How ‘Contagious’ Can Eating Disorders Be in The Eyes of Patients?

It has often been suggested that eating disorders might be the result of a “contagious” social learning experience, especially through imitation or mimicking behavior of others. This might be the case within the smaller context of family and peer relations, but this could also happen on a larger social level, when media disseminate and glamorize a picture of specialness around eating disorders. More specifically, social contagiousness may be more likely to occur within the high-density environments of eating disorder communities, such as special web sites or online forums and clinical settings in which eating-disordered patients are treated together.

Do patients themselves realize that a social learning process might be involved in the development of their own eating disorder? Since we could not find any prior studies of patients’ opinions on this subject, we designed and carried out a survey in Flanders, the Dutch-speaking part of Belgium.

The Online Survey

An anonymous internet survey was put online for 2 months through a special server of the University of Leuven, Belgium. A call for participation was sent by email to a large group of (ex) patients who had been admitted during the previous 2 years at the Eating Disorders Unit of the Alexian Brothers Psychiatric Hospital in Tienen, Belgium. Another call was posted on the web site of the largest Flemish self-help organization (www.anbn.be).

In the first part of the survey, participants were asked a few general questions, including self-diagnosis of the type of eating disorder they were showing most recently:

• ANR, or anorexia nervosa restrictive type (underweight, restricted food intake, no purging);
• ANM, or anorexia nervosa mixed type (underweight, purging and/or binge-eating);
• BN, or bulimia nervosa (binge eating, more or less normal weight); or
• EDNOS, or eating disorder not otherwise specified (different from previous ones).

The second part of the survey contained two series of questions about the possibility that social influence played a role in the development of an eating disorder: (a) an opinion on the degree to which “bad” examples might trigger an eating disorder, and (b) the degree to which the respondents themselves had experienced such social influence in their own lives.

We received a total of 203 completed questionnaires. According to their self-diagnosis, the participants were divided into the following subgroups: 64 ANR (31.5%), 65 ANM (32%), 33 BN (16.3%), and 41 EDNOS (20.2%). All respondents were females between 14 and 53 years of age, with an average age of 25 (SD =7.7). The mean age at onset was 15.4 years (SD=4.1) and at first treatment for an eating disorder was 19.4 years (SD=5.9).

Can ‘Bad’ Examples Trigger an Eating Disorder?

“Some people believe that girls may start controlling their body weight by means of restricted food intake or purging behavior under the influence of some ‘bad’ examples.” With this introduction, the participants were asked to rate their (dis)agreement about the following statements (Table 1):

• ‘This problem is seriously overrated and occurs far less than generally thought’: more than half (52.2%) of the participants agreed partially or totally.
• ‘This can happen only in girls who already have a problem or already show the first signs of an eating disorder’: only about a third (34.5%) supported this statement.
• ‘This is especially the case when someone in the close surrounding (family member or friend) is giving immediate cause’: opinions were equally divided (about 40% each way).
• ‘This happens predominantly through influences from the internet via web sites or chat sessions related to eating disorders’: about 30% had no opinion or disagreed, while 40% endorsed this statement.
• ‘This happens predominantly through stories about famous people (from television, movies, music, fashion) who have an eating disorder’: the majority (53.2%) endorsed this, while one-fourth (25.1%) expressed no opinion.
• ‘This can happen in treatments where one meets people who already have an eating disorder’: the majority (45.4%) agreed with this, while a third of respondents disagreed.

Table 1: Opinion on Social Influences*
	Totally agree (%)	Partially agree (%)	No opinion (%)	Partially disagree (%)	Totally disagree (%)
The problem is overrated.	13.8	38.4	14.8	28.6	4.4
Happens in girls with problems	8.9	25.6	10.3	41.4	13.8
Someone in close surroundings	12.3	27.1	19.7	34.0	6.9
Web sites or chat sessions	13.3	26.6	30.5	28.1	1.5
Stories of famous people	18.7	34.5	25.1	19.2	2.5
Contacts in therapy	15.8	29.6	21.6	28.6	4.4

*n=203

Have You Yourself Been Under the Influence of Such ‘Bad’ Examples?

The following part of the survey addressed the degree to which respondents themselves had experienced social influences linked to the development of their eating disorder (Table 2).

• ‘I have been influenced by someone in my close surroundings (family member, friend) who had an eating disorder’: more than half (56.2%) of the respondents did not report this, while 13.5% mentioned that it had a strong influence upon them.
• ‘I have been influenced by examples on the internet, such as special web sites or chat sessions about eating disorders’: again more than half (53.7%) did not experience this while a third (33.4%) reported some influence.
• ‘I have been influenced by stories about famous people (from television, movies, music, fashion) who themselves had an eating disorder’: the vast majority (71.3%) denied such personal experience, whereas one-fourth (25.7%) reported some influence.
• ‘I have been influenced by conversations or therapies together with other people who had an eating disorder’: a large majority (61.2%) did not report this kind of experience.

Table 2: Personal Experience of Social Influence*
	No influence (%)	Some influence (%)	Strong influence (%)
From close surroundings	56.2	30.3	13.5
Web sites or chat sessions	53.7	33.4	12.9
Stories of famous people	71.3	25.7	3.0
Contacts in therapies	61.2	30.3	8.5

* n=203

If the respondent in each of the examples mentioned either ‘some influence’ or ‘strong influence,’ she was asked to specify what happened to her in relation to the actual eating disorder: before any sign of an eating disorder; during the course of a yet untreated eating disorder; or during the treatment of their eating disorder (Table 3). Slightly more than half of the participants (50.5%) who mentioned influences from their close surroundings had experienced this before the start of their eating disorder, whereas more than a third (38.7%) already showed the signs of an eating disorder but weren’t yet in treatment. Two-thirds (63.2%) of the respondents mentioned influences from websites or chat sessions before they went into treatment. About the same result (57.9%) was found for the impact of stories about famous eating disordered people. Clearly influences from close contacts with fellow patients happened most often (80.3%) during treatment, although others had been influenced through conversations outside a treatment setting.

Table 3: Moment of Some/Strong Influence*
	Before disorder (%)	Disorder untreated (%)	During treatment (%)
From close surroundings (n=88)	50.5	38.7	10.8
Web sites or chat sessions n=93)	14.7	63.2	22.1
Stories of famous people (n=58)	31.6	57.9	10.5
Contacts in therapies (n=78)	6.5	13.2	80.3

*n=203

Finally, we calculated the correlations between expressed opinions on the possibility of social influence through ‘bad’ examples and the degree to which respondents themselves reported such influences in their personal life. We found a significantly negative relation with respect to internet influences (r= -.25, p=0.003), stories of famous people (r= -.47; p=0.000), and contacts with patients in treatment (r= -.47; p=0.000). No significant correlation was found between opinions and personal experiences regarding influences from the close social environment. So, with the exception of the direct surroundings, there is a remarkable reverse relationship: The more respondents endorse the statement that ‘bad’ examples can trigger an eating disorder, the less they report such influences in their personal life! Whether this might express some lack of self-criticism, we leave it up to the reader to judge because we have no convincing explanation for this strange finding. One explanation might be that this may be an example of the so-called Third Person Effect, known in media studies, in which people claim not to be affected directly by media (e.g., advertising) but perceive others to be susceptible.