Abstract
Objectives: To estimate the prevalence of eating disorders among Imam Abdulrahman bin Faisal University (IAU) in Dammam, Kingdom of Saudi Arabia, female students and to investigate the associated factors.
Methods: A cross-sectional study was conducted at IAU, Dammam, Kingdom of Saudi Arabia, in 2016 with 670 participants. Data was collected by self-administered questionnaires, including the Eating attitude test 26 (EAT-26) and the sociocultural attitudes toward appearance questionnaire (SATAQ-4). Eating behavior questions were asked to determine whether the respondents should seek evaluation for an eating disorder (ED) from a professional.
Results: Eating attitude test 26 results showed that 29.4% of participants were categorized at a high level of concern for an ED, and 45.5% were categorized as at risk of having problematic feeding behaviors. Of the participants who scored above 20 on the EAT-26, 60.7% demonstrated high-risk feeding behavior attitudes, and approximately 11.5% who scored a body mass index (BMI) classified as underweight or very underweight required referral to a psychologist. A significant relationship was found between family stress to lose weight and EAT-26 scores above 20. The media was also found to represent external influence to lose weight.
Conclusion: This study concluded that early detection of EDs or problematic feeding behaviors and attitude is vital to minimize the risks to physical health and well-being of the sufferers’ experiences.
Eating disorders (EDs) feature disturbances which impair health and psychosocial functioning.1 With health risks ranging from low blood pressure, slow heart rate and severe dehydration in anorexia.1 And electrolyte and type 2 diabetes mellitus in bulimia to death.1
Both males and females experience EDs during adolescence and young adulthood.1 However, rates are higher in young and adult females2,3 Which can begin as young as 6 and continue into adulthood.3
The American Psychiatric Association’s Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) categorizes EDs into mutually exclusive categories based on observed symptoms (Appendix 1), with young adult females particularly at high risk.2
Western countries report rates of 0.3% for anorexia nervosa and 1% for bulimia among young females.4 Studies in Kingdom of Saudi Arabia,5 Arab countries,6,7,8 and in an international context,9,10 report rates for young adult females of 24%,11,12,13 in which the usual presentation is with a mental health problem including anxiety, depression, or a behavioral disorder.11 Eastern countries report that subclinical types of EDs are more prevalent.14,15
Studies of EDs and the associated risk factors largely focus on sociocultural factors, perceived body image, the media, and perceived pressure from family and peers, individual comparisons of one’s body with others, perceived pressure from peers and family members, media representations and sociocultural influences.16,17,18,19
Early recognition of mild forms of ED can help prevent inappropriate eating behaviors and attitudes leading to more severe forms of ED.2,20 Eating disorder prevalence rates and associated risk factors among women (and men) in Kingdom of Saudi Arabia have been particularly understudied. With a focus on young adolescent girls.
As Saudi women comprise 60% of all college students in the Kingdom of Saudi Arabia.21 This study aims to estimate the prevalence of EDs among orientation-year female students in Imam Abdulrahman bin Faisal University, Dammam, Kingdom of Saudi Arabia, and to determine the association between EDs and internal and external pressure affecting these females.
Methods
This study was performed with a cross-sectional study design conducted at Imam Abdulrahman bin Faisal University, Dammam, Kingdom of Saudi Arabia, academic year 2015-2016. This is the largest government university in the Eastern Region of the Kingdom of Saudi Arabia; students join a preparatory year of 8 tracks (health, design, literature, etc.) and then, according to their grades, join different colleges the following year.
A stratified, random sampling technique was used to recruit preparatory-year female students from 4 specialty colleges at Imam Abdulrahman bin Faisal University in Dammam, Kingdom of Saudi Arabia, namely, medicine, interior design, science and community college, using proportional allocation. Excluding non-Saudi and pregnant females. A sample size of 670 participants were included in this study, which exceeded the minimum required sample size 480, calculated using Epi Info 3.4, assuming 50% prevalence of EDs, at a 95% confidence level and precision of 5 and further expecting an 80% response rate.
Data collection was carried out using both a pre-scheduled questionnaire (Appendixes 1) and anthropometric measurements. All data were collected by the primary investigators with the assistance of 3 trained nurses.
The questionnaire comprised 3 main sections. The first section collected participants’ socio-demographic data. Part of this section was adopted from the original EAT-26 questionnaire developed by Garner and Garfinkel in 1979, and additional variables were included in the questionnaire by the researchers after thorough literature review and given the association identified in many studies between higher socioeconomic status and more extreme eating attitudes and behaviors.22
The second section measured abnormal eating and weight concerns, using the 26 items of the EAT-26 questionnaire. Each item was answered on a 6-point Likert scale, giving a possible range of 0-78. An additional 5 eating behavior questions were included in the questionnaire (Appendix 2).
The third section of the data collection instrument comprised the socio-cultural attitudes toward appearance questionnaire (SATAQ-4), a 22-item questionnaire, each question measured on a 5-point Likert scale. This supported the collection of data related to the potential external pressure on the individual that might lead to an increase or decrease in the prevalence of EDs. The SATAQ-4 focused on 4 sociocultural domains: self-pressure to be thin or muscular; family pressure (including parents, brothers, sisters, relatives) toward appearance; peer pressure (including close friends, classmates, and other social contacts) toward appearance; and media pressures (including television, magazines, the internet, movies, billboards, and advertisements) toward appearance (Appendix 3). The questionnaire achieved a Cronbach alpha score >0.8 for the 3 sections.
The questionnaire was originally developed in English and then double translated from English into Arabic then to English by 2 certified translators.
Classes were selected by simple random sampling. All students in the selected classes were invited to participate; Students whom did not want to participate in the study were given permission to leave the classroom. Yet, the response rate was 98%. Participants were given 15 minutes at the end of the class lectures to complete the questionnaire. Two anthropometric measures were taken by a trained nurse in a one-to-one setting to ensure participants’ privacy: height (m), using a digital telescopic wall-mounted stadiometer; weight (kg), using an electronic scale.
Fourteen incomplete questionnaires were excluded from the data analysis, resulting in 656 eligible for analysis (colleges of medicine [n=151], interior design [n=61], science [n=319], and community college [n=125]); the data was entered in SPSS version 20.0 software (IBM Corp., Armonk, NY, USA).
The EAT-26 items formed 3 subscales: dieting, Bulimia/food preoccupation, and oral control. The subscale scores were computed by summing all items assigned to that scale as follows: dieting (13 items); Bulimia and food preoccupation (5 items); oral control (7 items).
Items 1-25 were scored as follows: Always=3, Usually=2, Often=1; sometimes, rarely and never=0; item 26 was scored in the reverse direction. A score of 20 or more on questions 1-26 was deemed to indicate that the participant was at high risk for developing an ED. If a participant checked any of the 5 additional behavior questions or scored 20 or more in EAT-26, she was considered liable for referral to a trained mental health professional for an evaluation.
The SATAQ-4 items formed 5 subscales: Internalization thin/low body fat items 3-5, 8 and 9; Internalization muscular/athletic items 1, 2, 6, 7, and 10; pressure of family items 11-14; pressure of peers items 15-18; pressure of media items 19-22. Each of the test items was scored on a 5-point variable scale: definitely disagree=1; mostly disagree=2; neither agree nor disagree=3; mostly agree=4; and definitely agree=5. An average score percent was computed for each of the 5 subscales.
Any behavioral eating questions that yielded a positive answer indicated that the respondent should seek evaluation from a professional.
Body mass index was calculated as body weight (kg) divided by height (m2) and then categorized as underweight (BMI<18.5), normal (18.5≤25), overweight (25≤30), and obese (BMI of 30 or more).
Descriptive statistics were presented as frequency and percent and summary statistics (median and range). The chi-square test was used to test the association of categorical variables with EDs, referral, and BMI categories.
The SATAQ-4, 5 subscales score percent were tested using Shapiro-Wilk test and were found to be not following normal distribution. Relationships between SATAQ-4 items and the presence of eating disorders, referral and BMI categories were tested using Mann–Whitney U and Kruskal–Wallis tests. A p-value<0.05 was considered significant.
The local Ethical Committee approved this study, and permission was also sought and obtained from the IAU, Dammam, Kingdom of Saudi Arabia.
Results
The study comprised 656 female students aged 18-23 years old, with a mean of 18.7±0.74 years; other demographic data are presented in Table 1.
Distribution of study participants according to sociodemographic characteristics.
As shown in Table 2, 29.4% were categorized at a high level of concern for an ED based on EAT-26 scores. Notably, 45.5% of participants were categorized to be at risk of having problematic eating behaviors. More than half the participants 61.5% had a BMI within normal range, 19.8% were overweight, 11.5% were underweight and 7% were obese. Their distribution among different specialty colleges was not statistically significant (p=0.326, p=0.347, and p=0.903).
Distribution of participants in each specialty college, according to level of concern, BMI, and need for referral.
Participants who scored above 20 on the EAT-26 formed 54.5% of obese participants (OR=3.8, 95% CI=2.02-7.28) versus 43% of overweight participants (OR=2.4, 95% CI=1.56-3.71), 23% of normal-weight participants [reference group], and 20% of underweight participants (OR=0.8, 95% CI=0.4-1.5) (X2=32.7, p=0.0001). High-risk eating attitudes were highest among obese 70% (OR=3.3, 95% CI=1.68-6.57) followed by overweight 48% (OR=1.28, 95% CI=0.85-1.94), underweight 46% (OR=1.17, 95% CI=0.7-1.96), and then normal weight 42% [reference group], (X2=13.5, p=0.003). For students who needed referral, they were 100% of underweight, 76.6% of obese [3.37,1.66-6.8], almost 61% of overweight (OR=1.6, 95% CI=1.06-2.38), and about half of normal-BMI students [reference group], (X2=74.2, p=0.0001) (Figure 1).
Distribution of high level of concern of Ed, risk of problematic ED, and need for referral among different BMI categories. BMI - body mass index, ED- Eating disorders
Table 3 shows that normal weight was at a higher percent among married students 65% than single 61.3%, divorced or widowed participants 50% [2 cases]), p=0.002. Moreover, normal weight was higher among those living with their parents 62.5% than among those living with their husbands 53%, or with another relative 40% or living alone one case of 4, p=0.002. High level of concern for an ED and need for referral were of higher percent among those whom had lower grades in skills test (general abilities) grades <90% (OR=1.78, 95%CI=1.14-2.77), (OR=1.65, 95% CI=1.16-2.35), compared to higher grades 90-100%, p=0.032, p=0.018.
Distribution of sociodemographic data of orientation-year students according to level of concern, BMI, and need for referral.
Otherwise, no statistically significant associations were found between the studied demographic and socioeconomic characteristics and any high level of concern for an ED, problematic eating behaviors, or the need for referral to a psychologist to address eating behavior issues.
Figure 2 shows that score percentages of participants’ internalizations for thin body shape or for a muscular body ranged from 20-100%, with median scores of 75% and 72%. Higher scores were calculated among those with a high level of concern for an ED compared to participants not to be of concern. A higher median score for internalization of thin body shape was also found in participants at risk for problematic eating behaviors compared to participants not found to be at risk (Figure 3).
Internal and external pressures on participants to maintain their ideal body image or to lose weight.
Relation between SATAQ-4 and high level of concern, BMI and need for referral. BMI - body mass index, SATAQ-4 - socio-cultural attitudes toward appearance questionnaire
Similarly, higher median scores for internalization of thin body shape or of muscular body shape were also found for participants in need of referral compared to those not in need of referral. Internalization of thin body shape and muscular body shape were highest in obese participants, followed by overweight participants, normal participants, and then underweight participants. Median scores are shown in Figure 3. The differences between the BMI and internalization for thin body shape groups was found to be statistically significant (p=.0001), but not for muscular body shape (p=0.203).
Regarding external pressure from family, the media, and peers to lose weight, the median score percent was higher for family pressure 50%, followed by the media 35%, and peers 35%.
The external pressure from family, media, and peers had higher scores among those at high level of concern for an ED, at high risk of problematic eating behaviors, and those in need for referral than others with no risk or need of referral. The differences in these scores were statistically significant (p=0.0001). Median score percentages of the SATAQ-4 data for each group are demonstrated in Figures 2&3.
Discussion
The main finding in this study is that 29.4% of preparatory-year Saudi female students studying at the Imam Abdulrahman bin Faisal University in Dammam, Kingdom of Saudi Arabia, were categorized at high level of concern for an ED, having achieved a score of 20 or more on the EAT-26. This prevalence rate is slightly higher than the 24% reported by Bano, et al.11 From a regional perspective, the 28.8% prevalence rate reported in this study is below the finding reported by Madanat et al,23 45.2% of female college students in Jordan required screening for EDs but above the finding by Schulte et al,7 20% prevalence rate among female students studying at 2 other universities in Abudabi and Sharga in the United Arab Emirates.
Explaining the differences in ED prevalence rates among young women living in the Kingdom of Saudi Arabia or the Middle East is broadly more difficult due to the lack of research studies of the epidemiology of EDs in the region. From a cultural perspective, Miller et al24 claims that ED prevalence rates will naturally vary across different races and ethnicities and change over time due to the evolution of national culture. This claim is supported in the findings of a systematic review conducted by Keel et al5 that “weight concerns may be a culturally bound phenomenon, restricted to socio-historical contexts that idealize thinness and denigrate fatness.”
The findings in this study highlight that EDs are not culture-specific. Much has been written in recent decades to dispute the conceptualization that EDs among women are primarily a Western phenomenon.5 In fact, Campbell et al25 have suggested, at the very least, “risk of an ED is part genetic and part environmental.” It may be concluded that the prevalence rate of EDs among young Saudi women may be linked to unique sociocultural factors and different levels of vulnerability toward EDs the study cohort experiences.
The higher prevalence of concerns for an ED among single adult female university students in this study is similar to the finding by Madanat et al23 that single, college-aged women studying in Jordan were more prone to external pressure regarding eating than their married counterparts. Eating disorders are complex phenomena in which eating behavior problems are linked to behavioral, cognitive, and emotional factors.26 Within this dynamic, marital status and interpersonal relationships more broadly are identified as contributing factors to maintaining EDs. Martínez-González et al,27 claimed that solitary eating can be a strong independent risk factor for developing an ED, and this may explain in part the findings in this study.
Family pressure represented the strongest external pressure to lose weight that Saudi female orientation-year students experienced. The influence of family (and peers) on eating behaviors and body-image dissatisfaction among females and males was also reported by Quiles et al,13 in their meta-analysis of studies examining this relationship. This finding highlights the importance of developing a deeper understanding of how family influence within specific cultural contexts act as a moderating effect for the development of an ED. For instance, Zeeni et al17 pointed to the role of social customs and cultural expectations as the pressure from family members exert the person to engage in certain eating behaviors.
The media also represented a significant external pressure to lose weight for Saudi female orientation-year students with EAT-26 scores indicating high level of concern about an ED. A number of studies from different fields of research, including EDs, health psychology, and mass communication, pointed to the role of mass media as a pivotal source of information and reinforcement regarding the nature of the thin beauty ideal, the level of importance it is assigned, and how to achieve it.27 One of the key implications of the prominent role of the media is that the globalization of media networks and the increasing prevalence of Western ideals promoted through these media outlets will result in a greater proliferation of the Western ideal body image in countries all around the world.27,28 The finding that the media might function as a leading external pressure source on adult Saudi female university students suggests that, similar to parents, the media presents a principal social agent for this cohort. That is, the media is a platform for the transmission of ideas, values, norms, attitudes, and behaviors that socialize and construct the social reality.29
Finally, this study found that peers are also a source of external pressure on Saudi adult female university students to lose weight. To explain this finding, the case might be that some cultural contexts can lead young women to be more sensitive to and influenced by comments from peers regarding body image than to some other cultural contexts. This implies that the sociocultural context might influence the level of importance placed on social judgments which, in the context of EDs, can translate into the acceptance of certain eating behaviors.
Limitations
It should be noted that the questionnaire instrument designed and implemented for data collection did not consider different scoring if the participants had a chronic disease, were already diagnosed as having an ED, or were on any medications for mood stabilization. Such conditions among the study participants may influence their eating behaviors and BMI measures.
In conclusion, eating disorders continue to be a problem among populations in most societies around the world. In turn, important implications emerge from the findings in this study of a high rate of concern for EDs among Saudi adult female university students and the potential influence of external sociocultural pressure on them to lose weight. Overall, more than one-third of the female students in this study received scores on the EAT-26 that qualified them to be concerned about having an ED. Almost one-half demonstrated problematic eating behaviors, more than one-half required referral to a psychologist to address eating behavior issues, and approximately one-third had a BMI outside of normal weight. Based on these results, this study recommends that suitable screening tools (namely, EAT-26 and SATAQ-4) be used in the university clinic for early detection of an ED and for timely referral to support intervention and health education programs regarding healthy nutrition addressing students and provided in the media.
Acknowledgment
The authors would like to thank https://www.elsevier.com/en-xm for English language editing.
Appendix 1
DSM-5 diagnostic criteria for eating disorders.
Appendix 2
EAT-26 Questionnaire (English).
Appendix 3
SATAQ-4 Questionnaire (English).
Footnotes
Disclosure. Authors have no conflict of interests, and the work was not supported or funded by any drug company.
- Received June 10, 2018.
- Accepted July 10, 2018.
- Copyright: © Saudi Medical Journal
This is an open-access article distributed under the terms of the Creative Commons Attribution-Noncommercial-Share Alike 3.0 Unported, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.