MD. Hadjigeorgiou Charalampos, PhD. Tornaritis Michael, MD PhD Savvas Savva, MA Solea Antonia, PhD Kafatos Antonis


Objective: the aim of the present study is the evaluation of secular trends in the eating behaviours of Cypriot adolescents assessed in 2003 and 2009-2010. Method: a representative sample of children and adolescents aged 10 to 18 were invited to complete the questionnaires EAT-26 and EDI-3, at their school. Results: the percentage of participants scoring above normal on the EAT-26 (>20), the drive for thinness (DT) and body dissatisfaction (BD) subscales of the EDI have remained on the same high levels for both genders over 6 years. Participants scoring above normal on the bulimia (B) subscale of the EDI have increased since 2003 for both genders. Females consistently scored higher on most scales. Discussion: given the evidence that disordered eating attitudes have remained common with an increasing bulimic trend, intervention in Cyprus is greatly warranted.



Eating disorders have received much public, medical and research attention over the years,1 and rightfully so since they lead to serious physical, psychological and emotional health problems,2 as well as having a high mortality rate.2-4 Eating disorders generally include anorexia nervosa (AN), bulimia nervosa (BN), binge-eating disorder (BED) and other eating disorder not otherwise specified (EDNOS)2,5.

Research has supported the theory that the development of eating disorders is determined by a number of predisposing factors which can be grouped into three large clusters: sociocultural, familial and individual factors.6,7 Concerning the sociocultural cluster, epidemiological research has consistently pointed out the higher prevalence of eating disorders in Western, industrialized and developed countries.8,9 One such country is Cyprus where in a period of thirty years it has transformed itself from a traditional, rural society to a European one as far as way of life, eating and clothing habits, entertainment and role models towards young people are concerned. Furthermore, although the importance given to weight and size varies between cultures no difference in the prevalence of eating disorders were found when Mediterranean countries were compared to other Western countries10. Familial factors found to be relevant in the development of eating disorders include a family history of an eating disorder, obesity, parental mood disorder or substance abuse to name a few.6, 11 The present study confines itself to measuring individual factors linked to eating disorders such as age, gender, dieting habits. 6 ,12, 13

A growing body of research is accumulating in other countries highlighting a number of important facts concerning the distribution of eating disorders among the population. Firstly, females are at a higher risk for eating disorders as opposed to males, 14 in the general population the female to male ratio of individuals with AN and BN is 10 to 1.12 Secondly, different age groups are at a different risk for eating disorders6,12 with adolescent girls and young females being the most vulnerable.13

The distribution of eating disorders is either measured by incidence or prevalence rates. Incidence rates are the actual number of new cases in a population per year whereas prevalence rates are the actual number of cases in a population at a certain time point. Although numbers vary from study to study, the average figure for the prevalence of AN is 0.28% and that of BN is 1%.8 The need for epidemiological data on eating disorders in Cyprus became evident with the doubling of cases (incidence) that were monitored at Makarios Hospital at the Child Psychiatric Ward over the period 1992 until 2002. Furthermore, there was a feeling amongst the pediatric community that they were diagnosing eating disorders a lot more frequently than they had been in the past.

Owing to the fact that no data existed in Cyprus pertaining to eating attitudes and behaviours the first epidemiological study took place in 2003 aiming to do just that. Since eating disorders usually onset and are first diagnosed during the adolescent years and even in late childhood between the ages of ten and twenty3,8,13 this is the age group that the study targeted. Approximately 2000 students aged between 10-18 attending primary and secondary schools from four districts in Cyprus were invited to participate in the study. The students were asked to complete two questionnaires, the Eating Attitudes Test-26 (EAT-26) and the Eating Disorder Inventory-3 (EDI-3) which are the most widely used instruments in eating disorders.15 The EAT-26 was selected because it is a short questionnaire containing only 26 items that can be easily administered and scored and the EDI-3 for its emphasis in assessing the behavioural and psychological characteristics of individuals with eating disorders.6

Studies concerning the prevalence of eating disorders over time have shown mixed results. Some studies have found that the prevalence has remained stable.16, 17 In two nationally representative surveys in Norway in 1991 and 2004 for example, the prevalence of obesity doubled whereas the prevalence of eating disorders remained constant.16 Currin et al. (2005) found that in the years 1994-2000 the incidence of AN remained consistent over the period while regarding BN there was an increase with rates declining after a peak in 1996.18 Others have found that the prevalence is clearly increasing19, 20, 21 whilst the age of onset of these disorders is decreasing.22 Survival analysis of data from the National Comorbidity Replication in 2001-2003 in Massachusetts suggests that the risk of BN, BED increased with successive birth cohorts.21 Figures from research however vary widely depending on methodology and sample selection. To determine whether in Cyprus eating disorders are becoming more frequent with the passage of time, six years after the original epidemiological study, the same methodology and age range was used to compare the present results with the previous.


Both in 2003 and in 2010 a representative sample of students was recruited to take part in the study. The National School Registry was consulted and the number of different schools and students in these schools counted. Schools in a given district were selected randomly from a list of all regional schools. Children above the age of ten and secondary school students were chosen randomly from the school register. These students were then given two questionnaires, the EAT-26 and the EDI-3 to complete at school. All students at a given school completed the questionnaires simultaneously under the supervision of a trained professional. The participation in the study was voluntary and students were ensured that their answers would be kept confidential.

In 2003, a total of 2033 participants took part in the study (mean age 14.5, SD 2.2), 955 males and 1068 females (10 participants did not state their gender). In 2010, a total of 1128 students completed the questionnaires (mean age 12, SD 3.5), of which 731 were female and 393 male (4 participants did not state their gender).


The Eating Attitudes Test (EAT-26) is a self-report questionnaire of 26 items assessing the range of symptoms of anorexia and bulimia nervosa. Each item is rated on a 6-point Likert scale ranging from “never” to “always”. Response are then assigned a score from 0-3 with non-symptomatic answers being “sometimes”, “rarely” and “never” which are given a score of 0. Symptomatic answers are responses “always”, “usually” and “often” and are scored 3, 2, and 1 respectively. Total EAT scores (the sum of all 26 items) was calculated for each participant. The clinical cut-off point for eating disturbances is a score of 20 or above, which is indicative of serious eating disturbances or weight concerns. Previous validations of the EAT-26 have found good reliability scores across studies ranging from .79 to .94. In this study the alpha coefficient was found to be .77.

The EDI-3 is also a self-report questionnaire consisting of 91 items measuring psychological domains that have relevance in the understanding and treating of eating disorders.6 It is intended mainly for individuals 13 years of age and older however it has also been used with younger adolescents 11-12 years of age. The EDI-3 presents items on a 6-point Likert scale using a 0-4 point scoring system (e.g. 001234) with null scores for the non-symptomatic answers. The 91 items of the questionnaire are organized into 12 primary scales: three eating-disorder specific scales and nine general psychological scales relevant but not specific to eating disorders.6

The three eating disorder specific scales are of most relevance to this research, and these include: (a) Drive for thinness (DT) which consists of 7 items measuring an individual’s preoccupation with dieting and fear of weight gain. (b) Bulimia (B) comprises of 8 items and assesses an individual’s tendency to think about and engage in periods of overeating. (c) Body Dissatisfaction (BD) scale consists of 10 items assessing the extent of people’s dissatisfaction with the overall shape and size of their body.6 Raw scores were calculated for each of these scales and using the EDI-3 interpretive guidelines the percentage of participants scoring in the elevated, typical and low clinical ranges was assessed. Reliability scores for the three Eating Disorder Risk scales were found to be in the high .80s to low .90s across normative groups. In this study Cronbach’s alpha coefficient for the DT, B and BD scales was .76, .62 and .86 respectively.

Statistical Analysis

All descriptive statistics including mean, frequencies and percentages were carried out using SPSS 18.0 statistical software. T-test analyses were performed to assess gender differences and differences in 2003 and 2010. For all analyses α < 0.05 was used as statistically significant difference between groups.


Table 1. Age distribution of the sample


Number %


Number %

10-13 29.2 11.8
14-16 31.2 25.5
17-19 17.6 16.4
Unspecified age 22.0 46.3

The range of EAT-26 scores for the entire sample was from 2 to 58 with mean score of 15.4 (SD= 8.9). The EAT scores of the participants are presented in Table 2.

Table 2. Mean EAT scores and the percentage of participants scoring above cut-off 20 on the EAT-26

  2003 2009
n SD >20 EAT-26

N (%)

n SD >20 EAT-26

N (%)

Males 804 13.4 8.2 151 (18.8) 112 13.2 8.1 21 (18.8)
Females 913 16.9 9.3 314 (34.4) 237 17.6 9.1 85 (35.9)

VThere are significant differences between participants in 2003 and 2009 (p<.05)

T-test analyses showed that the difference in the mean EAT-26 scores between males and females was statistically significant both in 2003 (t (df = 1715) = -8.13, p < .000) and in 2009 (t (df=347) = -4.44, p < .000). In other words, females consistently scored higher on the EAT-26 when compared to males.

Table 3. Mean and standard deviation of EAT-26 subscales

Males Females
2003 2009 2003 2009
n SD n SD n SD n SD
Dieting 809 7.54+5.5 112 8.05+5.7 922 11.03+6.9 242 11.29+6.9
Bulimia and Food Preoccupation 818 2.28+2.5 113 2.15+2.4 925 2.07+2.3 239 2.35+2.5
Oral Control 816 3.61+3.2* 112 2.96+2.8 926 3.80+3.4 240 3.97+3.3

*p < .05

Concerning gender comparisons statistical analyses revealed that in 2003 females scored higher on the Dieting subscale when compared to males (t (df= 1729) = -11.7, p< .000). A tendency was found for males to have higher scores on the Bulimia and Food Preoccupation subscale (t (df = 1741) = 1.85, p< .06). No statistically significant difference was found between males and females on the Oral Restriction subscale. In 2009, a different pattern is observed; significant differences were found whereby females scored higher on the Dieting (t (df = 352) = -4.65, p< .000) and Oral Restriction (t (df= 350) = -2.78, p< .006) subscales but not on the Bulimia and Food Preoccupation subscale (t (df = 350) = -0.69, p < .489).

Table 4. The scoring of participants on the DT scale of the EDI-3

Drive for Thinness MALE FEMALE

(n= 843)

n (%)


(n= 278)

n (%)


(n= 930)

n (%)



n (%)

Normal (n) 774 (91.8) 258 (92.8) 697 (74.9) 349 (71.4)
At Risk (n) 66 (7.8) 19 (6.8) 207 (22.3) 118 (24.1)
High (n) 3 (0.4) 1 (0.4) 26 (2.8) 22 (4.5)


T-test analyses showed that the difference in mean DT scores between males and females was significant for 2003 (t (df= 1771) = -14.71, p< .000) and for 2009 (t (df= 765) = -9.08, p< .000). Females are more preoccupied with dieting and weight gain when compared to males. There is a tendency for participants to have higher DT scores in 2009 when compared to 2003 (t (df = 2544) = -1.945, p< .052.

Table 5. The scoring of participants on the B scale of the EDI-3


(n= 864)

n (%)


(n= 280)

n (%)


(n= 955)

n (%)



n (%)

Normal (n) 627 (72.6) 177 (63.2) 621 (65) 320 (63.6)
At Risk (n) 228 (26.4) 99 (35.4) 329 (34.5) 170 (33.8)
High (n) 9 (1) 4 (1.4) 5 (0.5) 13 (2.6)

T-test analysis indicates that females have higher mean scores on the B subscale of the EDI-3 in comparison to males in 2003 (t (df = 1817) = -3.31, p < .001 but not in 2009 (t (df = 781) = -.959, p < .338). Further analysis illustrate that mean scores in 2009 are higher than in 2003 for both females (t (df= 1456) = -2.047, p < .041) and for males (t (df= 1142) = -2.588, p < .01).

Table 6. The scoring of participants on the BD scale of the EDI-3

Body Dissatisfaction MALE FEMALE

(n= 826)

n (%)


(n= 269)

n (%)


(n= 934)

n (%)



n (%)

Normal (n) 749 (90.7) 244 (90.7) 674 (72.2) 354 (73)
At Risk (n) 69 (8.4) 24 (8.9) 231 (24.7) 119 (24.5)
High (n) 8 (1) 1 (0.4) 29 (3.1) 12 (2.5)

Females have higher mean scores than males in 2003 (t (df = 1758) = -13.33, p < .000) and in 2009 (t (df = 752) = -5.48, p < .000) on the BD subscale of the EDI-3. No differences in mean scores were found between the years 2003 and 2009.


The present study examines secular trends in the patterns of eating disorders among the Cypriot adolescent population from 2003 until 2009, six years apart. In 2003, the first epidemiological study of the prevalence of eating disorders was undertaken with a nationally representative sample of adolescents. In 2009, using the same sampling procedure with a smaller number of adolescents, this was repeated in order to assess changes if any in the eating habits and beliefs in a new generation of ten to eighteen year olds.

The first results of the study back in 2003 confirmed the overall feeling among professionals that Cypriot adolescents struggle with distorted eating attitudes and beliefs to a much greater extent than was originally thought. Nearly 20% of adolescent boys and almost double this (34%) of adolescent girls have high scores of 20 or above on the EAT-26. Although the EAT-26 alone does not yield a specific diagnosis of either anorexia nervosa or bulimia nervosa, it is a useful screening tool to assess ‘eating disorder risk’. Practically, this means that in 2003, 465 students out of the 1700 who answered the EAT-26 meet the criteria to be interviewed by a qualified professional to determine if they meet the diagnostic criteria for an eating disorder.23, 24 This translates into 27% of the adolescent population; a much higher percentage than was initially expected.

Studies in other Mediterranean countries vary in the percentage of participants who have high scores on the EAT questionnaire. In South Africa roughly 19% of adolescent females have high scores, 25 in Israel 19.5%, 26 in Turkey 11.5% of the adolescent population had disturbed eating behaviour, 27 Spain 12.3% of girls and 3.2% of boys.28 Greece is perhaps the best comparison for Cyprus’ results as it is the closest Mediterranean country with which we share the same language, religion and influences. Studies have found that around 25% of girls and 5-8% of boys score at or above cut-off 20 on the EAT-26.29 One can observe that the prevalence of disordered eating habits and beliefs are higher among the Cypriot youth a finding also discovered in a cross-cultural study comparing British and Cypriot women.30

With the Eating Disorder Risk Scale of the EDI-3 one can gather a more complex view of the eating habits and beliefs of teenagers. The DT construct assesses a preoccupation with dieting and fears of weight gain and is associated with the onset and maintenance of eating-disorder symptoms in clinical samples. Among the non-clinical population a raw score in the high range is rare affecting around 1% of adolescents.6 Only 1.6% of Cypriot adolescents had scores in the high range, however 15.4% had scores in the typical clinical range which has been labeled ‘at risk’ in our study and is higher than what should be the case in a nonclinical sample (which is expected to be around 8% according to Garner.6 This finding highlights the fact that Cypriot teenagers are indeed preoccupied with their weight and perhaps this is also the reason that such a high percentage of participants had high scores on the EAT-26.

Bulimia is the tendency towards incidences of uncontrollable overeating that could be followed by the impulse to engage in self-induced vomiting. 31 The B scale measures the presence of thoughts and behaviours consistent with binge eating and vomiting to lose weight. In non-clinical adolescent populations 1% have scores in the high range6 and among Cypriot adolescents 0.8% have scores in this range indicative usually of a high level of psychopathology and the presence of a clinical eating disorder. ‘At risk’ adolescents comprised of 30.6% of the sample and typically among those 18 years and younger such scores are rare (12%) while they are more common for adults (30%). 6 Overeating tendencies of Cypriot adolescents are perhaps more similar to those of adults.

Body dissatisfaction reflects the belief that certain parts of the body that are associates with shape are too large.31 The BD scale measures a person’s displeasure with the overall shape and size of areas of the body of concern to those with eating disorders such as the stomach, hips and thighs.6 To those who are vulnerable it can be a risk factor responsible for initiating and maintaining weight-controlling behaviours that lead to eating disorders. In the adolescent population raw scores in the high ranges are rare (3%) in the Cypriot adolescent population this was found to be 2.1% and in the ‘at risk’ range 17%. Those in the latter range usually do not need any further evaluation since preoccupation with weight and shape are common to women in today’s society and those who are heavier are likely to have higher scores.6, 19

The fact that on all scales females scored higher than males came as no surprise and is a finding that has been consistently established throughout countless research. Females are at a higher risk for eating disorders as opposed to males and perceive more social pressure to be thin; 32 more girls answered that they would like to be thinner and are more dissatisfied with their shape and size.

In 2009-2010, a new group of adolescent participants were selected. Over six years one can see that the patterns of disordered eating attitudes and behaviours have remained relatively unchanged for both genders. One difference however was apparent on the B subscale of the EDI-3. In 2009 there was no difference between males and females in their scores, a finding that is replicated on the Bulimia and Food Preoccupation subscale of the EAT-26. Furthermore there is an increase in the number of adolescents at risk and vulnerable for bulimia in comparison to 2003. This indicates that it is likely that adolescents of both genders hold similar beliefs concerning over-eating tendencies and they both are equally likely to eat large amounts of food in secrecy in response to being emotionally upset. Furthermore, as a culture it can be said that bulimic trends are more accepted than anorexic ones.

A number of socioeconomic changes have taken place in Cyprus over a period of six years. On the 1st of May 2004 Cyprus joined the European Union and in 2008 the Cyprus pound was replaced by the Euro. A large influx of European immigrants has been observed in addition to the already existing tendencies of urbanization among Cypriots themselves. These changes mean that an increasing number of young people are being exposed to European cultures and attitudes. The media, films, comedies, have promoted the idea that when upset we comfort eat and this may explain why it is such a popular belief among adolescents today.

One limitation of the study was that it included children between the ages of 10 and 12 and although it has been used successfully with younger adolescents, 6 primarily it is for use with people of 13 years of age and older. There may have been an issue with children’s understanding of the questionnaires. Secondly, a smaller sample size was used in 2009 in comparison to 2003. The next step for researchers in Cyprus would be to identify those children with a clinical eating disorder or use stricter criteria such as the inclusion of BMI to recognize the very high risk population.

This study highlighted that a very large percentage of teenagers in Cyprus hold disordered and maladaptive eating behaviours and thoughts. The implications are that in schools some kind of intervention is greatly warranted as the number of students with borderline scores is too great to be ignored. It is unfortunate that a country that was once so heavily reliant on a Mediterranean diet that is healthy, desirable and enviable should have its youth troubled by detrimental eating habits.


  1. Vandereycken W. History of Anorexia Nervosa and Bulimia Nervosa. In: Fairburn CG, Brownell KD, editors. Eating disorders and obesity: A comprehensive handbook (2nd ed). New York: The Guilford Press, 2002, p. 151-154.
  2. Golden NH, Katzman DK, Kreipe RE, Stevens SL, Sawyer SM, Rees J, et al. Eating disorders in adolescents: Position paper of the society for adolescent medicine. Journal of Adolescent Health 2003: 33: 496-503.
  3. Lock J, Fitzpatrick KK. Advances in psychotherapy for children and adolescents with eating disorders. American Journal of Psychotherapy 2009: 63 (4): 287-303.
  4. Steinglass JE, Sysko R, Glasofer D, Albano AM, Simpson HB, Walsh BT. Rationale for the application of exposure and response prevention to the treatment of anorexia nervosa. International Journal of Eating Disorders 2010: 1-8.
  5. Keel PK, Brown TA. Update on course and outcome in eating disorders. International Journal of Eating Disorders 2010: 43: 195-204.
  6. Garner DM. Eating disorder inventory-3: Referral form manual. Lutz, Fl: Psychological Assessment Resources; 2004.
  7. Tylka TL, Subich LM. Examining a multidimensional model of eating disorder symptomatology among college women. Journal of Counseling Psychology 2004: 3: 314-328.
  8. Hoek HW. Distribution of eating disorders. In: Fairburn CG, Brownell KD, editors. Eating disorders and obesity: A comprehensive handbook (2nd ed). New York: The Guilford Press, 2002, 233-237.
  9. Kovacs TK. An epidemiological study of eating disorders among high school students in Romania. Journal of Cognitive and Behavioural Psychotherapies 2010: 10 (1): 77-86.
  10. Santonastaso P, Scicluna D, Colombo G, Zanetti T, Favaro A. Eating disorders and attitudes in Maltese and Italian Female Students. Psychopathology 2006: 39: 153-157.
  11. Striegel-Moore RH, Fairburn CG, Wilfley DE, Pike KM, Dohm FA, Kraemer HC. Toward an understanding of risk factors for binge-eating disorder in black and white women: a community-based case-control study. Psychological Medicine 2005: 35: 907-917.
  12. Grabarek C, Cooper S. Graduate students’ social and emotional functioning relative to characteristics of eating disorders. The Journal of General Psychology 2008: 135 (4): 425-451.
  13. Becker AE, Grinspoon SK, Klibanski A, Herzog DB. Current concepts: Eating disorders. New England Journal of Medicine 1999: 340: 1092-1098.
  14. Lewinsohn PM, Seeley JR, Moerk KC, Striegel-Moore RH. Gender differences in eating disorder symptoms in young adults. International Journal of Eating Disorders 2002: 32: 426-440.
  15. Gross J, Rosen JC, Leitenberg H, Willmuth ME. Validation of the eating attitude test and the eating disorders inventory in bulimia nervosa. J Consul Clin Psychol 1986: 54: 875-876.
  16. Zachrisson HD, Vedul-Kjelsas E, Götestam KG, Mykletun A. Time trends in obesity and eating disorders. Intenational Journal of Eating Disorders 2008: 41 (8): 673-680.
  17. Hoek HW. Incidence, prevalence and mortality of anorexia nervosa and other eating disorders. Current Opinion in Psychiatry 2006: 19 (4): 389-394.
  18. Currin L, Schmidt U, Treasure J, Herschel J. Time trends in eating disorder incidence. The British Journal of Psychiatry 2005: 186: 132-135.
  19. Bungam TJ, Lofton S. The association of body mass index, perceived body mass index, and predictors of eating disorders among a sample of college students. American Journal of Health Education 2003: 34 (6): 337-342.
  20. Latzer Y, Shatz S. Comprehensive community prevention of disturbed attitudes to weight control: A three-level intervention program. Eating Disorders: The Journal of Treatment and Prevention 1999: 7 (1): 3-31.
  21. Hudson JI, Hiripi E, Pope Jr HG, Kessler RC. The prevalence and correlates of eating disorders in the national comorbidity survey replication. Biological Psychiatry 2007: 61 (3): 348-358.
  22. Jones JM, Bennett S, Olmsted MP, Lawson ML, Rodin G. Disordered eating attitudes and behaviours in teenaged girls: a school based study. CMAJ 2001: 165 (5): 547-552.
  23. Dotti A, Lazzari R. Validation and reliability of the Italian Eat-26. Eating and Weight Disorders 1998: 3: 188-194
  24. Garner DM. Self-report measures for eating disorders. Current content, Social and Behavioral Sciences 1993: 8.
  25.  Szabo CP, Allwood CW. A cross-cultural study of eating disorders in adolescent South African females. World Psychiatry 2004: 3 (1): 41-44.
  26. Latzer Y, Tzischinsky O. Eating attitudes in a diverse sample of Israeli adolescent females: a comparison study. Journal of Adolescence 2005: 28 (3): 317-323.
  27.  Bas M, Asci FH, Karabudak E, Kiziltan G. Eating attitudes and their psychological correlates among Turkish adolescents. Adolescence 2004: 39: 593-599.
  28. Alonso A, Rodriguez M, Alonso J, Carretero G, Martin M. Eating disorders. Prevalence and risk profile among secondary school students. Social Psychiatry Psychiatric Epidemiology 2005: 40: 980-987.
  29. Yannakoulia M, Matalas AL, Grivetti LE. From frugality to disordered eating behaviour: The case of Greece. In: Ruggiero GM, editor. Eating Disorders in the Mediterranean Area. USA: Nova Science Publishers, Inc, 2003, p. 151-158.
  30. Katsounari, I. Self-esteem, depression and eating disordered attitudes: A cross-cultural comparison between Cypriot and British young women. European Eating Disorders Review 2009: 17 (6): 455-461.
  31. Garner DM, Olmstead MP, Polivy J. Development and validation of a multidimensional eating disorder inventory for anorexia nervosa and bulimia. International Journal of Eating Disorders 1983: 2 (2): 15-34.
  32. Anderson AE, Holman JE. Males with eating disorders: Challenges for treatment and resear