Hanan El-Sayed Badr

Correspondance
Department of Community Medicine and Behavioral Sciences
Faculty of Medicine
Kuwait University
P.O. Box 24923 Safat
13110 Kuwait
Email: hanan@hsc.edu.kw , hanan29@yahoo.com

A thorough understanding of weight stigma and its impact may be important to document the social and psychological consequences of obesity, and may be central to revealing the totality of effects of excess weight on body image satisfaction and the consequent behavioral tendencies. The terms 'obesity' and 'overweight' are frequently used interchangeably to describe the condition of excess weight, but in a health context are distinguished using body mass index (BMI) definitions. Although this distinction is important for identifying health risks, it is less apparent whether BMI cutoffs are meaningful with respect to stigma and bias resulting from excess weight. It is important to acknowledge that obese individuals face several layers of bias and self-fulfilling prophesies then may occur in which obese individuals behave in ways consistent with stereotypes.
Obesity comprises a major health problem worldwide and in Kuwait it is one of the most crucial health concerns. With the increase in oil wealth of the country the socioeconomic status of the nationals has improved remarkably. The government has shared the oil wealth with nationals through providing free education, housing, health care and other subsidies. With the increase in personal wealth, one feature that has become very widespread is the increased prevalence of sedentary lifestyle. At the same time, fast food consumption has become quite common. The above changes are accompanied by a rise in overweight and obesity as well as chronic illness. A recent study showed that among females in the age group of 20-24, about 51% were overweight and obese⁽¹⁾. The problem starts very early during childhood; another study found that 45% of the girls aged 13-20 years had BMI =25 kg/m²⁽²⁾.
Taken together, research documents a widely held perception that overweight people have multiple negative characteristics, ranging from flaws in personal effort (such as poor self-discipline or laziness), to central attributes of competence, attractiveness, and morality^(3,4). This stereotype perception is translated in several behavioral tendencies such as avoiding approaching clothes shopping, restaurants, and social gatherings where the issue of weight can be focused. Less work has examined why this population has become increasingly denigrated, and why it is socially acceptable to hold negative stereotypes about obese individuals.
Accordingly, the aim of the present study is to assess the association between obesity and body image behavioral tendencies among Kuwaiti female University students and the predictors of such associations using the Body Image Avoidance Questionnaire (BIAQ).

Data for this paper were collected through a cross-sectional survey of Kuwaiti female university students conducted during October 2007 to January 2008. Six faculties (practical and theoretical) were randomly selected for the study. About 90 students were randomly selected from each of three practical faculties (engineering, science and pharmacy) and three theoretical ones (Law, Arts and Commerce). A total of 553 female students were approached and those who completed the interview after getting a verbal consent were 525 female university students with a response rate of 94.9%.
The questionnaire was developed in English and then translated into Arabic. It consisted of two parts; the first was some socio-demographic personal data like age, social standard assessed by the level of education of both student's father and mother together with the family monthly income. The second was the Body Image Avoidance Questionnaire (BIAQ)⁽⁵⁾. It was used to measure behavioral tendencies that frequently accompany body-image disturbance. The questionnaire is a 19-item (one item was omitted as it was inconvenient with the students) instrument designed to assess avoidance situations that provoke concern about physical appearance, such as avoidance of social outgoings and tight-fitting clothes. Each item was measured on a 4 Likert scale ranging from "Never" to "Always". For example, a question was asked on whether the respondent wears baggy clothes, and whether this occurred always, sometimes, seldom or never. A weight of 4 was given to the item if it occurred always and a weight of 1 was given if it never occurred. The 18 items were summed to arrive at a composite score of body image avoidance behavior experience, with a range between 18 and 72.
Anthropometric measurements were carried out where weight was measured to the nearest kg and the height was measured to the nearest cm. Body mass index (BMI) was then calculated using the formula BMI=weight in kg/height in m². BMI was divided to normal (<25kg/m²) and overweight and obese for BMI = >25kg/ m².
Comparisons between normal BMI and overweight and obese according to the socio-demographic features were done by calculating the Chi-square test. Comparisons between the mean body image avoidance score and different socio-demographic factors as well as BMI were computed using student-t test and ANOVA. Multivariate analysis using binary logistic regression was then conducted in order to assess the factors that significantly predicted the prevalence of higher than median body image avoidance score.
Factor analysis using Principal Component Analysis (PCA) and Varimax rotation method was performed on the 18 items of the body image avoidance questionnaire. This analysis reflects the variance related uniquely to specific subsets of variables. Reliability Coefficient was performed to estimate the internal consistency between the studied stressors. The level of significance was p < 0.05 and CI = 95%.

The study was conducted among 525 female university students in Kuwait. The mean age of the students was 19.5±1.5 and that ranged from 15-27 years. The distribution of BMI among the Kuwaiti female university students according to different socio-economic factors is illustrated in Table 1. About 30% of the students were overweight and obese. More than half of this group were 20 years old and above that significantly is the reverse picture among the younger age group (p<0.01). They also were represented more in the theoretical faculties than the practical ones although the difference was not significant. Students who belonged to fathers with a level of education less than university were significantly more frequent among overweight and obese students than the students within the range of normal BMI (40.9% and 30.5% respectively). The same applies to the level of mother's education but the difference was not significant. The highest percentage of the overweight and obese students significantly reported monthly income of <1000 KD.
The PCA with Varimax rotation of the 18 items of the body image avoidance questionnaire generated four factors with Eigen values greater than unity as illustrated in Table 2. The four components explained about 43% of the total variances. The first factor explained about 13% of the total variation in the sample. The highest loadings on the factor were rejection of social gatherings if weight will be checked, discussed, in presence of thin persons, or if eating is involved and avoidance of clothes shopping. These items relate to "social activities".
The second component accounted for 12% of the total variances. The highest loadings on the factor were wearing baggy clothes, disliked clothes, wore dark colored and special set of them. These behaviors represented "clothing". The third component explained 9.4% of the total variances. The highest loadings on the factor were behaviors related to food such as restricting the amount of eaten food, only eating fruits, vegetables and low calorie diet and fasting for a day or longer. These grouped behaviors relate to "eating at restaurants".
The last component described 8.2% of the total variances. It relates to "grooming and weight" behaviors such as self-weighing, self looking in the mirror and get dressed or made up.
The mean score of the body image avoidance questionnaire of the students was 42.5 ± 6.1. The bivariate association between the score and different socio-demographic background of the students was demonstrated in Table 3. The mean score was significantly higher in the students who belonged to low level of educated mothers than those with higher level of education (43.1 vs. 42 respectively). The overweight and obese students scored significantly higher than their counterparts of normal BMI (45.9 and 41.1 respectively). Students who reported low monthly income <1000 KD showed significantly the highest mean score (44.2) among different monthly income categories.
The multivariate analysis was computed to determine the predictors of the body image avoidance behaviors among the female university students in Kuwait as shown in Table 4. The students' total score represented the dependent variable and was divided into two groups, those who scored less than the median (42) were scored as (0) and those with scores = the median were scores as (1). The six variables in Table 3 represented the independent variables. The stepwise binary logistic regression analysis revealed that BMI was the first predictor for the body image avoidance behaviors. Overweight and obese students were 3.7 times more likely to practice these behaviors than their normal BMI counterparts (p<0.001). The monthly income accounts for the second predictor for these behaviors where students with lowest level of income were at twice risk relative to those with the highest monthly income level (p<0.01). The father's level of education ranked the third and last predictor although it was insignificant in the bivariate analysis. Students whose fathers were of level of education less than university were at higher risk of 63% to practice body image avoidance behaviors than their co-respondents whose fathers were educated up to the level of university or more (p<0.05).

Rotation method: Varimax with Kaiser Normalization
Most loadings equal to almost 0.5 or more are bold

Variables		Adjusted Odds Ratio	p value	CI (95%)
BMI Normal (RG) Overweight and obese	1.304	3.682	<0.001	2.433-5.573
Monthly income (KD) <1000 1001-1500 > 1500 (RG)	0.763 0.298	2.145 1.347	0.009 0.002 0.208	1.321-3.482 0.847-2.143
Father education Less than university University and above (RG)	0.457	1.633	0.042	0.408-0.984

Six independent variables entered the model: age, type of faculty, father education, mother education, monthly income and BMI.
RG: Reference Group
KD: Kuwaiti Dinar

The present cross-sectional study of 525 Kuwaiti female university students aged 15-27 revealed that overweight and obesity, lower monthly income and father's level of education were positively correlated with reported negative body image behavioral tendencies. The main four behaviors: social activities, clothing, eating at restaurants and grooming and weight.
Overweight and obesity was the major predictor for practicing one or more of these behavioral propensities. Female students whose BMI was = 25 kg/m² were 3.7 times more likely to behave negatively than their counterparts with normal BMI (<25 kg/m²). Monthly income ranked as the second predictor for negative body image behavioral tendencies where those who reported low monthly income were at 2 times higher risk to perform these adverse behaviors than their correspondents with high monthly income. The same applies to students' father's level of education, as students who belonged to high educated fathers were at 63% lower risk to carry out these behavioral trends than their equivalents who belonged to low educated fathers.
It has been documented that obesity is more prevalent among lower socioeconomic groups. Sobal & Stunkard's (1989) review of 144 published studies demonstrated that persons of lower socioeconomic status (SES) are at increased risk of obesity across industrialized nations, and that this pattern is more consistent for obese women than men⁽⁶⁾. More recent studies reveal that SES has a linear relationship with obesity^(7,8). These findings are consistent with the results of our study as the fathers' level of education mostly reflects the level of monthly income. This can explain the finding that although there was a significant positive relationship of students' mother's level of education and obesity in the bivariate analysis- as mothers are usually responsible for preparing and planning for the style of food intake of the whole family and are responsible for the quality of food consumed by the family, and hence appeared this direct association with overweight and obesity, but in the multivariate analysis the role of mothers was diluted by the stronger effect of monthly income highlighted by the fathers' level of education which was insignificant on the bivariate level. This shows that although mothers education had great influence on the weight of their children the socio-economic level has the upper hand and is outweighed in controlling this factor powered by the father's level of education.
Perceptions about the causes of obesity may contribute to weight stigma and bias. Assumptions that obesity can be prevented by self-control, that individual non-compliance explains failure at weight-loss, and that obesity is caused by emotional problems, are all examples of attitudes that contribute to negative bias⁽⁹⁾. Poor body image and low self-esteem are main psychological outcomes that direct the individual to practice several undesirable behavioral tendencies as a method of coping or as a self-defending mechanism against these negative perceptions and attitudes towards body image dissatisfaction. Most of these behavioral tendencies originate from social rejection by peers, poor quality of interpersonal relationships and in addition it might have a potential negative impact on academic outcomes among these university female students.
It has been supported by other work that even parents hold negative stereotypes about overweight, which they communicate to their children^(10,11), and which may have serious emotional and health consequences for overweight youngsters. Other work indicates that some
overweight people have negative weight attitudes and react to stigma by applying negative stereotypes to themselves⁽¹²⁾. Some work indicates that weight bias occurs irrespective of an individual's own body weight, and that overweight people themselves tend to express bias^(13-16).
In order to determine the reasons for social disadvantages of obesity, more research is needed to examine whether variables like SES or intellectual performance become associated with obesity because of a potential common genetic or environmental link, or whether these consequences are created by stigma. Identification the underlying causes of these unfavorable behavioral tendencies is necessary to advance the understanding of this fascinating topic.
It is documented that obesity itself can open the gate to a series of health problems. Its co-morbid disorders represent significant public health concerns; they are considered the leading causes of morbidity and premature mortality around the world^(17-20). Those most exposed to stigma, for instance, may be vulnerable to psychological effects such as depression and social effects for example restricted social activities, and isolation. Consequences of bias such as isolation or social withdrawal could contribute to the exacerbation of obesity through psychological vulnerabilities that increase the likelihood of over-eating and sedentary activity⁽²¹⁾.
It is essential to guide the stigma reduction efforts among youngsters and cultural pervasive negative attitudes of the entire community towards obesity, a major public health concern all over the world.

1. Ministry of Health, State of Kuwait; 2008. Eastern Mediterranean Approach for control of non-communicable diseases (EMAN). Survey of risk factors for chronic non-communicable diseases.
   
2. Ministry of Health, State of Kuwait; 2005. Kuwait Nutrition Surveillance system.
   
3. Puhl R, Brownell KD. Obesity, bias and discrimination. Obes Res 2001; 9: 788-805.
   
4. Teachman BA, Gapinski KD, Brownell KD, Rawlins M, Jeyaram S. Demonstrations of implicit anti-fat bias: the impact of providing causal information and evoking empathy. Health Psychol 2003; 22: 68-78.
   
5. Rosen JC, Srebnik D, Saltzberg E and wendt S. Development of Body Image Avoidance Questionnaire, Psychological Assessment 1991; 3:32-37. Instrument reprinted by permission of authors and publisher.
   
6. Sobal J, Stunkard AJ. Socioeconomic status and obesity: a review of the literature. Psychol Bull 1989; 105: 260-275.
   
7. Goodman E. The role of socioeconomic status gradients in explaining differences in US adolescents' health. Am J Public Health 1993; 89: 1522-1528.
   
8. Gortmaker SL, Must A, Perrin JM, Sobol AM, Dietz WH. Social and economic consequences of overweight and young adulthood. New Engl J Med 1993; 329: 1008-1012.
   
9. Obesity Action Coalition (OAC). Understanding obesity stigma. An educational resource provided by the obesity action coalition.
   
10. Kraig KA, Keel PK. Weight-based stigmatization in children. Int J Obesity 2001; 25: 1661-1666.
    
11. Adams GR, Hicken M, Salehi M. Socialization of the physical attractiveness stereotype: parental expectations and verbal behaviors. Int J Psychol 1988; 23: 137-149.
    
12. Crocker J, Cornwell B, Major B. The stigma of overweight: affective consequences of attributional ambiguity. JPSP 1993; 64:60-70.
    
13. Teachman BA, Brownell KD. Implicit anti-fat bias among health professionals: is anyone immune? Int J Obesity 2001; 25:1525-1531.
    
14. Neumark-Sztainer D, Story M, Harris T. Beliefs and attitudes about obesity among teachers and school health care providers working with adolescents. J Nut Educ 1999; 31: 3-9.
    
15. Crandall C, Biernat M. The ideology of anti-fat attitudes. J Appl Soc Psychol 1990; 20: 227-243.
    
16. Harris MB, Waschull S, Walters L. Feeling fat: motivations, knowledge, and attitudes of overweight women and men. Psychol Rep 1990; 67: 1191-1202.
    
17. Lopez R. Urban sprawl and risk for being overweight or obese. Am J Public Health 2004; 94:1574-1579.
    
18. James PT, Rigby N, Leach R. The obesity epidemic, metabolic syndrome and future prevention strategies. Eur J Cardiovasc Prev Rehabil 2004; 11:3-8.
    
19. Moran R. Evaluation and Treatment of Childhood Obesity. Am Fam Physician 1999; 59:871-873.
    
20. Orzano AJ, Scot JG. Diagnosis and Treatment of Obesity in Adults: An Applied Evidence-Based Review. J Am Board Fam Pract 2004; 17:359-369.
    
21. R. M. Puhl and K. D. Brownell Psychosocial origins of obesity stigma: toward changing a powerful and pervasive bias. Obesity reviews 2003; 4,213-227.