Prevalence and characteristics of eating disorders among college students of a nonmetro city of Gujarat.

Context: There is a relative paucity of prevalence data about eating disorders (EDs) in India among young population. Aims: We aimed to estimate the prevalence and characteristics of EDs and abnormal eating behaviors among college students of a nonmetro city of Gujarat. Setting and Design: A cross-sectional survey was done among five colleges of a nonmetro city in Gujarat from February to September 2019. Subjects and
Methods: Total 790 college students were assessed using a semi-structured format, Eating Attitudes Test-26, and Bulimic Investigatory Test Edinburgh followed by structured clinical interview as per DSM-5 criteria for EDs. Statistical Analysis: Outcomes were expressed in frequency, proportion, mean, and standard deviation. P values were calculated by Pearson Chi-square test or Fisher's exact test to determine the significance of the result.
Results: The prevalence of abnormal eating behaviors was 25.2% (n = 199). Anorexia nervosa (AN) was not detected. The prevalence of bulimia nervosa (BN) was 0.2% and other specified feeding or eating disorder (OSFED) was 0.6%. "Being aware of calorie content" (53.7%) and "preoccupation with desire of thinness" (46.3%) were commonly found. "Impulse to vomit after meals" (2.5%) was least common. Lower body mass index was found among subjects with abnormal eating behavior. None of the subjects had amenorrhea. Conclusions: The prevalence of disordered eating behaviors, BN, and OSFED was 25.2%, 0.2%, and 0.6%, respectively. AN was not detected. OSFED was the most common ED and the characteristic "body image disturbance" was the most common symptom. Copyright:

Eating disorders (EDs) are characterized by abnormal eating patterns which may be either too less intake or excessive eating. This leads to physical and emotional health problems to individual. The common forms of EDs are anorexia nervosa (AN), bulimia nervosa (BN), and binge-ED (BED). People with AN have morbid fear of fatness, self-induced starvation, and medical signs of starvation. BN is characterized by episodes of binge-eating and compensatory behaviors like purging or doing excessive exercise. BED is characterized by recurrent episodes of binge eating in the absence of the inappropriate compensatory behaviors of BN. In a review[1] published in 2015, the authors have noted that the estimated lifetime prevalence of AN, BN, and BED is 0.9%, 1.5%, and 3.5%, respectively, in women and 0.3%, 0.5%, and 2.0% in men according to the National Comorbidity Survey. The authors who explored the prevalence of EDs in India have reported that the amount of research in the area of ED in India is relatively sparse.[1] Approximately 20% of college women experience transient bulimic symptoms at some points during their college years.[2]

Within the broader field of EDs, it is important to recognize that the study of EDs in Asia is still in its early stages and that research is limited, albeit quickly expanding.[3] There is a lack of systematic studies which examine the nature and prevalence of EDs in the non-Western world including India as compared to the Western world.[4] Thus, EDs still remain an under-researched area in India.

Hence, considering the relative paucity of data on prevalence of EDs among vulnerable population in India, this study was planned to estimate the prevalence and characteristics of EDs among college students of a nonmetro city of Gujarat state in India.

SUBJECTS AND METHODS

The study was approved by the Institutional Ethics Committee affiliated with Government Medical College located in the concerned city. A cross-sectional descriptive observational study was done between February and September 2019 among undergraduate students of five colleges of a nonmetro city of Gujarat having population of 7.5 lacks. The study sample comprised 834 college students. Subjects were recruited from five randomly selected colleges of the city. Systematic random sampling was used for selection of subjects from each college. The minimum sample size of 771 was calculated using the OpenEpi version two-open source calculator. We assumed prevalence of any ED including AN, BN, BED, other specified feeding or ED (OSFED), and unspecified feeding or EDs (UFED) 2.05% among population and absolute precision 1%. The confidence interval was kept at 95% and design effect was one.

The study was carried out in two steps including screening step followed by a structured clinical interview. Total 834 college students were approached after obtaining permission from the heads of respective institutions. Twenty-seven participants were unwilling to participate in the study. Written informed consent was obtained from each participant after explaining the objectives and method of the study. Subjects were explained how to fill study case record form. The questions in the case record form were explained to subjects, and the meaning of terms that were found difficult was clarified up to their satisfaction. Case record forms were filled in the classrooms under supervision of investigator(s).

The case record form included three sections, namely demographic data including height, weight, and menstrual history; Eating Attitudes Test-26 (EAT-26); and Bulimic Investigatory Test Edinburgh (BITE). EAT-26 and BITE were translated in the vernacular language Guajarati using the standard dictionaries. The translation was validated by back translation to English and comparison with original form by a group of experts who were well versed with both languages. Gujarati version was used for nonprofessional course including arts and commerce course students. English version was used for professional course students.

EAT-26 is a valid, reliable, and cost-effective self-rating questionnaire useful as an objective screening tool for AN.[5] It includes 26 items assessing respondents’ attitude and behaviors related to food and eating. Each item is rated on a six-point Likert scale ranging from “always” to “never.” The total score of EAT-26 may vary from 0 to 78. There are separate behavioral questions with dichotomous responses assessing binge-eating or binge-purging behavior and past history of treatment for EDs. It examines feelings and behaviors associated with AN. If a given subject has EAT-26 score more than or equal to 20 or has “yes” answer on any one dichotomous behavior question or subject is significant underweight, then he or she would be considered to have “possible ED.”

BITE is a 33-item self-report questionnaire designed to identify subjects with symptoms of BN or BED. It has been used as a screening tool to identify binge eaters in given population.[6] The subjects were asked to complete BITE based on their feelings and behavior over the past 3 months. BITE has two subscales, namely symptom subscale and severity subscale. Both provide a degree of symptoms and an index of the severity of binge-eating and binge-purging behavior as defined by their frequency. The maximum possible symptom score is 30. Based on the score obtained on symptom scale of BITE, the subjects were divided into three groups as high (>20), medium (10–19), and low scores (<10). In BITE severity subscale, a score of 5 or more was considered clinically significant and 10 or more indicated a high degree of severity of binge-eating and binge-purging behavior. Those subjects with high score have a high probability of BN.[6]

The EAT-26 or BITE screen-positive subjects were assessed within 15 days after screening by a psychiatrist who was unaware about the screen status of subject. The psychiatrist made individual structured clinical interview according to DSM-5 diagnostic criteria for AN, BN, BED, other specified feeding or eating disorder (OSFED), or UFED and for all screen-positive subjects.

The data were analyzed with GraphPad InStat statistical analysis software version 3.06. Qualitative data were expressed in frequency and proportion whereas quantitative data were expressed in mean and standard deviation. The prevalence of EDs and high-risk abnormal eating behaviors was calculated with 95% confidence interval. P values were calculated by Pearson Chi-square test of independence or Fisher's exact test when applicable to determine the significance of the result and P < 0.05 was considered statistically significant.

RESULTS

Out of 807 participants who returned filled case record form, 790 subjects returned complete case record form giving a response rate of 97.8%. Twenty-seven participants did not agree to participate in the study. The mean age of all subjects was 20.4 years (range: 17–35) and standard deviation (SD) of 1.6. The mean body mass index (BMI) of all subjects was 19.5 and SD was 3.4 (range: 11.0–37.4). The demographic and mean values of BMI, EAT-26, and BITE scores are shown in Table 1.

Table 1

Mean values of variables

Variable	Mean (SD)		P*
	Male (n=34)	Female (n=756)
Age	20.97 (1.05)	20.43 (1.69)	0.0108
BMI	22.32 (4.42)	19.41 (3.38)	<0.0001
EAT-26 score	10.58 (7.33)	7.75 (6.91)	0.0050
BITE score (symptom)	6.35 (3.10)	4.65 (2.94)	0.0011
BITE score (severity)	1.41 (2.19)	1.46 (1.63)	0.3130

*Mann-Whitney test. EAT-26 – Eating Attitudes Test-26; BITE – Bulimic Investigatory Test Edinburg; SD – Standard deviation; BMI – Body mass index

Males had higher BMI, EAT-26, and BITE (symptom) scores compared to females. BITE severity scale did not significantly differ between males and females (1.41 vs. 1.46, P = 0.3130).

Out of 790 screened subjects, the prevalence of EAT-26-positive screens was 20.5% (162 subjects) with risk of ED and the prevalence of BITE-positive screens was 4.7% (37 subjects). Hence, overall 25.2% (n = 199/790) of the subjects were found with disordered eating behaviors on either EAT or BITE. Total 199 subjects (162 + 37) were called for structured clinical interview. Structured clinical interview detected two subjects with BN (prevalence: 0.2%) and five subjects with OSFED (prevalence: 0.6%). All seven subjects who received ED diagnosis on structured clinical interview were female. Out of 162 high-risk subjects screened by EAT-26, 150 (92.6%) were female and 12 (7.4%) were male.

Almost half (n = 84, 51.8%) of the subjects screened by EAT-26 at high risk for EDs had normal BMI. As shown in Table 2, severely underweight (BMI <16) and obese (BMI >30) subjects were more among the EAT-26 screened positive group compared to the EAT-26-negative group.

Table 2

Body mass index of Eating Attitudes Test-26 screened positive and negative subjects

BMI	EAT-26 positive (n=162), n (%)	EAT-26 negative (n=628), n (%)	P*
Severely underweight	47 (29)	66 (10.5)	<0.0001
Underweight	12 (7.4)	218 (34.7)	<0.0001
Normal	88 (54.3)	307 (48.9)	0.4672
Overweight	10 (6.2)	33 (5.3)	0.8999
Obese	5 (3.1)	4 (0.6)	0.0324

*Chi-square test for independence. BMI – Body mass index; EAT-26 – Eating Attitudes Test-26

The responses of all subjects on EAT-26 questionnaire are shown in Table 3 (dieting subscale), Table 4 (bulimia and food preoccupation subscale), and Table 5 (oral control subscale). Subjects were divided into two groups: high risk screen positive (162 subjects) and screen negative (628 subjects). Out of 162 high-risk subjects, the most frequently reported items were “aware of calorie content of food I eat” (53.7%, n = 87), “preoccupied with desire to be thinner” (46.29%, n = 75), “preoccupied with having fat on body” (38.27%, n = 62), and “think about burning up calories when I exercise” (38.27%, n = 62). However, 27.77% (n = 45) were engaged in dieting behavior and only 2.46% (n = 4) had impulse to vomit after meals.

Table 3

Analysis of dieting subscale of Eating Attitudes Test-26

EAT-26 items	Yes response among EAT-26- positive subjects (n=162), n (%)	Yes response among EAT-26- negative subjects (n=628), n (%)	P*
Am terrified about being overweight	53 (32.7)	59 (9.4)	<0.0001
Aware of the calorie content of foods that I eat	87 (53.7)	280 (44.6)	0.0422
Particularly avoid food with a high carbohydrate content	25 (15.4)	31 (4.9)	<0.0001
Feel extremely guilty after eating	11 (6.8)	7 (1.1)	0.0002
Am preoccupied with a desire to be thinner	75 (46.3)	65 (10.4)	<0.0001
Think about burning up calories when I exercise	62 (38.3)	85 (13.5)	<0.0001
Am preoccupied with the thought of having fat on my body	62 (38.3)	63 (10.0)	<0.0001
Avoid foods with sugar in them	40 (24.7)	44 (7.0)	<0.0001
Eat diet foods	27 (16.7)	11 (1.8)	<0.0001
Feel uncomfortable after eating sweets	23 (14.2)	19 (3.0)	<0.0001
Engage in dieting behavior	45 (27.8)	29 (4.6)	<0.0001
Like my stomach to be empty	42 (25.9)	44 (7.0)	<0.0001
Have the impulse to vomit after meals	4 (2.5)	3 (0.5)	0.0357

*Fisher’s exact test. EAT-26 – Eating Attitudes Test-26

Table 4

Analysis of bulimia and food preoccupation subscale of Eating Attitudes Test-26

EAT-26 items	Yes response among EAT-26-positive subjects (n=162), n (%)	Yes response among EAT-26-negative subjects (n=628), n (%)	P*
Find myself preoccupied with food	23 (14.2)	13 (2.1)	<0.0001
Have gone on eating binges where I feel that I may not be able to stop	14 (8.6)	13 (2.1)	0.0002
Vomit after I have eaten	2 (1.2)	5 (0.8)	0.6368
Feel that food controls my life	48 (29.6)	44 (7.0)	<0.0001
Give too much time and thought to food	39 (24.1)	72 (11.5)	0.0001
Enjoy trying new rich foods	88 (54.3)	273 (43.5)	0.0168

*Fisher’s exact test. EAT-26 – Eating Attitudes Test-26

Table 5

Analysis of oral control subscale of Eating Attitudes Test-26

EAT-26 items	Yes response among EAT-26-positive subjects (n=162), n (%)	Yes response among EAT-26-negative subjects (n=628), n (%)	P*
Avoid eating when I am hungry	14 (8.6)	21 (3.34)	0.0083
Cut my food into small pieces	50 (30.9)	72 (11.5)	<0.0001
Feel that others would prefer if I ate more	8 (4.9)	15 (2.4)	0.1113
Other people think that I am too thin	44 (27.2)	106 (16.9)	0.0047
Display self-control around food	58 (35.8)	95 (15.1)	<0.0001
Feel that others pressure me to eat	44 (27.2)	62 (9.9)	<0.0001

*Fisher’s exact test. EAT-26: Eating Attitudes Test-26

As shown in Table 6, BMI did not differ significantly among BITE screened positive and BITE-negative subjects. Tables 7-9 show the results of BITE individual item analysis divided into symptom subscale 1, symptom subscale 2, and binge-eating subscale, respectively.

Table 6

Body mass index of Bulimic Investigatory Test Edinburg screened positive and negative individuals

BMI†	BITE positive (n=37), n (%)	BITE negative (n=753), n (%)	P*
Severely underweight	1 (2.7)	77 (10.3)	0.3258
Underweight	13 (35.1)	251 (33.3)	0.9746
Normal	20 (54.1)	375 (49.8)	0.8802
Overweight	2 (5.4)	41 (5.5)	0.9999
Obese	1 (2.7)	8 (1.1)	0.3587

*Chi-square test for independence. BMI – Body mass index; BITE – Bulimic Investigatory Test Edinburg

Table 7

Bulimic Investigatory Test Edinburg symptom scale item analysis for subscale 1

BITE item	Yes response among BITE-positive subjects (n=37), n (%)†	Yes response among BITE-negative subjects (n=753), n (%)†	P*
Are you a strict dieter?	17 (45.9)	160 (21.2)	0.001
Do you feel a failure if you break your diet once?	16 (43.2)	123 (16.3)	0.002
Do you count the calories of everything you eat, even when not on a diet?	8 (21.6)	59 (7.8)	0.0092
Do you ever fast for a whole day?	30 (81.1)	382 (50.7)	0.0003
Does your pattern of eating severely disrupt your life?	10 (27.0)	53 (7.0)	0.0003
Would you say that food dominated your life?	24 (64.9)	266 (35.3)	0.0004
Do you ever eat and eat until you are stopped by physical discomfort?	5 (13.5)	21 (2.8)	0.0055
Are there times when all you can think about is food?	15 (40.5)	120 (15.9)	0.0005
Do you eat sensibly in front of others and make up in private?	7 (18.9)	51 (6.8)	0.0143
Do you ever experience overpowering urges to eat and eat and eat?	7 (18.9)	43 (5.7)	0.0063
When you are feeling anxious, do you tend to eat a lot?	1 (2.7)	44 (5.8)	0.7162
Does the thought of becoming fat terrify you?	21 (56.8)	188 (24.9)	<0.0001
Do you ever eat large amounts of food rapidly?	19 (51.3)	179 (23.8)	0.0006
Are you ashamed of your eating habits?	6 (16.2)	37 (4.9)	0.0119
Do you worry that you have no control over how much you eat?	13 (35.1)	60 (7.9)	0.0001
Do you turn to food for comfort?	10 (27.0)	114 (15.1)	0.0630
Do you deceive other people about how much you eat?	6 (16.2)	22 (2.9)	0.0012

*Fisher’s exact test. BITE – Bulimic Investigatory Test Edinburg

Table 9

Bulimic Investigatory Test Edinburg binge-eating item analysis

BITE item	Yes response among BITE-positive subjects (n=37), n (%)	YES response among BITE-negative subjects (n=753), n (%)	P*
Do you ever binge on large amounts of food?	6 (16.2)	56 (7.4)	0.0618
If yes, do such binges leave you feeling miserable	7 (18.9)	20 (2.6)	0.0001
If you do binge, is this only when you are alone?	5 (13.5)	15 (1.9)	0.0016
Would you go to great lengths to satisfy an urge to binge?	5 (13.5)	26 (3.4)	0.0119
If you overeat do you feel very guilty?	15 (40.5)	137 (18.1)	0.0021
Do you ever eat in secret?	4 (10.8)	37 (4.9)	0.1190
Would you consider yourself to be a compulsive eater?	6 (16.2)	35 (4.6)	0.0094
Does your weight fluctuate by more than 5 lbs in a week?	14 (37.8)	67 (8.8)	<0.0001

*Fisher’s exact test. BITE – Bulimic Investigatory Test Edinburg

DISCUSSION

The present study is comparable in sample size with other similar studies done among the same age group populations from the non-Western world.[4] This study observed that overall 25.2% (n = 199) of the subjects were found with disordered eating behaviors on either EAT or BITE. The prevalence of positive screens for EDs among college students ranges from 8% to 17% and up to 20% in a national survey.[7] A recent study from Mysore city of South India among students of 15–25 years’ age group, reported that 26.6% of students displayed abnormal eating attitudes and eating discrepancies.[8] Hence, the results of the current study are consistent with previous similar studies.

We did not find any case of AN in study sample. A study among 662 MBBS course college students of Madras also reported that they could not find any case of AN or bulimia among their subjects.[4] The authors also noted that many experienced mental health professionals shared that they have hardly seen a case of AN or BN during their clinical practice.[4] A extensive review[9] from Western countries has examined studies which used similar two-stage screening methods reported that the average prevalence of AN is 0.3% among young female population.[9] According to literature, the prevalence of bulimia ranges from 1% to 4% among young women and BN is more prevalent than AN.[2] Our study also supports this trend.

We detected two subjects with BN (0.2% prevalence). This is lesser than the prevalence reported by a study among 504 North Indian college girl students from an industrial town in the foothills of the Himalayas which reported a 0.4% prevalence of BN using BITE screening tool.[10] The lower prevalence of BN in our study can be explained due to study design where screened participants were also subjected to structured clinical interview.

The striking finding in our study is that we found five subjects (0.6% prevalence) with other specified feeding or EDs (OSFED). Previous studies have shown that the majority of the patients with EDs fall into the category of ED not otherwise specified.[1112] Hence, the subthreshold forms of ED were most prevalent among our subjects. A study from Madras in South India reported that subsyndromal EDs what they described as “eating distress syndrome” are more prevalent in India as compared to classical EDs.[4]

All detected subjects with any ED were females in the current study. However, total male subjects in the study were 34 (4% of sample) and none of them were detected with any ED. Less is known about prevalence among males, but AN is far less common in males than in females, with clinical populations generally reflecting approximately a 10:1 female-to-male ratio.[13]

“Enjoy trying new rich foods” was a commonly endorsed item by all subjects in both screen-positive and screen-negative subjects. This suggests that the presence of “trying new rich foods” is a good predictor of having ED. “Being aware of the calorie content of food” was the most common reported item among EAT-26-positive subjects. The characteristic body image disturbance expressed as “preoccupation with a desire to be thinner” was the second most commonly endorsed item. “Thought of burning up calories when exercise” and “preoccupation with thought of having fat on body” both were the third most common items among them. More than half (56.7%) of the BITE-positive subjects also reported “thought of becoming fat terrify them.” This suggests that distorted perception of body shape and obsessive preoccupations about thinness were highest among subjects. This finding is in contrast to the notion that distorted body image is predominantly seen in the Western world and not in non-Western population.[14] Studies from non-Western countries reported that patients often present without prominent expression of weight concerns.[15] Younger individuals with AN may have atypical presentation including denial of “fear of fatness.”[13] A recently published review article[1] also reported that nonfat phobic variant of AN is frequently reported in various countries and hence retained in ED NOS category. The earliest published case reports and case series of AN from India do not show body image disturbance. Although, the relentless desire for slimness is often not considered important for attractive appearance in many areas of Gujarat and heftiness and is accepted a sign of attractive and “healthy look.” The finding of our study indicates that the Western theme of attractive appearance in slimness has been spread with time in this region too.

Current or recent history of amenorrhea was not reported by any of the screen-positive subjects in our study. The presence of amenorrhea suggests strong biological disturbance. Amenorrhea is often associated with severely undernutritional status. DSM-IV criterion that required an individual to exhibit amenorrhea for minimum 3 months to qualify for a diagnosis of AN was removed in DSM-5.[16] The present study also supports this change. In a review article in 2009, the authors concluded that amenorrhea is a physiologically important disturbance often seen in association with AN and a useful indicator of clinical severity. However, they have suggested that it should not be included as essential core criteria in DSM-5 for the diagnosis of AN.[17]

“Impulse to vomit after meals” (1%, n = 7) and “vomit after meals” (1%, n = 7) were the least prevalent items among subjects and did not differ between screen-positive and screen-negative subjects. A study conducted among 2489 Brazilian college students has reported that 3.3% of subjects had vomiting behavior.[18]

Although the weight fluctuation of more than 5 lbs/week was more among BITE-positive subjects compared to BITE-negative subjects (37.8% vs. 8.8%, P < 0.0001), BMI did not significantly differ among BITE-positive and BITE-negative subjects. Literature suggests that most of the patients with BN maintain normal weight throughout the duration of disorder.[16] Severely underweight (BMI <16) and obese (BMI >30) subjects were more among EAT-26 screen-positive groups compared to EAT-26 screen-negative ones. This suggests that lower BMI and obesity are associated with disordered eating behavior.

This is the first descriptive original research study emerging from Gujarat in Western India which has examined relatively large sample compared to previous studies on EDs among similar population in the region. However, the study was based on college students’ sample, hence the findings cannot reflect the general population trends of EDs. Future research is recommended involving large-scale community sample to confirm the findings of this study.

CONCLUSIONS

The prevalence of disordered eating behaviors, BN, and OSFED was 25.2%, 0.2%, and 0.6%, respectively. AN was not detected. OSFED was the most common ED, and the characteristic “body image disturbance” was the most common symptom.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.

Table 8

Bulimic Investigatory Test Edinburg symptom scale item analysis for subscale 2

BITE item	Yes response among BITE-positive subjects (n=37), n (%)	Yes response among BITE-negative subjects (n=753), n (%)	P*
Do you have a regular day-to-day eating pattern?	29 (78.4)	630 (83.7)	0.3702
Can you always stop eating when you want to?	29 (78.4)	496 (65.9)	0.1528
Are you able to leave food on the plate at the end of a meal?	26 (70.3)	535 (71.0)	1
Does how hungry you feel determine how much you eat?	22 (59.5)	464 (61.6)	0.8630
Are your eating habits what you would consider to be normal?	19 (51.4)	558 (74.1)	0.0040

*Fisher’s exact test. BITE – Bulimic Investigatory Test Edinburg