Self-Care Behavior and Associated Factors of Nursing Students with Dysmenorrhea: A Structural Equation Model

Objective . To assess the factors infuencing the self-care behavior among nursing students with dysmenorrhea. Background . Te practice of self-care behavior for dysmenorrhea has gradually attracted immense attention from society; however, thus far, only a few studies have been conducted to predict this behavior and analyze the associated factors by creating a structural equation model. Methods . A cross-sectional multistage cluster sampling study was conducted among nursing students within six universities in Shaanxi province, China. A model was constructed, and structured questionnaires were adopted to measure model variables, including e-health literacy, negative emotion, self-efcacy, self-care agency, degree of dysmenorrhea, and self-care behavior for dysmenorrhea. Descriptive data analysis was performed using SPSS 23.0 software, and AMOS 23.0 was used to verify and analyze the structural model. Results . In total, 1851 valid questionnaires were collected; the efective recovery rate was 93.15%, and the prevalence of dysmenorrhea was 64.51%. e-Health literacy ( B � 0.171, P < 0.001), self-efcacy ( B � 0.416, P < 0.001), self-care agency ( B � 1.177, P < 0.001), and degree of dysmenorrhea ( B � 0.310, P < 0.001) signifcantly infuenced self-care behavior for dysmenorrhea. Te total, direct, and indirect efects of e-health literacy and self-efcacy on self-care behavior for dysmenorrhea were 0.158 and 0.492, 0.128 and 0.248, and 0.030 and 0.244, respectively. Conclusion . Te self-care behavior for dysmenorrhea is afected by several factors and self-efcacy has the greatest efect on it. To promote girls to actively implement self-care behavior for dysmenorrhea, educators should strengthen the training of self-efcacy and self-care agency of the nursing students to alleviate the uncomfortable experience brought by dysmenorrhea and decrease the harm of dysmenorrhea. Implications for Nursing Management . Nursing managers should work with constant eforts to explore and optimize the management model for dysmenorrhea, encouraging young women to actively engage in self-care behavior for dysmenorrhea, to alleviate the discomfort experienced by individuals and improve women’s overall health.


Background
Dysmenorrhea is defned as recurrent spasmodic pain in the lower abdomen during menstruation and is one of the most common causes of pelvic pain and menstrual disorders [1].According to relevant reports, 56.4%-90% of women have dysmenorrhea [2][3][4], which is the main cause of their absenteeism from school and work [5,6].Dysmenorrhea is clinically divided into primary and secondary dysmenorrhea [7].From the viewpoint of individual physical and mental development, severe dysmenorrhea can interfere with their daily activities as well as cause varying degrees of harm to their life and quality of life [8,9], thus adding to a huge public health burden [10].Conversely, from the social and economic viewpoint, absenteeism caused by dysmenorrhea [11] causes an annual loss of 140 million working hours in the United States and economic losses of over $4.2 billion in Japan, thus making dysmenorrhea one of the important issues that need to be urgently addressed by researchers globally.Te level of self-care agency of patients with dysmenorrhea is closely associated with their dysmenorrhea status [12].Terefore, one of the most important approaches for alleviating dysmenorrhea symptoms and reducing the harm of dysmenorrhea is whether individuals can fully take the initiative and be enthusiastic to implement self-care behaviors benefcial for their health.
According to the latest revision of the WHO Guidelines on self-care interventions [13], self-care is defned as the ability of individuals, families, and communities to promote health, prevent disease, maintain health, and cope with illness and disability, with or without the support of health workers.As a new primary healthcare approach [14], selfcare has the potential to save health system resources and reduce patient care and follow-up costs [15,16]; furthermore, it is cost-efective and easy to learn [17].Self-care has been widely used in postpartum women, those with spinal cord injury, those who underwent cancer surgery, and those with chronic diseases and plays an important role in the clinical outcome and prognosis of patients with chronic diseases [18][19][20][21][22][23][24][25].Meanwhile, nursing students, as a special group, have been shown that the implementation of self-care behavior can not only help nursing students relieve pressure and make them better transition to nurse status but also help nurses and patients jointly create a safe clinical environment [26].However, there are few research studies on the combination of nursing students' dysmenorrhea and self-care behavior as the research direction, the existing research studies are mostly single-factor or cross-sectional investigations, and the holistic research based on the scientifc nursing theory is relatively lacking.Terefore, in this study, taking Orem's self-care theory as the theoretical framework, a structural equation model was created to conduct a multidimensional investigation and analysis of the self-care behavior of nursing students, discussed the status quo of nursing students self-care behavior for dysmenorrhea and related infuencing factors, and provided reference and basis for medical personnel to frame efective intervention measures and guide these nursing students in actively implementing self-care behavior.
1.1.Literature Review.Dysmenorrhea is one of the most common gynecological disorders afecting the quality of life and social activities of women [27].However, only 20.8% of the students choose to seek medical help for alleviating the discomfort and pain caused by dysmenorrhea; the vast majority of students prefer self-care [28].Te self-care behavior for dysmenorrhea is afected by many factors, among which self-care agency has the greatest impact on it, demonstrating a positive correlation and direct efect [29,30].Wong et al. [30] pointed out that previous knowledge regarding menstruation has direct and indirect efects on self-care behaviors for dysmenorrhea and that there is a direct efect between the mother's education level and degree of dysmenorrhea and self-care behaviors for dysmenorrhea.Te higher the e-health literacy [31,32] and the better the individuals' self-efcacy [33,34], the more willing they are to adopt self-care behaviors that can improve their health status and alleviate their pain.However, thus far, the existing studies on self-care for dysmenorrhea are mostly single factor, and there is a lack of systematic research on the relationship between multiple variables based on scientifc theories and models.Terefore, to help medical staf identify the potential or existing wrong self-care behavior of patients with dysmenorrhea as early as possible and improve self-care agency, it is important to assess the self-care behavior of these patients.

Structural Equation Model. Tis study is guided by
Orem's self-care theory, which considers that each individual can independently implement self-care behavior [35] and that nursing intervention aims to help individuals improve their self-care agency [36].
Orem's self-care theory comprises self-care, self-care defect, and nursing system theories [37].Te self-care theory can provide the foundation for explaining self-care behavior, assessing the relationship between relevant variables, and predicting the efect of self-care intervention.Tis theory points out that self-care agency is an important requirement for individuals to implement self-care behavior and that its strength is closely associated with basic condition factors (BCFs).BCFs include age, sex, developmental status, health status, sociocultural context, health reasons (e.g., medical diagnoses and therapeutic interventions), family factors, lifestyle, environmental factors, and availability [38].All BCFs can afect the level of self-care agency and are used in self-care behavior, that is, BCFs afect selfcare behavior through self-care agency [37].
Resource availability and its adequacy are one of the BCFs that have a direct impact on self-care agency [37].e-Health literacy as a refection of the ability to acquire and apply network resources demonstrates that [39,40] the level of ehealth literacy of individuals directly afects their ability to care for poor health.Te pain intensity during dysmenorrhea is a distress signal released by the body after an individual's health condition is compromised; the pain degree of women with dysmenorrhea is signifcantly associated with their selfcare behavior [30,41].Terefore, in the model framework of the present study, from Orem's self-care theory, resource availability and the health status were selected as exogenous variables and self-care agency and self-care behavior as media and outcome variables, respectively.
Furthermore, the path relationship between the factors was established based on the fndings from the existing research.Negative emotional experiences can decrease the pain threshold and increase the pain degree of individuals, leading to a vicious cycle of negative emotional responses to dysmenorrhea [42]; nonetheless, negative emotions have a signifcant negative relationship with self-care agency [43,44].Individuals with high self-efcacy are frequently able to actively learn relevant knowledge, take correct and positive nursing measures, improve their self-care agency, and alleviate their pain and discomfort through continuous learning [45].Tese individuals also display higher self-care abilities and implement self-care behaviors that are favorable to symptom relief or improvement of their condition [46].Although Orem's self-care theory does not highlight the association among negative emotion, self-efcacy, and degree of dysmenorrhea, previous studies have confrmed the signifcant association between the abovementioned variables; therefore, 2 Journal of Nursing Management the path generated by the abovementioned variables was included in the present model.To sum up, the following hypothesis model was proposed after integrating Orem's selfcare theory with the fndings of previous studies (Figure 1).

1.3.
Purpose.Tis study aimed to create a structural equation model and use it to clarify the infuencing factors and the relationship between the variables of the self-care behavior of nursing students with dysmenorrhea in China to provide a theoretical basis and reference for medical staf and educators to better provide dysmenorrhea-related diagnosis and develop treatment strategies and scientifc protective behaviors.

Overview.
In this study, a cross-sectional multistage cluster sampling survey was conducted among nursing students from six universities in Shaanxi province, China.Te structural equation model was used to assess the relationship between various variables and self-care behaviors for dysmenorrhea.

Participants and Data Collection. From February 2023 to
April 2023, a multistage cluster sampling method was used for recruiting students.First, Shaanxi province in northwest China was chosen as the sampling unit among the seven major geographical regions in China.Second, according to the size and category of the university, six medical colleges were chosen from 15 universities ofering undergraduate nursing majors in Shaanxi province and using the convenience sampling method to select eight classes in frst-and second-year nursing undergraduate students from each medical college, resulting in a total of 48 classes.Finally, all nursing students from the selected classes who met the inclusion criteria were taken as subjects.Inclusion criteria were as follows: (1) consistent with the diagnosis of dysmenorrhea [47] and (2) not having a psychiatric illness.Exclusion criteria included communication difculties, poor compliance, and inability to complete the questionnaire.
According to the calculation method of the sample size of the structural equation, the baseline sample size was 15-20 times the predictive variables [48,49] and the number of measured variables in this study was 23.By calculating 20 times the measured variables, 460 subjects should have been selected.In this study, it was calculated by the following equation: (3 + 3+4 + 4+3 + 6) * 20 � 460.A total of 1,987 questionnaires were sent to six schools in this study, and 1,851 valid questionnaires were fnally recovered after excluding those that did not meet the inclusion criteria.Te efective recovery was 93.15%.

e-Health Literacy Scale
. Te e-Health Literacy Scale [50], which includes eight items and wherein each item adopts the Likert 5-level scoring method, including three dimensions, application ability, evaluation ability, and decision-making ability of online health information and services, was used as an assessment tool.Te Cronbach's α coefcient of the Chinese version of the scale was 0.913 [51], and Cronbach's α was 0.960 in this study.

Depression Anxiety Stress Scale-Simplifed Version.
A simplifed version of the Depression Anxiety Stress Scale [52], which includes depression, anxiety, and stress with a total of 21 items, was used to measure the negative emotions of the study subjects.Te Likert 5-level scoring method was used.In this study, the Cronbach's α coefcient of the Chinese version of the scale was 0.890 [53] and Cronbach's α was 0.969.

Self-Rated Abilities for the Health Practices Scale.
Te Self-Rated Abilities for Health Practices Scale compiled by Becker et al. was used to measure self-efcacy [54].Tis tool was translated from English into Chinese with a total of 28 items and four dimensions that represented the ability to promote healthy behavior in four aspects, namely, nutrition, exercise, psychological well-being, and health responsibility.Te Likert 5-level scoring method was used.In this study, the Cronbach's α coefcient of the Chinese version of the scale was 0.950 [55] and Cronbach's α was 0.976.

Self-Care Agency Scale. Te Exercise of Self-Care Agency Scale translated into Chinese by Kearney and
Fleischer and [56] and Taiwan scholars [57] as a research tool with a total of 43 items and four dimensions, including the health knowledge level, self-concept, self-care responsibility, and self-care skills, was used.Te Likert 5-level scoring method was used.In this study, the Cronbach's α coefcient of the Chinese version of the scale was 0.890-0.920[57] and Cronbach's α was 0.931.

Degree of Dysmenorrhea Scale. Te Menstrual Distress
Questionnaire, developed by Moos [58], which includes two subscales of physical dysmenorrhea symptoms and mental Journal of Nursing Management and psychological symptoms before menstruation, each with three dimensions and a total of 30 items, was used.Te subscale of physical dysmenorrhea symptoms in the Menstrual Distress Questionnaire, including pain-related symptoms, autonomic nervous disorder symptoms, and water and sodium retention symptoms, with a total of 14 items, was selected.Te Likert 5-level scoring method was used.Te higher the score, the more severe the dysmenorrhea.In this study, the Cronbach's α coefcient of the Chinese version of the scale was 0.937 [59] and Cronbach's α was 0.943.
2.3.6.Adolescent Dysmenorrhea Self-Care Scale.Te Adolescent Dysmenorrhea Self-Care Scale developed and compiled by Hsing et al. [60], and translated into Chinese and Cantonese by Wong et al. [61] (Hong Kong, China), is a research tool for measuring self-care behavior of nursing students with dysmenorrhea.Te scale includes six dimensions of knowledge acquisition, emotional expression, seeking help, control of external factors, resource utilization, and self-control, with a total of 35 items.Te Likert 6-level scoring method was used.In this study, the Cronbach's α coefcient of the Chinese version of the scale was 0.940 [61] and Cronbach's α was 0.973.

Data Analysis.
Te data collected in this study were directly exported from the questionnaire platform.IBM SPSS Statistics version 23 software and IBM SPSS AMOS version 23 software were used to analyze.Te measurement data were represented by (x ± s); the counting data were statistically described by frequency and probability.Te structural equation model was composed of two major parts as follows: the measurement model and the structural model.In the measurement model, confrmatory factor analysis (CFA) and the Pearson correlation coefcient were used to test the reliability and validity.In the structural model, the comparative ft index (CFI > 0.90), normed ft index (NFI > 0.90), Tucker-Lewis index (TLI > 0.90), and root mean square error of approximation (RMSEA < 0.06) were analyzed for model ft [62][63][64].Bootstrapping analysis was used for the indirect efects, total efects, and statistical signifcance of the model, and the bootstrap ML method was utilized; the results of this method are more stable [65] and more accurate than other methods in studies with a sample size greater than 200 [66].Te 95% bias-corrected confdence intervals were used and the number of repeated samples was 5000 times.Two-sided P values less than 0.05 were considered signifcant.

Ethical Considerations.
Tis study has been approved by the Ethics Committee of the Xi'an Medical University (number: XYYJSLS2022067).Te participants voluntarily enrolled in this investigation and provided signed informed consent before the study.

Characteristics.
Te number of valid questionnaires in this study was 1,871, of which 1207 cases (64.51%) had dysmenorrhea.Te average age of the subjects was 19.97 ± 1.27 years and approximately half of the girls had menarche at the age of 13-14 years (49.3%).Among the nursing students with dysmenorrhea, 186 (15.4%) had very irregular menstrual periods and the average number of menstrual days was 5-7 days in 826 (67.6%).Other general information is shown in Table 1.

Measurement Model.
First, the original measurement model was verifed, and the measurement variable with a standard factor load value below 0.7 was deleted [67,68], that is, the self-control dimension of dysmenorrhea self-care behavior, resulting in a fnal measurement model comprising six latent variables and twenty-two measured variables.
Second, confrmatory factor analysis was conducted on the fnal measurement model.Te results indicated that all factors had an average variance extracted (AVE) greater than 0.5 and construct reliability (C.R) higher than 0.7 [69,70], demonstrating good convergent validity of the data in this study (Table 2).
Finally, based on Pearson correlation analysis, the selfcare agency for dysmenorrhea was signifcantly positively correlated with e-health literacy, self-efcacy, self-care agency, and degree of dysmenorrhea (r � 0.318, P < 0.001; r � 0.503, P < 0.001; r � 0.524, P < 0.001; r � 0.058, P < 0.05) and negatively correlated with negative emotion (r � −0.098, P < 0.01) (Table 3).Te results of the discriminant validity test in the measurement model demonstrated that the correlation coefcients between variables were all lower than the square root of AVE, indicating good discriminant validity of the model (Table 3).

Structural Equation Model.
Te theoretical framework (Figure 1) drew a total of nine paths; the hypothetical model is constructed based on the framework, and the result of the model ft showed that RMSEA � 0.062, CFI � 0.961, NFI � 0.953, and TLI � 0.954.According to the results of the software correction index, the model was further modifed, the path between self-efcacy and dysmenorrhea in the hypothetical model was removed, and the fnal model with a total of eight paths was formed and analyzed, as shown in Figure 2. Te ft index of the modifed model showed the following: RMSEA � 0.058, CFI � 0.967, NFI � 0.959, and TLI � 0.962.
Te parameter estimation table of the modifed model for the self-care behavior of nursing students with dysmenorrhea is shown in Table 4. Tis study showed that eight out of nine pathways were statistically signifcant, whereas the path from self-efcacy to dysmenorrhea showed no statistical signifcance.Among them, the explanation rate of the degree of selfcare behavior explained by self-care agency, degree of dysmenorrhea, e-health literacy, and self-efcacy was 41.7%, that of self-care agency explained by e-health literacy, negative emotions, and self-efcacy was 53.7%, and that of the degree of dysmenorrhea explained by negative emotions was 32.1%.

Efects.
In the present study, self-care agency had the greatest efect on the outcome variable self-care behavior for dysmenorrhea and self-efcacy had the greatest efect on the media variable self-care agency.Te direct and overall efects 4 Journal of Nursing Management Journal of Nursing Management of the eight pathways and the indirect efects of e-health literacy and self-efcacy on self-care behavior for dysmenorrhea were statistically signifcant (Table 5).

Current Situation of the Self-Care Behavior of Nursing
Students with Dysmenorrhea.According to the general characteristics of the population included in this study, approximately half of the girls had menarche at the age of 13-14 years (49.3%) and the average number of menstrual days was 5-7 days in 826 (68.4%).Te incidence of dysmenorrhea was about 64.51% (1207/1871), which was similar to the fndings of a previous study [71], indicating that dysmenorrhea is still a major problem that should be paid more attention to during the healthy development of female body and mind.Most individuals with dysmenorrhea have menarche at the age of 13-14 years, accounting for 49.5%.It is suggested that schools should form a relationship education mechanism with local primary healthcare departments and parents of students and conduct activities including lectures or exchange forums related to menstruation to establish correct cognition of normal physiological phenomena such as menstruation in individuals and encourage them to take positive and correct self-care behaviors for dysmenorrhea spontaneously, thus helping them alleviate the discomfort caused by dysmenorrhea.

Fit Degree of the Structural Equation Model for the Self-Care Behavior of Nursing Students with Dysmenorrhea.
Tis study took Orem's self-care theory as the core framework.
After the verifcation and modifcation of the structural model, the ft degree of the structural equation model for the self-care    Journal of Nursing Management behavior for dysmenorrhea had indirect efects.Te variables that directly afected self-care agency were e-health literacy, negative emotion, and self-efcacy; the explanation rate of these variables was 53.7%.Furthermore, negative emotion was a factor that directly afected the degree of dysmenorrhea; the explanation rate of this factor was 32.1%.

Mediating Efect of Various Variables on the Self-Care
Behavior of Nursing Students with Dysmenorrhea.Individual e-health literacy can not only directly afect the self-care behavior for dysmenorrhea but also indirectly afect self-care agency.Similarly, self-efcacy not only has a direct efect on the self-care behavior for dysmenorrhea but also indirectly afects the self-care behavior of individuals through self-care agency.Tis fully demonstrates that the self-care agency plays a critical role in e-health literacy, selfefcacy, and self-care behavior for dysmenorrhea.Terefore, healthcare departments or universities should be fully aware of self-care agency when conducting menses-related education for nursing students and boost self-care agency by improving individual e-health literacy and guiding students to improve self-efcacy to achieve the aim of actively implementing self-care behaviors for dysmenorrhea.

4.4.
Limitations.Tis study has two limitations.First, the sample size is confned to nursing students exclusively from select medical colleges in Shaanxi province, China, which may restrict the generalizability and diversity of the fndings.
To enhance the universality and persuasiveness of the research results, it is recommended to expand the types and range of samples in future studies.Second, due to the inherent constraints of cross-sectional research regarding time relationships and causality, establishing associations between variables in this study relies on theoretical frameworks and previous literature.Terefore, conducting longitudinal or interventional studies on factors infuencing self-care behaviors for dysmenorrhea among nursing students would be advantageous for further exploration of predictive factors impacting such behaviors.

Conclusion
Based on Orem's self-care theory and the fndings of previous studies, this study created a scientifc structural equation model that can explain the factors afecting selfcare behavior for dysmenorrhea.Te model verifed and analyzed the explanation degree of e-health literacy, negative emotion, and self-efcacy on self-care behavior for dysmenorrhea, and the mediating role and efect size of the selfcare agency and degree of dysmenorrhea.

Implications for Nursing Management
Dysmenorrhea is a signifcant factor afecting women's health.It is not merely a temporary issue for individuals, schools, or families but also a major concern related to students' academic performance and future fertility.It profoundly impacts their overall well-being, future life, and work.Implementing proactive and efective self-care behaviors can alleviate the discomfort and enhance their quality of learning and life.Schools and families should ofer proper guidance to students with dysmenorrhea, foster the development of correct cognitive attitudes, and establish positive coping strategies.Nursing managers can refne the management model for self-care behaviors associated with dysmenorrhea.With the advancements in the Internet and big data, they can create authoritative and scientifc information platforms for students to access and reference, thereby reducing the pressure on students' e-health literacy.
In addition, considering individual personality traits and lifestyle habits, tailored self-care recommendations can be provided, thereby reducing menstrual discomfort and furthering women's health.

Table 4 :
Parameter estimates of variables for the modifed model.