Body perceptions and weight control behaviors among Palestinian University female students: a cross-sectional study

This study aimed to assess weight control behaviors and their relationship with body weight and image perceptions among female university students. A cross-sectional study was conducted among female students. The study included 420 female students aged 17–27 years using a convenient sample. Students were invited from all faculties. The data collection tools included a questionnaire, body image perceptions scale, and measurement scale for weight and height. Of the participants, 154 (36.9%) female students (95% CI 32–41%) had risky eating behaviors and reported a significant level of concern about diet and weight. In general, most of the study sample were dissatisfied with their bodies; in particular, 55.7% of participants were suffering from negative body image, while 24.5% showed a positive body image. Approximately one-third (31.8%) of underweight university students participated in risky eating behavior compared to 36.3% and 44.4% in normal weight and overweight and obese groups, respectively. The logistic regression analysis revealed that weight control behaviors were mainly associated with the perception of body image. The results showed a high level of concern about food and weight by the female students. Additionally, it showed that they are prone to risky eating behaviors. Therefore, interventions are needed to promote healthy eating behaviors and body image among university students.


Introduction
Obesity is a complex condition, one with serious social and psychological dimensions, that affects virtually all age and socio-economic groups and threatens both developed and developing countries (World Health Organization 2019). The prevalence of obesity has doubled in 73 countries worldwide and steadily increased in others since 1980, and health problems resulting from being overweight or obese now affect more than 2 billion people (Friedrich 2017). On the other hand, weight stigma has been rising in prevalence (Tomiyama et al. 2018). A growing body of evidence reported psychological, physiological, and social consequences of weight stigma (Lancet and Health 2019). Eventually, the concept of meeting an ideal weight and image has been increasingly noticed, especially in the feminine field (Kilpela et al. 2015).
Weight control behaviors are precipitated by the perception of being "too fat" regardless of body weight (Yang et al. 2014). Self-perception of body image determines the personal evaluation of one's body weight as "underweight" or "normal weight" or "overweight" regardless of the actual body mass index (BMI) (Grogan 2008). Disagreements between actual weight and weight perception may have significant implications; they can lead to unhealthy eating and exercise practices and potentially raise the risk of developing eating disorders (Haynes et al. 2018). The complicated relationship between weight status and body image is essential because it influences adolescent health behaviors. Adverse health consequences may result from the unique interaction between weight status and body image. These include physical inactivity, eating disorders, and dysfunctional exercise (Voelker et al. 2015).
With the growing obsession toward the thin ideal body, unrealistic body weight perception and weight loss practices have been increasing among adolescents with significant gender differences in favor of females (Bhurtun and Jeewon 2013). Studies have shown that female university students struggle with thinness (Sirang et al. 2013;Haynes et al. 2018). A cross-cultural study in five Arabian countries reported a drastic change in women's body size preferences from plumpness to thinness with general intensity among university students (Musaiger 2014). Additionally, girls are more prone to adopt various forms of eating behaviors than boys as they are more worried and sensitive to their changing body size, beauty, and shape, raising the importance of studying disordered eating, particularly among females (Smink et al. 2012).
Low body satisfaction was identified as one of the most significant predictors of diet, in addition to perceived media pressure and perceived weight effects on social interactions (Bayyari et al. 2013). People who perceive their weight status as overweight are more likely to have poor eating habits, such as failing to consume enough fruit and vegetables (Southerland et al. 2013;Haynes et al. 2018).
Research shows that attitudes toward body size have changed dramatically due to globalization in developed countries, where more people tend to be slim as a symbol of beauty and health (Musaiger 2014). Mass media, social networking, and photo sharing present many fitness, beauty, and ideal body shapes. As a result, an increasing prevalence of body dissatisfaction, weight stigma, and disordered eating was reported, especially among adolescents (Voelker et al. 2015).
The current study aims to examine the relationship between weight control behaviors and body image perceptions, actual body weight status, and other background variables among female university students in Palestine and address whether body image perception is associated with risky eating behaviors. In addition, laying the foundation stone for future studies on eating disorders is a significant mental health problem in Palestinian society. Specifically, the study aims to answer the following research questions: (1) what is the level of risky weight control behaviors, (2) what is the level of body image and weight perception, and (3) what is the relationship between weight control behaviors and body image and weight perceptions among female university students.

Study design and population
Cross-sectional research was performed at a national university, the largest provider of higher education in Palestine, comprising students from all parts of the West Bank and a wide variety of socio-economic identities. The sample size was calculated using a 95% confidence level and a 0.05 absolute precision. Based on the anticipated proportion of 50% with risky eating attitude and an expected 20% non-response rate (Madanat et al. 2011a). A total sample of 420 was obtained who was selected using a convenient sampling technique. Students from all faculties were invited to participate voluntarily, doing our best to achieve a representative sample by including all campuses and selecting students from all faculties equal to their university proportion. However, pregnant and physically disabled female students were excluded due to difficulties in anthropometric measurements. The study was approved by the institutional review board (IRB) at An-Najah National University. Students were informed about the study's aim and objectives. Voluntary participation was assured, as well as privacy and confidentiality.

Measurement tool
The main variables explored in this study were students' weight control behaviors and their body image perceptions. Weight control behaviors were classified into healthy eating behaviors and risky eating behaviors based on the Eating Attitudes Test (EAT-26) total score. EAT-26, a short form of the original EAT-40, is the standardized self-reported questionnaire used to recognize the risk of eating behaviors in some non-clinical settings. It has been a beneficial screening instrument to assess risky and abnormal eating behaviors in high school and universities, and other samples. It is used to evaluate various types of behaviors and attitudes associated with eating disorder tendencies. It is composed of 26 items divided into three subscales: Dieting (13 items; 1, 6, 7, 10, 11, 12, 14, 16, 17, 22, 23, 24& 25), Bulimia and Food Preoccupation (six items;3,4,9,18,21 & 26), and Oral Control (seven items; 2, 5, 8, 13, 15, 19 & 20). Scoring is done on a 6-points Likert scale from always to never; (0 = sometimes, rarely, or never; 1 = often; 2 = usually; 3 = always). A score ≥ 20 on the EAT-26 indicates risky eating behaviors and a high level of concern about dieting and body weight. While scores <20 indicate healthy eating behaviors (Garner et al. 1982).
The EAT-26 was translated into Arabic then backtranslated to English by two independent translators. The original and the back-translated English versions were reviewed and found to be conceptually comparable. The final draft of EAT-26 was pre-tested on 40 female students to improve its validity. Cronbach's alpha was measured and found to be 0.78, which indicated good reliability. Cronbach's alpha was also computed for the EAT-26 subscales, dieting, bulimia, and food preoccupation, and oral control and found as 0.77, 0.62, and 0.67, respectively. The Stunkard Figure Rating Scale assessed body image perception (Stunkard et al. 1983). Nine silhouette figures with increasing body weight from 1 to 9 were shown to the students. Experts consider these silhouette figures as; underweight (one and two), normal weight (three and four), overweight (five through seven), and obese (eight and nine). The participants indicated which figure best represented how they currently looked (feel) and how they wanted to look (ideal). The difference between an actual figure and an ideal figure was calculated and interpreted into body image satisfaction and body image dissatisfaction; with negative perceptions, the desire to be fatter and positive perceptions means the desire to be thinner and body satisfaction for zero score results.
One of the authors measured the students' height nearest to 0.1 cm and weight nearest to 0.1 kg using the SECA weighing scale and SECA body meter, where students wore lightweight clothes and no shoes. Body mass index (BMI) was calculated as body weight/height 2 (kg/m 2 ) and was used to assess each student's weight status. Other variables were participant's age, faculty of study, place of current residence, participants' selfperceived health status, marital status, social influence on weight status, and participants' actual body weight.

Data analysis
Data were coded and analyzed using the Statistical Package of Social Sciences (SPSS) version 20.0. Descriptive statistics (means, standard deviations for continuous variables and frequency distributions, and proportions for categorical variables) were employed to describe the participants' demographic and clinical characteristics. Chi-squared tests and logistic regression were used to detect significant relations between different groups. Significance levels were set at p ≤ 0.05.

Background characteristics
A total of 420 female college students were recruited. The mean age of the participants was 19.8 (SD = 1.5). Approximately 93% of participants were single, and 89.5% of them live with their families. Their mean BMI was 21.8 kg/ m 2 (SD = 2.97). Most of them were average weight (76.7%), while 12.9% fell within the overweight category. Table 1 provides detailed information about participants' demographic characteristics.

Weight control behaviors
For the whole sample, the mean of the EAT-26 score was 17.2 ± 9.6. Using a cut-off point of 20 and more to be considered as risky eating behavior (Garner et al. 1982), 155 female participants (36.9%) had risky eating behaviors and reported a high level of concern about dieting and eating behaviors (95% CI 32-41%) as seen in Fig. 1.
Based on the participants' responses to the question "How would you describe your weight?" we found that more than half (65.7%) consider themselves to be a healthy weight, 12% underweight, 22% overweight, and obese. Participants generally had a negative perception of their bodies and appearance. More than half of the participants (55.7%) were dissatisfied with their body shape, which means they had a negative body image, and only 24.5% had a positive body image. The rest (19.8%) were satisfied with their bodies' shapes (Fig. 2).
Weight control behaviors were studied with the background variables. The results did not show a significant relationship between weight control behaviors and students' age, place of current residents, and BMI (P value >0.05). For the body image and weight control behaviors, it was found that body image dissatisfaction was positively associated with risky eating behaviors (P value <0.05), as seen in Table 2.

Multivariable analysis of factors associated with weight control behaviors
Multivariable logistic regression was used to determine predictors of risky weight control behaviors and control the confounding factors. It revealed that participants with negative body image perception are 2.3 times more likely to be engaged in risky eating behaviors (P value =0.002), as seen in Table 3.

Discussion
This study aimed to draw attention to dieting behaviors, weight control practices, and the current body image perceptions among female students in Palestine. Moreover, it aimed to guide health education messages delivered to females in colleges about healthy attitudes and practices toward diet and body weight.
Weight control behaviors were assessed based on EAT-26, a widely used instrument that provides information about the symptoms and the risk of eating disorders in non-clinical settings. It is a useful screening instrument to assess "eating disorder risk" in high school and college. It has been used in many countries, including regional ones (Madanat et al. 2011b;Musaiger 2014;Momeni et al. 2020).
The results showed that 36.9% of female university students had risky eating behaviors. Comparable to other neighboring countries, the risky eating behaviors among female participants in Palestine are higher. This indicates a higher tendency to develop eating disorders (Momeni et al. 2020) and a high level of concern about dieting, body weight, and being prone to risky eating behaviors compared to other Arab countries. This worrying result may be due to rapid socio-cultural and lifestyle change, which includes modernization and urbanization. In addition, it may be due to social media's spread, representing a thin body as beauty. Compared with Jordan, the results were almost close, with a slight increase among Jordanian females. These similar results may due to the convergence of cultural and social norms between the two neighboring countries (Madanat et al. 2011b). This result is a cause for concern, as risky eating behaviors are associated with eating disorders. It is also related to other health concerns, including depression, anxiety, nutritional and metabolic disorders (Grogan 2008;Hatata et al. 2009;National Eating Disorders Collabration N 2015).
Regarding body image, the results showed that most participants preferred a thin figure for themselves, and most of them were dissatisfied with their body shape, where 55% of participants had a negative body image. This agrees with the findings reported by several studies (Grogan 2008;King et al. 2013).
Moreover, the results on body image perception were compared with another previous study conducted in Palestine, where the percentage of females dissatisfied with their body image increased by 10% (Bayyari et al., 2013). This increase could be attributed to the spread of social media and photo sharing and the media pressures and the model of beauty imposed by modern society, reflecting the thin woman body as the ideal body, regardless of women's health.
Regarding BMI and weight control behavior, no significant association was found between risky eating behaviors and BMI. This indicates that participants were engaging in risky eating behaviors regardless of their BMI classification. In regard to weight control behaviors and weight perception, no significant relationship was observed. This result is a risk indicator, as students who were following risky eating behaviors had accurate weight estimation and healthy BMI, which may lead to difficulty convincing them that there is no need for such actions.
Body image dissatisfaction is strongly associated with female students' risky eating behaviors; 45.3% of participants who engaged in risky eating behaviors had negative body image. This supports the hypothesis that body image dissatisfaction would present higher levels of disordered eating attitudes Swierkosz and Clutter 2010;Yeng and Sedek 2012). Moreover, this indicates that body image dissatisfaction, rather than actual BMI, is a better predictor of dieting behaviors among study participants, supported by the literature (McAlonan et al. 2007).

63.1%
Riskry eaƟng behviors EAT-26≥ 20 Healthy eaƟng behaviors EAT-26<20 Fig. 1 Distribution of weight control behavior among female university students (n = 420) Finally, we have to admit that risky eating behaviors are significantly associated with eating disorders. It is also related to other health concerns, including depression, anxiety, selfhatred, nutritional and metabolic problems, obsessivecompulsive issues, and suicide (National Eating Disorders Collaboration 2015). Therefore, participants who scored ≥20 were asked to seek health evaluation by a health professional specializing in the treatment of eating disorders to determine their health status (Agras 2010).
Some limitations of this study should be considered when interpreting its results. This study's cross-sectional design makes it difficult to determine the temporal relationships and  causality between the weight control behaviors and the studied independent variables. In addition, using a convenient sample may not represent the study population. We could not have the list of students enrolled in the university (sample frame) as the regulations did not allow it. However, great care has been taken to select students from both campuses and all faculties equal to their university proportion.

Conclusion and recommendations
Risky eating behaviors were shown to be strongly associated with dissatisfaction with body image among female students. Those with negative body image were twice as likely to be engaged in risky eating behaviors than those who were satisfied with their body image. This indicated that female students with body dissatisfaction are more likely to be involved in risky eating behaviors. Given the spread of risky eating habits among female students and the fact that body image dissatisfaction motivates risky eating behaviors, the university needs to plan health promotion programs to promote positive body image perception and healthy eating behaviors.
More importantly, it is to clarify to students that eating disorders are a severe mental health problem and not a modern diet or lifestyle. Establishing a nutrition clinic in the university would be essential to prevent and control such health problems.
Authors contribution Zaher Nazzal and Watan Nazzal conceptualized and designed the study. Watan Nazzal collected data and carried out the analysis of data and writing. Zaher Nazzal was responsible for quality control and offered guidance on the study design, the analysis of the data, and the manuscript's drafting. Beesan Maraqa contributed to the analysis, interpretation, and writing of the manuscript. All the authors read the manuscript, gave their final approval, and agreed to be responsible for all aspects of the work.
Funding The authors disclose that they have no funding source for this work.
Availability of data and material The data used to support the findings of this study were provided as supplementary material.

Declarations
Ethical approval All procedures performed in studies involving human participants were per the institutional research committee's ethical standards and the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Institutional Review Board (IRB) Committee of Al-Najah National University.
Consent to participate Informed consent was obtained from all individual participants included in the study.

Conflict of interest
The authors declare that they have no conflicts of interest.