Participants and study design
In this study, we used a secondary analysis of data from a more extensive study that identified the menopausal pattern among a sample of Iranian women (Miandoab, West Azerbaijan, Iran). Participants were menopausal women who were randomly selected and recruited form health centers using the SIB (an abbreviation for Persian integrated health system) of household health files from 1 September to 30 November of 2018. The SIB uses the Electronic Health Record (HER) created for all Iranian people [15]. The SIB database was screened for menopausal women aged 45 to 65. The search criteria excluded women with a) mental and cognitive disorders, b) musculoskeletal disabilities, and c) surgical conditions. Menopausal status was defined based on the classification of stages in the Reproductive Aging Workshop (STRAW) [16], including that it had been 12 months since the last menstruation. Women were randomly selected from the original screened group and contacted by telephone to ask participate in the study and confirm eligibility. During the phone interview, the interested eligible women were invited to refer the health centers and participate in the study. A trained researcher conducted the interviews.
The study questionnaire
Menopause Rating Scale (MRS) is an 11-item instrument consisting of three dimensions: somatic symptoms (4 items), psychological symptoms (4 items), and urogenital symptoms (3items) [11]. The somatic symptoms include hot flushes, heart discomfort, joint and muscular discomfort, and sleep problems. The psychological symptoms include depressive mood, irritability, anxiety, and physical and mental exhaustion. The urogenital symptoms include sexual problems, bladder problems, and dryness of the vagina. Possible answers were based on the severity of symptoms, using a five-point Likert scale with 0=none, 1=mild, 2=moderate, 3=severe, and 4=very severe. The total severity ranged from a minimum of 0 to a maximum of 44 and was determined by adding the scores of the three subscales. In addition to the MRS response data, demographic information was collected and included age, education, occupation, marital status, menstruation age, menopausal age, and obstetrics history.
Translation
Because of the potential for the questionnaire to be influenced by the cultural context in which it was administered, a backward translation was applied [17]. Two bilingual health professionals, one Persian (the Iranian language) and the other English, independently translated the English version of the MRS into Persian. Then, a member of the research team (MM) produced a consolidated version for use in the survey. If there were differences between the two translated versions, the question was resolved through discussion with the translators to obtain a provisional unified translation. In cases where there was substantial disagreement, a third independent translator was engaged for additional review. Next, two independent English translators without previous knowledge of the questionnaire reviewed and translated the survey back to English to assess the comparability with the original English version and ensure that there were no discrepancies.
Face and content validity
As part of this study, qualitative face validity was implemented. A sample of menopausal women (n=8) was asked to assess the scale and give feedback for improvement. This process led to some changes in the wording of the scale. An expert panel evaluated the provisional Persian model of the MRS. The relevance and appropriateness of items to Iranian women and their cultural context were reviewed by three professors in midwifery, three in health education, one gynecologist, and one psychologist. The survey’s Content Validity Index (CVI) was evaluated by the panel using a four-point scale: 4 = very relevant, 3 = relevant with some revisions to wording, 2 = only relevant if the text is significantly revised, and 1 = irrelevant. They also suggested changes to improve the wording of each question. If a panel member rated any question less than 4, they were asked to recommend changes. According to the World Health Organization (WHO) recommendations, a CVI score greater than 0.79 confirmed content validity [16]. For the face validity and to improve clarity, the pre-final version of the questionnaire was evaluated by 10 menopausal women with the same study eligibility. In the end, no questions were deleted, meaning that the length of the Persian model of the MRS was similar to that of the original MRS.
Sample size
To obtain an optimal sample size, a ratio of 15 respondents to one question was used [19]. The sample size was calculated by multiplying the number of questions (11) in the MRS survey by the number of respondents (15); this resulted in a sample size of 165. Two sample sizes were considered for analysis in the validation process. The survey data from the first sample (n1=165) were used for a factor analysis (EFA). The second sample (n2=165) was used for cross-validation of the confirmatory model derived from the sample n=1 data. As a result, 330 eligible participants were invited to the study, and a total of 325 menopausal women completed the questionnaires (response rate of 96.9%). Table 1 summarizes the characteristics of the participants in the two samples.
Statistical analysis
The analyses were performed using the statistical program SPSS for Windows version 23.0 and Amos 24.0. To assess the sampling adequacy of the factor analysis, the Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s test of sphericity were applied. Any factor with an eigenvalue equal to one or above was considered significant for factor extraction. Where the loading criterion was 0.4 or more, a principal component analysis using varimax rotation was used for extraction in the factor analysis. The fit indices included the Comparative Fit Index (CFI), the Tucker-Lewis Index (TLI), the Root Mean Square Error of Approximation (RMSEA), and the Standardized Root Mean Square Residual (SRMSR). Cut-off points for inferring adequate fit indices were set at (CFI > 0.95; TLI > 0.95; Root Mean Square Error of Approximation (RMSEA), and Standardized Root Mean Square Residual (SRMSR) with acceptable values of zero to one.