Close
Home
Collections
Login
USC Login
Register
0
Selected
Invert selection
Deselect all
Deselect all
Click here to refresh results
Click here to refresh results
USC
/
Digital Library
/
University of Southern California Dissertations and Theses
/
Bulimia nervosa: A response to trauma associated with parental alcoholism and sexual abuse in college sorority women
(USC Thesis Other)
Bulimia nervosa: A response to trauma associated with parental alcoholism and sexual abuse in college sorority women
PDF
Download
Share
Open document
Flip pages
Contact Us
Contact Us
Copy asset link
Request this asset
Transcript (if available)
Content
BULIMIA NERVOSA: A RESPONSE TO TRAUMA
ASSOCIATED WITH PARENTAL ALCOHOLISM AND SEXUAL
ABUSE IN COLLEGE SORORITY WOMEN
by
Lynn Suzanne Tracy
A Dissertation Presented to the
FACULTY OF THE GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(Counseling Psychology)
May 1994
©1994 Lynn Suzanne Tracy
UMI Number: DP71340
All rights reserved
INFORMATION TO ALL USERS
The quality of this reproduction is dependent upon the quality of the copy submitted.
in the unlikely event that the author did not send a complete manuscript
and there are missing pages, these will be noted. Also, if material had to be removed,
a note will indicate the deletion.
UMI
Dissertation Rjblishing
UMI DP71340
Published by ProQuest LLC (2015). Copyright in the Dissertation held by the Author.
Microform Edition © ProQuest LLC.
All rights reserved. This work is protected against
unauthorized copying under Title 17, United States Code
uesf
ProQuest LLC.
789 East Eisenhower Parkway
P.O. Box 1346
Ann Arbor, Ml 48106- 1346
UNIVERSITY OF SOUTHERN CALIFORNIA
THE GRADUATE SCHOOL
UNIVERSITY PARK
LOS ANGELES, CALIFORNIA 90007
This dissertation, written by
Lynn Suzanne Tracy
under the direction of h Æ Dissertation
Committee, and approved b y all its members,
has been presented to and accepted by The
Graduate School, in partial fulfillm ent of re
quirements for the degree of
D O C TO R OF PH ILOSOPH Y
Dean of Graduate Studies
Date
April 15, 1994
DISSERTATION COMMITTEE
Chairperson
11
Acknowledgments
I would first like to acknowledge the people who have
survived oppressive and abusive environments in their own homes
and in greater society. Their struggles and strengths made this paper
possible, and I hope it gives their experiences a voice.
I would also like to thank the many people who have given me
the support to complete this project:
My mentors and role models deserve my sincere appreciation
for their belief and support of my intellectual abilities. Braddie
Dooley gave me my first opportunity to speak my truth, and
encouraged me to believe I could succeed. Laurie Mintz encouraged
my academic and professional endeavors wholeheartedly, and has
served as a professional role model. Joe White has been a truly
supportive career advocate, despite the poor job market in
California. Jeff Prince has unflinchingly advocated for me on the
job, and shown me that you can have a sense of humor while still
being the consummate professional. Lynn Winsten has taught me
that success, while not painless, is deserved.
I am immensely grateful to my husband, Jeff Allenstein, who
has supported my academic and professional ambitions with
generosity and pride. Through his own courage and growth, he has
taught me that anything is possible.
Ill
Deborah Brown has been a personal and professional
inspiration to me. She has wisdom and passion greater than anyone
I've known. I have gained immeasurably by her being alive.
Jim Robinson, thank you: For thinking I’m smart, even if I
looked like a surfer chick; for the best conversations.
I also want to thank my sister, Anne Tracy, whose belief in me
has never faltered. She rises to the challenge every time.
Finally, I would like to thank my dissertation committee:
Scott Whiteley, Chalmer Thompson, and John Briere. Scott was
eternally patient and reassuring, even when I called for Sunday
morning phone advisement. John Briere’s brilliant formulations
about child abuse inspired this paper.
IV
Table of Contents
Acknowledgments.......................................................................................... ii
List of Tables and Figures ............................................................................vi
A b stract........................................................................................................... vii
CHAPTER ONE: INTRODUCTION AND REVIEW
OF THE LITERATURE ......................................................................... 1
Description of the D isorder............................................................... 4
Etiological M o d e ls..............................................................................5
Biological T heories..................................................................5
Sociocultural Explanations . .................................................7
Psychological Factors................................................................. 11
Personality F actors..........................................................11
Personality D isorders..................................................... 13
Psychodynamic Interpretations...................................... 14
Familial Factors in Bulimia.........................................................17
Family Demographics..................................................... 17
Family Relationships....................................................... 18
Family L egacies...............................................................19
Intergenerational Transmission of Psychopathology....................... 22
Intergenerational Transmission of A lcoholism .....................22
Intergenerational Transmission of Affective
D isorder......................................... 25
Bulimia and Affective D isorder....................................25
Intergenerational Transmission of B ulim ia............................ 26
Bulimia and Substance A buse........................................ 28
Substance Abuse and A ddiction....................................29
Bulimia as a Type of Substance Abuse or
Addiction........................ 31
Coincidence of Bulimia and Other
Substance A buse.............................................................. 33
Substance Abuse in the Relatives of
Bulimics ...........................................................................35
Sexual Abuse and Eating D isorders.........................................37
Trauma Theories........................................................................44
Statement of the Problem....................................................................... 47
Research Questions and Hypotheses......................................................49
CHAPTER TWO: M E T H O D .................................................................54
Subjects...................................................................................................54
Instrum entation........................................................................ 55
The Bulimia Test-Revised..........................................................55
Children of Alcoholics Screening T c s t-6 ...............................57
Childhood Sexual Abuse Questions .........................................59
Trauma Symptom Inventory.....................................................59
Demographic Q uestions............................................................61
Procedure ............................................................................................61
Subject Recruitment................................................................... 61
Instructions..................................................................................63
Group Assignment......................................................................64
Data Analysis .............................................................................65
Limitations...............................................................................................67
Design and Internal V alidity.....................................................67
External V alidity........................................................................ 69
Measurement................................. 69
CHAPTER THREE: R E S U L T S ............................................................71
Demographic Characteristics of Respondents....................................71
Frequencies of V ariables......................................................................72
Analyses of Research H ypotheses.......................................................75
Hypothesis 1 ............................. 75
Hypothesis 2 ............................................................................... 76
Hypothesis 3 ............................................................................... 77
Hypothesis 4 ............................................................................... 79
Hypothesis 5 ................................................................................81
CHAPTER FOUR: DISCUSSION.............................................................83
Interpretation of R e su lts......................................................................83
Hypothesis 1 ................................................................................83
Hypothesis 2 ................................................................................85
Hypothesis 3 ............................................................................. 87
Hypothesis 4 ............................................................................... 90
Hypothesis 5 ................................................................... 91
Additional F indings............................... 94
Limitations/Delimitations........................ 96
VI
Implications of F in d in g s.....................................................................98
Theoretical Implications...........................................................98
Clinical Im plications................................................................101
Directions for Future R esearch...............................................103
References ...................................................................................................... 106
Appendices
Appendix A: The Bulimia Test-Revised.......................................... 120
Appendix B: The Children of Alcoholics Screening
T e st-6 .................................................................................................... 126
Appendix C: Sexual Abuse Questions . .................. 127
Appendix D: Trauma Symptom Inventory ....................................128
Appendix E: Letter to Sorority P residents......................................132
Appendix F: Results F o r m ................................................................ 133
Appendix G: Information and Consent F o r m ................................. 134
vu
List of Tables and Figures
Table Page
1 Differences Between Bulimic and Non-Bulimic 72
Subjects on Demographic Data
2 Frequencies of Variables 73
3 Descriptive Statistics of All Variables According to 74
Bulimia Status
4 Results of Multiple Regression Analysis of 75
Traumatic Symptomatology on Bulimia
5 Trauma Symptom Inventory Means and Standard 77
Deviations According to Parental Alcoholism Status
6 Results of Univariate F-Tests of Trauma Symptom 78
Inventory Subscales by Sexual Abuse History
7 Trauma Symptom Inventory Means and Standard 80
Deviations According to Sexual Abuse History
8 Means and Standard Deviations of BULIT-R Scores 80
According to Parental Alcoholism Status
9 Means and Standard Deviations of BULIT-R Scores 82
According to Sexual Abuse History
Figure Theoretical Model of the Intergenerational 100
1 Transmission of Bulimia
V lll
Abstract
While it is generally held that multiple factors contribute to
the development of bulimia nervosa, less is known about the specific
experiences which are risk factors for this eating disorder. Based on
research results documenting a higher incidence of parental
alcoholism among bulimics and a relationship between childhood
sexual abuse and later bulimia, the present study hypothesized that
parental alcoholism and sexual abuse were two specific risk factors
in the development of bulimia. In order to test this hypothesis, the
rates of parental alcoholism and childhood sexual abuse in bulimic
subjects were investigated. In addition, the present study
hypothesized that traumatic symptomatology would be present at a
higher rate among bulimics, adult children of alcoholics (ACAs),
and victims of childhood sexual abuse than among individuals who
have not had such experiences.
Three hundred fifty-five college sorority women from 15
universities across the United States completed the following
instruments: Bulimia Test-Revised (BULIT-R), Children of
Alcoholics Screening Test-6 (CAST-6), the Trauma Symptom
Inventory (TSI), as well as several questions related to childhood
sexual abuse.
Each of the research hypotheses studied was at least partially
confirmed: The incidence of parental alcoholism was significantly
IX
greater among bulimics in this sample, as was the incidence of
childhood sexual abuse, but not incest. Bulimics in this sample had
significantly higher rates of traumatic symptomatology. While
parental alcoholism was not significantly related to individual
traumatic symptomatology domains, adult children of alcoholics had
higher total levels of traumatic symptomatology. Finally, higher
rates of traumatic symptomatology were found among subjects who
reported sexual abuse history, but not incest history.
The findings support the hypothesis that parental alcoholism
and sexual abuse are risk factors in the development of bulimia. In
addition, these results lend evidence to the hypothesis that parental
alcoholism and sexual abuse are traumagenic experiences and that
bulimia may indeed be a means of coping with such trauma.
1
CHAPTER ONE
INTRODUCTION AND REVIEW OF THE LITERATURE
The eating disorder bulimia nervosa has been the focus of
much attention in the psychological literature over the last two
decades. The increased attention paid to bulimia may be a result of
the alarming prevalence of this disorder: estimates of the prevalence
of binge eating among women range from 24% to 90%; studies
operationalizing the DSM III criteria for bulimia have reported
prevalence rates from 5% to 20% in college women (Crowther,
Wolf, & Sherwood, 1992; Johnson & Connors, 1987; Pope, Hudson,
Yurgelun-Todd, & Hudson, 1984). In a study of 643 college
women’s eating patterns, only 33% of the subjects reported “normal”
eating habits; the majority of the subjects were classified as having
some intermediate form of eating behavior problem (Mintz & Betz,
1988). Thus, some form of disordered eating seems to be more
“normal” than normal eating for many women.
As is recently the case with many other psychological
disorders, bulimia has been hypothesized to be caused by biological,
psychological, sociological, and family factors (Johnson & Connors,
1987). Yet although the etiological factors contributing to bulimia
are multiple and interactive, there has been little consensus among
the research concerning the exact etiology (Carter, lyiboz, &
Greenockle, 1986; Swift & Letven, 1984). In an effort to better
2
explain the correlates of bulimia, two recent areas of focus have
been on bulimia’s association with substance abuse and sexual abuse
(Abramson & Lucido, 1991; Beckman & Bums, 1990; Bulik, 1987b;
Hatsukami, Mitchell, Eckert, & Pyle, 1986). Both familial
alcoholism and incest have been found to occur more frequently in
the families of bulimics (for example, Buhk, 1987a; Calam & Slade,
1989), although a few studies have not found these results (Finn,
Hartman, Leon, & Lawson, 1986). Given the retrospective and
correlational or descriptive nature of these studies, however, it is
impossible to draw causal conclusions. Yet by using research
methodologies which attempt to mle out competing hypotheses, it is
possible to gain important knowledge of the association of sexual
abuse and parental alcohohsm to bulimia.
To this author’s knowledge, no study has addressed the
coincidence of bulimia, incest, and parental alcoholism within the
same family. The present investigation is based on a combination of
object relations and trauma theories. Especially pertinent to this
study are the formulations of object relations theorists who propose
that bulimia is an attempt to soothe the self and regulate intolerable
tension states; according to these theorists, this need arises out of
failures in the bulimic’s early parental environment (Humphrey &
Stem, 1988; Sugarman & Kurash, 1982; Swift & Letven, 1984). In
addition, John Briere’s “tension reduction” theory of coping with
child abuse trauma provides a theoretical base for the present study
3
(Briere, 1992). Briefly, Briere (1992) hypothesized that what are
frequently called “addictive “ or “acting out” behaviors or “impulse
control disorders” such as substance abuse and buhmia are in fact
ways to moderate painful affective states stemming from earlier
abusive experiences in childhood. Based on Briere’s theory, the
present study proposed that buhmia is a tension-reduction or coping
mechanism, much like other addictive processes, which is employed
in response to marred or traumatic parenting experiences. In a
recent large-scale study of the long-term psychosocial effects of
childhood exposure to parental problem drinking, the authors found
that individuals exposed to parental problem drinking were at higher
risk for developing greater lifetime psychiatric problems and marital
instability (Greenfield, Swartz, Landerman, & George, 1993). In
this same study, adverse lifetime symptomotology was also associated
with sexual abuse. The traumatic childhood experiences which are
often associated with being a child of an alcoholic parent or a victim
of incest engender painful feelings of anxiety, fear, abandonment,
and isolation. The bulimic cycle may help to temporarily stave off
these painful affective states. The present study addressed this
hypothesis by examining the rates of incest or sexual abuse and
parental alcoholism in women with bulimia. In addition, traumatic
symptomatology was also measured in order to further examine its
association with parental alcoholism, sexual abuse, and bulimia.
Before discussing the role these factors play in bulimia, the disorder
4
will be described and the proposed etiological models for bulimia
reviewed.
Description of the Disorder
According to the DSM III-R (APA, 1987), the following
criteria must be met to diagnose bulimia:
a. Recurrent episodes of binge eating (rapid
consumption of a large amount of food in a discrete
period of time, usually less than two hours).
b. During the eating binges, a fear of not being able to
stop eating.
c. Regularly engaging in self-induced vomiting, use of
laxatives, or rigorous dieting or fasting, in order to
counteract the effects of binge eating.
d. A minimum average of two binge eating episodes per
week for at least three months. (APA, 1987)
Often the foods binged on are “junk foods”: foods high in
sugar and/or carbohydrates, e.g., cookies, chips, and popcorn.
Bulimics tend to be secretive about their binges, attempting to hide
their food and binge when no one else can see them. Once the binge
starts, the person fears that he/she cannot stop eating. There is a
pervasive feeling of being out of control. The binges are usually
stopped by abdominal pain, purging, sleep, or by being interrupted
by someone. Sometimes vomiting itself may be the desired activity
for its “cleansing” feeling; some researchers have hypothesized that
this is what perpetuates the cycle, not the binge. The emotional cycle
5
associated with bulimia is one of guilt, depression, and self-
deprecation. Promises and resolutions to stop the binge-purge cycle
are repeatedly broken, resulting in shame and feelings of loss of
control (Johnson & Connors, 1987).
Etiological Models
Biological Theories
Biological factors have been found to be etiologically
significant in bulimia (Fava, Copeland, Schweiger, & Flerzog, 1989;
Johnson & Connors, 1987; Morley & Blundell, 1988). Johnson and
Connors (1987) discussed the existing evidence for hypothalamic-
pituitary-adrenal (HPA) axis involvement in the syndrome of
bulimia. The finding that up to 67% of bulimic patients were
nonsuppressors on the dexamethasone suppression test (DST) was
evidence for this hypothesis. Weight loss, reduced caloric intake,
and catabolic state have a very powerful influence on the HP A axis
and other endocrine systems (Fichter & Perk, 1984, in Johnson &
Connors, 1987).
Morley and Blundell (1988) reported that the neuropeptide Y,
GAB A, and the biogenic amines (norepinephrine [NE], dopamine
[DA], and serotonin) are involved in feeding behaviors and thus may
be linked to anorexia and bulimia. They discussed the possibility of
a link between nutritional composition of the diet, brain
neurochemistry, and behavioral mental changes in which eating
disorders could be created, sustained, or even exacerbated through
6
neurobiological processes. Yet they conclude that the exact
mechanisms involved in feeding behavior is not clear and propose a
“multifactorial” etiology (Morley & Blundell, 1988).
Fava and his colleagues (1989) also concluded that although
neurochemical abnormalities exist in bulimic patients, the exact role
neurotransmitters play in the disorder is still unclear. The authors
discussed the altered noradrenergic functions in bulimia: patients
with bulimia had a blunted rise in plasma NE concentration on
standing even after they had stopped bingeing and purging. Bulimic
patients were also found to have a significantly lower resting pulse,
systolic blood pressure, and plasma NE than control subjects. In
addition, bulimic subjects had a blunted L-tryptophan-induced rise
of plasma prolactin levels as compared with controls. Because
greater central nervous system concentrations of L-tryptophan lead
to an increase in serotonin synthesis, and increased brain serotonin
turnover contributes to satiety, it seem possible that serotonin may
be implicated in the binge symptom of bulimics. Finally, Fava and
his colleagues (1989) summarized the findings of the studies on the
neurochemical abnormalities in bulimia: Two studies found lower
NE activity and two studies found less serotonin turnover.
Therefore, although the neurochemical changes can precede,
accompany, or follow bulimic symptoms, the research supports their
role in bulimia.
7
Sociocultural Explanations
Sociocultural forces have been identified as another
contributor to the development of bulimia (Boskind-Lodahl, 1976;
Schwartz, Thompson, & Johnson, 1982). Boskind-Lodahl (1976)
discussed bulimia from a feminist perspective based on her clinical
experience with bulimic patients. She proposed that women who
overidentify with the traditional female role in American culture
develop bulimic symptoms. This definition of the feminine role has
been characterized as placing an emphasis on being attractive. Thus,
according to Boskind-Lodahl, receiving validation from men is the
way in which women validate their self-worth. Bulimics never
questioned their assumptions that wifehood, motherhood, and
intimacy with men are the fundamental components of femininity.
Their attempts to control their physical appearance demonstrates a
disproportionate concern with pleasing others, particularly men. In
addition, bulimic/anorexic women are completely vulnerable to
rejection from men, with whom they have been socialized to believe
are the only meaningful relationships.
According to Boskind-Lodahl, the distorted concept of body
size characteristic of women with bulimia may have been related to
parental and societal expectations that emphasize physical appearance
as well. The dieting and purging phases of bulimia can be seen as
resulting from this attempt to be physically appealing and to gain
control and a sense of competence and effectiveness. Women have
8
been socialized to expect men to be attracted to them if they fulfilled
these physical criteria; when the expected results of being attractive
to men was not fulfilled, the binge behavior would ensue (Boskind-
Lodahl, 1976).
Another sociocultural explanation of eating disorders has been
offered by Schwartz et al. (1982). These authors suggested that the
cultural obsession of the relentless pursuit of thinness has been partly
responsible for the incidence of anorexia and bulimia. To
demonstrate this hypothesis, Schwartz and his colleagues looked at
Playboy magazine Playmate centerfolds and contestants and winners
of the Miss America Pageant from 1959 through 1978. Both
groups’ mean weights were significantly less than corresponding
population means each year. Thus, women selected as exemplars of
feminine beauty were consistently thinner than the actual norms for
comparable women in the population. In addition, weight declined
across the 20-year period which was studied. Ideal shapes for
women became progressively thinner over the 29 years. After 1970,
the Miss America Pageant winners’ weights were consistently less
than the average weight of all contestants. Meanwhile, the average
weight of women under 30 has become consistently higher than in
1959.
Society’s hatred and disgust towards the overweight was cited
by Schwartz et al. (1982) as the reason that some women have used
extreme methods of weight control such as anorexia or bulimia.
9
They cited the work of Wooley et al. (1979, in Schwartz, 1982),
who compiled studies documenting stigma of obesity in childhood.
These authors found that children whose builds are socially deviant
are regarded by others as responsible for their condition and
deserving of social disapproval. “. . . (f)ailure to (diet) is seen as
‘weakness’, ‘wanting to be fat’, or even as a masochistic desire for
rejection” (p. 23). These authors noted that this stigma prejudicially
affects women more than men.
Orbach (1978) discussed the societal contributions to
compulsive overeating rather than anorexia or bulimia. Yet her
discussion seems relevant to the problem of these eating disorders as
well. Similar to the studies cited in Schwartz et al. (1982), Orbach
(1978) posited that society blames overweight women for not being
able to control their impulses. Thus, like the children in Wooley et
al.’s 1979 research, women come to believe their problems with
dieting and weight are all their fault, rather than as attempts to meet
unrealistic physical standards. Consistent with this individualistic
view, Orbach (1978) noted that female fatness has been seen as an
obsessive-compulsive symptom related to separation-individuation,
narcissism, and insufficient ego development by psychology. But,
according to Orbach (1978), a feminist perspective correctly shifts
the focus of blame off of the victim and onto society:
The fact that compulsive eating is overwhelmingly a
woman’s problem suggests that it has something to do
10
with the experience of being female in our society. . ..
Getting fat can thus be understood as a definite and
purposeful act; it is a directed, conscious or
unconscious, challenge to sex-role stereotyping and
culturally defined experience of womanhood. Fat is a
social disease, and fat is a feminist issue. Fat is not
about lack of self-control or lack of will power. Fat is
about protection, mothering, substance, assertion, and
rage. It is a response to the inequality of the sexes, (pp.
4-5)
Orbach also pointed to the relegation of women to the social
roles of wife and mother as having consequences that contribute to
“the problem of fat” (p. 52). According to Orbach, in order to
become a wife and mother, one must first have a man. To get a
man, a woman has to leam to regard herself as an item, a
commodity, a sex object. Just as Boskind-Lodahl pointed out, much
of a woman’s experience and identity depends on how others see her.
Moreover, Orbach contended that women are extremely limited in
socially sanctioned ways to define themselves. To demonstrate this
point, she pointed to the two ways in which women are portrayed in
the media: mother or sex object. Extrapolating from Orbach’s
work on compulsive eating, anorexia, and bulimia can also be seen as
reactions to the conflict which arises out of placing women as
objects: Dieting is an attempt to fit this image, bingeing is a
rejection of this objectification, and purging is an attempt to fit back
in.
11
While it is held that sociocultural influences play a role in the
development of eating disorders, these influences are not enough to
explain why some women in our culture develop eating disorders
while others do not. Most likely, the combination of biological,
psychological, sociocultural, and family forces combine to produce
these disorders (Johnson & Connors, 1987). A review of the
psychological factors in bulimia will be presented next.
Psychological Factors
Personality factors. According to Wonderlich’s (1992) review
of the personality trait studies, bulimics tend to exhibit the following
personality characteristics: impulsivity, interpersonal sensitivity,
and low self-esteem. While there have not been empirical cluster
analyses to identify subtypes of bulimics as there have been for
anorexics, Johnson and Connors (1987) discussed personality
characteristics of bulimics based on the Eating Disorder Inventory,
an instrument designed by Garner, Olmsted, and Polivy (1983) to
delineate and measure certain psychological traits and symptoms
presumed to be relevant to eating disorders. The following is a
description of the eight constructs found to be more prevalent in
individuals with eating disorders:
1. Drive fo r Thinness indicates excessive concern with dieting,
preoccupation with weight, and entrenchment in an extreme pursuit
of thinness. Both restricting and normal-weight bulimics score
12
higher on this scale than anorexic, obese, and normal subjects
(Johnson & Connors, 1987).
2. Bulimia indicates the tendency toward episodes of
uncontrollable overeating (bingeing) and may be followed by the
impulse to engage in self-induced vomiting. Bulimics score
significantly higher on this scale than restricting anorexics, obese,
and normal subjects.
3. Body Dissatisfaction reflects the belief that specific parts of
the body associated with shape change or increased “fatness” at
puberty are too large. Bulimics report more body dissatisfaction
than normal controls (Mintz & Betz, 1988; Johnson & Connors,
1987).
4. Ineffectiveness assesses feelings of general inadequacy,
insecurity, worthlessness, and the feeling of not being in control of
one’s life. Ineffectiveness has also been found to be highly
correlated with low self-esteem. Bulimics report significantly lower
self-esteem than normal subjects; the self-esteem problems appear to
include high self-expectations, self-criticalness and guilt, high needs
for approval from others, external locus of control, low
assertiveness, and interpersonal sensitivity (Johnson & Connors,
1987).
5. Perfectionism indicates excessive personal expectations of
superior achievement. Bulimics tend to score slightly higher than
obese and normal control subjects on the perfectionism scale.
13
6. Interpersonal Distrust reflects a sense of alienation and a
general reluctance to form close relationships. It relates to an
inability to feel comfortable expressing emotions towards others.
This construct does not appear to differ significantly for bulimic
individuals.
7. Interoceptive Awareness reflects one’s lack of confidence in
recognizing and accurately identifying emotions or visceral
sensations of hunger or satiety (Garner, Olmsted, & Polivy, 1983).
Johnson and Connors (1987) reported that both bulimic anorexics
and normal-weight bulimics score significantly higher on this scale
than obese, restricting anorexic, and normal subjects. They attribute
this finding to bulimics’ greater overall intrapsychic disorganization.
Difficulty with interoceptive awareness appears to be related to
symptoms of ineffectiveness and body-image disturbance.
8. Maturity Fears measures one’s wish to retreat to the
security of the preadolescent years because of the overwhelming
demands of adulthood. Bulimic anorexics score higher on this scale
than normal-weight bulimics, obese, and normal controls (Johnson &
Connors, 1987).
Personality disorders. One focus of the literature on
personality factors in bulimia has been the coincidence of eating
disorders and personality disorders (Gartner, Marcus, Halmi, &
Loranger, 1989; Swift & Letven, 1984; Wonderlich, 1992). For
example, Gartner et al. (1989) interviewed 35 female patients with
14
eating disorders to examine the prevalence of personality disorders
in this population. They found that 57% of the subjects met the
criteria for at least one Axis II diagnosis; borderline, self-defeating,
and avoidant were the most common personality disorders assigned.
Forty percent of the patients were given two or more diagnoses, and
17% of the patients met criteria for five to seven diagnoses. No
differences were found between patients with anorexia and
restricting or normal-weight bulimia (Gartner et al., 1989).
Wonderlich (1992) found that several personality disorders
tend to co-occur most frequently with eating disorders. Similar to
Gartner et al.’s findings, borderline, self-defeating, avoidant, and
possibly obsessive-compulsive disorders occur across all types of
eating disorders, whereas histrionic personality disorder appears
more often in normal weight bulimics.
Psychodynamic interpretations. Psychodynamic theories add
an alternate explanation of the bulimic syndrome (Bruch, 1973;
Humphrey & Stern, 1988; Sugarman & Kurash, 1982; Swift &
Letven, 1984). From this perspective, bulimia is viewed as a
demonstration of some type of fault in the bulimic’s relationship
with her primary caretaker or love object (e.g., mother) at an
important developmental stage. Given that the feeding interaction is
one of the first and fundamental interactions between mother and
child, the nourishing process is emotionally charged. Thus, these
theorists speculate that the feeding process is a fertile breeding
15
ground for psychological deficits to be passed on from mother to
child.
For example, Bruch (1973) posited that an individual’s ability
to identify and articulate internal states is largely influenced by early
mother-child interactions around need gratification, such as the
feeding process. She posited that eating disordered individuals’
difficulty with interoceptive awareness, specifically relating to
hunger and satiety, is due to inappropriate responses (e.g.,
neglectful, overindulgent, or inhibiting behavior) on the part of the
mother.
Johnson and Connors (1987) proposed that anorexia and
bulimic behaviors are differential responses to maternal
overinvolvement and underinvolvement, respectively. They posited
four types of bulimic individuals based on different feeding
disturbances: (a) psychotic (the least common type), (b) borderline,
(c) false self/narcissistic, and (d) neurotic (see previous section for
discussion of eating disorders and personality disorders).
Sugarman and Kurash (1982) posited that a failure in the
separation-individuation phase of development occurred in the
bulimic’s childhood. This failure led to the body of the bulimic
being used as a vehicle for both representing (in the process of
bingeing) and rejecting (in the process of vomiting) the maternal
object, which was symbolized by food. Likewise, bulimia has also
16
been viewed as an expression of rage at the ambivilently held
maternal object (Johnson & Connors, 1987).
Swift and Letven (1984) proposed that bulimics suffered from
a basic fault in their ego structure which made them deficient in
their ability to tolerate high levels of tension. Through the binge-
purge behaviors, the bulimic attempted to rid herself of the
intolerable tension and to repair the basic ego fault. Unfortunately,
according to these authors, bulimic acts do not succeed in making up
for the deficient maternal environment, even though there might be
a temporary relief from anxiety.
In a similar vein, Humphrey and Stem (1988) posited a theory
of bulimia based on components of object relations theory, including
Winnicott’s (1965) mother-infant “holding environment”, a
metaphor for the total protective, empathie stance that a “good
enough” mother provides the infant. These authors proposed that all
members of a bulimic family have experienced various failures in
early parental holding environments. These failures involved (a)
nurturance, (b) soothing and tension regulation, and (c) empathy and
affirmation of separate identities between mother and child. The
bulimic’s compulsion to binge resulted from an emotional hunger
that was based on life-long feelings of deprivation and emptiness.
These authors viewed bulimia as an attempt to alleviate intolerable
underlying tension states which resulted from a failed parental
environment. Given that each member of the bulimic’s family have
17
been severely deprived of emotional nurturance in their own
development, the bulimic daughter inherits the legacy of nurturing
her mother rather than being mothered herself.
The psychodynamic literature has not yet been substantiated
through empirical research, so the above hypotheses have remained
theoretical in nature. The psychodynamic explanations have in
common a deficient parental environment as the cause of bulimia.
Another theme which recurred in this body of literature was the
bulimic’s inability to tolerate tension. To understand the context in
which bulimia develops, the discussion will now turn to a review of
the literature on familial factors in bulimia.
Familial Factors in Bulimia
Family demographics. Among the families of bulimics a
demographic profile emerges. Kog and Vandereycken (1985)
observed an over-representation of the higher social class among
eating disorder subjects’ families. Similarly, Herzog (1982) found
37% of the bulimic patients’ family income was over $45,000 per
year. In contrast, however, Johnson and Connors (1987) stated that
bulimic patients cluster more in the lower-middle-class range.
Family size, the individual’s birth order, and the incidence of
divorce among bulimics’ parents do not differ significantly from the
general population. With regard to ethnicity, it is important to note
the dearth of studies which have included a cross-culturally
representative sample in the study of eating disorders.
18
Family relationships. Strober and Humphrey (1987) discussed
the research concerning interactions of bulimics’ families. These
authors reviewed the family literature on bulimia and found that
bulimics and their parents perceived their family environments as
less cohesive and nurturant and more conflictual and disengaged than
normal control subjects. They also observed that bulimics’ family
relationships tended to be enmeshed, intrusive, hostile, and negating
of the child’s emotional needs. In addition, Strober and Humphrey
found that bulimia has been strongly associated with a lack of
parental affection. They hypothesized that this may have led to an
unfulfilled craving for nurturance, which then led to bulimia.
Similarly, in their review of the literature on bulimics’ family
relationships, Kog and Vandereycken (1985) revealed dysfunctional
interaction patterns in bulimics’ families. Specifically, they
discussed a pattern of control and interdependence in these families.
Another indicator of family dysfunction was bulimics’ families’
performance on the Family Environment Scale (FES; Moos, 1981, in
Kog & Vandereycken, 1985), which characterized these families as
having high levels of conflict.
More recently, Kog and Vandereycken examined the family
interactions in eating disorder patients and normal control subjects
(Kog & Vandereycken, 1989). They investigater) 30 families with an
eating disorder patient and 30 normal control families on a variety
of self-report and observational measures. These authors found that
19
the bulimic family showed strong interpersonal boundaries, a less
stable organization, and less avoidance of disagreements than
anorexic and normal families; the bulimic patient described her
family as conflictual, uncohesive, and badly organized, which is
consistent with prior research.
Finally, Viaro (1990) conceptualized relationships in the
bulimics’ families as a “family game” (p. 289). The rules of this
game include each family member playing an interactive role in
which the participants hide and disguise their feelings, goals, and
intentions from one another.
Family legacies. Roberto (1986) posited that within bulimic
families, there exist interactional patterns and belief systems which
enable and perpetuate the bulimic cycle. She proposed a
tr ans generational model in which families pass down roles, values,
and customs (“family legacies”). In families with bulimia, the family
legacy revolves around weight, attractiveness, fitness, and success.
The bulimic’s weight becomes a barometer of her acceptance of her
familial values. Roberto hypothesized that the most destructive
belief in these families is that of “filial loyalty” (p. 233). These
families demand self-sacrifice to further the needs of the family and
visualize outsiders as competitors or threats to the family’s
cohesiveness; in other words, this family system is closed and has
rigid boundaries. The taboo against allying with outsiders extends to
the bulimic’s peers so that over time she becomes increasingly skilled
20
at responding to family problems while also increasingly self
alienated, self-sacrificial, and dysfunctional outside the family.
Both Humphrey and Stem (1988) and Schwartz and Grace
(1989) also discussed the family legacy in bulimic families. The
legacy in these families consists of rigid adherence to high standards
of achievement and physical appearance. These families, which
Schwartz and Grace (1989) labeled “Hyper-Americanized” (p. 93),
place extreme importance on appearance, self-control, and success
while also being self-absorbed and competitive.
Integrating object relations and family systems theory,
Humphrey and Stern (1988) hypothesized that all members of
bulimic families have experienced various failures in their early
parental holding environment, and that these failures are transmitted
as an intergenerational legacy. The primary failures in families with
a bulimic member include problems in providing: (a) nurturance,
(b) soothing and tension regulation, and (c) empathy and affirmation
of separate identities. In this model, the bulimic member is seen as
no more “sick” than the rest of the family; each member of the
system is needy and dysfunctional.
The literature on the family environments of bulimics has
revealed a dysfunctional pattern which includes rigid boundaries,
enmeshment, and hostility. Interestingly, the psychodynamic
theories seem to converge with the empirical literature on the
bulimics’ families. For example, Johnson and Connors (1987)
21
believed that bulimics’ developmental histories reflect a pattern of
maternal underinvolvement; in the empirical study by Kog and
Vandereycken mentioned above (Kog & Vandereycken, 1989), the
authors characterized bulimic families as “interpersonal distance-
sensitive” (p. 21). There seems to be support, with the caveat of
much inadequate research design, for the importance of family
factors in the development of bulimia. Indeed, transgenerational
research has demonstrated that parental mental health is an important
determinant of children’s psychopathology (Broderick, 1991).
Specific to the transmission of bulimia, Broderick (1991), in
his review of the intergenerational transmission of psychopathology,
discussed a number of variables common to the bulimic family
summarized above. For example, Broderick states, “Of all of the
different aspects of parental behavior that have been found to
influence the way children grow up, their emotional warmth is
consistently found to be the most important” (p. 3). Strober and
Humphrey (1987) found that lack of affection was frequently found
in the families of bulimics, as well as parents negating the emotional
needs of the children, which may be translated as lack of emotional
warmth.
In addition, Broderick (1991) cited style and frequency of
control attempts as important variables in parenting. Kog and
Vandereycken (1985) found that families of bulimics were
struggling with conflicts over control and autonomy. In a study
22
using videotaped family interactions (Humphrey, 1986), the authors
observed that “The parents of the bulimic were highly controlling
and intrusive towards their daughter during family interactions” (p.
171). Although this study did not discuss control attempts in terms
of style or frequency, it seems logical to assume that these control
attempts were not egalitarian, and therefore, not effective in
promoting self-esteem.
Although control and warmth were only two factors cited in
Broderick’s review as influencing the transmission of
psychopathology from parents to children, they lend support to a
familial role in the development of an eating disorder. Moreover,
taking into account the elevated rates of substance abuse in bulimics’
families, some aspect of bulimia appears to be “passed on” in the
family. The literature on the intergenerational transmission of
alcoholism may represent a model for the transmission of bulimia.
Intergenerational Transmission of Psychopathology
Intergenerational Transmission of Alcoholism
Twin studies, adoption studies, and longitudinal studies have
shown that alcoholism runs in families (Crabbe, McSwigan, &
Belknap, 1985). For example. West and Prinz (1987) documented a
relation between parental alcoholism and their adolescent
offspring’s’ alcohol abuse. Perhaps the most well-known of these
studies are those conducted by Goodwin and his colleagues in
Denmark (Goodwin, Schulsinger, Knop, Mednick, & Guze, 1977a,
23
1977b; Goodwin, 1979). They studied children of alcoholics
extensively in an attempt to find a genetic factor in alcoholism. A
series of large adoption studies were performed to test whether
individuals separated at birth from their alcoholic parents would
develop alcoholism at the same rate as those individuals who were
raised by their alcoholic parents. It was hypothesized that if both
groups developed alcoholism at the same rate, a genetic factor could
be indicated.
Children of alcoholics were divided into the following
categories: a) males raised by their alcohohc biological parents, b)
females raised by their alcoholic biological parents, c) males raised
by non-alcoholic foster parents, and d) females raised by non
alcoholic foster parents. Age-matched controls were paired with
each group. The results showed that both adopted and non-adopted
sons of alcoholics were significantly more likely than controls to
develop alcohohsm (18% versus 3.5%), which indicated some
genetic component in the familial transmission of alcoholism in
males.
The findings for the daughters of alcoholics studies were less
definitive. Both groups of daughters (adopted and non-adopted) as
well as the control group evidenced significantly higher rates of
alcoholism than the general population. In addition, the females had
significantly lower rates of heavy drinking than the males.
24
However, 30% of daughters raised by their alcoholic parents
had been treated for depression. The authors concluded that being
raised by an alcoholic parent increased the risk for depression in
women but not for men; if the daughters were raised by non
alcoholic foster parents, this risk was not present. Research has
supported the higher incidence of affective disorder in bulimics and
their relatives as well (for example, Hatsukami, Eckert, Mitchell, &
Pyle, 1984).
According to Goodwin and his colleagues, counteractive
biological factors (e.g., decreased tolerance to alcohol) and
environmental factors (e.g., societal restrictions on female drinking)
may limit the expression of alcoholism in females even though they
possess a genetic susceptibility to this disorder. Bulik (1987a)
posited that depression and bulimia might be expressions of this
genetic susceptibility which are not subject to societal restrictions.
In their observation that similarities exist between bulimics and
alcoholics, various authors have also cited societal restrictions as
accounting for the difference in the substance abused (Brisman &
Seigel, 1984; Eilstead, Parrella, & Ebbitt, 1988). Thus, as
Broderick (1991) has suggested, there may be intervening variables
in the transmission of psychopathology in families. Among the
factors he discussed was the importance of the “meso-system”, or the
child’s extended network. In the case of bulimia, one message the
25
meso-system may transmit is the sociocultural pressure for women
to be thin.
Intergenerational Transmission of Affective Disorder
Like alcoholism, research has shown that both bipolar
disorder and unpopular depression have heritable components
(Davison & Neale, 1990). Estimates on the frequency of affective
disorders among first-degree relatives of individuals with bipolar
illness range from about 10% to 20%. Studies of unipolar
depression in twins usually report monozygotic concordance rates of
about 40% and dyzygotic concordance rates of about 11% (Allen,
1976). As was noted above, there have been differences noted in
male and female heritability patterns: both male and female first-
degree relatives of men whose depression occurs for the first time
rather late in life, after age 40, are at about equal risk for
depression. But for male and female first-degree relatives of women
whose first episode of depression begins earlier, the pattern is
different. The female relatives are at greater risk for depression
than male relatives, who are at greater risk for alcoholism and
sociopathy (Winokur, 1979). The incidence of bulimia among the
females in these studies has not been observed.
Bulimia and affective disorder. There has been extensive
discussion over the link between affective disorder and bulimia
(Johnson & Connors, 1987; Kassett, Gershon, Maxwell, & Gurroff,
1989; Pope & Hudson, 1986). Research has supported an elevated
26
rate of affective disorder in the first- and second-degree relatives of
bulimics (Kassett, et al., 1989). Moreover, major depression is
frequently seen in bulimic patients (Pope & gudson, 1986). As was
discussed above, bulimics have abnormal neuroendocrine finding
similar to those of depressed patients (scores on the DST) and
respond well to antidepressant medication. These findings have been
the basis on which Pope and Hudson (1986), for instance, have
argued for the same pathogenesis of bulimia and affective disorder.
Yet Johnson and Connors (1987) reported that symptoms of
depression seem to remit on improvement of disturbed eating after
specific psychological interventions. Mitchell (1988) concluded that
depression is a secondary phenomenon to bulimia, although the
relationship is interactive.
Intergenerational Transmission of Bulimia
The research on the incidence of eating disorders (e.g.,
anorexia nervosa or bulimia nervosa) in bulimics’ relatives has been
equivocal. For example, Kassett and associates (1989) found
significantly more eating disorders among first-degree relatives of
patients with anorexia nervosa or bulimia when compared to a
control group of medical patients. Yet Garfinkel and coworkers
(1983) found no difference among restricting anorexics, bulimic
anorexics, and normal controls on parents’ actual weights, body-size
estimates, and body satisfaction. Several studies, however, have
found a higher incidence of maternal obesity among the bulimic
27
subgroups compared to the restricting anorexics (Herzog, 1982;
Strober, 1981).
Although the research findings on the direct intergenerational
transmission of eating disorders in general, and bulimia specifically,
have been mixed, other family factors have been found to be
important in bulimia. Much of the literature on bulimics’ families
thus far has focused on the global family environment of bulimics
rather than specific family factors, such as sexual abuse. Wonderlich
(1992) suggests the following:
. . . unique ‘nonshared’ family experiences may be
extremely significant in shaping personality. This is
consistent with personality theories that suggest that
specific types of environmental (possibly familial)
experiences influence and shape the development of
specific types of personality disturbance, which may in
turn increase the chance of developing a clinical
syndrome, such as bulimia. Theories of this type, which
I describe as family predispositional theories, generally
acknowledge some organismic vulnerability factor such
as temperament (e.g., Benjamin, in press) or affective
instability (e.g., Johnson & Connors, 1987), but the
emphasis is more clearly on the role of interpersonal
(family) disturbance. . . . These family processes would
be considered etiologically critical and conceptually
before the development of both personality disorder and
any associated eating disturbance; furthermore, the
personality disturbance would be considered a
reasonable and expected adaptation to the family
environment, (pp. 119-120)
From this discussion, the study of specific family processes, such as
familial substance abuse and familial sexual abuse, would be
28
important to study as etiologically significant in the development of
bulimia. These two factors (familial substance abuse and sexual
abuse) will now be discussed as they relate to bulimia.
Bulimia and substance abuse. In an attempt to explain bulimia,
one vein of research has focused on the relationship between bulimia
and substance abuse (Brisman & Seigel, 1984; Jones, Cheshire, &
Moorhouse, 1985; Lacey & Moureli, 1986; Zweben, 1987).
Brisman and Seigel (1984) posited that alcoholism and bulimia may
be a different expression of the same underlying addictive disorder.
In support of this contention of symptom substitution, much of the
research has shown elevated rates of substance abuse in the bulimic
population (Beary, Lacey, & Merry, 1986; Bulik, 1987a; Jones et
al., 1985; Kassett et al., 1989; Killen et al., 1987; Mitchell,
Hatsukami, Pyle, & Eckert, 1988). Research also has indicated that
there is an elevated rate of psychiatric illness in the families of
bulimics, including eating disorders (for example, Strober &
Humphrey, 1987) and substance abuse (Bulik, 1987b; Jones et al.,
1985; Kassett et al., 1989). In sum, although parents’ alcoholism
appears to play a role in the development of bulimia in children, the
exact nature of the role is more illusive.
Separately, two lines of research have added to the
understanding of this possible link between parental alcoholism and
the development of bulimia: First, the intergenerational
transmission of alcoholism through genetic pathways has been
29
supported (Goodwin, 1979). Second, family systems literature has
posited and found evidence for intergenerational transmission of
psychopathology (e.g., Broderick, 1991) and specifically, familial
contributors to bulimia (Kog & Vandereycken, 1985; Wonderlich,
1992).
From these discussions it seems logical to explore the
possibility that addictive behavior, expressed either as addiction to
alcohol or addiction to food, may be a “transmitted” disorder. In
the following section, addiction will be discussed, including bulimia
as a type of substance abuse. Next, the research citing the
coincidence of bulimia and substance abuse will be summarized.
Third, the literature examining bulimics’ relatives’ substance abuse
will be reviewed. Finally, the family systems literature as it relates
to intergenerational transmission as well as the research on the
intergenerational transmission of alcoholism will be discussed in an
attempt to support the possible intergenerational transmission of
bulimia.
Substance abuse and addiction. The phenomenon commonly
known as “addiction” has been associated with many substances and
behaviors, from heroin to work. This term has been adopted from
the substance abuse literature which defines addiction as “. . . a
behavioral pattern of drug use that is defined as a preoccupation with
the acquisition of the drug, compulsive use of a drug, and a
propensity to relapse to the use of drugs” (Miller & Gold, 1989, p.
30
224). Preoccupation with the acquisition of the drug is illustrated in
DSM III-R by the following criteria: (a) “a great deal of time spent
in activities necessary to get the substance, or recovering from its
effects” (p. 68) and (b) “important social, occupational, or
recreational activities given up or reduced because of substance use”
(p. 69; APA, 1987).
Compulsive use, defined as continued use of a drug in spite of
adverse consequences, is often associated with regular use, but not
always, as a binge might be compulsive (Miller & Gold, 1989). The
DSM III-R criteria which illustrate this feature are (a) “substance
often taken in larger amounts or over a longer period of time than
the person intended”; (b) “frequent intoxication or withdrawal
symptoms when expected to fulfill major role obligations at work,
school, or home”; and (c) “continued substance use despite
knowledge of having a persistent or recurrent social, psychological,
or physical problem that is cause or exacerbated by the use of the
substance” (pp. 68-69).
Relapse is connoted in DSM III-R by the criteria “persistent
desire or one or more unsuccessful efforts to cut down or control
substance use” (p. 68). DSM III-R uses the terms “Psychoactive
Substance Dependence” or “Psychoactive Substance Abuse” (the
former requires more criteria for diagnosis) rather than addiction to
apply to the above criteria.
31
Bulimia as a type of substance abuse or addiction. Bulimia has
been described as having addictive features similar to those described
by DSM III-R above (Filstead et al., 1988; Zweben, 1987). Filstead
and his colleagues (1988) described the common symptoms of
addiction in bulimia as including: preoccupation with food,
continuance of the binge-purge cycle despite negative physical
consequences, and loss of control over the binge. These symptoms
fall under the three components of addiction described by Miller and
Gold (1989): preoccupation, compulsive use, and relapse. In his
theoretical discussion of substance abuse and eating disorders,
Zweben (1987) added feelings of shame, the need to hide the
behavior, and the compulsive quality of bulimia as other addictive
components of the disorder.
Other researchers have discussed the conunonalties between
bulimia and substance abuse (Brisman & Seigel, 1984; Lacey &
Moureli, 1986). Lacey and Moureli (1986) investigated the
incidence of eating disorders among alcoholics. These authors
interviewed 27 female inpatients who had been diagnosed as
alcoholic. Lacey and Moureli noted the similarities in the behavior
between binge eating in bulimics and bout drinking in alcoholics.
They found that both groups used bingeing (on either food or
alcohol) for the same purposes: to diminish anxiety pertaining to
interpersonal problems, and to reduce feelings of degradation, self
disgust, anger, and depression.
32
Commensurately, in their theoretical review of bulimia and
alcoholism, Brisman and Seigel (1984) posited that all substance
abusers (including bulimics) have a core personality structure in
common that is deficient in emotional and self-regulatory
functioning. In their opinion, the function the substance serves is
more crucial than the specific substance used. One would binge on
food or cocaine, for example, in an attempt to minimize the impact
of ego deficits. These authors hypothesized further that the choice
of substance abused may be a result of exposure to particular
cultural pressures or factors.
Similarly, Filstead et al. (1988) investigated whether high-risk
situations for engaging in heavy drinking and binge eating were
similar or different in 85 inpatients being treated for both
alcoholism and an eating disorder. These authors found that three
out of eight risk situations were statistically significant for both
alcohol use and binge eating according to their risk situations
questionnaire. “Negative emotional state” was cited as the most
frequent reason for engaging in both behaviors. These authors
concluded that although differences existed between the high-risk
situations for engaging in binge drinking and binge eating, the
motivation may have been the same with the behavior varying
depending on social context.
Lastly, Johnson and Berndt (1983, in Kog & Vandereycken,
1985) compared the results on the Social Adjustment Scale of
33
bulimia patients with those of depressed, alcoholic, and
schizophrenic patients. The bulimics’ general pattern of social
adjustment was most comparable to that of the alcoholic sample.
The consensus among the above literature was that bulimia and
substance abuse share common addictive features. These similarities
included preoccupation with a substance, continuing to abuse the
substance when such use results in negative consequences, and loss of
control over the consumption of a substance. Feelings that were
shown to be common among bulimics and substance abusers included
shame, guilt, and obsession with the substance. Both alcoholics and
bulimics reported using their chosen substance for the same reasons.
To support the hypothesis that bulimia and substance abuse may be
different expressions of the same underlying phenomenon further,
the coincidence of these two disorders has been explored.
Coincidence of buhmia and other substance abuse. A high
coincidence of bulimia and other substance abuse has been found in
the literature (Beary et al., 1986; Bulik, 1987b; Hatsukami et al.,
1984; Killen et al., 1987; Pyle, Mitchell, & Eckert, 1981). In
particular, Beary et al. (1986) investigated the associations between
eating disorders and alcohol abuse. These investigators conducted a
semi-structured interview with 20 alcohohc women, 20 bulimic
women, and 17 age-matched controls. These authors found that 50%
of their bulimic sample used alcohol excessively.
34
Similarly, Bulik (1987b) studied patterns of drug and alcohol
abuse in 35 bulimic women and 35 normal controls. These subjects
were not inpatients; rather, they were recruited through media
announcements. This author found that one of the most common
concomitant diagnosis in her sample of bulimic women was alcohol
abuse. Compared to the control subjects, the bulimic subjects were
also ranked significantly higher on measures of alcohol dependence,
drug abuse, and drug dependence according to the Diagnostic
Interview Schedule.
Hatsukami et al. (1984) examined the occurrence and severity
of alcohol and other drug use among 108 normal-weight bulimic
inpatients by means of a semi-structured interview. Like the above
researchers, Hatsukami and her colleagues found that 18.5% of these
bulimic women were diagnosed as having alcohol or drug abuse
problems and 13.9% had received treatment for alcohol or drug
abuse.
The findings of Killen et al. (1987) also supported the high
coincidence of bulimia and substance abuse. In an attempt to assess
the prevalence of bulimia in tenth grade females, these researchers
administered a questionnaire which included questions regarding
alcohol, cigarette, and marijuana usage. Killen and colleagues
reported that the subjects who had been classified as bulimic
reported higher rates of drug use and more frequent “bouts of
drunkenness” than normal subjects.
35
Finally, Pyle et al. (1981) also found a high rate of chemical
dependency problems within their outpatient sample of bulimic
women. An interview was performed which included questions
based on the DSM III criteria for bulimia. Of the 34 subjects , eight
reported receiving prior treatment for chemical dependency and 19
subjects reported using alcohol at least several times a week. Eleven
subjects reported using amphetamines at least intermittently.
The above studies examined the contemporaneity of bulimia
and other substance abuse and revealed that this phenomena was
quite prevalent. Bulimics frequently abused alcohol and/or other
drugs. To elucidate the relationship between substance abuse and
bulimia further, the research on incidence of substance abuse in the
relatives of bulimics will be presented next.
Substance abuse in the relatives of bulimics. Recent research
has cited increased substance abuse among the relatives of bulimics
(Bulik, 1987a; Claydon, 1987; Hudson, Pope, Jonas, & Yurgelun-
Todd, 1983; Hatsukami et al., 1986; Herzog, 1982; Jones et al.,
1984; Stern et al., 1984; Strober, Salkin, Burroughs, & Morrell,
1982).
Bulik (1987a) studied patterns of drug and alcohol abuse in
bulimic women and their first- and second-degree relatives. She
found that the most frequently occurring psychopathology in the
relatives of bulimics was alcoholism. Nearly half of bulimics’ first-
36
degree relatives were diagnosed with alcoholism using the Diagnostic
Interview Schedule.
Claydon (1987) found similar results when investigating the
relationship between college students with alcoholic home
backgrounds and eating problems. He found that those students who
were children of alcoholics were twice as likely to report having an
eating disorder.
While investigating characteristics of 114 bulimic outpatients,
Hatsukami et al. (1986) also found a high rate of alcoholism among
the parents of her sample. Among the different bulimic subgroups
(bulimia only [N=40], bulimia with a history of affective disorder
[N=34], and bulimia with a history of substance abuse [N=34]), the
rate of alcoholism for fathers ranged from 18% to 32%; for
mothers, 8% to 12%.
Herzog (1982) reported on 30 cases of bulimia from whom he
gathered family histories. Of the 30 bulimic patients, 10 had a
history of alcoholism in at least one first-degree family member, a
finding which was convergent with the above research.
Jones et al. (1984) found that 7 of the 27 patients with an
eating disorder had an alcoholic parent. Similarly, Strober et al.
(1982) examined family psychiatric morbidity of bulimic anorexics
and found that a positive family history for alcoholism characterized
83% of these bulimics. Other drug use disorders tended to occur
more frequently in the relatives of bulimic subjects as well.
37
This research suggests that an elevated rate of substance abuse,
particularly alcoholism, is present in the families of bulimics.
Taking into account these findings, it seems evident that substance
abuse, in the bulimic and in her family, is somehow important in the
development of bulimia. One way in which the two problems
(familial substance abuse and bulimia) may be related is that
alcoholic parents may abuse a substance (alcohol) as a means of
coping with distress, which sets a model for their child to do the
same. Similar to how an alcoholic uses alcohol, the bulimic uses
food. Indeed, the binge phase of bulimia is often described as
“numbing” or an “escape.” Thus, the alcoholic parent may model a
coping strategy of using substances to anesthetize pain.
Sexual Abuse and Eating Disorders
An interesting and relatively new hypothesis that is being
discussed in the literature is that eating disorders are related to
sexual abuse, both in childhood and later sexual assaults such as rape
(Beckman & Burns, 1990; Bulik, Sullivan, & Rorty, 1989; Calam &
Slade, 1989; Goldfarb, 1987; Hall, Tice, Beresford, Wooley, & Hall,
1989; O’Halloran, Bostwick & Jacobs, in press; Oppenheimer,
Howells, Palmer, & Chaloner, 1985; Palmer, Oppenheimer,
Chaloner, & Howells, 1990; Wonderlich, et al., in press). Eor the
purpose of the present paper, adulthood sexual victimization as it
relates to eating disorders will not be discussed here. Instead, the
38
research on the coincidence of eating disorders and childhood sexual
abuse, both intra- and extrafamilial, will be reviewed.
One line of literature has addressed the theoretical link
between eating disorders and sexual abuse. For example. Root and
Fallon (1989) posited that eating disorders are a reaction to earlier
sexual abuse. These authors discuss the functions bulimia may play
for women who have been the victims of sexual abuse. These
functions include the following:
1. Bulimia may act as an anesthesia for negative feelings
associated with victimization of women (e.g., rage, pain, fear,
powerlessness). It may serve to repress the pain through
dissociation.
2. Purging may be an attempt to cleanse the person from
feelings of being “dirty” or ashamed from the sexual abuse.
3. Bulimia may act as an outlet for anger in a currently
abusive relationship. If a woman feels it is not safe to express anger,
she may use the binge-purge behaviors to express it symbolically.
Or, if a woman was abused after she displayed direct anger, she may
feel responsible for the abuse.
4. Bulimia justifies that she is worthless and deserved the
abuse. A woman might think, “I am bad for engaging in such a
disgusting, ugly, out-of-control habit. If I were only a better
person, smaller, thinner, I will not be (would not have been)
abused.”
39
5. Bulimia is an attempt to establish physical and
psychological space. Sexual abuse is an invasion of privacy; bulimia
is hidden and private.
6 Bulimia can be viewed as an attempt to control the victim’s
environment and body. Through restrictive eating, the bulimic (or
anorexic) gains control through willpower. The purging aspect of
bulimia may also be seen as a way for the bulimic to bring her body
back under control after an out-of-control binge.
7. The bulimic’s body becomes the object of hatred rather
than her perpetrator. She transfers her rage towards the abuser to
her own body. This “body betrayal” phenomenon is often expressed
by victims of sexual abuse, especially if they felt positive sexual
feelings during the abuse (Finklehor, 1979).
8. Bulimia is a predictable experience in a chaotic world.
9. Bulimia is a way to cope with stress and relieve tension,
like alcohol and drugs.
Similarly, Sloane and Leichner (1986) hypothesized that
anorexia is a logical outcome of early sexual trauma. In contrast to
earlier psychodynamic formulations of the disease that linked it to
oral impregnation fantasies, these authors state that anorexia can
instead be seen as true fear of impregnation from sexual abuse. As
was mentioned above, anorexia can be a way to avoid secondary sex
characteristics and maintain control over one’s body, which are both
40
ways of denying the unbearable and guilt-ridden sexual past (Bemis,
1978, in Sloane & Leichner, 1986).
Another theoretical link between eating issues and sexual abuse
was discussed by Maybury and Cameron (1986) who stated that the
reasons women use (abuse) food include power, denial or avoidance
of sexuality, and alleviation of guilt. They note that these are the
same feelings so commonly found among women who have been
sexually abused.
Empirical studies have also examined the coincidence of eating
disorders and childhood sexual abuse. Recently, Mallinckrodt,
McCreary, and Robertson (1993) conducted a meta-analysis of the
co-occurrence of eating disorders and childhood sexual abuse.
Problems in performing a meta-analysis of these types of studies
included a wide variation in samples and methods used to determine
eating disorders, childhood sexual abuse, and incest, as well as the
failure of many studies to break down intra- versus extrafamilial
abuse and type of eating disorder (e.g., bulimia versus anorexia
versus subclinical eating disorders). Excluding studies of
nonempirical nature and those using small sample sizes (less than 10
subjects), Mallinckrodt et al. (1993) reviewed 27 articles. They then
categorized the studies into five types: (a) single sample comparison
studies using samples of clients or recruited women with eating
disorders; (b) single sample comparison studies using samples of
clients in treatment for incest; (c) comparisons of clients in treatment
41
by diagnosis, e.g., comparisons of women with eating disorders
versus multiple personality disorder; (d) nonclinical samples
comparing those who reported eating disorders to those who did not;
and (e) comparisons of clinical samples and control groups.
The results of the studies reviewed in the first three categories
yielded eating disorder and childhood sexual abuse coincidence rates
of 7% to 49%. The fourth category of studies generally yielded a
significantly higher rate of eating disorders among those reporting
childhood sexual abuse histories than those not reporting such
history. The fifth category of studies also reported higher incidence
of childhood sexual abuse among eating-disordered and psychiatric
clients than controls, but not significantly different from each other.
For example, Steiger and Zanko (1990) compared rates of sexual
abuse (both intra- and extra-familial) among women with eating
disorders (N=73), women with other psychiatric diagnoses (N=21)
and normal controls (N=24). The women with eating disorders and
other psychiatric diagnoses had significantly higher rates of
childhood sexual abuse than normal controls, but similar rates to
each other (eating disorders group = 30% childhood sexual abuse;
psychiatric diagnosis group = 33% childhood sexual abuse; normal
controls = 4% childhood sexual abuse). From this category of
studies, Mallinckrodt et al. (1993) hypothesized that
". . . given the possible high prevalence of incest in the general
population, perhaps the only significant co-occurrence elevation
42
above expected base rates is that incest may be more likely in clients
who present with eating disorders, and eating disorders may be more
likely in clients who are incest survivors” (p. 3). To summarize, in
their meta-analysis, when one study of questionable methodology
which yielded a 7% coincidence rate was excluded, these authors
found a 20% coincidence rate of eating disorders and incest among
the chnical sample, which is significantly higher than the 16% rate
of sexual abuse for the general population of women suggested by
Russell’s large-scale epidemiology study (x^[l] = 4.10, p <
.05)(Russell, 1983). Yet among non-clients, the coincidence rate of
eating disorders and incest obtained in their meta-analysis was 9%,
which is lower that Russell’s figures.
Pope and Hudson (1992), in their meta-analysis of eating
disorders and sexual abuse studies, concluded that childhood sexual
abuse is not significantly higher in women with eating disorders than
in other women in part because the rate of sexual abuse is so high in
the general population. One of the reasons these authors give as
explanation for why eating disordered clients report childhood
sexual abuse more often than nonclinical populations is that eating
disorders and sexual abuse both may be more likely in chaotic
families (although this does not necessarily imply causality).
Mallinckrodt et al. (1993) posited that while eating disorders are
multiply determined, one of the developmental pathways may indeed
be childhood sexual abuse. More generally, they hypothesized that
43
eating disorders . . may be linked to sexual abuse, especially
incest, by a set of common dysfunctional family dynamics such as
role confusion and unhealthy modes of coping with family conflict”
(p. 4).
These particular dysfunctional family dynamics that
Mallinckrodt and his colleagues cited are interestingly similar to the
dysfunctional family dynamics which occur in alcoholic homes. For
example, one of the most frequently mentioned dynamics in the ACA
literature is role confusion. In this, the child of an alcoholic parent
takes on the role of the parent while the alcoholic parent takes on the
child role (for example, Woititz, 1983). Moreover, sexual abuse
often occurs within the context of an alcohol-intoxicated perpetrator
(Briere, 1992). Many authors agree that the dysfunctional family
dynamics which are associated with parental alcoholism are
important etiological factors in the development of later
psychological disturbance of the child (for example, Broderick,
1991). Yet, similar to the multifactorial model proposed in the
present study, parental alcoholism is generally not held to be the sole
contributor to later psychological maladjustment, nor does it
necessarily result in a homogenous disorder. Indeed, there is
increasing evidence in the literature against the “uniformity myth” of
adult children of alcoholics (for example, Mintz, Kashubeck, &
Tracy, in press). Given that parental alcoholism and incest
frequently occur within the same family, and that bulimic individuals
44
often have an increased rate of each of these dysfunctions in their
families of origin, these two factors may be related to the
development of bulimia.
Trauma Theories
Various authors have developed theories about child abuse
symptomatology. Relevant to the present study are the trauma
theories of Judith Herman (1992) and John Briere (1991, 1992).
Both Herman and Briere discuss the aftermath of child abuse trauma
which can occur in alcoholic families and sexually abusive families.
Herman (1992) proposed that child abuse, including child
sexual abuse, result in traumatic aftermath much like wartime
veterans’ experiences. She discussed ways in which child abuse
victims learn to cope with their trauma, including altering their
states of consciousness and abusing their bodies. Herman described
two scenarios which often evoke terror-filled memories in child
abuse victims: bedtime and mealtime, both of which are generally
thought to be times of comfort and safety. In the case of a child who
is being sexually abused within her own family, bedtime may be a
frightening time in the anticipation of sexual intrusion by her
parental perpetrator. In the case of a child of an alcoholic, mealtime
may be a time of drunken chaos. In both of these scenarios, the
child’s basic bodily functions (sleeping, eating) have been disrupted
by fear. Out of these painful experiences grows the coping
mechanism of dissociation to avoid the affective pain. Herman
45
hypothesized that one way abused children deal with the feelings of
terror, rage, grief, and abandonment is by hurting their bodies:
“Abused children discover at some point that the feeling can be most
effectively terminated by a major jolt to the body” (p. 109). She
listed purging and vomiting as one way which abused children
regulate their emotional states. It is an attempt to “. . . obliterate
their chronic dysphoria and to simulate, however briefly, an internal
state of well-being and comfort that cannot otherwise be achieved”
(p. 110). If, as Herman proposed, victims of parental alcoholism
and/or sexual abuse are vulnerable to relying on the coping
mechanisms of altered consciousness (dissociation) and bodily abuse,
it seems likely that bulimia would be a method of choice as it is
hypothesized to be a dissociative mechanisms as well as clearly
painful to the body.
Similarly, Briere (1992) in his book on child abuse trauma,
theorized that, “. . .in the face of extreme abuse-related
abandonment or rejection, the survivor may engage in any of a
number of external activities that anesthesize, soothe, interrupt, or
forestall painful affect” (p. 63). Bulimia may be one of the external
activities child abuse victims use to reduce their painful affect.
Briere listed reasons that an abuse survivor may engage in tension
reduction activities including temporary distraction, interruption of
dissociative or dysphoric states, anesthesia of psychic pain,
restoration of control, temporary filling of perceived emptiness.
46
self-soothing, and specific relief from guilt or self-hatred. While
many bulimics often do not link their binge-purge behavior to
specific events or feelings, Briere explains that many of child abuse
victims’ disturbing memories and painful feelings have been denied
or dissociated. Thus, the tension-reduction responses may be seen as
impulsive, or from “out of nowhere” (p. 64). Interestingly, this
model fits perfectly with bulimic individuals’ descriptions of what
precedes the binge-purge cycle. For example, in their study asking
bulimics to describe the factors that precipitated a binge-purge
episode, Abraham and Beaumont (1982) found that tension (91%),
feeling bored and lonely (59%), physical symptoms of anxiety
(80%), and feelings of depersonalization and derealization (79%)
were endorsed as factors precipitating a binge. Similarly^ Carroll
and Leon (1981) reported that the most common affective states
preceding a binge were anger, anxiety, rejection, and loneliness.
Hatsukami, Owen, Pyle, and Mitchell (1982) found that feeling tense
and anxious was the most commonly cited reason for binge eating in
bulimics.
The binge portion of the bulimic cycle, as mentioned earlier,
is often described as numbing and has been hypothesized to be a
dissociative mechanism (Root & Fallon, 1989). For example, in the
study mentioned above, 34% of respondents described relief from
anxiety during the binge, 66% described freedom from anxiety at
the conclusion of a binge, and 72% reported being free of negative
47
mood states while bingeing (Abraham & Beaumont, 1982).
However, Johnson and Larson (1982) suggested that the purge rather
than the binge portion of the bulimic cycle is in fact sought over
time for its tension-reducing effect; purging is often described as a
way for the bulimic to feel “back in control.” Clearly, the bulimic
cycle closely matches Briere’s description of an abuse-related
tension-reducing behavior cycle.
The present study seeks to lend support to this tension-
reduction/coping theory of child abuse trauma by investigating
whether traumatic symptomatology will be present at a higher rate
in those individuals who have experienced alcoholic parenting and
sexual abuse in childhood as well as in those individuals who are
bulimic. The present study also relies on the object relations and
family systems formulations of Humphrey and Stem (1988)
discussed above, who, like Briere, posited that bulimia is a tension
regulation method; these authors, however, state that the deficits in
tension regulation bulimics experience is related to the failure of
bulimics’ parents to create a sufficient holding environment. While
not implying causality, if the results of this study are in the proposed
direction, they may lend credence to the tension-reduction model of
bulimia.
Statement of the Problem
The exact etiology of the eating disorder bulimia is still
unclear. While it is generally held that multiple factors such as
48
sociocultural influences, psychological factors, and family dynamics
contribute to the development of bulimia, less is known about the
specific family experiences which are risk factors for the
development of bulimia. From the above literature review, it is
apparent that familial alcoholism often plays a role in the
development of bulimia. Specifically, research has shown a higher
incidence of parental alcoholism among bulimics (for example,
Bulik, 1987a). Familial sexual abuse (incest) may also be another
important factor in the development of bulimia, although the link
between childhood sexual abuse and later bulimia is less clear (for
example, Calam & Slade, 1989). In addition, adult children of
alcoholics as well as survivors of childhood sexual abuse are at
higher risk for later substance abuse. Theoretically, bulimia has
been labeled as a type of substance abuse (Zweben, 1987). Thus,
parental alcoholism and childhood sexual abuse may be contributors
to the later development of the eating disorder bulimia. The present
study seeks to investigate whether and to what degree parental
alcoholism and incest are two specific famihal risk factors in the
development of bulimia.
The theoretical model on which the present investigation relies
is the tension-reduction model of bulimia as formulated by, among
others, Briere (1992) and Humphrey & Stem (1988). The present
study seeks to test this model by proposing that bulimia is a tension-
reduction mechanism related to childhood trauma. The specific
49
traumatic experiences which are hypothesized to be related to
bulimia are parental alcoholism and childhood sexual abuse. This
study, then, will investigate whether traumatic symptomatology is
present at a higher rate in those individuals who have experienced
alcoholic parenting and sexual abuse in childhood and who develop
bulimia than those who have not had such childhood experiences.
Research Questions and Hypotheses
1. Will subjects who have higher rates of traumatic
symptomatology have higher rates of bulimia than subjects with
lower rates of traumatic symptomatology?
Hvpothesis 1: Subjects with higher rates of traumatic
symptomatology will have higher rates of bulimia than subjects with
lower rates of traumatic symptomatology.
The present study attempted to lend support to the theory that
bulimia is an expression of an addictive or tension-reducing
behavior. Theorists have proposed that bulimia may be a coping
mechanism in response to traumatic childhood experiences or
(Briere, 1992; Herman, 1992) or insufficient tension-reduction
capacities (Bruch, 1973; Humphrey & Stem, 1988; Swift & Letven,
1984). In order to investigate this hypothesis, the rates of post-
traumatic symptomatology in individuals with bulimia were
examined.
2. Will subjects with alcoholic parents have higher rates of
traumatic symptomatology than subjects without alcoholic parents?
50
Hypothesis 2: Subjects with alcoholic parents will have higher
rates of traumatic svmptomatology than subjects without alcoholic
parents.
Much of the literature on the effects of parental alcoholism on
adult children concludes that adult children of alcoholics (ACAs) are
prone to greater negative psychological sequelae than their non-ACA
counterparts. For example, the following are among the difficulties
for which ACAs have been found to be at greater risk: psychosocial
maladjustment (Russell, Henderson, & Blume, 1985), increased self-
deprecation (Berkowitz & Perkins, 1988), and psychiatric symptoms
and impaired marital relationships (Greenfield et al., 1993). In the
clinical literature, it has been theorized that ACAs experience
problems such as inability to trust, low self-esteem, eating disorders,
and alcohol and substance abuse (Black, 1990). One explanation for
the greater negative psychological symptoms that some ACAs
experience is that being a child of an alcoholic parent can be a
traumagenic experience. While various personality characteristics
and psychological disorders of ACAs have been examined in the
research and clinical literature, traumatic symptomotology per se in
ACAs has never been measured systematically to this author’s
knowledge. In order to examine whether being an ACA is indeed a
traumagenic experience, the present study examined whether those
subjects who are ACAs have a greater incidence of traumatic
symptomatology than those subjects who are non-AC As.
51
3. Will subjects with sexual abuse histories have higher rates
of traumatic symptomatology than subjects without sexual abuse
histories?
Hypothesis 3: Subjects with incest or sexual abuse histories
will have higher rates of traumatic symptomatology than subjects
without incest or sexual abuse histories.
The child abuse literature has documented the traumagenic
nature of child sexual abuse by showing an increased risk for
traumatic symptomatology among those who are victims of child
sexual abuse in clinical and college populations (for example, Briere,
Elliott, Harris, & Cotman, 1993; Smiljanich & Briere, 1993). This
study attempts to replicate these findings by showing an increased
rate of traumatic sequelae in college sorority women who have been
sexually abused.
4. Will subjects with alcoholic parents have higher rates of
bulimia than subjects with non-alcoholic parents?
Hypothesis 4: Subjects with alcoholic parents will have higher
rates of bulimia than subjects with non-alcoholic parents.
Most researchers have found a high coincidence of parental
alcoholism and bulimia (for example. Clay don, 1987; Hatsukami et
al., 1986; Hudson, Pope, Jonas, & Yurgelun-Todd, 1983b) although
others have not (for example, Mintz et al., in press). Two related
hypotheses that can be used to explain the elevated rate of alcoholism
in bulimics’ families are: (a) The alcoholic parent modeled
52
substance (alcohol) abuse as a way of coping with psychological
distress and thus the bulimic abuses food to do the same, and (b) the
bulimic uses the binge-purge cycle as a way to cope with the
traumatic experience of being a child of an alcoholic parent. In an
attempt to replicate the majority of research findings that there is an
increased risk for bulimia among adult children of alcoholics, and to
lend support to the above explanations of this phenomenon, the rate
of parental alcoholism among bulimics was examined.
5. Will subjects with incest or sexual abuse histories have
higher rates of bulimia than subjects without incest or sexual abuse
histories?
Hypothesis 5: Subjects with incest or sexual abuse histories
will have higher rates of bulimia than subjects without incest or
sexual abuse histories.
Results of the recent research examining childhood sexual
abuse and bulimia have been mixed. For example, the general
findings of the meta-analyses of eating disorders and sexual abuse
studies have concluded that, among nonclinical populations,
childhood sexual abuse was not significantly higher in subjects with
eating disorders than in other women, although a higher than
average coincidence rate of eating disorders and incest among
clinical samples was found (Mallinckrodt et al., 1993; Pope &
Hudson, 1992). One methodological issue which may help to explain
the findings was that in most of these studies, bulimia nervosa and
53
anorexia nervosa were not distinguished as distinct disorders. In
fact, research has shown that the two disorders are discrete, although
related, phenomena (Johnson & Connors, 1987). Moreover, it has
been hypothesized that bulimia, but not anorexia, is associated with
childhood sexual abuse (for example, Steiger & Zanko, 1990).
Based on these findings, and the theory that bulimia is a tension-
reduction mechanism in response to childhood trauma, a higher rate
of bulimia was anticipated in those subjects with childhood sexual
abuse histories.
54
CHAPTER TWO
METHOD
This chapter describes the subjects who participated in the
present study, the measures and procedures employed to carry out
the study, the statistical procedures used to analyze the data, and its
limitations. In order to examine the proposed research hypotheses, a
factorial design was employed. Because the criteria under study
(bulimia, parental alcoholism, child sexual abuse, and traumatic
symptomatology) cannot be manipulated, the study was
nonexperimental.
Subjects
Subjects were 355 undergraduate female college students who
were members of a national Panhellenic sorority at colleges and
universities across the United States. Forty-eight percent of
respondents were from midwestem schools (N = 171), 26% were
from western schools (N = 91), 19% were from eastern schools (N =
6 8 ), 7% were from southern schools (N = 23), and the geographic
region of 2 respondents was undetermined. The age of participants
ranged from 17 to 30 years of age; 98% of the subjects were
between 18 and 22 years of age. Ninety-one percent of the sample
was Caucasian/Euro-American (N = 326), 3% was Asian/Pacific
Islander (N = 9), 3% was Hispanic/Mexican-American/Latina (N =
9), 1% was Native American (N = 5), .03% Black or African-
55
American (N = 1), .03% was ethnicity other than those listed (N =
4), and .03% did not respond (N = 1). Ninety-eight percent of
respondents were heterosexual (N = 349), 1% did not know their
sexual orientation (N = 4), .03% were bisexual (N=l), and .03% did
not respond (N = 1).
Instrumentation
The Bulimia Test-Revised
The Bulimia Test-Revised (BULIT-R; Thelen, Farmer,
Wonderlich, & Smith, 1991; see Appendix A) is based on the
original BULIT, which is a 32-item self-report questionnaire
designed to identify individuals with bulimia in the normal
population based on DSM-111 criteria (Smith & Thelen, 1984).
Welch and Hall (1989) determined that the BULIT was the most
reliable in discriminating between bulimic and non-bulimic
individuals in an inpatient sample of the three measures of bulimia
that they compared. They found that the BULIT was more reliable
than the Conroy-Healy Eating Questionnaire and the Bulimia
sub scale of the Eating Disorder Inventory with an internal
consistency coefficient of .92 for the total test. These authors
determined that “. . . the BULIT is a promising screening measure
of bulimic behaviors. . .”(p. 131).
The BULIT-R, a 28-item, self-report, multiple choice test,
was recently developed in order to measure bulimia in nonclinical
populations based on the more stringent DSM 111- R criteria for
56
bulimia nervosa. From their revision of the original BULIT, Thelen
et al., 1991 determined that the BULIT-R is a reliable and valid tool
by which to measure bulimia nervosa. For example, total BULIT-R
scores showed a significant difference between bulimic and control
groups {t [46] = 16.41; p < .0001). To determine the BULIT-R’s
predictive validity, the BULIT-R was administered to independent
samples of bulimic (N = 23) and non-bulimic controls (N = 157).
Items showed significant differences between bulimic and control
group means in the expected direction (ts ranged from 3.84 to 17.68;
77s < .001). Similar to the original BULIT, high internal
consistency was indicated by a Cronbach’s coefficient alpha of .97.
To test its applicability to nonclinical populations, the BULIT-R was
administered to 1,739 college women and was found to have high
test-retest reliability after a 2-month period (.95; p < .0001).
Predictive validity of this instrument was also demonstrated by the
correspondence of the BULIT-R cutoff scores and diagnoses based
on independent clinical interviews. For example, of 28 college
females who scored above the cutoff score of 104, 23 were judged to
be bulimic on the basis of a clinical interview.
To score the BULIT-R, the 28 items are scored and totaled.
For the purposes of this research study, a BULIT-R score of 104 and
above categorized subjects as bulimic. In addition, as recommended
by Thelen, Kanakis, and Farmer, (1993), scores of 82 to 104 (one
57
standard deviation below the cutoff score) categorized subjects as
subclinical bulimics.
Children of Alcoholics Screening Test- 6
The Children of Alcoholics Screening Test- 6 (CAST-6 ;
Hodgins, Maticka-Tyndale, El-Guebaly, & West, 1993; see Appendix
B) is based on the original CAST, a 30-item inventory designed to
aid in the identification of children of alcoholics. The questionnaire
measures children’s attitudes, feelings, perceptions, and experiences
related to their parents’ drinking behavior. The CAST consists of 30
one-sentence items arranged in a true-false format with the total
number of true answers determining whether a person is categorized
as a child of an alcoholic. Maxwell (1985) reported a split-half
reliability Spearman-Brown coefficient of .948 on the CAST.
Several validity studies have reported that 100% of self-identified
children of alcoholics and children of parents diagnosed as alcoholic
scored 6 or higher on the CAST (Mintz et al., in press). The CAST
has also been used as a measure of disruption in the family home and
has been reported to have a high degree of content validity as a
measure of distress about parental drinking (Roosa, Sandler,
Gehring, Beals, & Capp, 1988). When used with cutoff scores, the
CAST categorizes individuals into adult children of alcoholics
(ACA) status.
Recently, a short-form of the CAST was developed by
identifying a subset of the 30 CAST items that discriminated between
58
adult children of alcoholics (ACA’s) and nonalcoholics (ACNA’s)
among three distinct samples: outpatient substance abusers,
outpatient psychiatric patients, and medical students (Hodgins et al.,
1993). Through principal components analysis, seven items were
selected for further assessment for the short form of the CAST. Six
of these items addressed perception of parental drinking problem or
interactions associated with alcohol consumption (e.g., arguments).
The internal consistency of the shortened CAST was assessed by
comparing item-total correlations for this and the full CAST and
with Cronbach’s alpha. Item-total correlations for the CAST- 6
ranged from .62-.89 across the three samples. Correlations between
CAST- 6 and full CAST scores ranged from .92-.94 for the three
samples. Cronbach’s alphas were .8 6 , .91, and .92 for the student,
substance abuse treatment, and outpatient samples, respectively. The
authors used a cut-point of 3 for the CAST- 6 (i.e., scores of 3 or
higher categorize a respondent as an ACA) which resulted in 2%
false-positives and 4% false-negatives relative to CAST full scale
categorization. The authors concluded that the CAST- 6 compared
favorably with the full CAST and provides a more efficient way to
identify adult children of alcoholics.
In the present study, the CAST- 6 was used to identify ACAs
from non-ACAs by employing a cutoff score of 3 and above.
59
Childhood Sexual Abuse Questions
Three questions inquiring about childhood sexual abuse (see
Appendix D) were adapted from the Childhood Maltreatment
Interview Schedule (Briere, 1992). The first question asks about
sexual contact with someone 5 or more years older than the
respondent before she was 17 years old; the second question asks
whether this sexual contact was with a family member; the third asks
about the respondents’ perception of sexual abuse.
Trauma Symptom Inventory
The Trauma Symptom Inventory (TSI; Briere, 1991; see
Appendix H) is a 104-item instrument designed to measure long
term traumatic symptomatology associated with childhood abuse and
adult trauma. The measure consists of nine clinical sub scales:
Anxious Arousal, Depression, Anger/Irritability, Dissociation,
Sexual Concerns, Dysfunctional Sexual Behavior, Intrusive
Experiences, Defensive Avoidance, and Impaired Self-Reference.
Respondents are asked to indicate the frequency with which they
have experienced various traumatic symptoms in the last 6 months
on a Likert-type scale from 0 (never) to 3 (often).
The TSI grew out of the Trauma Symptom Checklists (TSC-
33, TSC-40; Briere & Runtz, 1989) which are research measures
designed to reliably assess long-term impacts of child abuse.
Prompted by the inappropriate use of these instruments as clinical
screening tools, Briere (1991) developed the TSI as a clinically
60
useful, psychometrically sound measure of posttraumatic
symptomatology. The TSI items were developed on a
rational/intuitive basis stemming from clinical experience and
relevant literature.
Based on large scale studies using clinical, university and
general population samples, the TSI has been determined to possess
good predictive and concurrent validity (for example, Briere,
Elliott, & Smiljanich, in press). In addition, the clinical subscales
have been demonstrated to be highly reliable among these
populations. For the purpose of the present use of the TSI, the
findings of the study employing a university sample will be
reviewed.
The TSI was administered to 279 university students (99 males
and 180 females) (Smiljanich & Briere, 1993). Subjects’ TSI scores
were significantly associated with their childhood histories of sexual
abuse (F[9,257] = 2.08; p < .031). The specific TSI subscales which
were significantly associated with childhood sexual abuse were
Anxious Arousal (F = 5.91; p < 016), Anger/Irritability (F = 8.64;
p < 004), Sexual Concerns (F = 3.72; p < 055), Dysfunctional
Sexual Behavior (F = 12.61; p < 001), and Intrusive Experiences (F
= 3.85; p < 051). The reliability coefficients of the TSI subscales
for these subjects was as follows: Anxious Arousal .8 8 ;
Depression a^= .89; Anger/Irritability a = .91; Dissociation a = .8 8 ;
Sexual Concerns a = .83; Dysfunctional Sexual Behavior a = .78;
61
Intrusive Experiences a = .91; Defensive Avoidance a = .89; and
Impaired Self-Reference a = .87, indicating that these scales were
highly reliable.
To score the TSI, sums of individual items are calculated to
create scores for the nine clinical subscales.
Demographic Questions
Seven demographic questions pertaining to sex, age, ethnicity,
sexual orientation, present height, present weight, and ideal weight
were included.
The above measures were presented in three different
counterbalanced orders to alleviate order effects.
Procedure
Subject Recruitment
Subjects were recruited from 15 chapters of a national
Panhellenic sorority at colleges and universities across the United
States (west coast, east coast, south, midwest). The executive office
of this sorority was contacted for permission to use their members in
this research project. The national director of this sorority granted
written permission to conduct research on eating disorders with their
members. The names, addresses, and telephone numbers of 15
chapter presidents were attained from a national sorority
representative. The investigator then contacted each chapter
president by mail requesting their chapters’ participation in the
research study (see Appendix E). Letters were sent to the chapter
62
presidents instructing them about distribution of the questionnaires
and individual packets. Each member’s packet contained the
questionnaires, a consent form, a return envelope, and a form
indicating whether she would like the results of the study to be
mailed to her (this form was not linked to their questionnaires in
order to protect anonymity). The letter to the presidents instructed
them to distribute the packets to each member, to ask each member
to complete the questionnaires, and to ask each member to mail the
completed packet to this researcher directly to protect anonymity.
The president was also reminded that their members’ participation
was voluntary and was instructed to communicate this to their
members (see Appendix E for sample letter to presidents). Two
weeks after the packets were mailed to each president, the
investigator contacted them by telephone to ask if they had any
questions or concerns about the research.
To summarize, the individual members received a detailed
information and consent form (see Appendix G for consent form) in
addition to the group of questionnaires (see Appendices E, G, H, and
I for questionnaires). To protect anonymity, each individual subject
was also provided with a return envelope addressed to the
investigator in which to mail the completed questionnaires. The
subjects were offered a copy of the results of this dissertation at their
request. In addition, psychological referrals to relevant agencies
(e.g., university counseling centers) were provided upon request
63
(one subject contacted the investigator for this purpose; referrals to
her university counseling center were provided).
Of the 1,113 packets of questionnaires that were mailed to the
15 universities, 355 women responded, which indicates a return rate
of 32%. This return rate, which is lower than the expected 50% in
similar studies, assumes that all 1,113 packets were distributed
among the chapter members. It is likely that a smaller number of
individuals actually received the packets. For example, the number
of packets sent to each sorority was based on their total chapter
membership in the fall semester; it is possible that a portion of the
members had graduated, were inactive in the sorority that semester,
or were not present at the time the questionnaires were distributed
by the chapter president. If this was the case, the return rate would
be higher than 32%, and possibly closer to the expected 50% rate.
Instructions
All subjects were given a packet of the assessment instruments
with a consent form that described the contents (see Appendix G).
Subjects were informed that they had a right to discontinue the
research at any time and that anonymity was maintained. The packet
included the following instruments: The Bulimia Test-Revised
(BULIT-R), the Children of Alcoholics Screening Test-6 (CAST-6 ),
the Trauma Symptom Inventory (TSI), three childhood sexual abuse
questions, and a brief demographic questionnaire. Also included in
the packet were complete written instructions regarding each
64
instrument. The principal investigator had written contact with the
subjects by mail or oral contact if the subjects telephoned the
investigator for counseling, referrals, or questions regarding the
study.
Group Assignment
Random assignment was not used in this study. For the
purposes of some analyses, subjects were classified into one of two
eating disturbance categories based on their responses to the Bulimia
Test-Revised (BULIT-R): (a) bulimic or (b) non-bulimic. As this
instrument is measured on a continuous scale, subjects’ scores were
also examined as “more” or “less” bulimic in some analyses.
Subjects were classified into one of two adult child of
alcoholic parent categories based on their scores on the Children of
Alcoholics Screening Test- 6 (CAST-6 ): (a) adult child of an
alcoholic (ACA) or (b) adult child of a non-alcoholic (non-AC A)
(control). A CAST- 6 score of 3 or higher categorized a respondent
as an ACA; a CAST- 6 score of 2 or lower categorized a respondent
as a non-ACA. As was recommended in a similar recent study
(Mintz, et al., in press), the CAST- 6 was also used as a continuous
measure in some analyses.
Subjects were classified into one of three sexual abuse history
categories based on their scores on three sexual abuse questions: a)
no sexual abuse history, (b) sexual abuse history, no incest history
and (c) incest history. Subjects were classified into the “incest
65
history” category if their history of sexual abuse involved a family
member.
The TSI was not used to classify subjects into groups; rather,
the purpose of administering this instrument was to investigate
whether traumatic symptomatology is differentially present within
bulimic versus non-eating-disordered individuals, ACAs versus non-
AC As, and individuals with or without sexual abuse or incest
histories.
Data Analysis
To measure the first hypothesis, that subjects with higher rates
of traumatic symptomatology would have higher rates of bulimia
than subjects with lower rates of traumatic symptomatology, a
multiple regression analysis was performed with traumatic
symptomatology, measured by the nine TSI subscale scores, as the
independent variables and bulimia, measured by BULIT-R scores, as
the dependent variable.
The second hypothesis, that subjects with alcoholic parents
would have higher rates of traumatic symptomatology than subjects
without alcoholic parents, was investigated by performing a Multiple
Analysis of Variance (MANOVA) using ACA status, measured by
subjects’ CAST- 6 scores, as the independent variable and traumatic
symptomatology, measured by subjects’ scores on the nine TSI
subscales, as the dependent variables. In addition, a one-way
6 6
ANOVA was performed using ACA status as the independent
variable and the total TSI scores as the dependent variable.
The third hypothesis, that subjects with incest or sexual abuse
histories would have higher rates of traumatic symptomatology than
subjects without incest or sexual abuse histories, was measured by
conducting a MANOVA using responses to three sexual abuse and
incest history questions as the independent variables and traumatic
symptomatology, measured by the nine TSI subscale scores, as the
dependent variables. In addition, two one-way ANOVAs were
performed using incest or sexual abuse history as the independent
variables and the total TSI scores as the dependent variable.
The fourth hypothesis, that subjects with alcoholic parents
would have higher rates of bulimia than subjects with non-alcoholic
parents, was tested by conducting a one-way ANOVA using ACA
status, measured by CAST- 6 scores, as the independent variable and
bulimia, measured by BULIT-R scores, as the dependent variable.
The fifth hypothesis, that subjects with incest or sexual abuse
histories would have higher rates of bulimia than subjects without
incest or sexual abuse histories, was measured by performing two
one-way ANOVAs using sexual abuse and incest history as the
independent variables and bulimia, as measured by BULIT-R scores,
as the dependent variable.
67
Limitations
The limitations of the present study will be presented in the
following categories: (a) design and internal validity, (b) external
validity, and (c) measurement issues.
Design and Internal Validity
One of the most important limitations of the present study is
that it is correlational in nature, and thus causality may not be
implied from the findings. Although the theory on which the
present study was based holds that sexual abuse and parental
alcoholism are likely to cause traumatic symptoms in adulthood, and
that bulimia is a way to cope with these traumatic experiences, the
means used to test this model do not allow it to be proved or
disproved.
In his article delineating methodological issues in studying
sexual abuse effects, Briere (1992) presented many issues relevant to
the limitations in the present study as the research of both parental
alcoholism and child sexual abuse present similar methodological
difficulties. For example, as was the case in the present study,
Briere mentioned that subjects are often asked simultaneously about
past abuse and current functioning in sexual abuse research and that
child abuse is considered the independent variable and psychological
measures are considered the dependent variable. Likewise, the
present study assigned childhood sexual abuse and parental
alcoholism as the independent variables and traumatic
6 8
symptomatology and bulimia as the dependent variables. Yet due to
the correlational and retrospective nature of this design, causality
may not be implied. In fact, it may be true that bulimia, for
example, produces traumatic symptomatology instead of the other
way around, as hypothesized.
Moreover, given the nonrandom design of the present study, it
is unlikely that all subjects were equivalent in all ways except on
those variables studied. Thus, it cannot be concluded that the
differences in bulimia and traumatic symptomatology were due to
solely parental alcoholism and child sexual abuse. It is more likely
that, while these variables may have some impact on traumatic
symptomatology and bulimia, they do not account entirely for these
phenomena. For example, other forms of child abuse (e.g., physical
abuse, psychological abuse) which commonly coexist with child
sexual abuse and parental alcoholism were not included in the
present model. Thus, the present results may have been influenced
by these potentially confounding variables.
Another potential problem in the present design is that “time-
specific abuse sequelae” (Briere, 1992, p. 196) were examined in this
study. In other words, traumatic symptomatology and bulimia can
be differentially present at various points in one’s life, and therefore,
by examining these variables only within a 6 -month time span (as is
specified in these questionnaires), true expression of these symptoms
may have been skewed. An attempt was made to address this issue
69
with regard to bulimia in the selection of a sample; bulimia is most
likely to express itself in college-aged women (Johnson & Connors,
1987).
In sum, even though theory supports the present model,
the limitations of internal validity and design in the present study
make it impossible to know unequivocally whether the variables
under study caused the results.
External Validity
Given that a sample of college sorority women was used in
this study, the generalizability beyond similar populations is limited.
As will be discussed in chapter four, there is evidence to suggest that
college students, and more specifically, sorority women, represent a
biased sample with regards to the variables under study. For
example, ACAs who are college students have been shown to be
more resilient and higher functioning than a non-college sample
(Mintz, et al., in press).
Measurement
Another limitation in the present study was that each measure
used was based only on self-report; most importantly, both the
CAST- 6 and child sexual abuse questions inquire about events that
may have occurred in the distant past. Briere (1992) discussed some
important caveats about the validity of such measures. Both
repressed memories (amnesia brought on by psychological
adaptation to painful experiences) and the passage of time may
70
impede subjects’ responses to the CAST- 6 and child sexual abuse
questions. The bias would, presumably, be in the direction of
underestimation of parental alcoholism and sexual abuse. In the case
of the latter variable, Briere noted that amnestic abuse victims are
more symptomatic than their nonamnestic cohorts, which may have
had significant impact on the present findings. Yet, Briere also
pointed out that younger subjects are more likely to report abuse
than older subjects, which may make up for some of the
underreporting due to amnesia. Given that all measures used in the
present study were self-report, it cannot be assured that the
responses were in fact accurate.
Another limitation in the measurement of sexual abuse that
was used in the present study is that it did not ask specifically with
whom the sexual contact occurred (relationship to the abuser), at
what age(s) the abuse occurred, the frequency and duration of the
abuse, whether physical force was used, or extent and nature of
sexual abuse. This information may have shed additional light on the
findings as it has been shown that severity, force, duration, and
relationship to the abuser are all important factors in long-term
abuse effects (Beitchman et al., 1992).
71
CHAPTER THREE
RESULTS
The results of the data analyses will be presented in the present
chapter. First, differences between the bulimic, subclinical bulimic,
and nonbulimic groups on basic demographic characteristics will be
presented. Next, the sample will be described in terms of their
distribution on the variables of parental alcoholism, sexual abuse
history, and traumatic symptomatology in general and according to
bulimia status (bulimic, subclinical bulimic, and nonbulimic).
Lastly, the research hypotheses will be restated followed by the
results of the statistical analyses used to test each hypothesis.
Demographic Characteristics of Respondents
Preliminary analyses were conducted to determine if
differences existed between bulimic, subclinical bulimics, and
nonbulimic subjects on basic demographic data (see Table 1). The
results of chi-square tests indicated that no significant differences
existed between these groups with regard to geographic region
(x2 [6 ] = 11.07, p < .08), ethnicity [10] = 8.37, p < .59), sexual
orientation (% ^ [6 ] = 5.05, p < .54), and age (x^ [18] = 23.31,
p < .18).
72
Table 1
Differences Between Bulimic and Nonbulimic Subjects
Demographic Variable df
P
Geographic Region 11.07 6 .08
Ethnicity 8.37 1 0 .59
Sexual Orientation 5.05 6 .54
Age 23.31 18 .18
Frequencies of Variables
Frequencies were calculated for each of the variables included
in the study (see Table 2). Eighteen (5.4%) subjects were classified
as bulimic, 24 (6 .8 %) subjects were classified as subclinical bulimic,
and 294 (82.8%) subjects were classified as nonbulimic according to
the following classification system: Participants were designated as
bulimic if their BULIT-R scores were 103 and above, as subclinical
bulimics if their BULIT-R scores were between 82 and 102, and as
nonbulimic if their BULIT-R scores were lower than 82. In terms
of ACA status, 19.8% (N = 67) of the respondents had CAST- 6
scores of three or higher, indicating that they were adult children of
alcoholics. The present sample reported incidence rates of 2.9% (N
= 10) for incest history, 3.8% (N = 13) for sexual abuse history, and
93.2% (N = 317) for no sexual abuse history. To determine
traumatic symptomatology in the present population, means and
73
standard deviations of subjects’ TSI scores were calculated (see Table
2).
Table 2
Frequencies of Variables
Variable % N
Bulimia
Nonbulimic 82.8% 294
Subclinical 6.8% 24
Bulimic 5.4% 18
Parental Alcoholism
Non-ACA 80.2% 271
ACA 19.8% 67
Sexual Abuse History
No Sexual Abuse 92.9% 317
Sexual Abuse 3.8% 13
Incest 2.9% 10
Traumatic Symptomatology
TSI Subscales Mean SD
Anxious Arousal 11.6 6.6
Depression 10.4 7.1
Anger/Irritability 11.2 7.0
Dissociation 8.1 6.0
Sexual Concerns 5.6 5.9
Dysf. Sexual Behavior 3.5 4.2
Intrusive Experiences 8.4 7.3
Defensive Avoidance 10.0 7.4
Impaired Self-Reference 10.0 6.8
Total TSI Score 77.7 48.2
74
In addition, descriptive statistics were calculated for each
variable included in the study according to bulimia status (bulimic,
subclinical bulimic, and nonbulimic) (see Table 3).
Table 3
Variable Bulimic
Subclinical
Bulimic Nonbulimic
Parental Alcoholism
Non-ACA (N=255) N=10; 3.9% N=15; 5.9% N=230; 90.1%
ACA (N=65) N=6; 9.2% N=7; 10.8% N=52; 80%
Sexual Abuse History
No Sexual Abuse (N=301) N=14; 21.5% N=24; 7.8% N=263; 87.4%
Sexual Abuse (N=12) N=3; 4% N=0 N=9; 75%
Incest (N=10) N=l; 10% N=0 N=9; 90%
Traumatic Symptomatology Mean(SD) Mean(SD) Mean(SD)
Anxious Arousal 19.22 (6.23) 16.36 (5.19) 10.85 (6.35)
Depression 22.17 (8.37) 13.91 (6.93) 9.40 (6.17)
Anger/Irritability 20.35 (7.34) 15.78 (7.28) 10.39 (6.48)
Dissociation 18.53 (7.45) 10.96 (5.67) 7.37 (5.27)
Sexual Concerns 12.78 (8.76) 6.81 (6.46) 5.13 (5.41)
Dysfunctional Sexual Behavior 8.28 (6.96) 4.57 (4.84) 3.13 (3.75)
Intrusive Experiences 16.17 (9.64) 12.82 (9.48) 7.76 (7.00)
Defensive Avoidance 19.06 (7.67) 13.52 (6.79) 9.13 (6.94)
Impaired Self-Reference 20.50 (6.44) 13.55 (5.57) 9.29 (6.33)
Total TSI Score 159.06 (53.62) 107.33 (42.67) 71.25 (43.08)
75
Analyses of Research Hypotheses
Hypothesis 1
Hypothesis 1, that subjects with higher rates of traumatic
symptomatology will have higher rates of bulimia than subjects with
lower rates of traumatic symptomatology, was confirmed.
A multiple regression analysis was performed using the nine
TSI sub scale scores as independent variables and BULIT-R scores as
the dependent variable. The results indicated that traumatic
symptomatology accounted for a significant amount of the change in
these subjects’ BULIT-R scores (F[9,294] = 15.23, p < .0005, R^ =
.32). However, of the nine TSI subscales, only Depression was
significant at the .05 level (beta = .296, p < .004) (see Table 4).
Table 4
Results of Multiple Regression Analysis of Traumatic
TSI Subscale b wt SE b Beta t Sig. t
Anxious Arousal .26 .30 .08 .86 ^ 92
Depression .92 .32 .30 Z88 .004
Anger/lrritability .04 .27 .01 .16 .874
Dissociation .57 .34 .16 1.71 .089
Sexual Concerns .42 .25 .12 1.68 .095
Dysf. Sexual Behavior .57 .36 .11 1.58 .116
Intrusive Experiences -.31 .26 -.11 -.118 .238
Defensive Avoidance .21 .29 .07 .71 A l l
Impaired Self-Reference -.24 .35 -.08 -.69 .489
Note: Multiple R = .564, r 2 = .318, Standard Error = 18.47
76
In addition, the trend of the bulimics’ traumatic
symptomatology levels was in the expected direction: the mean TSI
sub scale scores of bulimics and subclinical bulimics were higher than
their nonbulimic peers (see Table 3).
Hvpothesis 2
Hypothesis 2, that subjects with alcoholic parents will have
higher rates of traumatic symptomatology than subjects without
alcoholic parents, was partially confirmed.
A MANOVA was performed using parental alcoholism as the
independent variable and TSI subscale scores as the dependent
variables. Using Hotteling's F, the results indicated there was not an
overall significant difference between these adult children of
alcoholics and adult children of nonalcoholics on the measure of
traumatic symptomatology (F[ 1,302] = .037, p < .298).
A one way ANOVA was also performed using parental
alcoholism status (ACA, non-ACA) as the independent variable and
the total TSI scores as the dependent variable. There was a
significant difference between ACA status and their total levels of
traumatic symptomatology (F[I,302] = 5.48, p < 02).
In addition, the trend of these adult children of alcoholics’
traumatic symptomatology levels was in the expected direction: the
mean TSI subscale scores of ACAs were higher than their non-ACA
peers (see Table 5).
77
Table 5
Trauma Symptom Inventory Means and Standard Deviations
According to Parental Alcoholism Status
TSI Subscale ACAs
Mean (SD)
Non-ACAs
Mean (SD)
Anxious Arousal 13.0 (6.2) 11.0 (6.6)
Depression 11.8 (7.3) 9.9 (6.8)
Anger/Irritability 12.1 (6.5) 10.9 (7.0)
Dissociation 9.7 (6.4) 7.7 (5.8)
Sexual Concerns 7.0 (6.5) 5.3 (5.7)
Dysf. Sexual Behavior 4.3 (4.5) 3.3 (4.0)
Intrusive Experiences 10.1 (7.7) 7.9 (7.1)
Defensive Avoidance 11.3 (7.0) 9.6 (7.3)
Impaired Self-Reference 11.1(6.6) 9.8 (6.8)
Total TSI Score 90.0 (47.4) 74.2 (47.0)
Hypothesis 3
Hypothesis 3, that subjects with incest or sexual abuse histories
will have higher rates of traumatic symptomatology than subjects
without incest or sexual abuse histories, was partially confirmed.
Two MANOVAs were conducted using sexual abuse history
and incest history as the independent variables and TSI sub scale
scores as the dependent variables. The results indicated that there
was a significant overall effect for sexual abuse (F[ 1,298] = .074,
p < 01), but not for incest (F[1,296] = .036, p < 33), meaning
sexual abuse history was significantly associated with more traumatic
symptomatology, but incest history was not. For sexual abuse
78
history, each of the TSI subscales was significant at or below the .04
level except for Sexual Concerns (See Table 6).
Table 6
Results of Univariate F-Tests of Trauma Symptom Inventory
TSI Subscale F
P
Anxious Arousal 4.41 .370
Depression 7.70 .006
Anger/Irritability 9.44 .002
Dissociation 10.38 .001
Sexual Concerns 2.08 .150
Dysfunctional Sexual Behavior 5.81 .017
Intrusive Experiences 14.88 .0005
Defensive Avoidance 9.25 .003
Impaired Self-Reference 9.77 .002
Two one-way ANOVAs were also performed using sexual
abuse history and incest history as the independent variables and the
total TSI score as the dependent variable. There was a significant
difference between subjects with sexual abuse histories on their total
levels of traumatic symptomatology (F[ 1,298] = 11.71, p < .0007),
suggesting that sexually abused subjects had a higher overall level of
traumatic symptomatology than subjects without sexual abuse
histories. There was not a significant difference between subjects
with incest histories on their total levels of traumatic
symptomatology (F[1,296] = .644, p < .42), which suggests that
79
subjects with incest histories did not have a higher overall level of
traumatic symptomatology than subjects without incest or sexual
abuse histories.
Although there was no overall significant relationship between
subjects’ incest histories and traumatic symptomatology, the trend of
the subjects with incest histories TSI scores was in the expected
direction: The mean TSI sub scale scores of subjects with incest
histories were higher than those without sexual abuse histories,
except on the Anger/Irritability and Impaired Self-Reference
subscales (see Table 7).
Hypothesis 4
Hypothesis 4, that subjects with alcoholic parents will have
higher rates of bulimia than subjects with nonalcoholic parents, was
confirmed.
A one-way ANOVA was conducted using parental alcoholism,
measured by CAST-6 scores of 3 and above as ACA and CAST-6
scores of less than 3 as non-ACA, as the independent variable and
bulimia, measured by BULIT-R scores, as the dependent variable.
As hypothesized, these adult children of alcoholics were more likely
to be bulimic than these adult children of nonalcoholics (F[l,318] =
6.414, p < ,01).
80
Table 7
Trauma Symptom Inventory Means and Standard Deviations
TSI Subscale
No Sexual
Abuse
Mean (SD)
Sexual Abuse
Mean (SD)
Incest
Mean (SD)
Anxious Arousal 11.4 (6.6) 15.7 (8.3) 11.1 (6.7)
Depression 10.1 (6.9) 16.1 (9.2) 10.7(6.4)
Anger/Irritability 10.9 (6.7) 17.3 (8.1) 10.7 (7.6)
Dissociation 7.8 (5.7) 13.5 (9.5) 10.9 (9.2)
Sexual Concerns 5.6 (5.9) 8.2 (5.3) 6.9 (6.6)
Dysf. Sexual Behavior 3.4 (4.0) 6.4 (5.7) 5.0 (5.1)
Intrusive Experiences 8.0 (7.2) 16.5 (8.4) 10.7 (7.4)
Defensive Avoidance 9.5 (7.1) 16.2 (7.6) 12.9 (8.3)
Impaired Self-Reference 9.9 (6.7) 16.4 (7.7) 10.4 (9.1)
Total TSI 76.4 (46.8) 126.3 (62.3) 89.2 (55.8)
Means and standard deviations of BULIT-R scores according
to ACA status were calculated, and are presented in Table 8. As
expected, the mean BULIT-R scores of the ACAs are higher than
their non-ACA counterparts.
Table 8
Means and Standard Deviations of BULIT-R Scores
BULIT-R
Parental Alcoholism Status Mean (SD)
ACA 52.4 (20.8)
Non-ACA 60.0 (24.3)
81
Hvpothesis 5
Hypothesis 5, that subjects with incest or sexual abuse histories
will have higher rates of bulimia than subjects without incest or
sexual abuse histories, was partially confirmed.
One-way ANOVAs were conducted using sexual abuse history
and incest history as the independent variables and bulimia as the
dependent variable. There was a significant difference between
sexually abused and nonsexually abused individuals on the variable
bulimia (F[l,311] = 3.98, p < .05), but there was not a significant
difference between the subjects with incest histories and nonsexually
abused subjects on the variable bulimia (F[l,309] = .013, p < .91).
Thus, individuals who had been sexually abused outside of their
family had significantly higher rates of bulimia than control subjects,
but individuals who had been sexually abused within their families
did not have significantly higher rates of bulimia than control
subjects.
Means and standard deviations were calculated for BULIT-R
scores according to sexual abuse history, and are presented in Table
9. Again, contrary to this hypothesis, the mean BULIT-R score of
the subjects with incest histories is lower than subjects with no
history of sexual abuse. The percentage of bulimics who reported
incest histories in the present study (5.6%, N =l) was higher than the
nonbulimic subjects who reported incest histories (3.2%, N=9),
8 2
although this difference was not statistically significant (x^ = 1.14,
p<.57).
Table 9
Means and Standard Deviations of BULIT-R Scores
BULIT-R
Sexual Abuse History Mean (SD)
No Sexual Abuse 54.0 (21.8)
Sexual Abuse 66.9 (28.2)
Incest 53.2 (27.1)
Surhmary
Each of the research hypotheses studied was at least partially
confirmed. To summarize: (a) traumatic symptomatology was
significantly related to bulimia; (b) parental alcoholism was not
significantly related to individual traumatic symptomatology
domains, but ACA status was significantly related to total levels of
traumatic symptomatology; (c) sexual abuse history was significantly
associated with rates of traumatic symptomatology, but incest history
was not significantly associated with rates of traumatic
symptomatology; (d) parental alcoholism was significantly related to
bulimia; and (e) sexual abuse history was significantly related to
bulimia, but incest history was not significantly related to bulimia.
83
CH A PTER FO U R
D ISC U SSIO N
In this chapter, the results of the present study will be first
interpreted and discussed in terms of prior research, then, its
delimitations of the present study will be presented. Finally,
implications of these results will be discussed, and future directions
for research suggested.
Interpretation of Results
Overall, the present study found partial support for the
hypothesis that bulimia is associated with prior sexual abuse and
parental alcoholism as well as traumatic symptomatology. The
results of the study will now be discussed in order of hypothesis.
Hypothesis 1
The results of the first hypothesis support the idea that
traumatic symptomatology is related to bulimia. In this sample of
predominately Caucasian sorority women, significantly higher levels
of bulimia were associated with higher levels of traumatic
symptomatology. One conclusion that might be drawn from these
findings is that bulimia is a method of coping with posttraumatic
symptomatology. While this hypothesis has never been tested
directly, various authors have proposed such an association. For
example, researchers have discussed bingeing and purging as a
response enacted to cope with traumatic experiences such as
84
childhood sexual abuse (Briere, 1992; Herman, 1992; Root & Fallon,
1990). As discussed above, in chnical report bulimics often describe
distressing feelings including tension, anxiety, and derealization
preceding a binge-purge cycle (Abraham & Beaumont, 1982).
Given that these experiences (tension, anxiety, and derealization) are
subsumed under the domains assessed in the Trauma Symptom
Inventory, and that the women with bulimia in this study reported
more overall traumatic symptomatology, this clinical report is
supported.
The individuals with bulimia in this sample were experiencing
high levels of traumatic symptomatology both in comparison to their
nonbulimic counterparts as well as to other university females
(Smiljanich & Briere, 1993). This raises the question of how and
how well these women are coping with these disturbing levels of
traumatic symptomatology? Although there are many alternatives,
bulimia may indeed be a way to soothe, anesthetize, or reduce the
disturbing feelings and thoughts which these women report
experiencing. Interestingly, the only TSI subscale elevation for
bulimics was Depression; if bulimia is being used as a defense against
traumatic symptomatology as hypothesized, it appears to be
successfully warding off other traumatic sequelae. As discussed in
chapter one, depression is commonly found to co-occur with
bulimia. Thus, these findings are consistent with prior research and
support the present model as well. Although tension-reduction
85
behavior was not measured per se in the present study, it would
follow that these women may be using some means of tension-
reduction or soothing behavior to cope with their high levels of
traumatic symptomatology. As has been proffered in the
psychoanalytic literature, the bingeing and purging behavior of these
women may be regulating and alleviating intolerable inner states,
albeit temporarily. The bulimia may offer the traumatized
individual a way out of her painful internal state by externalizing the
conflict onto eating issues: Rather than experiencing the pain of
depression, for example, these women may turn to an obsession with
body image, dieting, bingeing and purging as a way to escape
difficult feelings. Bingeing and purging may serve as a comfort, a
distraction, or tension reducer; it detaches the traumatized woman
from her painful internal experience.
Hypothesis 2
The results of the second hypothesis lend support to the
conception that adult children of alcoholics experience higher levels
of traumatic symptomatology than their non-AC A peers. Indeed, in
this study the individuals with alcoholic parents have higher rates of
some trauma-related factors than those with nonalcoholic parents.
The specific traumatic symptoms which were significantly associated
with parental alcoholism in the present study are Depression,
Dissociation, Intrusive Experiences, and Defensive Avoidance.
Similar to the first hypothesis, the incidence of post-traumatic
8 6
symptomatology in adult children of alcoholics has to this author’s
knowledge never been studied. Thus, it is impossible to compare it
directly to prior research findings. The results of the present study
are not surprising, however, given the increased rates of other
negative psychological sequelae that have been associated with being
an adult child of an alcoholic in the research and clinical literature.
For example, consistent with the finding of increased scores on the
Depression subscale of the TSI among ACAs in this sample, a higher
incidence of depression has been found among female ACAs (for
example, Bulik, 1987a).
The present findings also support the clinical literature which
contends that families with an alcoholic parent are often the contexts
for emotional and psychological neglect. According to Black
(1990), for example, alcoholic families are characterized by chaos,
inconsistency, and fear. When a child cannot depend on the
predictability of a parent’s behavior or availability (emotionally
and/or physically) due to the effects of alcohol, this leaves the child
vulnerable to feelings of terror, abandonment, and anxiety. These
observations are consistent with the present findings that ACAs in
this sample have higher rates of Defensive Avoidance and Intrusive
Experiences, which include flashbacks and painful memories.
In addition, if a parent is alcoholic, it seems safe to assume
that this would put significant limitations on his or her ability to
provide an appropriate environment for the child. As proposed by
87
Winnicott (1965), it is necessary for a parent to provide a protective,
empathie “holding environment” to enable a child to self-soothe and
regulate his or her own affective states, including tension. When an
alcoholic parent is under the influence of a mind-altering substance,
his or her capacity to mirror the infant’s needs and provide
emotional nurturance and soothing when the infant experiences pain
or excessive tension is marred. Instead, the child is likely to
encounter an emotionally unavailable or self-preoccupied parent and
thus does not achieve the necessary criteria to develop his or her
own tension-reduction and soothing strategies. One way in which
these adult children of alcoholic parents may reduce internal tension
states and soothe themselves is through bulimic behavior (Humphrey
& Stem, 1988; see Hypothesis 4). Thus, the present finding that
children of alcoholic parents are at greater risk for some
posttraumatic symptomatology is compatible with past clinical and
research findings. Also, given that lower rates of negative
psychological symptomatology have been found among nonclinical
and university ACA samples (Kashubeck & Christensen, 1992), it
may be that the rates of posttraumatic symptomatology are in fact
greater than documented in the present study among clinical ACA
populations.
Hvpothesis 3
The third and fifth hypotheses must be interpreted with
caution given the small sample size in the sexual abuse group (N =
8 8
13) and incest group (N = 10). Yet, as hypothesized, the women in
the present study who reported sexual abuse histories evidenced
higher rates of traumatic symptomatology than individuals without
histories of sexual abuse. This finding is consistent with prior
research, which has documented a significant relationship between
childhood sexual abuse and TSI scores in university students
(Smiljanich & Briere, 1993). This finding lends further support to
the increasingly common understanding that childhood sexual abuse
is often a traumatic experience that has long-lasting effects on the
victim (for example, Beitchman et al., 1992). Some long-term
effects that have been cited as more prevalent in women with
histories of childhood sexual abuse include sexual dysfunction,
anxiety, depression, suicidal ideation and behavior and marital
disturbance. The present findings lend support to this body of
knowledge, and reveal that college sorority women are similarly
affected by childhood sexual abuse.
Contrary to the third hypothesis, though, individuals in this
study who reported incest histories did not have significantly higher
rates of traumatic symptomatology than their peers who reported no
sexual abuse, although their mean scores were higher on seven of
nine domains of traumatic symptomatology. This finding, that
women in this study who reported familial sexual abuse were not at
risk for significantly higher levels of traumatic symptomatology, is
contrary to the research literature on this topic. Many studies
89
examining traumatic effects of childhood sexual abuse have not made
the delineation between intrafamilial and extrafamilial sexual abuse
(e.g., Smiljanich & Briere, 1993), so the comparability of this
finding is limited. Some explanations for the present findings
include the following: Given that there were only 10 respondents
who reported incest histories, if a few of these respondents were
“asymptomatic” in terms of traumatic after-effects of their childhood
sexual abuse, the mean scores would be significantly lower for the
entire incest group. Also, the present study did not take into account
factors which have been shown to increase the risk for traumatic or
negative psychological sequelae associated with sexual abuse and
incest including (a) age at time of abuse, (b) duration of abuse,
(c) use of threat or force, (d) abuse involving penetration, (e)
internal attribution or self-blame, or (f) disclosure of the abuse
(Beitchman et al., 1992; Wyatt & Newcomb, 1990). In the present
study, incest was determined by the question, “Did this (sexual
abuse) ever happen with a family member?” with no specification of
the relationship of that family member. Research has shown that
abuse involving a father or stepfather is associated with greater long
term negative effects. It is plausible that the victims of incest in the
present study were abused by a family member that was not a father
or stepfather and thus had a lower incidence of traumatic
symptomatology than expected for incest survivors.
90
Hypothesis 4
The adult children of alcoholics in the present sample are
more likely to be bulimic than their non-AC A counterparts,
supporting the fourth hypothesis. These results are consistent with
much of the past research, which has found a higher incidence of
bulimia among AC As (for example, Bulik, 1987b), but contrary to
the findings of Mintz, et al., (in press), who likewise used a college
female sample. One explanation for the divergence from these latter
findings may be that the present study employed the BULIT-R,
which specifically looks at bulimic behavior, versus the Eating
Disorder Inventory, which looks at eating disorder related
characteristics versus symptomatology and was originally designed
to trace anorexia nervosa, not bulimia. Given that bulimia more
similarly resembles a tension-reduction cycle than anorexia, it would
follow that bulimia is more common among AC As than anorexia.
This is consistent with the divergence in the findings between the
present study and the Mintz et al. study. In addition, it has been
borne out in the literature that bulimia is more commonly associated
with parental alcoholism than anorexia (for example, Claydon,
1987). Earlier it was proposed that the disproportionate amount of
bulimia among ACAs could be explained by the following
hypotheses: (a) the alcoholic parent was unable to provide a
sufficient holding environment for the future bulimic and thus
created a deficit in the child’s ability to self-soothe and reduce
91
tension; (b) the bulimic uses the binge-purge cycle as a way to cope
with the traumatic and deficient experience of being a child of an
alcoholic parent; and (c) the alcoholic parent modeled substance
(alcohol) abuse as a way of coping with psychological distress and
thus the bulimic abuses food to do the same. The results of
Hypothesis 2 in the present study would give further credence to the
first proposed explanation for the coincidence of parental alcoholism
and bulimia: these ACAs are at greater risk for some traumatic
symptomatology than their non-AC A peers. Together, these two
findings imply that ACAs are traumatized and may use bulimia as a
way to reduce the painful feelings and thoughts associated with this
trauma.
Hypothesis 5
As predicted in the fifth hypothesis, the women in this study
who reported sexual abuse histories had significantly higher rates of
bulimia than the women who did not report such abuse. This finding
is consistent with most of the previous studies which compared rates
of eating disorders among individuals with and without sexual abuse
histories using university samples (Calam & Slade, 1989; Smolak,
Levine, & Sullins, 1990). Both Smolak and colleagues (1990) and
Calam and Slade (1989) found elevated rates of eating disturbances
among those subjects reporting sexual abuse histories. It is difficult
to compare these findings directly to the results of the present study
due to the differences in methods used to measure eating disturbance
92
and sexual abuse. For example, Calam and Slade (1989) used the
Eating Attitudes Test to measure eating problems; Smolak et al. used
the Eating Disorder Inventory to measure eating problems. Neither
looked specifically at bulimia, as was measured in the present study.
Both of these authors used different criteria to measure sexual abuse
as well: Calam and Slade (1989) used the Sexual Events
Questionnaire; Smolak and colleagues adapted questions used in
Finkelhor’s 1979 study of childhood sexual abuse (Finklehor, 1979;
in Smolak et al., 1990).
The results of the present study lend support to the theoretical
hypotheses linking sexual abuse and bulimia. For example, Root and
Fallon’s model of bulimia as a posttraumatic response to sexual abuse
would be supported by the results of both the third and fifth
hypotheses. This suggests that for these sexually abused women,
bulimia may “serve” them in the following ways: (a) act as an
analgesic for negative feelings associated with victimization; (b)
serve to repress the pain of sexual abuse through dissociation, a
posttraumatic response; (c) in the purging phase, may be an effort to
cleanse the sexual abuse victim from feelings of being “dirty” or
ashamed; (d) be a way to avoid or express symbolically, anger; (e)
justify that the sexually abused bulimic is worthless and deserved the
abuse; (f) establish physical and psychological space as a response to
the invasion of sexual abuse; (g) be an attempt to control the sexually
abused woman’s environment and body; (h) be a way in which the
93
sexual abuse victim projects her hatred and rage toward her
perpetrator on her own body; (i) be a way to feel consistency and
predictability in contrast to the often chaotic and powerlessness
associated with childhood sexual abuse; and finally, (j) be a way to
temporarily stave off tension states.
Contrary to the fifth hypothesis, the women in the present
study who reported childhood incest histories (intrafamilial sexual
abuse) did not have significantly higher rates of bulimia. This
finding conflicts with the results of Abramson and Lucido (1991)
who found that bulimics reported significantly more intrafamilial
sexual abuse than nonbulimics. One explanation for this divergence
may be that the bulimics in the Abramson and Lucido study were
predominately in some type of treatment and were a mixture of
students and non-students as compared to the exclusively college
population here. As has been found in other studies, the rate of co
occurrence between incest an eating disorders reported among
clinical populations is far higher than among nonclinical or
university populations (Mallincrodt et al., 1993). The results of the
present study more similarly match the findings of Beckman and
Bums (1990) and Calam and Slade (1989). Beckman and Bums
(1990) found that in their sample of 340 college females, bulimic
women reported significantly more sexual abuse after age 12 by a
non-relative than control subjects, but not significantly more incest
after the age of 12; however, the bulimic subjects did not differ
94
significantly in their reports of intrafamilial or extrafamilial sexual
abuse before age 12. The present study did not specify an age range
for when sexual abuse occurred, so the comparability of these
findings is somewhat limited. The study performed by Calam and
Slade (1989) similarly did not find that bulimia was associated with
intrafamilial sexual experience in their sample of undergraduate
females and bulimics in treatment, although bulimia was related to
sexual experience involving force. Finally, the percentage of
bulimics who reported incest histories in the present study was
higher than the nonbulimic women who reported incest histories,
although this difference was not statistically significant.
Similar to the lack of significantly higher trauma scores
among incest survivors in the present study, alternate reasons may
exist for the lack of significantly higher bulimia scores among incest
survivors. For example, the absence of support for increased
negative symptomatology among incest survivors may be attributed
to the low number of subjects in this sample (N = 10) and variables
not accounted for in this study (e.g., relationship to familial
perpetrator).
Additional Findings
The levels of bulimia found in the present sample (7.2%
subclinical bulimic, 5.4% bulimic) is consistent with some estimates
of rates of bulimia among college females (for example, Beckman
and Bums, 1990: 12.94% subclinical and bulimics), but higher than
95
others (for example, Mintz & Betz, 1988: 3% bulimic; Thelen et ah,
1987: 3.1% bulimic).
The rate of parental alcoholism found in the present sample
was 19.8% (N=67), which is generally consistent with those found
among similar college female populations (for example, Mintz et al.,
in press)
The incidence of sexual abuse (3.8%; N=13) and incest (2.9%;
N=10) reported by the women in this sample is far lower than the
commonly reported incidence rate of sexual victimization, which has
been estimated to occur at a rate closer to 38% for extrafamilial
sexual abuse and 16% for incest (Russell, 1983). This finding may
be explained in a few ways. First, it is possible that sorority women
are at a substantially lower risk for sexual abuse than the general
population. Second, it is possible that the questions used in the
present survey were not valid measures for childhood sexual abuse.
Third, it is also possible that a proportion of these women were
repressing memories of childhood sexual abuse. Estimates for
repression of memories of sexual abuse range from 19% to 59%
(Loftus, Polonsky, & Fullilove, 1994). Thus, if the lowest estimate
for repressed memories of childhood sexual abuse was applied to the
present sample, the prevalence rates would be closer to 22.8% for
sexual abuse and 21.9% for incest. Fourth, given the lower than
expected return rate, it is possible that the responders had a lower
incidence of sexual abuse than the non-responders ; in other words, it
96
is possible that the individuals who had sexual abuse histories
selected themselves out of the study at a greater rate than those
without sexual abuse histories.
The rates of traumatic symptomatology for the total sample
was consistent with similar university student samples (for example,
Smiljanich & Briere, 1993).
In the present study, the effect of parental alcoholism on
bulimia was approximately 62% larger than the effect of sexual
abuse on bulimia, implying that being an AC A is a greater risk
factor for bulimia than sexual abuse. This finding is consistent with
prior research, which has more unanimously found parental
alcoholism in the histories of bulimics than sexual abuse (for
example, Bulik, 1987b; Mallincrodt et al., 1993).
Limitations/Delimitations
Some delimitations of the present study warrant mention.
First, the study did not include an ethnically diverse sample. Thus,
the present study falls prey to a common problem in eating disorder
research and psychological research in general: it failed to include a
representative sample of ethnically diverse individuals, and does not
include ethnicity as an independent variable, as is recommended by
many cross-cultural researchers (for example, Ponterotto & Casas,
1991). By leaving ethnicity out of the present model, an important
variable was overlooked. As will be recommended later, these issues
should be addressed in future research in this area. Second, the
97
findings that college students generally have a lower level of
symptomatology concerning parental alcoholism (for example,
Mintz, et ah, in press), traumatic symptoms related to child sexual
abuse (for example, Elliott & Briere, 1993), yet a higher level of
bulimia than the population in general or clinical samples (for
example, Johnson & Connors, 1987), may have affected the present
findings. Third, sorority members are a generally homogeneous
group in terms of demographics: they tend to be from higher
socioeconomic backgrounds and are predominately Caucasian
(Wilder, Hoyt, Surbeck, Wilder, & Camey, 1986). In addition,
research has documented a significantly higher incidence of purging
after eating in sorority women (Meilman, von Hippel, & Gaylor,
1991). Fourth, although the present sample was assumed to be
nonclinical, there was no assessment of this factor; hence, there may
have been some “clinical” subjects in this nonclinical sample. Given
that clinical populations generally report more abuse-related
symptoms and greater severity of abuse, the rates of traumatic
symptoms and bulimia found in the present study may not, for
example, be representative of other clinical or non-college
populations. Fifth, all subjects were volunteers; thus, the responders
may have differed significantly from the non-responders in unknown
ways. For example, one way in which responders may have differed
from non-responders may be that subjects who completed and
returned the surveys may have done so because the questionnaires
98
did not require that they disclose any uncomfortable information nor
did they engender disturbing feelings because the experiences
inquired ahout were unfamiliar (hulimia, parental alcoholism, sexual
ahuse, traumatic symptomatology); on the other hand, those who
failed to respond may have hegun the survey and found it too
disturbing to continue when finding their experiences reflected in the
questions asked. Finally, the present study only addressed a few
elements of the known multifactorial model of bulimia; as reviewied
in chapter one, hulimia occurs in a sociocultural and biological
context as well as a familial and psychological one.
Implications of Findings
Theoretical Implications
Although there have heen a number of significant theoretical
formulations which address the etiology of bulimia, there has been
an acute paucity of quantitative studies to support these theorizations.
The present study attempted to operationalize some of the theoretical
concepts that have heen offered as explanations for hulimia. For
example, hulimia has heen hypothesized to he a response to and a
method of coping with traumatic effects of childhood abuse (Briere,
1992; Herman, 1992; Root & Fallon, 1989). This study attempted to
operationalize this model by examining rates of posttraumatic
symptomatology and incidence of sexual abuse in women who were
bulimic as well as by examining levels of traumatic symptomatology
in women who reported sexual abuse histories. Correspondingly,
99
another theoretical model on which the present study was based is
that deficient parental experiences are causal to bulimia, and that
bulimia is an attempt to alleviate intolerable underlying tension states
which results from these failed parental environments (Bruch, 1973;
Humphrey & Stem, 1988; Swift & Letven, 1984). Specifically, the
present study hypothesized that bulimia would be more common
among adult children of alcoholics because these parents would be
more likely to fail to meet the child’s emotional needs due to their
addiction to substances, which would leave the adult child prone to
maladaptive soothing and tension reduction behaviors, such as
bulimia. The way in which this study attempted to test such a model
was by examining the rates of parental alcoholism in bulimic women
as well as by examining traumatic symptomatology in adult children
of alcoholics. Thus, by operationalizing important theoretical
formulations of bulimia, the present study made significant
contributions to the understanding of bulimia, trauma, sexual abuse,
and substance abuse.
Extrapolating from these models, this author proposes the
theoretical model for the intergenerational transmission of bulimia
presented in Figure 1. This model depicts parental alcoholism and
sexual abuse as risk factors for bulimia. As shown, these factors are
hypothesized to be traumagenic experiences which, combined with
intervening biological, sociological, and psychological variables,
predispose an individual to bulimia.
100
O )
u
3
b X )
E
c 3
3
pq
'o
I
c /3
C/3
I
1
I
I
I
a
hH
hH
iiil
13 o
■11
1 1 1 ill I
s 1 d
3 ë 1 f -a
t-lll
D .
IÎIÎÎ
IlilillH
I lil t
C O
0 4 C O
(N
101
Clinical Implications
Some important clinical guidelines can be gleaned from the
results of the present study. First, when performing assessments
with clients, it would be important to assess for the co-occurrence
traumatic symptomatology, sexual abuse history, bulimia, and
parental alcoholism if individuals present with any one of these
issues. For example, when screening for membership of an adult
children of alcoholics group at a university counseling center, it
would be important to assess for presence of bulimia and traumatic
symptomatology as well.
Given the developmental deficits and traumatic experience of
bulimics, adult children of alcoholics, and sexual abuse survivors
proposed and differentially supported in the present model, long
term therapy may be the treatment of choice. Through an extended
nurturing therapeutic relationship, the client would be allowed to
develop healthy object relationships and learn healthy self-soothing
and tension reduction techniques. The therapeutic setting could
become an appropriate “holding environment” for the client to heal.
Equally important, clinicians need to respect the strengths of these
clients; As can be seen from the present sample, many of these
individuals have persevered through difficult, and often traumatic
experiences; their bulimia is a way to cope with these experiences.
The therapist can use these clients' survival skills as leverage in the
healing process. The strengths in these women are evident:
102
Although the bulimics in this sample are experiencing high levels of
trauma, they are attending college and are active members of a social
organization.
Evidence against the uniformity myth is also available from
the results of the present study: not all ACAs, bulimics, and sexual
abuse survivors are similarly troubled. There are many mediating
variables for which the present model did not account which might
influence the effects of parental alcoholism and childhood sexual
abuse. These differences are important to recognize in a clinical
setting so that all sexual abuse survivors, for example, are not
treated for posttraumatic stress disorder. A recent example in the
author’s clinical experience may help to illustrate this point: A client
came in with limited time (three months) for individual therapy
presenting with issues of low self-esteem and identity. Mid-way
through the treatment, the client revealed that she had some
memories of being sexually abused by an older male relative. Given
that the client had limited time available for treatment, and that her
level of functioning in work and social areas were high, this clinician
tailored treatment to meet the short-term needs of the client. As an
alternative to the longer-term treatment option suggested above,
other treatment modalities may be effective in treating clients with
bulimia, alcoholic parents, and sexual abuse history. For example,
topic-focused group therapy is often effective for each of these
103
populations (for example, Briere, 1991; Johnson & Connors, 1987;
Woititz, 1983)
Finally, for preventative purposes, counseling psychologists
could perform outreach and psycho-education to various at-risk
populations regarding the consequences of childhood sexual abuse,
parental alcoholism, and eating disorders.
Directions for Future Research
The present study performed a much needed empirical
examination of some theoretical models of bulimia. This research
could be extended in the following ways: The generalizability of the
present study was limited by its sample; while it yielded valuable
information for college sorority women and similar populations, it
would be important to test this model using other populations such as
clinical and general populations. Moreover, it would be valuable to
study more heterogeneous groups with various ethnic, sexual
orientation, and socioeconomic backgrounds.
Additional intervening or mediating variables might also be
included in future research. For example, physical and
psychological abuse often co-occur with sexual abuse and parental
alcoholism and would be important variables to include in future
research. Another variable not directly tested in the present study
was tension reduction behavior; since the implementation of this
study, a new scale has been added to the Traumatic Symptom
Inventory called Tension Reduction Behavior which might more
104
directly assess an area that this study attempted to examine
indirectly. Finally, a hypothesis on which the present study relied
was that of insufficient holding environments and problematic object
relationships; this hypothesis might be elucidated further by
including variables of attachment and family environment in future
research.
An ideal study of bulimia as a response to trauma associated
with prior sexual abuse and parental alcoholism would include
qualitative as well as quantitative elements. A longitudinal study of
families who are evaluated before sexual abuse, alcoholism and
bulimia occurs would be particularly effective. These families could
be followed over a period of time and compared to examine the
differences within and between the families who express these
disorders. Advanced statistical methods such as path analysis might
be effectively employed to accommodate the complex phenomena
examined.
It is vital that counseling psychologists continue to study the
problem of eating disorders. According to the American Anorexia
and Bulimia Association, 150,000 American women die of anorexia
nervosa each year, which translates to one woman dying every four
minutes from an eating disorder (Wolf, 1991). The present study
focused on important familial and traumagenic factors of the eating
disorder bulimia; it will be important to continue to do research on
105
all contributors to this problem, including biological, sociological,
and psychological factors.
106
R eferences
Abraham, S., & Beaumont, P. J. V. (1982). How patients describe
bulimia or binge eating. Psychological Medicine. 12, 625-635.
Abramson, E. E., & Lucido, G. M. (1991). Childhood sexual
experience and bulimia. Addictive Behaviors. 16. 529-532.
Allen, M. G. (1976). Twin studies of affective illness. Archives of
General Psychiatry. 33. 1476-1478.
American Psychiatric Association (1987). Diagnostic and Statistical
Manual of Mental Disorders (3rd edition, revised). American
Psychiatric Association, Washington, D. C.: Author.
Beary, M. D., Lacey, J. H., & Merry, J. (1986). Alcoholism and
eating disorders in women of fertile age. British Journal of
Addiction. 81151. 685-689.
Beckman, K. A., & Bums, G. L. (1990). Relation of sexual abuse
and bulimia in college women. International Journal of Eating
Disorders. 915). 487-492.
Beitchman, J. H., Zucker, K. J., Hood, J. E., DaCosta, G. A.,
Akman, D., & Cassavia, E. (1992). A review of the long-term
effects of child sexual abuse. Child Abuse and Neglect. 16. 101-
118.
Berkowitz, A., & Perkins, H. W. (1988). Personality characteristics
of children of alcoholics. Journal of Consulting and Clinical
Psychology. 5612). 206-209.
Black, C. (1990). Double Duty. New York: Ballantine Books.
Boskind-Lodahl, M. (1976). Cinderella's stepsisters: A feminist
perspective on anorexia nervosa and bulimia. Signs: Journal of
Women in Culture and Society. 212). 342-356.
107
Briere, J. (1989). Therapy for Adults Molested as Children:
Beyond Survival. New York: Springer Publishing Company.
Briere, J. (1992a). Child Abuse Trauma: Theory and Treatment of
the Lasting Effects. Newbury Park, CA: Sage Publications.
Briere, J. (1992b). Methodological issues in the study of sexual
abuse effects. Journal of Consulting and Clinical Psychology.
6012). 196-203.
Briere, J. Elliott, D. M., Harris, K., & Cotman, A. (1993, August).
The Trauma Symptom Inventory: Trauma history, gender, and
clinical status. Paper presented at the 101st meeting of the
American Psychological Association, Toronto, Ontario, Canada.
Briere, J., Elliott, D. M., & Smiljanich, K. (in press). The Trauma
Symptom Inventory: Reliability and validity in clinical and
nonclinical groups.
Briere, J., & Runtz, M. (1989). The Trauma Symptom Checklist
(TSC-33): Early data on a new scale. Journal of Interpersonal
Violence. 4. 151-163.
Briere, J., & Runtz, M. (in press). Childhood sexual abuse: Long
term sequelae and implications for psychological assessment.
Brisman, J., & Seigel, M. (1984). Bulimia and alcoholism: Two
sides of the same coin? Journal of Substance Abuse Treatment.
112). 113-118.
Broderick, C. B. (1991). From Generation to Generation.
Bruch, H. (1973). Eating Disorders: Anorexia Nervosa. Obesitv.
and the Person Within. New York: Basic Books.
Bulik, C. M. (1987a). Alcohol use and depression in women with
bulimia. American Journal of Drug and Alcohol Abuse. 1313).
343-355.
108
Bulik, C. M. (1987b). Drug and alcohol use by bulimic women and
their families. American Journal of Psychiatry. 144112). 1604-
1606.
Bulik, C. M., Sullivan, P. F., & Rorty, M. (1989). Childhood sexual
abuse in women with bulimia. Journal of Clinical Psychiatry.
50(12), 460-464.
Calam, R. M., & Slade, P. D. (1989). Sexual experience and eating
problems in female undergraduates. International Journal of
Eating Disorders. 814). 391-397.
Carter, J. A., lyiboz, Y., & Greenockle, K. M. (1986). Bulimia:
Some physiological considerations. Journal of American College
Health. 35(3), 139-140.
Claydon, P. (1987). Self-reported alcohol, drug, and eating disorder
problems among male and female collegiate children of
alcoholics. Journal of American College Health. 36, 111-116.
Courtois, C. A. (1988). Healing the Incest Wound: Adult Survivors
in Therapy. Ontario, CA: Penguin Books Canada Ltd.
Crabbe, J. C., McSeigan, J. D., & Belknap, J. K. (1985). The role of
genetics in substance abuse. In Galizio, M., & Maisto, S. A.
Determinants of substance abuse: Biological, psychological, and
environmental factors. New York: Plenum.
Crowther, J. H. , Wolf, E. M., & Sherwood, N. E. (1992).
Epidemiology of bulimia nervosa. In J. H. Crowther, S. E.
Hobfoll, M. A. P. Stephens, & D. L. Tennebaum (Eds.), The
Etiology of Bulimia Nervosa: The Individual and Familial
Context (pp. 1-26). Washington, D. C.: Hemisphere Publishing
Corporation.
Davison, G. C., & Neale, J. M. (1990). Abnormal Psychology.
New York: Wiley.
109
Elliott, D. M., & Briere, J. (1993, August). Childhood
maltreatment, later re victimization, and adult symptomotology:
A causal analysis. Paper presented at the 101st annual meeting of
the American Psychological Association, Toronto, Canada.
Fava, M., Copeland, P. M., Schweiger, U., & Herzog, D. B. (1989).
Neurochemical abnormalities of anorexia nervosa and bulimia
nervosa. American Journal of Psychiatry. 146(8), 963-971.
Filstead, W. J., Parrella, D. P., & Ebbitt, J. (1988). High risk
situations for engaging in substance abuse and binge-eating
behaviors. Journal of Studies on Alcohol. 49(2). 136-141.
Finkelhor, D. (1979). Sexually Victimized Children. New York:
Collier Macmillan.
Finkelhor, D., & Browne, A. (1985). The traumatic impact of child
sexual abuse: A conceptualization. American Journal of
Orthopsychiatry. 55(4), 530-541.
Finn, S. E., Hartman, M., Leon, G. R, & Lawson, L. (1986). Eating
disorders and sexual abuse: Lack of confirmation for a clinical
hypothesis. International Journal of Eating Disorders. 516).
1051-1060.
Garfinkel, P., Gamer, D., Rose, J., Darby, P., Brandes, J. S.,
O’Hanlon, J., & Walsh, N. (1983). A comparison of
characteristics in the families of patients with anorexia nervosa
and normal controls. Psychological Medicine. 13. 821-828.
Gamer, D. M., Olmsted, M. P., & Polivy, J. (1983). Development
and validation of a multidimensional eating disorder inventory
for anorexia nervosa and bulimia. International Joumal of
Eating Disorders. 2, 15-34.
110
Gartner, A. F., Marcus, R. N., Halmi, K., & Loranger, A. W.
(1989). DSM-III-R personality disorders in patients with eating
disorders. American Journal of Psychiatry. 146. 1585-1591.
Goldfarb, L. A. (1987). Sexual abuse antecedent to anorexia
nervosa, bulimia, and compulsive overeating: Three case reports.
International Journal of Eating Disorders. 615). 675-680.
Goodwin, D. W. (1979). Alcoholism and heredity: A review and
hypothesis. Archives of General Psychiatry. 36. 57-61.
Goodwin, D. W., Schulsinger, F., Knop, J., Mednick, S., & Guze, S.
B. (1977a). Alcoholism and depression in adopted-out daughters
of alcoholics. Archives of General Psvchiatrv. 34. 751-755.
Goodwin, D. W., Schulsinger, F., Knop, J., Mednick, S., & Guze,
S. B. (1977b). Psychopathology in adopted and nonadopted
daughters of alcoholics. Archives of General Psychiatry. 34.
1005-1009.
Gordon, M. (1989). The family environment of sexual abuse: A
comparison of natal and stepfather abuse. Child Abuse and
Neglect. 13. 121-130.
Greenfield, S. F., Swartz, M. S., Landerman, L. R., & George, L.
K. (1993). Long-term psychosocial effects of childhood exposure
to parental problem drinking. American Joumal of Psychiatry.
15014). 608-614.
Hall, R. C. W., Tice, L., Beresford, T. P., Wooley, B., & Hall, A.
K. (1989). Sexual abuse in patients with anorexia nervosa and
bulimia. Psychosomatics. 3011). 73-79.
Hatsukami, D., Eckert, E., Mitchell, J. E., & Pyle, R. (1984).
Affective disorder and substance abuse in women with bulimia.
Psychological Medicine. 14. 701-704.
I l l
Hatsukami, D., Mitchell, J. E., Eckert, E. D., & Pyle, R. (1986).
Characteristics of patients with bulimia only, bulimia with
affective disorder, and bulimia with substance abuse problems.
Addictive Behaviors. 11. 399-406.
Hatsukami, D., Owen, P., Pyle, R., & Mitchell, J. (1982).
Similarities and differences on the MMPI between women with
bulimia and women with alcohol or drug abuse problems.
Addictive Behaviors. 7, 435-439.
Herman, J. L. (1992). Trauma and Recovery. New York: Basic
Books.
Hinson, R. C., Becker, L. S., Handal, P. J., & Katz, B. M. (1993).
The heterogeneity of children of alcoholics: Emotional needs and
help-seeking propensity. Joumal of College Student
Development. 34. 47-52.
Hodgins, D. C., Maticka-Tyndale, E., El-Guebaly, N., & West, M.
(1993). The CAST-6: Development of a short-form of the
Children of Alcoholics Screening Test. Addictive Behaviors. 18.
337-345.
Hudson, J. I., Pope, H. G., Jonas, J. M., & Yurgelun-Todd, D.
(1983a). Family history study of anorexia nervosa and bulimia.
British Joumal of Psychiatry. 142. 133-138,
Hudson, J. I., Pope, H. G., Jonas, J. M., & Yurgelun-Todd, D.
(1983b). Phénoménologie relationship of eating disorders to
major affective disorder. Psychiatry Research. 9, 345-354.
Hudson, J. I., Laffer, P. S., & Pope, H. G. (1982). Bulimia related
to affective disorder by family history and response to the
dexamethasone suppression test. American Journal of Psychiatry.
139(5 ). 685-687.
112
Humphrey, L. L., & Stem, S. (1988). Object relations and the
family system in bulimia: A theoretical integration. Joumal of
Marital and Family Therapy. 14(4). 377-380.
Humphrey, L. L. (1986). Family relations in bulimic anorexic and
nondistressed families. International Joumal of Eating Disorders.
5, 223-232.
Johnson, C., & Connors, M. E. (1987). The Etiology and Treatment
of Bulimia Nervosa: A Biopsychosocial Perspective. New York:
Basic Books, Inc.
Johnson, C. L., & Larson, R. (1982). Bulimia: An analysis of
moods and behavior. Psychosomatic Medicine. 44. 333-345.
Jones, D. A., Cheshire, N., & Moorhouse, H. (1985). Anorexia
nervosa, bulimia and alcoholism: Association of eating disorders
and alcohol. Joumal of Psychiatric Research. 19(2-3). 377-380.
Kashubeck, S., & Christensen, S. A. (1992). Differences in distress
among adult children of alcoholics. Joumal of Counseling
Psychology. 39(3). 356-362.
Kasset, J. A., Gershon, E. S., Maxwell, M. E., & Gurroff, J. J.
(1989). Psychiatric disorders in the first-degree relatives of
probands with bulimia nervosa. American Joumal of Psychiatry.
146(11). 1468-1471.
Killen, J. D., Taylor, C. B., Telch, M. J., Robinson, T. N., Maron,
D. J., & Saylor, K. E. (1987). Depressive symptoms &
substance use among adolescent binge eaters and purgers: A
defined population study. American Journal of Public Health.
77(12). 1539-1541.
Kog, E., & Vandereycken, W. (1985). Family characteristics of
anorexia nervosa and bulimia: A review of the research
literature. Clinical Psychology Review. 5(2). 159-180.
113
Kog, E., & Vandereycken, W. (1989). Family interaction in eating
disorder patients and normal controls. International Journal of
Eating Disorders. 8. 11-23.
Lacey, J. H., & Moureli, E. (1986). Bulimic alcoholics: Some
features of a clinical sub-group. British Journal of Addiction.
81(3). 389-393.
Loftus, E. F., Polonsky, S., & Fullilove, M. T. (1994). Memories of
childhood sexual abuse: Remembering and repressing.
Psychology of Women Quarterly. 18. 67-84.
Mallinckrodt, B., McCreary, B. A., & Robertson, A. K. (1993,
August). Co-occurrence of eating disorders and incest: The role
of attachment, family environment, and social competencies.
Paper presented at the 101st meeting of the American
Psychological Association, Toronto, Ontario, Canada.
Maxwell, S. (1985). Review of Children of Alcoholics Screening
Test. In J. V. Mitchell, Jr. (Ed.), The ninth mental measurements
yearbook (Vol. 1, pp. 307-308). Lincoln: University of
Nebraska Press.
Maybury, S. M., & Cameron, M. S. (1986). The role sexual trauma
plays in the development of eating disorders. Unpublished paper.
Meilman, P. W., von Hippel, F. A., & Gaylor, M. S. (1991). Self
induced vomiting in college women: Its relation to eating,
alcohol use, and Greek life. Joumal of American College Health.
40, 39-41.
Miller, N. S., & Gold, M. S. (1989). Suggestions for changes in
DSM III-R criteria for substance use disorders. American
Joumal of Dmg and Alcohol Abuse. 15. 223-230.
Mintz, L. B., & Betz, N. E. (1988). Prevalence and correlates of
eating disordered behaviors among undergraduate women. _
Joumal of Counseling Psychology. 35(4). 463-471.
114
Mintz, L. B., Kashubeck, S., & Tracy, L. S. (in press). Relations
between parental alcoholism, eating disorders, and substance
abuse in nonclinical college women: Additional evidence against
the uniformity myth. Journal of Counseling Psychology.
Mitchell, J. E., Hatsukami, D., Pyle, R., & Eckert, E. (1988).
Bulimia with and without a family history of drug abuse.
Addictive Behaviors. 13(3). 245-251.
Mitchell, P. B. (1988). The pharmacological management of
bulimia nervosa: A critical review. International Journal of
Eating Disorders. 7(1). 29-41.
Morley, J., & Blundell, J. E. (1988). The neurobiological basis of
eating disorders: Some formulations. Biological Psychiatry.
23(1). 53-78.
O ’Halloran, M. S., Bostwick, T., & Jacobs, L. (in press). Critical
issues in women’s health: Sexual assault and eating disorders.
Oppenheimer, R., Howells, K., Palmer, R. L., & Chaloner, D. A.
(1985). Adverse sexual experience in childhood and clinical
eating disorders: A preliminary description. Journal of
Psychiatric Residency. 19(2/3). 357-361.
Orbach, S. (1978). Fat is a Feminist Issue. New York: Berkley
Books.
Palmer, R. L., Oppenheimer, A. D., Chaloner, D. A., & Howells, K.
(1990). Childhood sexual experiences with adults reported by
women with eating disorders: An extended series. British
Journal of Psychiatry. 156. 699-703.
Ponterotto, J. G., & Casas, J. M. (1991). Handbook of Racial/
Ethnic Minority Counseling Research. Springfield, IL: Charles
C. Thomas.
115
Pope, H. G., & Hudson, J. L. (1986). Antidepressant drug therapy
for bulimia: Current status. Journal of ClinicaLPsychiatry, 41,
339-345.
Pope, H. G., & Hudson, J. L. (1992). Is childhood sexual abuse a
risk factor for bulimia nervosa? American Journal of Psychiatry.
149, 455-463.
Pope, H. G., Hudson, J. I., Yurgelun-Todd, D, & Hudson, M.S.
(1984). Prevalence of anorexia nervosa and bulimia in three
student populations. International JoumaLof Eating Disorders, 3,
45-51.
Pyle, R. L., Mitchell, J. E., & Eckert, E. D. (1981). Bulimia: A
report on 34 cases. Journal of Clinical Psychiatry, 42(2), 60-64.
Roberto, L. G. (1986). Bulimia: The transgenerational view.
Journal of Marital and Family Therapy, 12(3), 231-240.
Roosa, M. W., Sandler, I. N., Gehring, M. Beals, J., & Capp, L.
(1988). The Children of Alcoholics Life-Events Schedule: A
stress scale for children of alcoholic-abusing parents. Journal of
Studies on Alcohol, 49, 422-429.
Root, M. P., & Fallon, P. (1989). Treating the victimized bulimic.
Journal of Interpersonal Violence, 4(1), 90-100.
Russell, G. (1979). Bulimia nervosa: An ominous variant of
anorexia nervosa. Psychological Medicine, 9, 429-448.
Russell, D. (1983). Incidence and prevalence of intrafamilial sexual
abuse of female children. Child Abuse and Neglect, 7, 133-146.
Russell, M., Henderson, C., & Blume, S. B. (1985). Children of
alcoholics: A review of the literature. Buffalo: New York State
Division of Alcoholism and Alcohol Abuse, Research Institute on
Alcoholism.
116
Schwartz, D. M., Thompson, M. G., & Johnson, C. L. (1982).
Anorexia nervosa and bulimia: The socio-cultural context.
International Journal of Eating Disorders. 1(31. 20-36.
Schwartz, R. C., & Grace, P. (1989). The systemic treatment of
bulimia. Journal of Psychotherapy and the Family. 613-41. 89-
105.
Sloan, G., & Leichner, P. (1986). Is there a relationship between
sexual abuse or incest and eating disorders? Canadian Journal of
Psychiatry. 31. 656-660.
Smiljanich, K., & Briere, J. (1993, August). Child abuse history
and trauma symptoms in a university sample. Paper presented at
the 101st meeting of the American Psychological Association,
Toronto, Ontario, Canada.
Smith, M. C., & Thelen, M. H. (1984). Development and validation
of a test for bulimia. Journal of Consulting and Clinical
Psychology. 52. 863-872.
Smolak, L., Levine, M. P., & Sullins, E. (1990). Are child sexual
experiences related to eating-disordered attitudes and behaviors in
a college sample. International Journal of Eating Disorders. 912).
167-178.
Steiger, H., & Zanko, M. (1990). Sexual traumata among eating-
disordered, psychiatric, and normal female groups: Comparison
of prevalences and defense styles. Journal of Interpersonal
Violence. 5(1). 74-86.
Stem, S. S., Dixon, K. N., Nemzer, E., Lake, M., Sansone, R. A.,
Smeltzer, D. J., Lantz, S., & Schrier, S. S. (1984). Affective
disorder in the families of normal weight bulimics. American
Journal of Psychiatry. 141. 1224-1227.
117
Strober, M. (1981). The significance of bulimia in juvenile anorexia
nervosa: An exploration of possible etiologic factors.
International Journal of Eating Disorders. 1(1). 28-43.
Strober, M., Salkin, B., Burroughs, J., & Morrell, W. (1982).
Validity of the bulimia-restrictor distinction in anorexia nervosa:
Parental personality characteristics and family psychiatric
morbidity. Journal of Nervous and Mental Disease. 170. 345-
351.
Strober, M., & Humphrey, L. L. (1987). Familial contributions to
the etiology and course of anorexia nervosa and bulimia. Journal
of Consulting and Clinical Psychology. 55(5). 654-659.
Sugarman, A., & Kurash, C. (1982). The body as a transitional
object in bulimia. International Journal of Eating Disorders.
1(4). 57-67.
Swift, W. J., & Letven, R. (1984). Bulimia and the basic fault: A
psychoanalytic interpretation of the binge-vomiting syndrome.
Journal of the American Academy of Child Psychiatry. 23. 489-
497.
Thelen, M. H., Farmer, J., Wonderlich, S., & Smith, M. (1991). A
revision of the bulimia test: The BULIT-R. A Journal of
Consulting and Clinical Psychology. 3(1). 119-124.
Thelen, M. H., Kanakis, D. M., Farmer, J., & Pruitt, J. (1993).
Bulimia and interpersonal relationships: An extension of a
longitudinal study. Addictive Behaviors. 18. 145-150.
Thelen, M. H., Mann, L. M., Pruitt, J., & Smith, M. (1987).
Bulimia: Prevalence and component factors in college women.
Journal of Psychosomatic Research. 31(1). 73-78.
Viaro, M. (1990). Session-intersession sequences in the treatment of
chronic anorexic-bulimic patients: Following the model of
“family games”. Family Process. 29. 289-296.
118
Welch, G., & Hall, A. (1989). The reliability and discriminant
validity of three potential measures of bulimic behaviors. Journal
of Psychiatric Residencv. 23(2). 125-133.
West, M. O., & Prinz, R. J. (1987). Parental alcoholism and
childhood psychopathology. Psychological Bulletin. 102. 204-
218.
Wilder, D. H., Hoyt, A. E., Surbeck, B. S., Wilder, J. C., & Camey,
P. I. (1986). Greek affiliation and attitude change in college
students. Journal of College Student Personnel. 32 510-519.
Winnicott, D. W. (1965). The maturational processes and the
facilitating environment. New York: International Universities
Press.
Winokur, G. (1979). Unipolar depression. Archives of General
Psychiatry. 36. 47-52.
Woititz, J. G. (1983). Adult Children of Alcoholics. Florida:
Health Communications, Inc.
Wolf, N. (1991). The Beauty Myth: How Images of Beauty Are
Used against Women. New York: William Morrow and Co.,
Inc.
Wonderlich, S. A. (1992). Relationship of family and personality
factors in bulimia. In J. H. Crowther, S. E. Hobfoll, M. A. P.
Stephens, & D. L. Tennebaum (Eds.), The Etiologv of Bulimia
Nervosa: The Individual and Familial Context (pp. 103-120).
Washington D. C.: Hemisphere Publishing Corporation.
Wonderlich, S., Donaldson, M. A., Carson, D. K., Staton, D., Gertz,
L., Leach, L., & Johnson, M. (in press). Eating disturbance and
incest.
119
Wyatt, G. E., & Newcomb, M. (1990). Internal and external
mediators of women’s sexual abuse in childhood. Journal of
Consulting and Clinical Psychology. 58(6). 759-767.
Zweben, J. E. (1987). Eating disorders and substance abuse.
Journal of Psychoactive Drugs. 19(2). 181-192.
120
Appendix A
The Bulimia Test-Revised (BULIT-R)
Thelen, Farmer, Wonderlich, and Smith, 1991
Answer each question by circling the appropriate number. Please respond to
each item as honestly as possible; remember all of the inform ation you provide
w ill be kept strictly confidential.
1. I am satisfied with my eating patterns.
1. agree
2. neutral
3. disagree a little
4. disagree
5. disagree strongly
2. Would you presently call yourself a “binge eater”?
1. yes, absolutely
2. yes
3. yes, probably
4. yes, possibly
5. no, probably not
3. Do you feel you have control over the amount of food you consume?
1. most or all of the time
2. a lot of the time
3. occasionally
4. rarely
5. never
4. I am satisfied with the shape and size of my body.
1. frequently or always
2. sometimes
3. occasionally
4. rarely
5. seldom or never
5. When I feel that my eating behavior is out of control, I try to take rather extreme measures to
get back on course (strict dieting, fasting, laxatives, diuretics, self-induced vomiting, or vigorous
exercise).
1. always
2. almost always
3. frequently
4. sometimes
5. never or my eating behavior is never out of control
121
6. I use laxatives or suppositories to help control my weight.
1. once a day or more
2. 3-6 times a week
3. once or twice a week
4. 2-3 times a month
5. once a month or less (or never)
7. I am obsessed about the size and shape of my body.
1. always
2. almost always
3. frequently
4. sometimes
5. seldom or never
8. There are times when I rapidly eat a very large amount of food.
1. more than twice a week
2. twice a week
3. once a week
4. 2-3 times a month
5. once a month or less (or never)
9. How long have you been binge eating (eating uncontrollably to the point of stuffing yourself)?
1. not applicable; I don’t binge eat
2. less than 3 months
3. 3 months - 1 year
4. 1-3 years
5. 3 or more years
10. Most people I know would be amazed if they knew how much food I can consume at one
sitting.
1. without a doubt
2. very probably
3. probably
4. possibly
5. no
11. I exercise in order to burn calories.
1. more than 2 hours per day
2. about 2 hours per day
3. more than 1 but less than 2 hours per day
4. one hour or less per day
5. I exercise but not to burn calories or I don’t exercise
12. Compared with women your age, how preoccupied are you about your weight and body
shape?
1. a great deal more than average
2. much more than average
3. more than average
4. a little more than average
5. average or less than average
122
13. I am afraid to eat anything for fear that I won’t be able to stop.
1. always
2. almost always
3. frequently
4. sometimes
5. seldom or never
14. I feel tormented by the idea that I am fat or might gain weight.
1. always
2. almost always
3. frequently
4. sometimes
5. seldom or never
15. How often do you intentionally vomit after eating?
1. 2 or more times a week
2. once a week
3. 2-3 times a month
4. once a month
5. less than once a month or never
16. I eat a lot of food when I’m not even hungry.
1. very frequently
2. frequently
3. occasionally
4. sometimes
5. seldom or never
17. My eating patterns are different from the eating patterns of most people.
1. always
2. almost always
3. frequently
4. sometimes
5. seldom or never
18. After I binge eat I turn to one of several strict methods to try to keep from gaining weight
(vigorous exercise, strict dieting, fasting, self-induced vomiting, laxatives, or diuretics).
1. never or I don’t binge eat
2. rarely
3. occasionally
4. a lot of the time
5. most or all of the time
19. I have tried to lose weight by fasting or going on strict diets.
1. not in the past year
2. once in the past year
3. 2-3 times in the past year
4. 4-5 times in the past year
5. more than 5 times in the past year
123
20. I exercise vigorously and for long periods of time in order to burn calories.
1. average or less than average
2. a little more than average
3. more than average
4. much more than average
5. a great deal more than average
21. When engaged in an eating binge, I tend to eat foods that are tiigh in carbohydrates (sweets
and starches).
1. always
2. almost always
3. frequently
4. sometimes
5. seldom, or I don’t binge
22. Compared to most people, my ability to control my eating behavior seems to be:
1. greater than others’ ability
2. about the same
3. less
4. much less
5. I have absolutely no control
23. I would presently label myself a “compulsive eater” (one who engages in episodes of
uncontrolled eating).
1. absolutely
2. yes
3. yes, probably
4. yes, possibly
5. no, probably not
24. I hate the way my body looks after I eat too much.
1. seldom or never
2. sometimes
3. frequently
4. almost always
5. always
25. When I am trying to keep from gaining weight, I feel that I have to resort to vigorous
exercise, strict dieting, fasting, self-induced vomiting, laxatives, or diuretics.
1. never
2. rarely
3. occasionally
4. a lot of the time
5. most or all of the time
26. Do you believe that it is easier for you to vomit than it is for most people?
1. yes, it’s no problem at all for me
2. yes, it’s easier
3. yes, it’s a little easier
4. about the same
5. no, it’s less.easy
124
27. I use diuretics (water pills) to help control my weight.
1. never
2. seldom
3. sometimes
4. frequently
5. very frequently
28. I feel that food controls my life.
1. always
2. almost always
3. frequently
4. sometimes
5. seldom or never
29. I try to control my weight by eating little or no food for a day or longer.
1. never
2. seldom
3. sometimes
4. frequently
5. very frequently
30. When consuming a large quantity of food, at what rate of speed do you usually eat?
1. more rapidly than most people have ever eaten in their lives
2. a lot more rapidly than most people
3. a little more rapidly than most people
4. about the same rate as most people
5. more slowly than most people (or not applicable)
31. I use laxatives or suppositories to help control my weight.
1. never
2. seldom
3. sometimes
4. frequently
5. very frequently
32. Right after I binge eat I feel:
1. so fat and bloated I can’t stand it
2. extremely fat
3. fat
4. a little fat
5. OK about how my body looks or I never binge eat
33. Compared to other people of my sex, my ability to always feel in control of how much I eat
is:
1. about the same or greater
2. a little less
3. less
4. much less
5. a great deal less
125
34. In the last 3 months, on the average how often did you binge eat (eat uncontrollably to the point of stuffing
yourself)?
1. once a month or less (or never)
2. 2-3 times a month
3. once a week
4. twice a week
5. more than twice a week
35. Most people I know would be surprised at how fat I look after I eat a lot of food.
1. yes, definitely
2. yes
3. yes, probably
4. yes, possibly
5. no, probably not or I never eat a lot of food
36. I use diuretics (water pills) to help control my weight.
1. 3 times a week or more
2. once or twice a week
3. 2-3 times a month
4. once a month
5. never
126
Appendix B
The Children of Alcoholics Screening Test-6 (CAST-6)
Hodgins, Maticka-Tyndale, El-Guebaly, and West, 1993
Instructions: Please check the answer below that best describes your feelings,
behavior, and experiences related to a parent’s alcohol use. Take your time and be
as accurate as possible. Answer all 6 questions by checking either “Yes” or “No”.
YES NO QUESTIONS
___ 1. Have you ever thought that one of your parents had a drinking problem?
___ 2. Did you ever encourage one of your parents to quit drinking?
___ ___ 3. Did you ever argue or fight with a parent when he or she was drinking?
^ ___ 4. Have you ever heard your parents fight when one of them was drunk?
___ ___ 5. Did you ever feel like hiding or emptying a parent’s bottle of liquor?
___ 6. Did you ever wish your parent would stop drinking?
127
Appendix C
Sexual Abuse Questions
1. Before age 17, did anyone 5 or more years older than you ever have any sexual
contact with you? (Sexual contact includes kissing you in a sexual way, touching
your body in a sexual way, making you touch their sexual parts, inserting a finger
or object in your anus or vagina, or oral, anal, or vaginal intercourse with you.)
Yes ____ N o ______
2. Did this ever happen with a family member?
Y es N o _____
3. To the best of your knowledge, before age 17, were you ever sexually abused?
Yes No_____
128
Appendix D
Trauma Symptom Inventory (TSD
Copyright 1991, 1992, John Briere, Ph.D.
Please indicate how often each of the following experiences have
happened to you in the last six months:
Never
1. Heart pounding or beating too fast 0
2. Nightmares or bad dreams 0
3. Trying to forget about a bad time in your life 0
4. Unwanted sexual thoughts 0
5. Irritability 0
6. Stopping yourself from thinking about the past 0
7. Feeling empty inside 0
8. Sadness 0
9. “Flashbacks” (sudden memories or images of upsetting 0
things)
10. Not being satisfied with your sex life 0
11. Not being able to say “no” when someone wanted to have 0
sex with you, but you didn’t want to have sex
12. Feeling like you were outside of your body
13. Sudden disturbing memories when you were not
expecting them
14. Wanting to cry
15. Bad feelings about sex
16. Becoming angry for little or no reason
17. Feeling like you don’t know who you really are
18. Feeling depressed
19. Being bothered by memories
20. Having sex with someone you hardly know
21. Thoughts or fantasies about hurting someone
22. Your mind going blank
23. Not enjoying things you used to enjoy
Often
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
3
129
24. Periods of trembling or shaking 0
25. Pushing painful memories out of your mind 0
26. Not understanding why you did something 0
27. Threatening or attempting suicide 0
28. Feeling like you were watching yourself from far away 0
29. Feeling guilty 0
30. Feeling tense or “on edge” 0
31. Getting into trouble because of sex 0
32. Not feeling like your real self 0
33. Wishing you were dead 0
34. Worrying about things 0
35. Not being sure of what you want in life 0
36. Feeling like you weren’t really yourself 0
37. Bad thoughts or feelings during sex 0
38. Being easily annoyed by other people 0
39. Starting arguments or picking fights to get your anger 0
out
40. Suddenly feeling afraid for little or no reason 0
41. Having sex or being sexual to keep from feeling lonely or 0
sad
42. Getting angry when you didn’t want to 0
43. Not being able to feel your emotions 0
44. Confusion about your sexual feelings 0
45. Using drugs other than marijuana 0
46. Feeling jumpy 0
47. Absent-mindedness 0
48. Needing other people to tell you what to do 0
49. Yelling or telling people off when you felt you shouldn’t 0
have
50. Flirting or “coming on” to someone to get attention 0
51. Sexual thoughts or feelings when you thought you Ü
shouldn’t have them
130
52. Intentionally hurting yourself (for example, by 0
scratching, cutting, or burning) even though you weren’t
trying to commit suicide
53. Having a feeling that something bad was about to happen 0
54. Sexual fantasies about being dominated or overpowered 0
55. High anxiety Ü
56. Problems in your sexual relations with another person 0
57. Nervousness 0
58. Getting confused about what you thought or believed 0
59. Avoiding things that you knew would upset you 0
60. Feeling mad or angry inside 0
61. Getting into trouble because of your drinking 0
62. Staying away from certain people or places because they 0
reminded you of something
63. Wishing you could stop thinking about sex 0
64. Suddenly remembering something upsetting from your 0
past
65. Feeling anxious or depressed because you were alone 0
66. Wanting to hit someone or something 0
67. Feeling hopeless 0
68. Suddenly being reminded of something bad 0
69. Getting into relationships that were bad for you 0
70. Sudden feehngs of anger 0
71. Trying to block out certain memories 0
72. Sexual problems 0
73. Using sex to feel powerful or important 0
74. Violent dreams 0
75. Acting “sexy” even though you didn’t really want sex 0
76. Just for a moment, seeing or hearing something upsetting 0
that happened earlier in your life
77. Using sex to get love or attention 0
78. Frightening or upsetting thoughts popping into your 0
mind
131
79. Getting your own feelings mixed up with someone else’s 0
80. Wanting to have sex with someone who you knew was 0
bad for you
81. Feeling down and unhappy 0
82. Feeling ashamed about you sexual feelings or behavior 0
83. Trying to keep from being alone 0
84. Trouble paying attention to people 0
85. Having the same (or nearly the same) bad dream over and 0
over again
86. Your feelings or thoughts changing when you were with 0
other people
87. Having sex that had to be kept a secret from other people 0
88. Taking drugs or alcohol to stop your feelings 0
89. Not letting yourself feel bad about the past 0
90. Feeling like things weren’t real 0
91. Feeling like you were in a dream 0
92. Drinking or taking drugs to stop certain thoughts or 0
memories
93. Trying not to have any feelings about something that 0
once hurt you
94. Painful or disturbing memories 0
95. Daydreaming 0
96. Trying not to think or talk about things in your life that 0
were painful
97. Feeling like life wasn’t worth living 0
98. Being startled or frightened by sudden noises 0
99. Trouble controlling your temper 0
100. Being easily influenced by others 0
101. Wishing you didn’t have any sexual feelings 0
102. Feeling afraid you might die or be injured 0
103. Feeling so depressed that you avoided people 0
104. Feeling worthless 0
132
Appendix E
Letter to Sorority Presidents
January 15, 1994
President
Gamma Phi Beta Sorority
Address
Dear Ms. President,
I have been given permission by Diane Thompson, the International
President of Gamma Phi Beta, to contact you regarding your chapter’s
cooperation in my dissertation research. The purpose of my study is to
find out more about the factors that contribute to women’s eating issues.
As an alumna of Gamma Phi, and as a therapist counseling college women,
I realize that eating disorders can be a serious concern for sorority
members. I believe your help in completing this research is very important.
To assist me, all you need to do is follow these steps:
1. Give each member an envelope.
2- Ask each member to complete the questionnaires in the
envelope.*
*The questionnaires should only take 10-20 minutes to complete.
3. Ask each member to mail the completed questionnaires to me by
February 15, 1994.
These questionnaires are completely anonymous, so please have the
members return them directly to me instead of collecting them yourself. If
you have any questions, please feel free to call me at (510) 284-0726 or
(510) 642-9494 or my dissertation chair. Dr. Scott Whiteley, at
(213) 740-3255.
Thank you in advance fo r helping a fellow Gamma Phi!
Gratefully,
T.ynn S. Tracy, M.S.
133
Appendix F
Results Form
Please check here if you would like written results of this dissertation._ _
134
Appendix G
Information and Consent Form
I would like to ask you to participate in a research study which is being
conducted in order to satisfy the requirements of the Ph.D. degree at the University
of Southern California in the School of Education, Division of Counseling and
Educational Psychology. The goal of this study is to investigate eating issues and
the family.
Please read the following information describing the study as well as the
potential risks and benefits to you as a participant. When you feel that you
understand the process of this study including the potentisd risks and benefits to
you as a participant and are willing to volunteer as a research subject, you may
begin the questionnaires.
You will be asked to answer various questionnaires describing family
behaviors, issues related to food and eating behaviors, and childhood experiences.
All of your questionnaires will be completely anonymous. No information which
could identify you will be requested. It will take approximately 10 to 20 minutes to
complete the questionnaires. Your participation is completely voluntary and you
may choose to skip particular questions or stop at any time while completing the
questionnaires.
To make an informed decision about participating in this study, you need to
consider the potential risks involved. Due to the sensitive nature of the questions
asked, the primary risk is the arousal of uncomfortable feelings and memories, as
well as some anxiety surrounding any negative experiences. Negative childhood
experiences, regardless of their nature or degree, can arouse a wide range of strong
emotions including anger, shame, guilt, depression, and abandonment. You may
experience some of these feelings as you read through and answer the
questionnaires. Also, you may be embarrassed to answer certain questions.
Again, all of the questionnaires are anonymous and the validity of this study rests
on as much honesty and openness as possible. If you become too uncomfortable,
you may discontinue your participation at any time without penalty.
I am a Registered Psychological Assistant and can provide you with
referrals for low-fee counseling in your area (at the college counseling center, for
example) if you need help in dealing with any feelings and experiences that may
arise during the study.
You will be providing valuable and greatly needed information which will
aid researchers and therapists in more effectively understanding and treating those
who have experiences similar to yours by honestly sharing your experiences in the
following questionnaires. Your contribution in this research study will benefit
many other similar individuals.
It is also likely that you will personally benefit from your participation in
this study. The detailed and sensitive nature of the following questionnaires will
give you the chance to explore your previous relationships and experiences with
others with the opportunity to gain greater insight into your underlying thoughts
and feelings. As you remember and explore past experiences, it is anticipated you
will gain a clearer understanding of your present feelings and behaviors.
135
If you would like to discuss any portion of this research or would like
counseling referrals, you may contact me, Lynn Tracy, at (510) 284-0726.
If you have any problems or questions regarding this research, you may
contact the Institutional Review Board at the University Park Campus of the
University of Southern California at (213) 740-6721.
Signature of Student Researcher Date
Signature of Faculty Advisor Date
Abstract (if available)
Linked assets
University of Southern California Dissertations and Theses
Conceptually similar
PDF
Alianza Hispano-Americana, 1894-1965: A Mexican American fraternal insurance society
PDF
The role of the qualified social worker in the skilled nursing facility in Los Angeles County
PDF
Latino children, child abuse, symptoms of posttraumatic stress disorder and psychiatric difficulties.
PDF
An experimental study of the effect of aikido training on the self-concept of adolescents with behavioral problems
PDF
The religious sector's role in the delivery of supportive services to elderly: Collaboration with Los Angeles City and County Area Agencies on Aging
PDF
The social meaning of women's literacy in nineteenth-century America
PDF
The relationship between human values and moral reasoning as components of moral behavior
PDF
Effects of parenting style and ethnic identity on European American and Asian Indian adolescents' academic competence and self esteem
PDF
SEC's enforcement actions and its effectiveness in an efficient market
PDF
Differential erection response patterns of sexual child abusers to stimuli describing activities with children
PDF
The critical technique of Sainte-Beuve considered in its relationship to the modern biography as exemplified by Lytton Strachey and Andre Maurois
PDF
Women's autobiography and national identity: Natalia Ginzburg, Anna Banti and Renata Vigano
PDF
Characteristics of clients and care rendered to clients within a county public guardianship program
PDF
The emotional attitudes experienced by children of divorce in relation to their parents
PDF
A three generational study: Traditional Korean value systems and psychosocial adjustment of Korean immigrants in Los Angeles
PDF
The dialyzing father: A study of adolescent coping responses
PDF
Crisis preparation in technical organizations: A study using a multi-dimensional approach
PDF
Reaffirmation and transformation of gender in popular film: A feminist approach to mythic rhetoric
PDF
Social correlates of women's post-retirement adaptation
PDF
The effects of intellectual functioning and mediation on conceptual learning for Black college students
Asset Metadata
Creator
Tracy, Lynn Suzanne
(author)
Core Title
Bulimia nervosa: A response to trauma associated with parental alcoholism and sexual abuse in college sorority women
Degree
Doctor of Philosophy
Publisher
University of Southern California
(original),
University of Southern California. Libraries
(digital)
Tag
OAI-PMH Harvest,Psychology,Social Sciences
Format
application/pdf
(imt)
Language
English
Contributor
Digitized by ProQuest
(provenance)
Permanent Link (DOI)
https://doi.org/10.25549/usctheses-c37-45682
Unique identifier
UC11632818
Identifier
DP71340.pdf (filename),usctheses-c37-45682 (legacy record id)
Legacy Identifier
DP71340.pdf
Dmrecord
45682
Document Type
Dissertation
Format
application/pdf (imt)
Rights
Tracy, Lynn Suzanne
Type
texts
Source
University of Southern California
(contributing entity),
University of Southern California Dissertations and Theses
(collection)
Access Conditions
The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...
Repository Name
University of Southern California Digital Library
Repository Location
USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA