this tool[13].
Since personal distress as a key factor is used in diagnosing HSDD, female sexual distress scale-revised version (FSDS-R) was used to assess it. This is a 13-item questionnaire with items marked on a 5-point Likert scale ranging from 0 (never) to 4 (always), with the higher score (11 and above) indicating greater sexual distress. The original version of FSDS-R has shown acceptable reliability with the Cronbach’s alpha falling within the range of 0.87–0.93% and a higher test-retest reliability (the internal correlation coefficient of 74–86%). For determining the reliability of the Persian version FSDS-R, internal consistency and reliability were assessed using test-retest method. Furthermore, Alpha coefficients were determined as a measure of internal consistency, while the reliability for correlation coefficients was assessed using test-retest measurement (repeatability index). The homogeneity of 70% and more was considered acceptable. Using test-retest method, the internal consistency and reliability were calculated to be > 0.70. Therefore, the Persian version FSDS-R is considered a valid and reliable tool for assessing distress related to sexual dysfunction in Iranian women and can be used for screening patients with HSDD[14].
Women with low Sexual desire (SIDI-F score ≤ 33.0) and sexually related personal distress (FSDS-R score ≥ 11) were classified as having HSDD. When combined with a requirement for personally related sexual distress, this provides a robust definition of HSDD for use in an epidemiologic context.
Statistical Methods
In the study on women aged 20–60 years, the reported prevalence of sexual desire problems using Female Sexual Functioning Index (FSFI) in 28 cities of Iran was 35% (p = 35%)[15]. With an estimated precision of 3% (d = 0.03) at 2- tailed 5% significance difference (α) (Z = 1.96), and a nonresponse rate of 10%, a sample size of 1000 women was obtained.
Statistical analyses were performed using SPSS software version 18. The distribution, mean, and SD values were initially obtained using descriptive statistics. The socio-demographic factors linked to LSD and HSDD were then obtained using the chi-square test.
A P value lower than 0.20 in chi-square test was adopted as representing the critical level for the selection of variables. Those variables with a P value greater than 0.20 were maintained as adjusting factors into the multivariate logistic regression analysis[16].
Multivariate regression analysis was conducted to estimate the strength of associations.
After adjustment in accordance with the variables in multivariate regression analysis, the associations between the variables investigated and HSDD and LSD for which P values of 0.05 or lower were considered statistically significant.