Prevalence of bulimic behaviors and trends in eating attitudes among Turkish late adolescents

Adolescence,  Winter, 2006  by Gul Kiziltan,  Efsun Karabudak,  Sibel Unver,  Emine Sezgin,  Ayse Unal

• E-mail
• Print
• Link

Subclinical forms of eating disorders have been investigated in terms of eating attitudes and behaviors (Button & Whitehouse, 1981; King 1989). Recently, Wichstrom (1995) described the importance of such research to determine potential causes of milder forms of eating disorders. Until now, epidemiological studies on eating problems have been conducted mostly among European and North American populations. However, emerging evidence suggests that eating disorders in fact occur among a wide range of ethnic, cultural, and socioeconomic groups. Unhealthful eating behaviors remain predominant among late adolescents, especially those in college (Koszewski, 1996). On the other hand, the onset of bulimia nervosa typically occurs in late adolescence or in the early twenties (Bruce & Wilfley, 1996). It is commonly believed that eating disorders are associated with Western societies due to their emphasis on thinness, and that anorexia nervosa (AN) and bulimia nervosa (BN) are much less common in non-Western countries (Nasser, 1988). A number of studies have identified the presence of these symptoms, sometimes reaching levels of disorder, in Asian countries such as Hong Kong (Lee, 1993), India (Khandelwal & Saxena, 1990), China (Song & Fang, 1990), Iran (Abdollahi & Mann, 2001), and Japan (Kuboki et al., 1996). Also, disordered eating attitudes and behaviors have been identified in the Middle East (Neumark-Sztainer, Palti, & Butler, 1997). However, studies related to eating problems and disorders are limited in Turkey. In developing countries in particular, modernization and industrialization have been implicated in changing individual subjectivity, self-determination, and instrumental agency, all of which are significant in the development of eating disorders (Littlewood, 1995). Studies have reported varying rates of prevalence of bulimia nervosa from 4.5% to 13% in community samples using a range of measuring tools (Halmi, Falk, & Schwartz, 1980; Pyle et al., 1983). Fairburn and Beglin (1990) suggest that the prevalence of bulimia nervosa among young adult women and adolescents is about 1%.

The present study sought to provide information regarding eating pathology by investigating the presence and severity of unhealthful eating habits and attitudes in female and male students in Turkey. The first objective for the study was to determine the prevalence of bulimic disorders in a group of late adolescents. The second objective was to survey the effect of gender on eating attitudes.

METHOD

Participants

The study population consisted of 18- to 24-year-old university students from three public and two private universities in Ankara, Turkey. The universities were chosen from a list of all public and private universities in Ankara using a proportional-size probability formula. In each university, 65 students were selected randomly. A total of 325 students were invited to participate, but 17 students declined. Of the 308 surveys distributed, 8 were not returned, leaving 300 surveys for analysis (response rate 94.0%). A total of 150 male and 150 female students participated in the study. Measurement and data collection were conducted during one semester between March and June 2004.

Instruments

Questionnaire. A self-administered questionnaire was used to gather demographic information, nutritional habits, and health status of the students.

BMI. Height and weight were measured and BMI was calculated. Since reference data on BMI for the Turkish population are not available, the World Health Organization (WHO) classification of BMI categories was used for estimating obesity in Turkish late adolescents. The students were grouped into four categories: underweight, normal-weight, pre-obese, and obese in accordance with the cut-off points of <18.5 kg/[m.sup.2], 18.5 to 24.9 kg/[m.sup.2], 25.0 to 29.9 kg/[m.sup.2] and >29.9kg/[m.sup.2], respectively (WHO, 2003).

The Bulimic Investigatory Test, Edinburgh (BITE) proposed by Henderson and Freeman (1987) was used to measure bulimic behaviors. Among the current sample, alpha coefficient for the test was 0.86. The BITE is a 33-item self-report measure, designed to identify subjects with symptoms of bulimia or binge eating. The BITE consists of two subscales: symptom scale, which measures the degree of symptoms present, and severity scale, which provides an index of the severity of bingeing and purging behaviors as defined by their frequency. The maximum possible score is 30 for the symptom scale. A symptom score of 20 or more indicates a highly disordered eating pattern and the presence of binge eating and also indicates a high probability for a diagnosis of bulimia; the medium range (10-19) suggests an unusual eating pattern; the low range (0-9) falls within normal limits. Such a score indicates the absence of both compulsive eating and binge eating. Henderson and Freeman (1987) suggest some additional questions in BITE which do not contribute to the subject's final score, but which are relevant to the study. In our study, additional questions were asked on issues such as dieting, regular periods, consultation with a professional about dieting/eating, suffering from an eating disorder, and membership in a slimming club. Participants were also asked whether they engaged in appropriate physical activities; if so, what they were and how often. Moreover they were asked if they ate main meals and/ or snacks, and if so, how often and what kind of foods.