The cross-cultural adaptation and psychometric properties of the menstrual health instrument (MHI) among Iranian adolescent females

Abstract

Background

Menstrual health encompasses a holistic state of physical, mental, and social well-being rather than simply the absence of illness associated with the menstrual cycle. It touches all aspects of human life. However, few tools are available to comprehensively examine the various areas of menstrual health. This study aimed to assess the psychometric properties of the menstrual health instrument (MHI).

Method

This study employed a cognitive method to psychometrically evaluate the English version of the Menstrual Health Questionnaire. This questionnaire consists of 29 items and 5 domains, including Affective symptoms, Somatic symptoms, school life, Daily habits for menstrual health, Menstrual cycle characteristics, Attitudes and perceptions on menstruation. It was conducted on 412 adolescent girls aged 13 to 18 from high schools in Babol City in 2023. Confirmatory and exploratory factor analyses were conducted in this study.

Results

A total of 412 adolescent girls from girls’ high schools in Babol City participated in this study. The mean age and mean age of menarche in participants were 14.20 ± 0.80 and 12.14 ± 0.99, respectively. The highest correlation among the 5 scales of the Menstrual Health Questionnaire was observed between emotional symptoms and physical symptoms and daily activities (r = 0.736, p < 0.001). In the content validity quantitative phase, all items of the original 29-question questionnaire had appropriate CVR and CVI and were retained in the study. The results of confirmatory factor analysis of the original questionnaire indicated that the hypothesized model has a reasonably good fit (Chi-square statistic = 846.09, CFI = 0.93, RMSEA = 0.057). The overall Cronbach’s alpha was 0.87. Finally, in the exploratory factor analysis within the same population for the initial 29 questions, 17 questions remained with 3 factors at the end of the structural validity phase, which explained 64.52% of the variance.

Conclusion

The Persian version of the menstrual health questionnaire, with three extracted factors, demonstrates appropriate validity and reliability for use in studies related to menstrual health among Iranian girls.

Introduction

In 2021, the World Menstrual Consortium defined menstrual health as “A condition characterized by comprehensive physical, mental, and social wellness, rather than simply the lack of illness, in relation to the menstrual cycle”. Attaining menstrual health necessitates a comprehensive framework that guarantees access to “information,” “health care,” a “stigma-free environment,” “involvement in all aspects of life,” and “WASH facilities and services” for women, girls, and all individuals who menstruate.” [1].

While menstrual hygiene concerns all adolescent girls, adult women, and individuals who experience menstrual cycles throughout their lives, adolescence marks the onset of menstrual hygiene needs and presents an opportunity to provide timely knowledge about menstruation and support self-care practices. Encouraging positive interactions with health care and educational services during this stage is crucial. On the present evidence, practice and policy initiatives have primarily focused on adolescents. As such, the need for age-specific indicators and measurements are required in order to further understand unmet needs at this life stage. Our focus underscores the potential for recommending evidence-based indicators targeting adolescents, kick-starting tracking of existing policies and programs, and immediately expanding efforts to better target additional populations [2]. Addressing adolescent menstrual health is essential for achieving sustainable development goals, including gender equality [3, 4].

There is growing evidence of the significance of menstrual experiences in the lives of women and girls, particularly highlighting the challenges faced, which are especially pronounced in low and middle-income countries [5, 6].

This self-perpetuating phenomenon can impact a woman’s mental well-being in multiple ways. To illustrate, many women suffer from certain physical conditions like dysmenorrhea, breast tenderness, and joint pains which are associated with their menstrual cycle [7]. These bodily manifestations often go with increased psychological distress, irritability, and lowered self-worth [8]. Additionally, a lot of women indicate an increase in interpersonal violence and a decrease in social engagement both in the pre-menstrual stage and during menses. This may exacerbate feelings of depression as well as social alienation [9]. Such negative affect is bound to be associated with increased impulsive behavior, substance use, and self-injurious behavior that is not aimed at death [10,11,12].

Menstrual Health issues are multi-dimensional– Right from the limited availability of hygienic absorbent products to inadequate clean and private spaces for cleaning/changing/ disposal facilities. Moreover, there is a paucity of well-rounded menstruation education and reproductive health; whilst menstrual disorders are under-diagnosed with less robust diagnosis and treatment options provided. Moreover, existing sociocultural taboos and traditional mindsets concerning menstruation have contributed to the stigmatization of menses which constrain social support systems [5, 13].

The 2021 progress report of this Joint Monitoring Programme, a partnership between the World Health Organization and UNICEF (the United Nations Children’s Fund), included a chapter on menstrual health that examined data from at least 42 countries measuring one or more of these four key indicators prioritized. These are, knowing menstruation and its time of onset, using sanitary products, private space for washing clean after the end letting nature take care without intervention supporting a healthy cycle to participate in activities during menstrual presence [14].

Even with the valuable insights provided by these initiatives, however, it is difficult to find a systematic methodology in their application of metrics that can cover all relevant factors and outcomes around menstrual health and hygiene needed for monitoring. Therefore, there is a need for a more comprehensive set of factors affecting menstrual health and hygiene that includes numerous and diverse elements. To accurately assess menstrual hygiene, the development and validation of new tools are needed to examine characteristics and address unmet needs [15, 16]. There are various perspectives on defining and implementing indicators for assessing menstrual health. Access to menstrual products alone is insufficient for achieving comprehensive menstrual health [17]. Generally, there are few tools available to comprehensively examine the various areas of menstrual health.

In 2023, Hengan et al. described a list of indicators for monitoring girls’ menstrual health and hygiene, including access to menstrual supplies, menstrual stigmas, and school hygiene and washing [2]. Moreover, other studies emphasized that well-designed psychometric surveys can provide more accurate information about menstrual patterns and related issues, aiding in therapeutic interventions and increasing public awareness [18, 19].

This tool will assess different aspects of menstrual health which certainly affect the quality of life of adolescents. Moreover, this instrument has 5 dimensions including affective symptoms, somatic symptoms and school life, daily habits for menstrual health, menstrual cycle characteristics, and attitudes and perceptions on menstruation. The detailed dimensions of this tool are one of the novelties. Evaluating individuals’ perceptions and attitudes regarding menstruation, as well as the effects of menstrual symptoms on both physical and emotional health, is essential for the effective screening and assessment of menstrual health [20]. Therefore, we decided to adopt the tool which was originally developed in Korea, to measure menstrual health and assess its validity and reliability in our own country.

Study design

This study employed a cognitive method to assess menstrual health Instrument among adolescent girls aged 13 to 18 from high schools in Babol City in Iran, 2023.

Translation process

The English version of the Menstrual Health Instrument (MHI) was developed by Shin et al. MHI is a self-report tool for adolescents in which the participants respond to questions related to their menstrual cycle in the past three months. The tool contains 29 items across 5 domains, which include affective symptoms (12 items), somatic symptoms (9 items), daily habits for menstrual health (3 items), menstrual cycle characteristics (3 items), attitude and perceptions on menstruation (2 items) anxiety and depression symptoms, somatic symptoms and school life, behaviors that contribute to menstrual health, characteristics of the menstrual cycle, and attitudes about menstruation and its effects. This questionnaire is divided into four categories based on the level of response to each item. One score ranges from 1 = Not at all to 4 = Very much. The somatic symptoms domain items and daily habits for menstrual health items are reversed. High points indicate better menstrual health than lower scores.

In this study, necessary correspondence was made with the MHI designer to obtain permission [21]. Then, the Persian version of the questionnaire was prepared according to the standard protocol of the World Health Organization [22]. This tool was translated using the forward-backward method. To begin with, two health-related translators translated the questionnaire into Persian. Taking the type of audience, their age, and cultural circumstances into account, the translators focused on conceptual translation rather than word-for-word translation. In the second stage, back-translation into English was done to validate the Persian version. Despite not being familiar with the questionnaire content, two proficient English translators translated the questionnaire back into English. In the third stage of the process, bilingual experts sat down with researchers to review translation quality. Alternative words were suggested when inconsistencies were found between translations. Finally, the resulting English version was sent to the author of the English version of MHI (Shin) for validation.

Item reduction

During this stage, MHI’s Persian version was examined for face and content validity (both qualitative and quantitative), structural validity, and reliability. This index was graded based on a 4-point Likert scale (a little, somewhat, considerable, and very much).

Face validity

Qualitative and quantitative methods were used to assess face validity. As part of the qualitative method, a questionnaire was distributed to 20 students from the target group who had similar characteristics as those of the target population. Additionally, some relevant experts were provided with the questionnaire. In order to assess the difficulty, relevance, and ambiguity of the items, participants were requested to provide comments. It was suggested that the items’ appearance be improved. For the quantitative methodology, the index was distributed to the same 20 individuals who participated in the qualitative evaluation of face validity. A scale of importance was presented to each student in which they had to rate the importance of each item using a 5-point system (5 = Very important to 1 = Not important at all). A formula was used to calculate the impact score for each item: Impact Score = Frequency (%) × Importance. A minimum impact score of 1.5 was required for an item to be retained [23].

Content validity

Both qualitative and quantitative methods were used to assess the content validity of the Persian MHI version. A qualitative method was used to assess the index by sending it to 10 experts in midwifery and women’s health. Moreover, they had experience translating tools and evaluating them psychometrically. In addition to checking for grammar errors, they checked that appropriate words were used, that items were positioned appropriately, and that scoring was accurate. Based on expert feedback, some items were edited.

An assessment of the content validity ratio (CVR) was made using the quantitative method. During this stage, a group of 10 experts was asked to rate the necessity of each item on a 3-point Likert scale (1 = Not Necessary, 2 = Useful but not necessary, and 3 = Necessary). Using the formula (ne - (N/2)) / (N/2), where “ne” represents the number of experts who rated the item as 3 (Necessary), and “N” represents the total number of experts, the CVR was calculated. In accordance with the Lawshe table, 0.62 was determined as the minimum acceptable value for the responses [24].

Following CVR determination, content validity indexes (CVIs) for the items were calculated. It was decided that the same 10 experts who were invited to determine CVR should rate the items on a three-point Likert scale (completely relevant, somewhat relevant, slightly relevant, and irrelevant) that was used for determining CVR. At least six experts must have a CVI value of 0.78 for it to be considered excellent, according to Owen et al. (2007) [25]. Content validity index had to be at least 0.78 to be considered acceptable in this study [26, 27]. Translation and back-translation were also used when cultural discrepancies were identified [28].

Construct validity

A descriptive study was conducted with a sample of 412 participants. Sampling was done through convenience sampling in girls’ high schools in Babol City. In this stage, stratified random sampling was employed. Accordingly, since Babol has 2 municipal areas, this city was divided into two strata (two municipal areas), and from each stratum’s list of eligible schools, two schools were randomly selected (a total of 4 schools were included in the study). Using the names of eligible students in each school, samples were then selected using a random selection method. The questionnaire was presented to them for completion.

Inclusion criteria included girls aged 13 to 18 who had menstrual bleeding for at least 3 years before the study, being healthy and not using hormonal medications or undergoing uterine or ovarian surgeries, and not having a known coagulation disorder, as stated by the individual at the time of the study. The informed consent form was obtained from the adolescent girls and their parents. The response to the questionnaire and the participation rate was 100%.

Before starting the study, the necessary approvals were obtained from the Department of Education and school administrators. Additionally, researchers have conducted training sessions for the administrators and school staff to ensure data protection.

Data on demographic variables, such as age, grade level, parents’ age and education level, age of menarche, and the Persian version of the MHI, were collected using the Cognitive and Clinical Population Questionnaire.

Confirmatory factor analysis (CFA)

The structural validity of this questionnaire was examined using Confirmatory Factor Analysis (CFA). CFA was conducted with 412 samples using maximum likelihood estimation. Model fit was evaluated by examining indices such as root mean square error of approximation (RMSEA) < 0.08, chi-square (χ2) test, chi-square/degree-of-freedom ratio (χ2/df) < 3.0, standardized root mean square residual (SRMR) < 0.10, comparative fit index (CFI) > 0.9, parsimonious comparative fit index (PCFI) and parsimonious normed fit index (PNFI) > 0.5, for close fit of the population RMSEA (PCLOSE) > 0.05, adjusted goodness-of-fit index (AGFI) > 0.8, incremental fit index (IFI) and Tucker-Lewis fit index (TFI) > 0.9 [29, 30].

Convergent and divergent validity

To determine whether the Persian version of the MHI was valid convergently and divergently, Fornell and Larcker’s approach was used. Averaging the extracted variations and calculating the Maximum Shared Squared Variance (MSV) were calculated to determine whether the extracted factors had convergent and discriminant validity. Convergent validity is established when the Average Variance Extracted (AVE) is greater than 0.5 and the Composite Reliability (CR) is greater than 0.7. A construct’s MSV must be lower than its AVE to satisfy discriminant validity criteria [31].

Reliability

The entire questionnaire and its scales were analyzed by Cronbach’s Alpha [32], and values above 0.70 were considered satisfactory [33]. Additionally, stability assessment was conducted using the test-retest method based on data collected from 30 students who met the study criteria. The questionnaire was completed twice by the students, separated by two weeks each time. Then, the scores obtained in these two stages were evaluated using the intraclass correlation coefficient (ICC). The interpretation of the figures is as follows: ICC values between 0.40 and 0.59 are considered acceptable, between 0.60 and 0.74 are considered good, and above 0.74 are considered excellent [34].

Exploratory factor analysis (EFA)

As an additional analysis to validate the structure of the Persian version of MHI, Exploratory Factor Analysis (EFA) was conducted. In order to carry out the EFA, Maximum Likelihood was used with Promax rotation. A Kaiser-Meyer-Olkin (KMO) test and the Bartlett Test of Sphericity were conducted in order to assess whether the sample was adequate and suitable for analysis. A KMO value of more than 0.70 was considered to be acceptable [30].

In this study, factors were extracted based on Eigenvalues and Scree Plot [35, 36]. It was determined that factors that had Eigenvalues greater than 1 were considered to be suitable in the study and retained. A minimum factor loading of 0.3 was considered adequate for each item to be included in the factors. The formula CV = 5.152 ÷ √ (n– 2) was used to determine the threshold. It should be noted that n is considered as sample size and CV is the minimum factor loading. A further exclusion from the EFA was made for items with commonalities that are less than 0.2 [30].

Data analysis

Statistical analysis was performed using SPSS 24. Using Amos version 24, CFA was conducted on the Persian HIP version in order to verify construct validity. The SPSS 24 was also used to calculate Cronbach’s alpha. Furthermore, exploratory factor analysis with the maximum likelihood method was used in SPSS 26.

At the end, the questions of each scale along with the total score of that scale were analyzed using Pearson correlation coefficient.

Ethical considerations

In accordance with the Ethics Committee of Babol University of Medical Sciences, this research has been approved (IR.MUBABOL.REC.1402.038). A written consent form was signed by all participants, and the participants’ rights were protected at all times (all data collected was kept anonymous and confidential).

Results

In this study, 412 adolescent girls from girls’ schools in Babol County participated. The average age of the participants was 14.20 ± 0.80 years, and the average age of onset of menstruation among participants was 12.14 ± 0.99 years. Table 1 shows the demographic characteristics of the participants. Additionally, the correlation between the 5 scales of the Menstrual Health Inventory (MHI) questionnaire was examined, showing moderate to strong and significant correlations between the total score and the 5 dimensions of the questionnaire, except for the Daily Habits dimension (p-value < 0.05)). Furthermore, the highest correlation among the 5 scales was observed between the Emotional Symptoms, Physical Symptoms, and Daily Activity scales (r = 0.736, p < 0.001) (Fig. 1).

Table 1 The participants’ characteristics

Fig. 1

Menstrual Health Instrument subscales correlation

Item reduction

Face validity AND content validity

In the qualitative method, based on the suggestions of experts, changes were made to the appearance of the items. The results of the face validity phase quantitatively showed that the score of all items was above 1.5. In the content validity assessment, qualitatively, based on the experts’ opinions, the items were edited. In the quantitative phase of content validity, all items were found to be appropriate in terms of CVR and CVI and were retained in the study.

Construct validity

Confirmatory factor analysis (CFA)

The model fit indices resulting from confirmatory factor analysis showed that the hypothesized model had a good fit (Fig. 2; Table 2). The results of the first-order factor analysis convergent validity showed that the factors had adequate convergent validity. However, the first to third factors did not have adequate discriminant validity, so a higher-order confirmatory factor analysis was needed (Table 3). The second-order confirmatory factor analysis was performed, and its model fit indices are presented in the second row of Table 2 (Fig. 3).

Fig. 2

Confirmative Factor Analysis Model for MHI Questionnaire (First order CFA)

Table 2 Fit model indices of the CFA of the Persian version of the MHI

Table 3 The indices of convergent and discriminate validity (First order confirmatory factor analysis)

Fig. 3

Confirmative Factor Analysis Model for MHI Questionnaire (Second order CFA)

The results of the confirmatory factor analysis showed that the chi-square statistic was 846.09, and the p-value was significant. Additionally, the CFI value above 0.9 was considered good (CFI = 0.93), and the RMSEA value was less than 0.08 (RMSEA = 0.057), thus indicating acceptable fit indices.

Reliability

The stability of the questionnaire was evaluated using the test-retest method to assess the intraclass correlation coefficient (ICC) for each dimension. The correlation values ranged from 0.62 to 0.93, which, based on standard interpretations, is considered to be in the good to excellent range.

The reliability results indicated that the scales and the overall questionnaire had an appropriate status in terms of stability. From the perspective of internal consistency, the overall questionnaire had an appropriate status but scales three and four did not have satisfactory internal consistency. This was also indicated by the CR results from the confirmatory factor analysis (Table 4).

Table 4 Psychological properties of the menstrual health instrument questionnaire

Exploratory factor analysis (EFA)

In this study, exploratory factor analysis was conducted with 412 samples as a secondary analysis. Before conducting construct validity, content analysis was performed. Six items were removed due to a correlation of less than 0.32. In this phase, 23 items were retained for construct validity.

To conduct exploratory factor analysis, the Kaiser-Meyer-Olkin (KMO) test was performed first. Then, the Bartlett Test of Sphericity was used to determine if the correlation between variables was not equal to zero. The results showed that the KMO measure of sampling adequacy was 0.947, and the Bartlett’s test (df = 136, p < 0.001, χ^2 = 3803.177) was significant.

In this model, three factors were extracted based on eigenvalues greater than one and the scree plot (Fig. 4). Items with Factor Loading ≤ 0.3 and communalities < 0.2, as well as those with cross-loading, were excluded from the analysis. At the end of the construct validity phase, 17 items remained in the index. These three factors ultimately explained 64.52% of the variance. Table 5 shows the results of construct validity and the variance of each factor.

Fig. 4

Scree Plot in exploration factor analysis to identify the number of factor loads (MHI Questionnaire)

Table 5 Factor loads for MHI questions in exploratory factor analysis using varimax rotation

Discussion

This study aimed to determine the validity and reliability of the Persian version of the menstrual health questionnaire for Iranian adolescent girls. The questionnaire demonstrated face and content validity. Factor analysis revealed that the remaining 17 items encompassed three factors: Affective symptoms, menstrual cycle characteristics, attitudes, and perceptions of menstruation as well which together explained 64.52% of the variance in menstrual health. In multidimensional instruments, a variance explanation above 30% is considered sufficient [37]. The fit indices of the model in the confirmatory factor analysis were also favorable.

The initial factor identified was psychophysical symptoms, encompassing affective symptoms, somatic symptoms, and daily habits for menstrual health. This factor represented 36.91% of the overall variance. Hormonal fluctuations throughout the menstrual cycle lead to various physical discomforts for many women, including dysmenorrhea, breast tenderness, and joint pain [7], during menstruation. Such physical discomfort can exacerbate mental distress and irritability, while also diminishing self-esteem. In addition, many women describe increased relationship difficulties and decreased social interaction just before the onset of menses; thus feelings of depression/rejection are further exacerbated [9], substance use is also elevated during this period [10], as well as any associated self-harming behaviors [11].

In addition to the emotional and behavioral effects of the menstrual cycle, there are several direct biological effects on mental health. For example, estrogen reduces dopamine transmission, mimicking the anti-dopaminergic action of many antipsychotic drugs. Higher levels of estrogen are hypothesized to have protective effects against symptoms such as psychosis. Consequently, increased vulnerability to psychosis has been observed when estrogen levels are low, such as during menstruation and the postpartum period [38]. Progesterone can have anti-anxiety effects by increasing allopregnanolone and subsequently enhancing GABA reinforcement [38,39,40,41]. However, other progesterone metabolites are not anxiolytic. Under stress, progesterone converts to cortisol, which increases stress responses and disrupts emotional processing. Therefore, progesterone may contribute to mood symptoms related to menstruation [42].

Several studies have shown that menstruation is associated with the worsening of symptoms of mental disorders [43] addictive behaviors [43], psychosis [38], suicide [44], and post-traumatic stress disorder [45].

The second extracted factor of concern was the menstrual cycle characteristics, which accounted for 2.85% of the total variance. This factor includes three items: regularity of menstruation, cycle length, and the individual’s evaluation of their menstruation. The regularity of menstrual cycles appears to significantly influence a person’s perception of menstrual health, affecting quality of life and ultimately impacting their overall sense of health [46].

Menstrual cycles in early adolescence are often irregular due to anovulatory cycles. Teenagers should pay attention to the regularity of their menstrual cycles [47]. The regularity and length of the menstrual cycle should be considered vital signs for evaluating menstrual health in girls and adolescents [48]. Causes of abnormal menstruation can vary among different age groups and include hormonal imbalances, pregnancy, cancers/chronic diseases, infections (sexually transmitted or otherwise), medical conditions outside female reproductive organs such as thyroid disease/wellness/disease states that may alter hormone metabolism and balance in the ovulatory endocrine system systemic involvement by male sex hormones too high either physiologically present condition results with clinical hirsutism acne breakout soft smooth fine hair growth dermal snaking [49]. Menstrual irregularities are also strongly associated with environmental factors and aspects of a modern lifestyle, like the consumption (versus non-consumption) of, coffee or alcohol, physical activity levels, stress Smoking -- cigarette smoking ultimately causes cancer age switch to likely; that follows natural aging obesity sudden increase body weight diet food choices [50].

The third factor extracted was attitude and perceptions on menstruation, accounting for 1.843% of the total variance. Items related to this factor include statements like “I experience pain and discomfort in the lower abdomen during menstruation” or “I have severe cramps during menstruation”, which are related to the concept. Women’s perceptions of their menstruation are often overlooked in discussions of menstrual health [51]. Attitudes and perceptions towards menstruation vary significantly across different cultures. Some social and cultural taboos and restrictions regarding menstruation still exist [52,53,54].

Diverse cultural interpretations of menarche include sexual maturation and femininity, both positive and negative symptoms like pain or interference with activities in daily life [54]. These attitudes can greatly affect how women process menstruation (how bad it is for them) [55]. Menstruation remains taboo in many countries, where discussing it is considered shameful [54, 56].

In terms of reliability, the extracted factors demonstrated good internal consistency. Cronbach’s alpha coefficient was 0.7 for all factors, and the AIC (average inter-item correlation) for all items fell within the appropriate range of 0.2–0.4 [57]. This adequate reliability suggests that the internal stability of the items within each factor adequately explains the underlying concept.

Limitations of the study

This study sampled participants exclusively from high schools in one city (Babol), which may not adequately represent the entire Iranian adolescent population. To enhance generalizability, future studies should include samples from various cities across Iran. Additionally, the exclusion of adolescents who do not attend school limits the study’s representation of society as a whole. Future research should aim to include a broader spectrum of teenagers from different societal backgrounds and regions.

Conclusion

The Persian version of the menstrual health questionnaire, with three extracted factors, demonstrates appropriate validity and reliability for use in studies related to menstrual health among Iranian girls.

Application of findings

Healthcare providers in school or primary care settings can utilize this questionnaire to assess adolescent menstrual health, identifying those with low scores as potential candidates for interventions focused on lifestyle changes and symptom management. The questionnaire also holds research potential, enabling evaluation of adolescent responses to interventions in clinical trials. Future studies using the MHI could explore its predictive validity and establish appropriate cut-off scores. However, further research is recommended to enhance generalizability and strengthen the psychometric validity of this instrument, including predictive validity assessments and determination of optimal cut-off scores. Extensive validation across diverse cultural settings would facilitate cross-cultural comparisons and inform targeted interventions.