Assessing disordered eating behaviours and attitudes: Factor structure and measurement invariance of the Arabic version of the Eating Attitudes Test (EAT-26) in Saudi Arabia

doi:10.21203/rs.3.rs-4918884/v1

Download PDF

Research Article

Assessing disordered eating behaviours and attitudes: Factor structure and measurement invariance of the Arabic version of the Eating Attitudes Test (EAT-26) in Saudi Arabia

https://doi.org/10.21203/rs.3.rs-4918884/v1

This work is licensed under a CC BY 4.0 License

You are reading this latest preprint version

Background

The factorial structure of the Eating Attitudes Test (EAT-26) has been found inconsistent across studies and samples. This study examined the factorial structure and measurement invariance of the Arabic version of the EAT-26 in a large non-clinical Saudi sample.

Methods

1,734 Saudi adults (M_age 26.88 and SD 9.13), predominantly female, completed an online survey. Confirmatory factor analysis evaluated several established factorial models to identify the best-fitting model for the Arabic version of the EAT-26.

Results

The original three-factor model and two other common models demonstrated poor fit. Instead, a 16-item, four-factor structure [(Self-Perceptions of Body Weight), (Dieting), (Awareness of Food Contents), and (Food Preoccupation)] showed acceptable fit ([CFI = 0.904; SRMR = 0.0554; RMSEA = 0.073 [90% CI: 0.068–0.077]). Internal consistency was good, and measurement invariance was supported across sex and BMI categories.

Conclusions

These findings suggest the need for culturally relevant validation of the EAT-26 among Arabic-speaking populations, as the factorial structure differed from previous models. Future research should further examine this revised 16-item, four-factor structure in clinical settings.

Eating attitudes test

EAT-26

eating disorder

disordered eating behaviours and attitudes

Anorexia Nervosa

Arabic

Disordered eating behaviours and attitudes are unhealthy behaviours and attitudes toward food and weight, which may include restrictive dieting, purging, excessive exercise, abuse of laxatives, distorted body image, and body dissatisfaction, all of which have been link to eating disorders .This study investigated the reliability and validity of the Arabic version of the Eating Attitudes Test (EAT-26) among Saudi adults. Contrary to the original factor structure, findings showed that the Arabic version has a four-factor structure with 16 items. This revised 16-items version showed good fit indices and reliability. Researchers and clinicians should avoid using the original EAT-26 and instead use this newly proposed 16-item version in this demographic.

Obesity is a significant global health issue. Saudi Arabia (SA) has been facing a rising prevalence of obesity over the past few decades, and this trend is likely to continue (World Obesity Federation, 2023). According to available data, the obesity rate in SA is among the highest in the world, with 23.7% of adults (aged ≥ 15 years) and 7.3% of children (< 15 years) being obese. Both females and males show comparable rates of obesity (General Authority for Statistics, 2024). The negative impact of obesity on health and well-being is recognised by various stakeholders, including individuals with obesity, caregivers, and healthcare professionals (Bin-Abbas et al., 2024). Urbanisation, sedentary lifestyles, unhealthy eating habits, and a lack of physical activity have been cited as contributing to the high rates of obesity (Memish et al., 2014).

Obesity is closely linked to disordered eating behaviours and attitudes (DEBAs) as well as eating disorders (EDs). DEBAs encompass a wide range of unhealthy behaviours and attitudes toward food and weight, including restrictive dieting, purging, excessive exercise, abuse of laxatives, distorted body image, and body dissatisfaction (Ghazzawi et al., 2024; Pereira & Alvarenga, 2007). These DEBAs may lead to the development of EDs such as anorexia nervosa (AN), bulimia nervosa (BN), and binge-eating disorder (BED), which are prevalent globally with a lifetime prevalence of 0.16%, 0.63% and 1.53% respectively (Qian et al., 2022). The Saudi National Mental Health Survey (SNMHS), a nationally representative population household survey, reported the 12-month prevalence of any of the three EDs at 3.2% and a lifetime prevalence of 6.1% (0.6% for AN, 2.8% for BN, and 2.6% for BED), which is higher than reported rates worldwide (AlHadi et al., 2022).

One commonly used screening tool to assess symptoms associated with DEBAs is the Eating Attitude Test (EAT). The EAT is a self-report measure that has two versions: the original EAT-40 (Garner & Garfinkel, 1979) and its abbreviated form EAT-26 (Garner et al., 1982). While the EAT was developed with clinical samples of adolescent females with AN, its use has since been expanded to diverse cross-cultural and non-clinical populations.

The EAT-26 consists of three factors: Dieting (i.e., 13 items related to avoidance of fatty foods and preoccupation with thinness), Bulimia and food preoccupation (i.e., 6 items related to thoughts about food and bulimia), and Oral control (i.e., 7 items related to self-control over food and societal pressure to gain weight) (Garner et al., 1982). The EAT-26 has been used in a range of populations including adults with EDs and non-clinical samples (Garfinkel & Newman, 2001), and has been translated into many languages including Chinese (Kang et al., 2017), French (Leichner et al., 1994), Italian (Dotti & Lazzari, 1998), Japanese (Mukai et al., 1994), Russian (Meshkova et al., 2023), Spanish (Rivas et al., 2010), and Urdu (Jamil et al., 2023).

Despite the widely acknowledged reliability of the EAT-26, its factorial structure has been found to be inconsistent across studies and populations (Rogoza et al., 2016). Although some studies replicated the same three-factor structure identified for the original English EAT-26 version (Dotti & Lazzari, 1998), more recent research using different statistical methodologies has shown different factor structures of the EAT-26. A study with two independent samples of female college students compared the three-factor structure with 26 items (Garner et al., 1982) and a four-factor structure with 20 items proposed by (Koslowsky et al., 1992). The results showed an unacceptable model fit for the three-factor EAT-26 and a poor fit for the four-factor EAT-20. Four items that presented low factor loadings were eliminated and the four-factor model with 16 items was found to have an acceptable fit (Ocker et al., 2007).

In a study conducted with a Russian non-clinical university female sample, the authors identified a five-factor model with 15 items that best fit the data (Meshkova et al., 2023). A six-factor model with 18 EAT items was found to be a reliable and valid measure of DEBAs for an Irish adolescent sample (McEnery et al., 2016). Other research also found a six-factor model with 18 EAT items has the best fit among a large sample of French-speaking, ethnically diverse European and African participants using exploratory structural equation modelling (ESEM) (Maïano et al., 2013). Research using Rasch analysis with an adult sample of university students and adults undertaking a behavioural weight loss program found that a 19-item EAT version demonstrated a satisfactory fit in accordance with the expectations of the Rasch model (Papini et al., 2022).

Similarly, although the EAT-26 has been widely used among Arabic-speaking populations in the Middle East region (Abu-Saleh et al., 2024; Al-Adawi et al., 2002; Alsheweir et al., 2023; Alwosaifer et al., 2018; Ebrahim et al., 2019; Fatima & Ahmad, 2018), fewer studies have reported on its psychometric properties. Al-Subaie et al. (1996) validated the EAT-26 among Saudi young female students (grades 7–12) against diagnostic clinical interviews. The authors found that the Arabic EAT-26 exhibited a lower positive predictive value compared to its negative predictive value (Al-Subaie et al., 1996). Furthermore, the factorial structure of the Arabic EAT-26 version is unstable and inconsistent, similar to the criticisms levelled against the English and other language versions of the EAT-26. A study using a large probability sample of predominantly young Qatari female university students identified a five-factor structure with 19 EAT items (Khaled et al., 2018). In a Lebanese community sample; however, a six-factor model had the best fit (Haddad et al., 2021). In this study, higher scores on the EAT-26 were associated with higher depressive symptoms, emotional eating, and starvation to reduce weight (Haddad et al., 2021).

Overall, the EAT-26 appears to have different factors in different ethnic and cultural groups, most of which did not correspond with the original EAT-26 three-factor structure in terms of the number of factors and the distribution of items within each factor. This underscores the importance of investigating the reliability and validity of the EAT-26 in the targeted population. Nonetheless, explanations for the varying factorial structure have been offered in the literature including the type of sample (e.g., clinical vs non-clinical) (Rogoza et al., 2016), cultural differences in lifestyle and eating habits and body image and perception (Spivak-Lavi et al., 2021, 2023).

Despite an extensive literature search, no publication was found that investigated the factorial structure of the EAT-26 specifically among SA populations. This is concerning since recent reviews have revealed that the EAT-26 is the most commonly utilised tool in research screening for EDs and DEBAs within this under-represented and under-researched population (AlHadi et al., 2022; Alsheweir et al., 2023; Melisse et al., 2020). Consequently, this research aimed to examine the factorial structure of the Arabic version of the EAT-26 and its measurement invariance across sexes and BMI categories in a non-clinical community sample.

Participants

This study formed a crucial part of a larger research project that delved into the intricate relationships between clinical perfectionism, depression, anxiety, and disordered eating behaviours among adults from the general population in SA. A total of 1734 Saudi participants from all 13 provinces completed an online survey. Participants had to be Saudi adults aged 18 years or older and native Arabic speakers to take part in this research. The age range was from 18 to 77 years (M_age 26.88 and SD 9.13), with 78.4% of participants being female. The average BMI was 24.78 (SD 6.25). Among the respondents, 48.6% had a normal weight, 17.2% were classified as obese, 22.7% were overweight, and 11.5% were underweight.

Procedure

A cross-sectional online study was conducted between October 2020 and January 2021. To recruit participants, a social media post with a link to the study on Google Forms was shared on social media platforms including X and Facebook, WhatsApp and Telegram student groups. All survey questions were made mandatory, meaning there could be no missing data. Participation was voluntary, and all participants provided electronic informed consent. No incentives were offered to the participants. The study received ethical approval from the university ethics review board.

Measures

Participants provided demographic information including age, sex, and area of residence, and self-reported their weight (kg) and height (cm), which were used to calculate Body Mass Index (BMI) according to the Center for Disease Control and Prevention guidelines (BMI < 18.5 underweight; BMI ≥ 18.5 and < 25 healthy weight; BMI ≥ 25 and < 30 overweight; and BMI ≥ 30 obesity) (CDC, 2022).

Depressive and anxiety symptoms were assessed using the Arabic versions of the Patient Health Questionnaire (PHQ-9) and the General Anxiety Disorder-7 (GAD-7) respectively. Higher scores on the PHQ-9 and GAD-7 indicate increased symptoms. The Arabic version of the PHQ-9 (AlHadi et al., 2017) and GAD-7 (Alghadir et al., 2020) have robust psychometric properties in Saudi samples. In the current study, the Cronbach's alpha was 0.88 for PHQ-9 and 0.91 for GAD-7. Participants also completed the Arabic version of the EAT-26, which is described in more detail below.

Eating Attitudes Test

The EAT-26 comprise three subscales according to the original English version: (1) dieting ( 13 items), (2) bulimia and food preoccupation (6 items), and (3) oral control (7 items) (Garner et al., 1982). The first 25 items are scored using a 4-point Likert scale where ranging from never, rarely, sometimes (0), often (1), usually (2), and always (3). Item 26 is reversed-scored, where never (3), rarely, (2), sometimes (3), and often, usually, and always (0). A total score is calculated by summing all items’ scores (range from 0 to 78). A score ≥ 20 indicates possible disordered food attitudes (Garner et al., 1982).

Data Analysis

The data was analysed using the SPSS 26.0 and Amos 24.0 software programs. To assess the factor structure of the Arabic EAT-26 version; a confirmatory factor analysis (CFA) was performed using the maximum likelihood estimation method. The model fit was evaluated using several goodness-of-fit indices: the comparative fit index (CFI), standardized root mean square residual (SRMR), and root mean square error of approximation (RMSEA). The model was considered a good fit if the CFI was greater than or equal to 0.90, and the SRMR and RMSEA were less than or equal to 0.08 (Browne & Cudeck, 1992; Byrne, 2016; Hu & Bentler, 1999). Descriptive statistics of the EAT were calculated and internal consistency was assessed using Cronbach’s alpha and McDonald’s omega coefficients.

Next, a multi-group confirmatory factor analysis (MGCFA), probably the most widely used approach to test measurement invariance between groups (Karl, 2023), was used to test the measurement invariance across sexes and BMI categories. Three levels of invariance were tested including configural, metric, and scalar invariance (Cheung & Rensvold, 2002). Configural invariance indicates that the factor structure is the same between the comparison groups. Metric invariance implies that the factor loadings for similar items are equivalent across groups. Scalar invariance means that the item intercepts are equivalent across groups (Karl, 2023). Evidence for invariance was determined if the changes in CFI (ΔCFI) were less than or equal to 0.01, the changes in RMSEA (ΔRMSEA) were less than or equal to 0.015, and the changes in SRMR (ΔSRMR) were less than or equal to 0.03 for tests of metric invariance. For tests of scalar invariance, the criteria were ΔCFI ≤ 0.01, ΔRMSEA ≤ 0.015, and ΔSRMR ≤ 0.01 (Byrne & van de Vijver, 2010; Chen, 2007).

Factorial Validity of EAT

CFA was conducted using the maximum likelihood estimation method to assess the original three-factor structure of EAT-26 (Garner et al., 1982). CFA results indicated that the model fit was unacceptable for the original three-factor model (CFI = 0.727; SRMR = 0.0911; RMSEA = 0.085 [90% CI: 0.082–0.087]).

Three alternative models from the literature were also tested. Second order 26-item three-factor model, Koslowsky et al.’s (1992) 20-item four-factor model, and Ocker et al.’s (2007) 16-item four-factor model. The results of the individual model fit indices are shown in Table 1. The results indicated that both second order three-factor and Koslowsky et al.’s (1992) models provided a relatively poor fit. Ocker et al.’s (2007) 16-item four-factor model demonstrated an acceptable fit (CFI = 0.904; SRMR = 0.0554; RMSEA = 0.073 [90% CI: 0.068–0.077]). The standardized factor loadings are shown in Table 2. All factor loadings were significant at p < 0.001 and ranged from 0.44 to 0.83. As a result, the psychometric properties of the 16-item four-factor model were assessed in all subsequent analyses. The revised Arabic EAT-16 version is in the Appendix.

Table 1

Fit indices of the original and alternative models of the Arabic EAT-26
Model	χ² (df)	p	CFI	RMSEA (90% CI)	SRMR
Original three-factor model	3959.44 (296)	.000	.727	.085 (.082-.087)	.0911
Second order three-factor model	3896.12 (297)	.000	.732	.084 (.081-.086)	.0913
20-item four-factor model	1786.20 (164)	.000	.847	.076 (.072-.079)	.0800
16-item four-factor model	992.16 (98)	.000	.904	.073 (.068-.077)	.0554
Note. df = degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval; SRMR = standardized root mean square residual.

Insert Table 1 here

Descriptive analysis, reliability, and correlations

Descriptive statistics and reliabilities coefficients of the EAT-16 are indicated in Table 2. The mean score of the EAT-16 was 11.20 (SD 10.10). The inter-factor correlations for the four identified factors (Self-Perceptions of Body Weight, Dieting, Awareness of Food Contents, and Food Preoccupation) were between 0.28 to 0.68. The revised Arabic EAT-16 showed good internal consistency (Cronbach alpha and McDonald’s omega were 0.88). Furthermore, the EAT-16 demonstrated convergent validity through significant positive correlations with measures of depression (r = 0.20, p < 0.001) and anxiety (r = 0.22, p < 0.001). These associations were within the expected magnitude and direction.

Insert Table 2 here

Table 2

Descriptive statistics, reliability coefficients, and standardized factor loading for the Arabic EAT-16 version
Factor	Item (number on the EAT-26)	Mean (SD)	Factor loading	α	ω
Self-perception of Body Shape	I am terrified about being overweight (1)	1.13 (1.22)	.66	.66	.71
	I am occupied with a desire to be thinner (11)	1.16 (1.30)	.83
	I am preoccupied with the thought of having fat on my body (14)	0.70 (1.12)	.44
Dieting	I feel extremely guilty after eating (10)	0.63 (1.07)	.66	.74	.74
	I think about burning up calories when I exercise (12)	1.16 (1.29)	.69
	I feel uncomfortable after eating sweets (22)	0.77 (1.14)	.56
	I engage in dieting behavior (23)	0.75 (1.10)	.58
	I Like my stomach to be empty (24)	0.67 (1.05)	.48
Awareness of Food Contents	I am aware of the calorie content of foods that I eat (6)	0.44 (0.91)	.61	.75	.75
	I particularly avoid food with a high carbohydrate content (i.e. bread, rice, potatoes, etc.) (7)	0.38 (0.85)	.66
	I avoid foods with sugar in them (16)	0.52 (0.94)	.62
	I eat diet foods (17)	0.43 (0.88)	.73
Food Preoccupation	I find myself preoccupied with food (3)	0.65 (1.04)	.69	.80	.80
	I have gone on eating binges where I feel that I may not be able to stop (4)	0.60 (1.02)	.69
	I feel that food controls my life (18)	0.60 (1.05)	.78
	I give too much time and thought to food (21)	0.59 (1.03)	.67
EAT-16 total Score		11.20 (10.10)	-	.88	.88
Note. SD, standard deviation; α, Cronbach alpha, ω, McDonald omega. All factor loadings were significant at p < .001.

Measurement invariance across gender and weight status

Results of invariance analyses across sexes and four BMI categories (underweight, normal weight, overweight, and obese) are shown in Table 3. The configural model to the data was acceptable between males and females (CFI = 0.903, SRMR = 0.0557, RMSEA = 0.052 [90% CI = 0.049–0.055]). The metric (ΔCFI = 0.001, ΔRMSEA = 0.002, ΔSRMR = 0.0000) and scalar (ΔCFI = 0.000, ΔRMSEA = 0.001, ΔSRMR = 0.0001) invariance models were also supported, suggesting measurement invariance between males and females.

With regards to the BMI categories, the configural model was supported (CFI = 0.902, SRMR = 0.0752, RMSEA = 0.037 [90% CI = 0.035–0.039]). The metric (ΔCFI = 0.000, ΔRMSEA = 0.002, ΔSRMR = 0.0002) and the scalar (ΔCFI = 0.001, ΔRMSEA = 0.001, ΔSRMR = 0.0001) invariance models were also supported. These results suggest that the factor loadings were equivalent across the different BMI categories and that responders, across the different BMI categories, interpreted and responded to the items in the same way. Taken together, all levels of measurement invariance were achieved across sexes and BMI categories.

Table 3

Measurement invariance of the EAT-16 across sexes and BMI categories
	Overall fit indices					Comparative fit indices
Model	χ² (df)	p	CFI	RMSEA (90% CI)	SRMR	Δχ² (df)	p	ΔCFI	ΔRMSEA	ΔSRMR
Sex¹
Configural	1102.68 (196)	.000	.903	.052 (.049-.055)	.0557
Metric	1125.11 (208)	.000	.902	.050 (.048-.053)	.0557	22.43 (12)	.032	.001	.002	.0000
Scalar	1137.25 (220)	.000	.902	.049 (.046-.052)	.0556	12.15 (12)	.433	.000	.001	.0001
BMI categories²
Configural	1315.80 (392)	.000	.902	.037 (.035-.039)	.0752
Metric	1353.06 (428)	.000	.902	.035 (.033-.037)	.0751	37.26 (36)	.411	.000	.002	.0002
Scalar	1374.86 (464)	.000	.903	.034 (.032-.036)	.0750	21.8 (36)	.970	.001	.001	.0001
Note. df = degrees of freedom; CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval; SRMR = standardized root mean square residual; ¹ Male vs. Female; ² Underweight, Normal weight, Overweight, and Obese.

Insert Table 3 here

This research explored the factorial structure, reliability, and measurement invariance across sexes and BMI categories of the Arabic version of the EAT-26 in a large non-clinical Saudi sample. By systematically testing existing factorial models of the EAT-26, this study adds to the literature by identifying the most robust and empirically supported structure among the Saudi general population. Successive CFAs identified an optimal 16-item, four-factor structure (Self-Perceptions of Body Weight, Dieting, Awareness of Food Contents, and Food Preoccupation) of the Saudi version of the EAT-26. This revised 16-item Arabic version had good internal consistency as assessed by Cronbach alpha and McDonald omega. Furthermore, measurement invariance analyses of the 16-item version supported configural, metric, and scalar invariance across sexes and BMI categories.

The revised EAT-16 Saudi version is consistent with Ocker’s et al. (2007) model, in terms of the number of factors and number of items. In their study, Ocker’s et al. (2007) evaluated the model fit of the 26-item original three-factor model (Garner et al., 1982) and a 20-item, four-factor model (Koslowsky et al., 1992) using two independent female college samples (a calibration sample and a cross-validation sample). Their analyses showed that even though the Koslowsky’s et al. (1992) model had a relatively better fit compared to the original model (Garner et al., 1982), the RMSEA and CFI values did not meet the acceptable standards. After removing the four problematic items, with low factor loadings, a four-factor model with 16 items was found to have an adequate fit (Ocker et al., 2007).

Measurement invariance, often neglected in applied research (Maassen et al., 2023), is important as it sheds light on whether group differences in a given psychological construct (a latent variable such as DEBAs) can be attributed to real differences in the underlying construct being measured, rather than being due to biases or problems with the measurement instrument itself (Karl, 2023). The EAT-16 Saudi version demonstrated configural, metric, and scalar invariance indicating that the factor structure, factor loadings, and item intercepts are equivalent across gender and the BMI categories (underweight, normal weight, overweight, and obese), and therefore measuring DEBAs in the same way across these groups. This finding is consistent with previous research providing evidence for measurement invariance for an 18-item EAT French version across gender, ethnicity, age and BMI groups (Maïano et al., 2013), a 21-item EAT Spanish version across gender (Montoro-Pérez et al., 2024), and for the same model tested in the current research (a 16-item, four-factor model English version) across Caucasian and Hispanic ethnicities (Belon et al., 2011). Overall, researchers using the Arabic EAT-16 in the Saudi general population can be assured that this measure is psychometrically equivalent across sexes and BMI categories.

The current study found small but significant positive correlations between total scores on the EAT-16, PHQ-9 (a measure of depression) and GAD-7 (a measure of anxiety), providing evidence for convergent validity. This finding is consistent with previous studies by Haddad et al. (2021) and McLean et al. (2023), which also reported positive correlations between measures of DEBAs, depression, and anxiety.

Given the widespread use of the EAT-26 to assess DEBAs among Arabic-speaking populations, especially Saudi samples (AlHadi et al., 2022; Alsheweir et al., 2023; Melisse et al., 2020), the current study has important implication. The original three-factors EAT-26 was developed in a Western context (Garner et al., 1982), which may not fully capture the nuances of DEBAs in non-Western contexts. The revised and culturally adapted Arabic EAT-16 would improve screen for and identify individuals at risk of EDs within the Saudi general population. This is crucial for enhancing early detection and facilitating timely intervention.

Limitations and future research

The present study has several strengths. First, it examined the psychometric properties of the Arabic version of the EAT-26 in Arabic-speaking general sample, and thus, provided empirical evidence for a revised version of the EAT-26 that is culturally relevant and empirically supported for use in this population. Second, previous research has primarily focused on female adolescents and adults (Belon et al., 2011; Kang et al., 2017; Khaled et al., 2018; Meshkova et al., 2023; Spivak-Lavi et al., 2023), precluding measurement invariance testing across genders. In contrast, the present study included both participants of both sexes and successfully established measurement invariance. Thus, providing evidence that the four-factor structure holds across sexes as well as BMI categories (underweight, normal weight and overweight and obese).

However, a few limitations should be noted. First, our sample consisted of Saudi adults from the general population and therefore our findings may not generalize beyond this population (e.g., young adults and individuals with mental health disorders including EDs). Second, height and weight were self-reported for BMI, which are subject to bias and inaccuracies. Future research should use objective measurements of height and weight. Future research should also test whether this revised 16-item EAT version holds in clinical samples and samples with age heterogeneity. The present study also did not assess other important psychometric properties including temporal stability, other forms of validity (e.g., criterion and discriminant validity). Future research should consider investigating these psychometric parameters to gain more insight into the measure's reliability and validity. Future work could also consider using item response theory models such as the Rasch analysis to evaluate the performance and unique contribution of individual items and identify those that do not fit the underlying latent construct well.

In summary, the present study provided evidence for a revised 16-item, four-factor model of the EAT-26 to measure disordered eating behaviours and attitudes among Saudi general population. This revised model demonstrated good internal consistency, convergent validity, and measurement invariance across sexes and BMI groups. Future research should examine the factorial structure of this revised model in clinical settings to further establish its utility for the identification and evaluation of eating-related concerns within Arabic cultural contexts.

EAT-26

The Eating Attitude Test 26-item version

GAD-7

The General Anxiety Disorder scale

PHQ-9

The Patient Health Questionnaire

BMI

Body Mass Index

DEBAs

Disordered Eating Behaviours and Attitudes

EFA

Exploratory Factor Analysis

CFA

Confirmatory Factor Analysis

ESEM

Exploratory Structural Equation Modelling

Ethics Approval

All procedures followed were in accordance with the Declaration of Helsinki. The study received ethical approval from XXXX.

Informed Consent

Electronic informed consent was obtained from all individual participants included in the study.

Availability of data and material

Data associated with this manuscript can be retrieved from https://doi.org/10.17605/OSF.IO/8ET5Q

Competing Interests

The authors declare no competing interests.

Funding

The authors have no funding to report.

Authors' contributions

M.M.A designed the study, collected data, interpreted the results and drafted the article; S.A.A conducted statistical analyses, interpreted the results and drafted and edited the manuscript. All authors reviewed and approved the final version of the manuscript.

Acknowledgements

The authors would like to acknowledge and thank all individuals who participated in this research.

Abu-Saleh, S. Y., Irsheid, W., Ghazzawi, H., Amawi, A. T., Pandi-Perumal, S. R., Trabelsi, K., Seeman, M. V., & Jahrami, H. (2024). The psychometric properties of the Arabic version of the Dieting Beliefs Scale (DBS). Heliyon, 10(7), e28842. https://doi.org/10.1016/j.heliyon.2024.e28842
Al-Adawi, S., Dorvlo, A. S. S., Burke, D. T., Moosa, S., & Al-Bahlani, S. (2002). A survey of anorexia nervosa using the Arabic version of the EAT-26 and “gold standard” interviews among Omani adolescents. Eating and Weight Disorders, 7(4), 304–311. https://doi.org/10.1007/BF03324977
Alghadir, A., Manzar, M. D., Anwer, S., Albougami, A., & Salahuddin, M. (2020). Psychometric properties of the Generalized Anxiety Disorder Scale among Saudi university male students. Neuropsychiatric Disease and Treatment, 16, 1427–1432. https://doi.org/10.2147/NDT.S246526
AlHadi, A. N., AlAteeq, D. A., Al-Sharif, E., Bawazeer, H. M., Alanazi, H., AlShomrani, A. T., Shuqdar, R. M., & AlOwaybil, R. (2017). An Arabic translation, reliability, and validation of Patient Health Questionnaire in a Saudi sample. Annals of General Psychiatry, 16(1), 32. https://doi.org/10.1186/s12991-017-0155-1
AlHadi, A. N., Almeharish, A., Bilal, L., Al-Habeeb, A., Al-Subaie, A., Naseem, M. T., & Altwaijri, Y. A. (2022). The prevalence and correlates of bulimia nervosa, binge-eating disorder, and anorexia nervosa: The Saudi National Mental Health Survey. International Journal of Eating Disorders, 55(11), 1541–1552. https://doi.org/10.1002/eat.23790
Alsheweir, A., Goyder, E., Alnooh, G., & Caton, S. J. (2023). Prevalence of eating disorders and disordered eating behaviours amongst adolescents and young adults in Saudi Arabia: A systematic review. Nutrients, 15(21), Article 21. https://doi.org/10.3390/nu15214643
Al‐Subaie, A., Al‐Shammari, S., Bamgboye, E., Al‐Sabhan, K., Al‐Shehri, S., & Bannah, A. R. (1996). Validity of the Arabic version of the eating attitude test. International Journal of Eating Disorders, 20(3), 321–324. https://doi.org/10.1002/(SICI)1098-108X(199611)20:3<321::AID-EAT12>3.0.CO;2-2
Alwosaifer, A. M., Alawadh, S. A., Wahab, M. M. A., Boubshait, L. A., & Almutairi, B. A. (2018). Eating disorders and associated risk factors among Imam Abdulrahman bin Faisal university preparatory year female students in Kingdom of Saudi Arabia. Saudi Medical Journal, 39(9), 910–921. https://doi.org/10.15537/smj.2018.9.23314
Belon, K. E., Smith, J. E., Bryan, A. D., Lash, D. N., Winn, J. L., & Gianini, L. M. (2011). Measurement invariance of the Eating Attitudes Test-26 in Caucasian and Hispanic women. Eating Behaviors, 12(4), 317–320. https://doi.org/10.1016/j.eatbeh.2011.07.007
Bin-Abbas, B., Al Sagheir, A., Taher, L., Elbadawi, H., Al Fares, A., Al Harbi, M., Refaat, M., & Ashmawy, A. (2024). ACTION Teens Saudi Arabia: Perceptions, attitudes, motivators, and barriers among adolescents living with obesity, caregivers, and healthcare professionals in Saudi Arabia. Clinical Obesity, e12674. https://doi.org/10.1111/cob.12674
Browne, M. W., & Cudeck, R. (1992). Alternative ways of assessing model fit. Sociological Methods & Research, 21(2), 230–258. https://doi.org/10.1177/0049124192021002005
Byrne, B. M. (2016). Structural equation modeling with AMOS: Basic concepts, applications, and programming (3rd ed.). Routledge. https://doi.org/10.4324/9781315757421
Byrne, B. M., & van de Vijver, F. J. R. (2010). Testing for measurement and structural equivalence in large-scale cross-cultural studies: Addressing the issue of nonequivalence. International Journal of Testing, 10(2), 107–132. https://doi.org/10.1080/15305051003637306
CDC. (2022, June 3). Defining Adult Overweight and Obesity. Centers for Disease Control and Prevention. https://www.cdc.gov/obesity/basics/adult-defining.html
Chen, F. F. (2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling, 14(3), 464–504. https://doi.org/10.1080/10705510701301834
Cheung, G. W., & Rensvold, R. B. (2002). Evaluating goodness-of-fit indexes for testing measurement invariance. Structural Equation Modeling, 9(2), 233–255. https://doi.org/10.1207/S15328007SEM0902_5
Dotti, A., & Lazzari, R. (1998). Validation and reliability of the Italian EAT-26. Eating and Weight Disorders - Studies on Anorexia, Bulimia and Obesity, 3(4), 188–194. https://doi.org/10.1007/BF03340009
Ebrahim, M., Alkazemi, D., Zafar, T. A., & Kubow, S. (2019). Disordered eating attitudes correlate with body dissatisfaction among Kuwaiti male college students. Journal of Eating Disorders, 7(1), 37. https://doi.org/10.1186/s40337-019-0265-z
Fatima, W., & Ahmad, L. M. (2018). Prevalence of disordered eating attitudes among adolescent girls in Arar City, Kingdom of Saudi Arabia. Health Psychology Research, 6(1), Article 1. https://doi.org/10.4081/hpr.2018.7444
Garfinkel, P. E., & Newman, A. (2001). The Eating Attitudes Test: Twenty-five years later. Eating and Weight Disorders, 6(1), 1–21. https://doi.org/10.1007/BF03339747
Garner, D. M., & Garfinkel, P. E. (1979). The Eating Attitudes Test: An index of the symptoms of anorexia nervosa. Psychological Medicine, 9(2), 273–279. https://doi.org/10.1017/s0033291700030762
Garner, D. M., Olmsted, M. P., Bohr, Y., & Garfinkel, P. E. (1982). The Eating Attitudes Test: Psychometric features and clinical correlates. Psychological Medicine, 12(4), 871–878. https://doi.org/10.1017/S0033291700049163
General Authority for Statistics. (2024). Health Determinants Statistics [Text]. Health Determinants Statistics. https://www.stats.gov.sa/en/1247
Ghazzawi, H. A., Nimer, L. S., Haddad, A. J., Alhaj, O. A., Amawi, A. T., Pandi-Perumal, S. R., Trabelsi, K., Seeman, M. V., & Jahrami, H. (2024). A systematic review, meta-analysis, and meta-regression of the prevalence of self-reported disordered eating and associated factors among athletes worldwide. Journal of Eating Disorders, 12(1), 24. https://doi.org/10.1186/s40337-024-00982-5
Haddad, C., Khoury, C., Salameh, P., Sacre, H., Hallit, R., Kheir, N., Obeid, S., & Hallit, S. (2021). Validation of the Arabic version of the Eating Attitude Test in Lebanon: A population study. Public Health Nutrition, 24(13), 4132–4143. https://doi.org/10.1017/S1368980020002955
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. https://doi.org/10.1080/10705519909540118
Jamil, A., Zahid, N., Imtiaz, M., Obaid, H. binte, Muhammad, A., & Jamil, A. (2023). Translation and cross-cultural adaptation of EAT-26 questionnaire in Urdu. MethodsX, 11, 102343. https://doi.org/10.1016/j.mex.2023.102343
Kang, Q., Chan, R. C. K., Li, X., Arcelus, J., Yue, L., Huang, J., Gu, L., Fan, Q., Zhang, H., Xiao, Z., & Chen, J. (2017). Psychometric properties of the Chinese version of the Eating Attitudes Test in young female patients with eating disorders in mainland China. European Eating Disorders Review, 25(6), 613–617. https://doi.org/10.1002/erv.2560
Karl, J. A. (2023). Multigroup Invariance Testing for Cross-Cultural Research. In C. U. Krägeloh, M. Alyami, & O. N. Medvedev (Eds.), International Handbook of Behavioral Health Assessment (pp. 1–17). Springer International Publishing. https://doi.org/10.1007/978-3-030-89738-3_7-1
Khaled, S. M., Kimmel, L., & Le Trung, K. (2018). Assessing the factor structure and measurement invariance of the Eating Attitude Test (EAT-26) across language and BMI in young Arab women. Journal of Eating Disorders, 6(1), 14. https://doi.org/10.1186/s40337-018-0199-x
Koslowsky, M., Scheinberg, Z., Bleich, A., Mark, M., Apter, A., Danon, Y., & Solomon, Z. (1992). The factor structure and criterion validity of the short form of the Eating Attitudes Test. Journal of Personality Assessment, 58(1), 27–35. https://doi.org/10.1207/s15327752jpa5801_3
Leichner, P., Steiger, H., Puentes-Neuman, G., Perreault, M., & Gottheil, N. (1994). Validation d’une échelle d’attitudes alimentaires auprès d’une population québécoise francophone. The Canadian Journal of Psychiatry, 39(1), 49–54. https://doi.org/10.1177/070674379403900110
Maassen, E., D’Urso, E. D., van Assen, M. A. L. M., Nuijten, M. B., De Roover, K., & Wicherts, J. M. (2023). The dire disregard of measurement invariance testing in psychological science. Psychological Methods. https://doi.org/10.1037/met0000624
Maïano, C., Morin, A. J. S., Lanfranchi, M.-C., & Therme, P. (2013). The Eating Attitudes Test-26 revisited using exploratory structural equation modeling. Journal of Abnormal Child Psychology, 41(5), 775–788. https://doi.org/10.1007/s10802-013-9718-z
McEnery, F., Fitzgerald, A., McNicholas, F., & Dooley, B. (2016). Fit for purpose, psychometric assessment of the Eating Attitudes Test-26 in an Irish adolescent sample. Eating Behaviors, 23, 52–57. https://doi.org/10.1016/j.eatbeh.2016.07.006
Melisse, B., de Beurs, E., & van Furth, E. F. (2020). Eating disorders in the Arab world: A literature review. Journal of Eating Disorders, 8(1), 59. https://doi.org/10.1186/s40337-020-00336-x
Memish, Z. A., El Bcheraoui, C., Tuffaha, M., Robinson, M., Daoud, F., Jaber, S., Mikhitarian, S., Al Saeedi, M., AlMazroa, M. A., Mokdad, A. H., & Al Rabeeah, A. A. (2014). Obesity and associated factors—Kingdom of Saudi Arabia, 2013. Preventing Chronic Disease, 11, Article 140236. https://doi.org/10.5888/pcd11.140236
Meshkova, T. A., Mitina, O. V., Shelygin, K. V., Aleksandrova, R. V., & Nikolaeva, N. O. (2023). The Eating Attitudes Test (EAT-26): Psychometric characteristics and factor structure in nonclinical sample of 876 female university students. Clinical Psychology and Special Education, 12(1), 66–103. https://doi.org/10.17759/cpse.2023120104
Montoro-Pérez, N., Montejano-Lozoya, R., Martín-Baena, D., Talavera-Ortega, M., & Rosario Gómez-Romero, M. (2024). Psychometric properties of the Eating Attitudes Test-26 in Spanish schoolchildren. Anales de Pediatría (English Edition), 100(4), 241–250. https://doi.org/10.1016/j.anpede.2024.03.042
Mukai, T., Crago, M., & Shisslak, C. M. (1994). Eating attitudes and weight preoccupation among female high school students in Japan. Journal of Child Psychology and Psychiatry, and Allied Disciplines, 35(4), 677–688. https://doi.org/10.1111/j.1469-7610.1994.tb01213.x
Ocker, L. B., Lam, E. T. C., Jensen, B. E., & Zhang, J. J. (2007). Psychometric properties of the Eating Attitudes Test. Measurement in Physical Education and Exercise Science, 11(1), 25–48. https://doi.org/10.1080/10913670709337010
Papini, N. M., Jung, M., Cook, A., Lopez, N. V., Ptomey, L. T., Herrmann, S. D., & Kang, M. (2022). Psychometric properties of the 26-item eating attitudes test (EAT-26): An application of rasch analysis. Journal of Eating Disorders, 10(1), 62. https://doi.org/10.1186/s40337-022-00580-3
Pereira, R. F., & Alvarenga, M. (2007). Disordered eating: Identifying, treating, preventing, and differentiating it from eating disorders. Diabetes Spectrum, 20(3), 141–148. https://doi.org/10.2337/diaspect.20.3.141
Qian, J., Wu, Y., Liu, F., Zhu, Y., Jin, H., Zhang, H., Wan, Y., Li, C., & Yu, D. (2022). An update on the prevalence of eating disorders in the general population: A systematic review and meta-analysis. Eating and Weight Disorders, 27(2), 415–428. https://doi.org/10.1007/s40519-021-01162-z
Rivas, T., Bersabé, R., Jiménez, M., & Berrocal, C. (2010). The Eating Attitudes Test (EAT-26): Reliability and validity in Spanish female samples. The Spanish Journal of Psychology, 13(2), 1044–1056. https://doi.org/10.1017/s1138741600002687
Rogoza, R., Brytek-Matera, A., & Garner, D. M. (2016). Analysis of the EAT-26 in a non-clinical sample. Archives of Psychiatry and Psychotherapy, 18(2), 54–58. https://doi.org/10.12740/APP/63647
Spivak-Lavi, Z., Latzer, Y., Stein, D., Peleg, O., & Tzischinsky, O. (2023). Differences in the factor structure of the Eating Attitude Test-26 (EAT-26) among clinical vs. Non-clinical adolescent Israeli females. Nutrients, 15(19), Article 19. https://doi.org/10.3390/nu15194168
Spivak-Lavi, Z., Peleg, O., Tzischinsky, O., Stein, D., & Latzer, Y. (2021). Differences in the factor structure of the Eating Attitude Test-26 (EAT-26) in different cultures in Israel: Jews, Muslims, and Christians. Nutrients, 13(6), 1899. https://doi.org/10.3390/nu13061899
World Obesity Federation. (2023). World Obesity Atlas 2023. https://data.worldobesity.org/publications/?cat=19

No competing interests reported.

AppendixEAT16Arabicversion.docx

Download PDF

Editorial decision: Revision requested
26 Sep, 2024
Reviews received at journal
24 Sep, 2024
Reviews received at journal
15 Sep, 2024
Reviewers agreed at journal
28 Aug, 2024
Reviewers agreed at journal
28 Aug, 2024
Reviewers invited by journal
28 Aug, 2024
Editor assigned by journal
19 Aug, 2024
Submission checks completed at journal
19 Aug, 2024
First submitted to journal
15 Aug, 2024

You are reading this latest preprint version

Assessing disordered eating behaviours and attitudes: Factor structure and measurement invariance of the Arabic version of the Eating Attitudes Test (EAT-26) in Saudi Arabia

Status:

Version 1

Abstract

Background

Methods

Results

Conclusions

Plain English summary

Background

Methods

Participants

Procedure

Measures

Eating Attitudes Test

Data Analysis

Results

Factorial Validity of EAT

Descriptive analysis, reliability, and correlations

Measurement invariance across gender and weight status

Discussion

Limitations and future research

Conclusion

Abbreviations

Declarations

References

Additional Declarations

Supplementary Files

Status:

Version 1