My sister is bulimic and has been in therapy for several months now. She seems to be making progress, but this eating disorder seems to rule her life nevertheless. Overwhelmed with conflicting desires, she is obsessed with food and her appearance. I see her suffer and wonder what has caused her to develop such behaviors. I know that there are several factors that can play a role in the inception of an eating disorder. Because of my sisterís problems, I have become interested in the interplay between familial relationships and bulimia. Is there a relationship between family interactions and bulimia?
There have been numerous studies about the characteristics of a bulimic's family. One of the earliest by Laurence Igoin-Apfelbaum (1985), studied 21 women who were diagnosed by the DSM-III as bulimics. In the group of patients, two patterns of family background could be found. Thirteen patients were from broken homes, and a common characteristic of these families was that the father virtually disappeared from the life of the daughter. The twelve other patients came from close knit families, in which the sacred union of these families against the outside world was a defensive organization hiding major tensions within the family unit.
The relationship between the bulimics and their mother is one of polarity. They feel that because their eating disturbances seem to worry their mother, she is the only one who cared, and as a result they do such things as calling their moms daily to make sure she is not worrying. At the same time they avoid their mother because they feel she can guess everything or demand so much from them that they would have no personal life left.
All the patients had harsh words for their fathers. They see him as an incompetent irresponsible man who, in 50% of the subjects, was a violent tyrant who flies into wild rages, beats his wife, lies, or is a seducer. The patients who were the most self-reproachful and depressed after binges were those who drew the worst images of their father.
According to this study, the patient is overcome by her desire for a loving family. The patients realize that their families are beyond any patching up, and to escape the reality, they become bulimics. They cannot give up the idea that their families are forever lost, and to have a sense of their own existence, they make up a fantasy. Bulimia then, is a secret behavior, that is a celebration. The patients regress to a time when the family was together, and this psychological state is accompanied by bingeing. The occurrence of bulimia may be related to the combination of a history of violent separations in the family, and the endless denial of these separations (Igoin-Apfelbaum, 1985).
I have several problems with this study. The first and foremost is that I have trouble understanding how the author was able to conclude from the evidence that bulimic patients binge because they are going through a regression. The evidence that led to that theory is not shown, and the theory itself is not explained very well. I can see how a bulimic might want their family to be together as it was when they were younger, but I don't see how that leads to bulimia.
Laura Lynn Humphrey (1986) studied 20 women who fulfilled the DSM-III criteria for bulimia. She used Benjamin's Structural Analysis of Social Behavior (SASB) model and methodology to assess the family relationships of the women. The SASB model is a circumplex model of interpersonal relationships and their intrapsychic representations based on Sullivan's interpersonal theory and prior circumplex models. She found that in comparison to normal women, the bulimics perceived the relationships with their parents as significantly more hostile and less supportive than did the controls. The bulimic women reported that they experienced deficits in parental nuturance and empathy, and felt they were abused and neglected by both parents.
A limitation in this study is that these patients were in therapy for at least a year and that they may have learned to conceptualize their relationships in the psychoanalytic perspective the way their therapists do. Their interpretations of their family situation may have been influenced by the therapists, and therefore the data is showing a construed version of the situation. The patients may not be telling experimenters everything either because they feel it is irrelevant or because they feel they are not conforming to the model their therapist subscribes to. A strength is that the SASB method of studying the patients is allows scientists to study the psychoanalytic hypotheses more directly and precisely and to examine the bulimic's current family relationships. Other methods focus on the developmental and intrapsychic factors, excluding current, ongoing, and interpersonal family relationships (Humphrey, 1986).
In a second study, Humphrey (1989) compared observations of family interactions among anorexic, bulimic-anorexic, bulimic, and normal families. A total of 74 triads participated, including father, mother, and teenage daughter. Each family was videotaped during a ten-minute discussion of the daughter's separation from the family. The tapes were encoded using the SASB model and observational schema. The analysis of the 34 bulimic families, showed that the bulimics and their parents were hostilely enmeshed. Bulimics and their mothers showed a hostile relationship which was manifested by a greater percentage of mutual belittling and blaming as well as sulking and appeasing than found in normal families. The fathers of bulimics were relatively more watching and managing as well as belittling and blaming their daughters. As a result, the daughters showed more sulking and appeasing toward them. In general, the parent-child relationships in bulimic families show minimal affection and minimal support in their interactions and are instead hostilely enmeshed. Both parents and daughters are mutually blaming and controlling of each other in their interactions. The daughters' efforts to assert their separate needs are undermined by hostility. The author concludes that the metaphor model of bulimia may be accurate in suggesting that bulimics and their families crave nurturance, protection, and empathy, and when they are unable to find them, they turn to inanimate objects or altered states for comfort, such as food or alcohol. Bulimic families project their hostilities outwardly and seem unable to modulate them internally, so the bulimic may learn to relieve her own feelings by expelling them through vomiting (Humphrey, 1989).
Stuart, et al. (1990) characterized the early family experiences of 30 women with bulimia and 15 women with depression and compares them with 100 female controls, with particular emphasis placed on parental rearing practices, family conflict resolution, sexual mistreatment, problematic childhood indicators, and childhood separation experiences. Standardized paper-and-pencil instruments were used to assess both attitudinal measures and behavioral indicators of the individual's early family experiences. Specifically, the Childhood Environment Questionnaire consisted of four subsections: EMBU, Memories of Child-Rearing Experiences Scale; Family Violence in Kentucky Questionnaire; and Victimization Inventory, and Early life events.
Women with bulimia described a family characterized by problems, tensions, threats, and physical coercion. They felt rejected by both parents with their mother lacking in warmth and caring and their father as overly controlling. They also experienced significant childhood separation anxiety, although they did not experience more actual losses or separations than the control group. However, they did display many problematic behaviors while growing up such as drug abuse, suicide attempts, and more general emotional problems. The profile that emerges is one of a child in a conflictual family environment that is not supportive or self-enhancing, and who expresses her anxiety and unhappiness in many maladaptive behaviors throughout her adolescence.
This study makes important contributions to the literature in several ways, although there are methodological problems. The study describes the early family experiences of women with bulimia and depression, which is an area that has been poorly researched. It includes for comparison a control group of women that allows one to better differentiate between early childhood experiences that are typical of most women and those that are associated with psychiatric illness. The inclusion criteria used for controls in this study is stricter than typical criteria in the literature, where it is common for the control population to be selected from a medically ill population or from among women with a psychiatric illness other than the one being studied.
There are two major problems with the study. Introspective and self-report data is used and it is difficult to know the extent to which the responses are representative of reality. Additionally, the patient populations used were diagnosed for their illnesses and in a treatment setting. Therefore, there is limited generalizability of the findings to a non-clinical population or to those that are not actively seeking treatment (Stuart, et al., 1990).
Kent and Clopton (1992) examined the relationship between bulimia and family variables. Eight hundred twenty female college students completed a personal history questionnaire, the Bulimia Test, and the Eating Disorders Inventory. Twenty-four subjects who met the DSM-III-R criteria for bulimia were compared with 24 subclinical bulimics and 24 symptom-free subjects on demographic and family variables including the Family Environment Scale and the Parental Bonding Instrument. Significant group differences indicated that the families of bulimics differ from other families. Bulimic families have a greater likelihood of father-daughter closeness and more weight and eating problems. However, in contrast done to the findings of research done in treatment settings, bulimics in this nonclinical setting did not report more family conflict or less caring from their parents than did symptom-free subjects.
This study does have a few limitations. Because there were only a few subjects who met the criteria for bulimia, some small effects may have not been detected, and on the other hand, the group may have not been representative. Finally, the use of self-report data is problematic.
I think it is interesting that this study's results differ from those who use patients from a clinical setting. The authors attribute that the pathological family interaction patterns found in previous research on bulimia may have been the result of the fact that the bulimics were in treatment for their bulimia (Kent and Clopton, 1992). I have trouble understanding this argument. Is it because the bulimics in treatment have more severe symptoms because of their inherited and environmental situations and therefore they have seeked treatment? Is the nonclinical bulimic's symptoms not as severe because there is no pathological family interaction, and as a result, they have not looked for treatment?
A review of eating disorders literature suggests that a number of variables at the level of individual, family, and societal level as important for the development of eating disorders. Specifically, these variables include a concern for weight and shape and a general concern for appearances. Laliberte, Boland, and Leichner (1999) conducted a two-part study investigating the variables believed more directly to disturbed eating and bulimia as contributing to the family conflict of eating disorders. In the first study, a nonclinical sample of 324 women who had just left home for college and a sample of 121 mothers evaluated their families. Analyses revealed the same factor structure for both students and mothers. Three factors were identified as conceptually distinct when comparing bulimic families versus nonbulimic and depressed families. These three factors where perceptions of the family's concern for weight and shape, perceptions of the families concern for social appearances, and perceptions of the family's emphasis on achievement and represented the hypothesized family climate for eating disorders. The combination of appearance and achievement variables identified in this research describes the perfect family -- a family that places great emphasis on appearance, family reputation, family identity, and achievement. The second study extended these findings into a clinical population, examining whether the family climate for eating disorders variables would distinguish individuals with bulimia from both depressed and healthy controls. Forty eating disordered patients, seventeen depressed patients, and twenty-seven healthy controls completed family measures. Family process variables distinguished clinical groups from healthy controls, but not one from another. In other words, the variables of conflict, cohesion, and expressiveness distinguished the population of bulimics and depressives from the normal group, but bulimia and depression were not distinguished. Controlling for depression removed the group differences on family process variables, but family climate variables continued to distinguish the eating-disordered group from both control groups.
This is an interesting study because it shows that there are certain factors that distinguish bulimic families from both depressive families and normal control families. Family process variables such as conflict, cohesion, and expressiveness differentiate bulimic families and depressive families from normal families, but does not distinguish the bulimia and depression from each other. The study goes further and says that there is a special family climate that is directly related to the development of an eating disorder, distinguishing bulimia and depression. There is a problem of conflict, cohesion, and expressiveness in both bulimic and depressive families, but with bulimic families there is the added variables of perceptions of the family's concern for weight and shape, perceptions of the families concern for social appearances, and perceptions of the family's emphasis on achievement. This is important because such characteristics of the family can be recognized early and quickly and the children of that family helped before they develop disturbances in eating such as bulimia (Laliberte, Boland, and Leichner, 1999).
With the above studies I have tried to establish the idea that there are differences between bulimic families and normal families. There seems to be dysfunctional relationships between the bulimic and each of her parents. The studies are inconsistent in identifying the characteristics of the relationships, but generally the bulimic does not feel she receives enough nuturance and care. There is also a strong trend for bulimic families to emphasize appearance and achievement. The daughter often feels she is treated hostilely and can not assert herself. There does in fact seem to be specific characteristics that distinguish bulimic families, and although they are slightly inconsistent across studies, there are certain patterns that can be identified. This identification can help with therapy and early recognition.
Humphrey, Laura Lynn (1986). Structural Analysis of Parent-Child Relationships in Eating Disorders. Journal of Abnormal Psychology, 95(4), 395-402.
Humphrey, Laura Lynn (1989). Observed Family Interactions Among Subtypes of Eating Disorders Using Structural Analysis of Social Behavior. Journal of Consulting and Clinical Psychology, 57(2), 206-214.
Igoin-Apfelbaum, Laurence (1985). Characteristics of Family Background in Bulimia. Psychother. Psychosom, 43, 161-167.
Kent, Jan S., and Clopton, J. R. (1992). Bulimic Women's Perceptions of Their Family Relationships. Journal of Clinical Psychology, 48(3), 281-292.
Laliberte, Michele., Boland, F. J. and Leichner, P. (1999). Family Climates: Family Factors Specific to Disturbed Eating and Bulimia Nervosa. Journal of Clinical Psychology, 55(9), 1021-1040.
Stuart, G. W., et al. (1990). Early Family Experiences of Women With Bulimia and Depression. Archives of Psychiatric Nursing, 4(1), 43-52.
Psychology DepartmentThe Health Psychology Home Page is produced and maintained by David Schlundt, PhD.
Vanderbilt Homepage | Introduction to Vanderbilt | Admissions | Colleges & Schools | Research Centers | News & Media Information | People at Vanderbilt | Libraries | Administrative Departments | Medical
|Return to the Health Psychology Home Page|
|Send E-mail comments or questions to Dr. Schlundt|