The association between body image, emotional health, relationships, and unhealthy dietary behaviors among medical sciences students: A structural equation modeling analysis

Abstract Aim Although it showed that poor body image may drive an individual to unhealthy dietary behaviors, but less is known about the role of emotional health and interpersonal relationships in this regard in different age group of population. This study aimed to investigate the association between body image, emotional health, relationships, and unhealthy dietary behaviors among medical sciences students. Methods In this cross‐sectional study, using a three‐stage stratified sampling method, 248 Iranian medical sciences students were recruited. The participants completed self‐administered questionnaires, including Youth Risk Behavior Survey Questionnaire, Adolescent Health Concerns Inventory, and Body Image Concern Inventory. The data were analyzed using SPSS‐20 and structural equation modeling in the Amos software version 24. Results The results indicate the acceptability of the goodness of fit model index. Direct association of body image with emotional health (β = 0.91, p = 0.005) and relationships (β = 0.88, p = 0.009) was significant, but any direct and indirect association between body and unhealthy dietary behaviors was not found. Conclusion This study underlines the importance of being aware of the association between body image, emotional health, and relationships Further studies are recommended to explore the relationship between these factors and unhealthy dietary behaviors in medical sciences students.

In this way it showed that over 40% of American students ignore government recommendations for healthy diet. Besides, approximately half of US adults are suffering from chronic diseases that are related to inadequate and low-quality diet, 5 as well as obesity has become a global phenomenon. 6 However, this issue is varied among various cultures and ethnicities. 7 Commitment to unhealthy dietary can be also influenced by several factors including sex, age, family and peers attitudes, and body image. 8 So, it is worth noting that owing to the establishment of unhealthy dietary behaviors during youth ages, understanding the relationship between contributing factors to the dietary behaviors is vital for youth future health. 9 Body image is a complex construction that comprises one's thoughts, believes, and perception of their own body. Body schema is associated with several items such as gender, age, culture, marital status, psychopathology, education level, spirituality, and so on. 1,10 Various studies have shown that poor body image could have a great influence on physical and psychological health, accordingly can lead to anxiety, depression, lower self-esteem, poor communication, social pressures, and so on. Consequently, these individuals are more likely to undertake various risky behaviors, such as unhealthy dietary behavior in order to relieve anxiety and worries about their appearance. 10 By considering the relationship between body image and unhealthy dietary behaviors, it is assumed that it is a multidimensional and complicated concept, which needs to be addressed in order to establish a better understanding of youth dietary behaviors. 11 Moreover, body schema can be influenced by youth emotional health and their personal and interpersonal relationships, including familial and societal relationships and feedbacks. 7 Numerous studies in developed countries have shown that body image disorder is seen alongside unhealthy eating, which is on the rise in countries such as China, Malaysia, and Japan. Our understanding of this issue is limited due to a little evidence about the relationship between body image and dietary health-threatening behaviors in non-Western populations. 12 Youngsters' body image could be affected by various social and environmental conditions. For instance, during college years, due to greater participation in the peer group, separation from family, being in more diverse social situations, they may be more likely to undertake unhealthy behaviors such as nutritional compensatory behaviors, dietary junk food to relieve psychological pressure. [13][14][15] Owing to university life being considered as the transition period, when medical sciences students face some challenges in their relationships and social expectation of ideal body, whereas, they are expected to be part of the loop of health circulation. 16 Given the importance of healthy dietary behaviors in medical students, the present study was designed to determine the path analysis fit of the conceptual model, which measures the mediating role of emotional health and relationships in the relationship between body image and unhealthy dietary behaviors among medical sciences students.

| Design and setting
This is a cross-sectional study, which was conducted among 248 male and female medical sciences students of all six faculties of Mazandaran University of Medical Sciences (MAZUMS), North of Iran, from September to December 2018.

| Inclusion and exclusion criteria
Medical sciences students aged between 18 and 25, who studied for a Bachelor's degree in Science or General Physician, were enrolled. Postgraduate students were not included in the study due to their different concerns as well as their older age than undergraduate students.

| Demographic information
It included some questions about participants' characteristics such as age, gender, marital status, body mass index, field of study, and their parents' education and job.

| Youth risk behavior survey questionnaire
To assess participants' dietary behaviors, seven relevant questions from this questionnaire were adapted. The dietary items included the following: during the 7 days before the survey, had eaten fruit or drunk 100% fruit juices, had eaten vegetables (lettuce, tomato, or other vegetables), had eaten breakfast, had drunk canned or carbonated drinks, and had drunk a few glasses of milk.
In each question, the number of times of each dietary behavior was reported from zero time/day to ≥3 times/day (except for dietary breakfast from zero to seven times). The more times an individual reported each dietary behavior implies the less unhealthy behavior (except for drinking canned or carbonated drinks which its score was reverse). The first version of this questionnaire was developed in 1997 by the Centers for Disease Control and Prevention by an acceptable reliability of 60.7 (kappas ranged from 23.6% to 90.5% with a mean of 60.7%). 18  The total score of the questionnaire was determined by calculating the mean total score. In this questionnaire, the more the total scores, the more positive body image. Its Cronbach alpha coefficient and its internal validity coefficient were reported as 0.93 and 0.92, respectively. Its reliability was established by Cronbach alpha coefficient for Iranian female and male students as 0.93 and 0.95 accordingly. 19

| Adolescent health concerns inventory
To assess participants' emotional health and their relationships, this inventory was used. The emotional health (17 items) refers to items such as: having a psychological balance, being anxious, dying, loneliness, feeling guilty, feeling good about one's self, being compared with others, and so on. The relationships included 14 items such as: getting alone with parents, siblings, friends, getting married, falling in love, having a boyfriend or girlfriend, and so on. This inventory was developed by Weiler in 1990 and each item is scored as either 0 or 1, and the higher scores indicated poor emotional health and poor interpersonal relationships. 20 The internal consistency of the Persian version of this inventory was determined by Cronbach alpha coefficient 0.96 and the Kappa coefficients 0.75, which demonstrate an acceptable compromise of the inventory. 21

| Data analysis
The data were then analyzed using SPSS-20. In describing numerical variables, mean and standard deviation were used and to describe categorical variables, frequency and percentage were used. The main variables studied in the model were unhealthy dietary behavior as the dependent variable, realationships and emotional, health as the mediating variables, and body image, concern, invertory as the Significance level in all tests was considered 0.05.   (Table 3).

| RE SULT
In this study, mean ± SD emotional health and relationship were As reported in Table 4, the direct and indirect associations of body image concern on unhealthy dietary behavior were inves-

F I G U R E 1
The final model of research and its parameters were investigated by structural equations modeling using Amos 24.0 software and the maximum likelihood. The χ 2 /df = 1.90, the comparative fit index (CFI) = 0.88, the parsimony comparative fit index (PCFI) = 0.65, and the root mean square error of approximation (RMSEA) = 0.06. Direct association of body image with emotional health (β = 0.91) and relationships (β = 0.88) was significant, but any direct and indirect association between body and unhealthy dietary behaviors was not found  6 There have been several studies on parents and peer relationships on body image, 31,32 . but no study has been done on the body image effect on parents and peer relationships, so it is suggested that more studies be done in different age groups.

| .Strengths and imitations
The structural equation modeling analysis as a multivariate statistical analysis technique was used to analyze the structural relationship between measured variables and latent constructs related to unhealthy dietary behaviors. Also, as entering the Iranian university is followed by a national entrance exam (named KOONKOOR), so the participants were students from all over the country and from different ethnic groups. This could increase the generalizability of the findings. Our findings must be interpreted in the light of some limitations. Due to the cross-sectional design of this study, it is impossible to determine the cause-and-effect relationships between variables because of temporality bias. Second, information bias is possible due to self-reported data gathering technique. As a result of culturally sensitive questions such as relationships with parents and opposite gender, especially in conservative societies such as Iran, social desirability bias is anticipated.
To reduce this bias, the participants were assured about the confidentiality of the information. Finally, the results may not be generalized to non-medical sciences students. It is suggested to conduct a similar study in non-medical students in different academic levels.