Perception of Family Functioning and Depressive
Symptomatology in Individuals With Anorexia Nervosa
or Bulimia Nervosa
Victor Fornari, Katarzyna Wlodarczyk-Bisaga, Michael Matthews, David Sandberg, Francine S. Mandel,
and Jack L. Katz
This study investigated the relationship between the
perception of family functioning and depressive symp-
tomatology in individuals with eating disorders (EDs).
Subjects were evaluated by diagnostic clinical inter-
view using DSM-III-R criteria for EDs, the Schedule for
Affective Disorders and Schizophrenia-Lifetime Ver-
sion (SADS-L), and two self-report measures, the Beck
Depression Inventory (BDI) and the Family Assess-
ment Device (FAD). A significant association was
found between self-reported depressive symptomatol-
ogy and perceived poor family functioning. Subjects
with bulimia nervosa (BN) reported a significantly
more dysfunctional family background than subjects
with anorexia nervosa (AN). In our sample, the pres-
ence of self-reported depressive symptomatology was
a more powerful predictive variable for perceived
family dysfunction than the diagnosis of affective
disorder. Also, the diagnosis of BN was a more consis-
tent predictor of dysfunctional family interaction than
the diagnosis of affective disorder. Depressive symp-
toms and EDs seem to play different roles in the way
in which they contribute to dysfunctional family pat-
terns.
Copyright© 1999by W.B. Saunders Company
I
NTEREST IN THE ROLE of family factors in
the development and maintenance of eating
disorders (EDs) dates back to the 1873 description
of anorexia nervosa (AN) by Lasegue. 1 More
systematic approaches to the investigation of fam-
ily factors in the etiology of EDs were initiated by
Minuchin et al, 2 and by Selvini-Palazzoli and the
Milan school of family therapy? Several studies
identified prevailing styles of functioning in fami-
lies with eating-disordered individuals and indi-
cated that interactions in these families differ from
those in families without eating-disordered mem-
bers. 4,5 Family interactions of subjects with anorec-
tic and bulimic symptoms have been found to be
dichotomous in that subjects with AN describe their
families as cohesive and organized with a lower
tolerance of conflict,6,7 as opposed to subjects with
bulimia nervosa (BN), who report their families to
be significantly more conflicted and less cohe-
sive. 8,9 However, it has also been suggested that
families of AN subjects are heterogenous, 1° and
that, for example, there is no single dominant
family pattern characteristic of AN. H Differences
described in the patterns of family functioning are
also possibly related to methodological limitations.
From the Departments of Psychiatry and Biostatisties, North
Shore University Hospital, Manhasset; and Division of Child
and Adolescent Psychiatry, Children's Hospital of Buffalo,
Buffalo, NY
Address reprint requests to Victor Fornari, MD, Division of
Child and Adolescent Psychiatry, North Shore University Hospi-
tal, 400 Community Dr, Manhasset, NY 11030.
Copyright © 1999 by W.B. Saunders Company
0010-440X/99/4006-0005510. 00/0
For example, there are theoretical differences in the
conceptualizations of family interactions, a2 and
findings are elicited by a wide variety of assess-
ment measures (self-report and observational mea-
sures). 13
In recent years, there has also evolved a consider-
able interest in the comorbidity of ED and other
psychiatric conditions. Halmi et al. TM and Fornari et
al.15 found that subjects with AN and BN have high
rates of both depressive and anxiety disorders.
Similarly, high rates of personality disorder (PD)
have been reported in ED subjects. 16,17 While the
family interactions of individuals with ED and the
possible comorbidity of ED with other conditions
have generated considerable research interest, there
has been little research concerning the relationship
between the functioning of these families and the
presence of other symptomatic behaviors. Steiger
et al.18 found no association between the ED
severity and the reported family functioning, but
found a significant relationship between the sever-
ity of the PD and the degree of dysfunctional
family interactions. Waller 19 found that in ED
subjects also diagnosed with borderline personality
disorder (BPD), specific features of pathologic
family interactions were associated with particular
symptoms of BPD. Thienemann and Steiner2° re-
ported that dysfunctional family interactions of ED
subjects were correlated with both the self-reported
level of depression and the clinical diagnosis of
major depressive disorder (MDD). Moreover, Miller
et al.21 found that families of patients with MDD
show severely impaired functioning, even more so
than families of subjects with alcoholism, schizo-
434 Comprehensive Psychiatry, Vol. 40, No. 6 (November/December), 1999: pp 434-441