Psychometric Properties of Binge Eating Scale Indonesian Version

This research aims to examine the psychometric properties of the Indonesian version of the Binge Eating Scale (BES) and to describe characteristics of binge eating among emerging adults aged 19 – 25 years old in 3 private universities in Indonesia. The Indonesian version of BES was translated forward and backwards, according to the second edition of the ITC guidelines to confirm conceptual and linguistic equivalence. The result provided factor structure evidence and showed good reliability of the BES Indonesian version. No significant difference between man and woman and Body Mass Index was observed.


INTRODUCTION
Binge Eating Disorder (BED) is a behavior characterized by eating large amounts, accompanied by feelings of loss of control in this episode, followed by repeatedly occurring sadness, without any anticipation of weight gain as in bulimia nervosa [1].
Binge eating disorder is associated with severe implications in both clinical and non-clinical populations [2,3]. Binge eating disorder is also associated with increased BMI, anxiety, depression, decreased quality of life, increased risk of death as well as the risk of suicide [2 -9].
Its prevalence in various countries seems to be increasing, especially in adolescence and emerging adults [10]. There are two peaks of binge eating onset, after puberty, on average at 14 years and 19-25 years [11,12]. Various studies show binge can occur in both men and women [12,13].
Binge eating was increasingly found in various parts of the world, Indonesia, as a country that has an increasing economy, and the rise of social media use, which tends to increase the external influence on the prevalence of binge eating.
The prevalence of binge eating withvarious problems had to be anticipated. So far in Indonesia, no research has found the prevalence of binge eating, therefore, valid and reliable measurements are undoubtedly needed. For this reason, it was important to have a valid measurement tool to measure binge eating in the Indonesian population.
Binge Eating Scale (BES) is a very well-known measurement for binge eating severity as well as a screening tool [3,14,15]. The purpose of this study is to examine the factor structure and reliability of the BES Indonesian version and relationships between BES with BMI and sex.

Measure
A self-report instrument, BES measures cognitive/ emotional symptoms and behaviors related to binge eating, consisting of 16 items, out of which 8 items describe behavioral manifestations and 8 items describe cognition/ feelings, each item consists of 3-4 choice statements/responses that indicate the severity of each binge eating characteristic being measured. Weighted scores range from 0-3 (0 indicate no Binge problems, 3 indicates a severe binge eating problem). The total score has a range of 0-46. Cut off score: non / mild ≤ 17, moderate 18-26, severe ≥ 27.
The Indonesian version of BES was carried out through a process of translational adaptation, namely forward and back translation, according to guidelines issued by the second edition of the ITC. The translation was carried out separately by three people, who were fluent in English and professionals and then reviewed by a bilingual research expert who knows the purpose of the research to get the translation appropriate. The panel results were then translated back into English by 3 professionals and one of them was a native speaker. The results of the translation were reviewed once again by bilingual researchers, who understood the purpose of the study. The Indonesian version of BES carried out trials on 40 students, then reviewed by a bilingual researcher who understood the purpose of the study to obtain a simple and easy-to-read measurement tool.

Statistical Analysis
Confirmatory Factor Analysis (CFA) was used to examine the factor structure of the BES Indonesian version. Some fit was used to test the CFA model with a fit or fit number as used by Escrivá-martínez (2019), chi-square has p < .05, CFI shows good fit if the value is more than .09, and root-mean-squared. The error of approximation (RMSEA) was declared fit if the value was less than .08, the Comparative Fit Index (CFI), the Goodness of Fit Index (GFI) and the Relative Fit Index (RFI) were declared fit if the value was greater or equal to 0.90. The analysis of BES was performed using SPSS and Lisrel Student 8.8 (Jöreskog & Sörbom). The reliability coefficient was calculated by Cronbach Alpha.

Factor Structure and Reliability of the BES Indonesian version.
The results of the translation of each item of BES are shown in Table 1. Summarized results of the analysis and loading factors of each item of BES are shown in Table 2. Each item showed a loading factor greater than 0.05 so that all items are significant. Match / fit is seen from X 2 = 169.49 (N = 553; p = .000); df = 86; RMSEA = 0.042; CFI = 1.00; CGI =, RFI = 0.99; GFI = .96. The results indicate that RMSEA, CFI, CGI, RFI, and GFI all show a good fit. Unidimension of BES Cronbach's alpha = 0.831, emotional/cognitive factors Cronbach's alfa = .754 and behavior factor Cronbach's alpha = 0.612. b. Sometimes when I eat "forbidden food" I eat. on a diet (regulating food intake), I feel as if I have "failed". and even eat more.
c. Often, when I overeat on a diet, I have the habit of telling myself, "I've failed now, why not just". When that happens, I eat even more.
d. I have the habit of regularly starting a strict diet for myself, but I break my diet by continuing to overeat. My life seems like "in a party situation (a lot of eating)" or in a "starvation" situation.
8. a. I rarely eat so much food that I feel uncomfortably full afterwards.
b. Usually around once a month, I eat large amounts of food, and eventually I feel very full c. I have regular periods or times throughout the month for eating large amounts of food, whether at mealtimes or at snack times or snacks.
d. I eat so much food that I regularly feel very uncomfortable after eating, and sometimes feel a little nauseous .
9. a. My calorie intake levels generally don't go up very high or fall very low b. Sometimes after I overeat, I will try to reduce my calorie intake to almost nothing (or eat almost nothing) to compensate for the excess calories I have eaten.
c. I have a regular habit of overeating at night. It seems that my routine is not feeling hungry in the morning, but overeating at night.
d. In my adult years I have had week long periods where I starved myself. This period occurs after a period I overeat. It seems that I live a life like the "in a party situation (a lot of eating)" or in a "starvation" situation. d. I feel so embarrassed by the overeating that I have, that I choose times to overeat when I know no one will see me. I feel like a "hidden eater".
13. a. I eat three times a day with only one snack once in a while.
b. I eat three times a day, but I usually also eat small meals between meals.
c. When I eat small meals that make me full, I get used to skipping my regular meals.
d. There are regular periods when I seem to be eating constantly, without a planned meal schedule or without eating 14. a. I didn't think much about controlling or controlling my unwanted appetites.
b. There are at least certain times when I feel my thoughts are occupied (preoccupied with) trying to control or control my appetite.
c. I feel that I often spend a lot of time thinking about how much food I've eaten or about trying not to eat anymore d. It seemed that most of the time I was awake I was occupied (preoccupied with) thoughts about eating or not eating. I felt like I was constantly struggling not to eat.

15.
a. I don't think much about food b. I have a strong desire to eat but it only lasts for a short time.
c. I have days when it seems I can think of nothing but food.
d. Most of my days, I seem to be occupied (preoccupied with) thoughts about food. I feel like I live to eat.
16. a. I usually know whether I am physically hungry or not. I take the right portion of food to satisfy my hunger b. Sometimes, I am not sure whether I know that I am physically hungry or not. At such times, it is difficult for me to know how much food I should take to satisfy my hunger.
c. Although I may know how many calories I should eat, I don't know the amount of "normal" food. for me.

Sex Difference in BES scores
Analysis of the differences in BES scores based on sex grouping, obtained in men (M = 7.756; SD = 7.271) and women (M = 8.836; SD = .348); obtained t (330) = -1.679, p > .05. It was found that there were no differences in BES scores in men and women.

The Relationship between BES Scores Based on BMI
The relationship between the Indonesian version of BES scores with BMI using Pearson's Moment coefficient analysis obtained r = 0.304; (p <0.05). There is no relationship between BMI and BES scores. The relationship between BMI with the two factors of BES, namely behavioral and cognitive/emotional manifestations, was carried out and the results obtained showed no relationship between BMI and behavioral manifestations (r = .308; p> 0.05) and BMI with factors cognitive / emotions (r = 0.422; p > 0.05).

DISCUSSION
Based on the research results obtained, the Indonesian version of BES looks valid to be used to measure the prevalence and severity of binge eating in the population of Indonesia.
The results of this study also show the validity of the two factors present in BES, namely, behavioral and cognitive/emotional manifestations. The loading factor was examined by the BES factors in accordance with the original version. The results of this study show similarities with studies conducted in populations of Saudi Arabia and Spain. There was a weak correlation between BMI and BES scores, the same results were obtained in studies conducted in populations of Spain and Portugal [3,14]. Another study also found there was no BMI difference [15].

CONCLUSION
Binge eating has high rates of persisting or worsening in early young adulthood [14]. This study support the existing literature that found binge eating in emerging adulthood. The Indonesian version of BES has good validity. The 2-factor BES was revealed to be important in Indonesia, cross-culturally. There were language and cultural differences among Indonesian ethnics, which required a scale that uses standard language.
This Indonesian version of BES can be used immediately. Thus, it can help improve planning of health promotion, prevention, and treatment of binge eating.

ETHICS APPROVAL AND CONSENT TO PARTICIPATE
Not applicable.

HUMAN AND ANIMAL RIGHTS
No Animals were used in this research. All human research procedures followed were in accordance with the ethical standards of the committee responsible for human experimentation (institutional and national), and with the Helsinki Declaration of 1975, as revised in 2013.

CONSENT FOR PUBLICATION
Informed consent was taken from all the patients when they were enrolled.

AVAILABILITY OF DATA AND MATERIALS
Not applicable.