Depressive tendencies and lower levels of self-sacrifice in mothers, and selflessness in their anorexic daughters.

Abstract

(1) To compare levels of selflessness (the tendency to ignore one's own needs and serve others') and asceticism of parents and daughters, in anorexic and control families. (2) To investigate the relationship between parents' depression and daughters' selflessness. Twenty-eight anorexic daughters and their 28 mothers and 23 fathers were compared to 29 control daughters and their 29 mothers and 28 fathers, participants were administered the Beck Depression Inventory, the Selflessness Scale, the asceticism scale of the Eating Disorder Inventory and the Structured Clinical Interview for DSM-IV. Anorexics' mothers showed significantly lower levels of selflessness and asceticism compared to control mothers; anorexic daughters showed significantly higher levels of selflessness and asceticism compared to control daughters. Depressive tendencies in anorexics' mothers were associated positively and significantly with their daughters' selflessness. The results support the clinical literature that depicts the anorexic daughters' readiness to sacrifice themselves for the family's needs. Clinical implications are drawn.

Full text

Page 1

Depressive Tendencies and Lower
Levels of Self-Sacrifice in Mothers,
and Selflessness in their Anorexic
Daughtersy
Eytan Bachar1,2*, Kyra Kanyas2, Yael Latzer3,
Laura Canetti2, Omer Bonne2and Bernard Lerer2
1Department of Psychology, Hebrew University of Jerusalem, Jerusalem, Israel
2Department of Psychiatry, Hadassah University Medical Center, Jerusalem,
Israel
3Eating Disorders Clinic, Psychiatric Division, Rambam Medical Center,
Israel
(1) To compare levels of selflessness (the tendency to ignore one’s
own needs and serve others’) and asceticism of parents and daugh-
ters, in anorexic and control families. (2) To investigate the relation-
ship between parents’ depression and daughters’ selflessness.
Twenty-eightanorexicdaughtersandtheir28mothersand23fathers
were compared to 29 control daughters and their 29 mothers and
28 fathers, participants were administered the Beck Depression
Inventory, the Selflessness Scale, the asceticism scale of the Eating
Disorder Inventory and the Structured Clinical Interview for
DSM-IV. Anorexics’ mothers showed significantly lower levels of
selflessness and asceticism compared to control mothers; anorexic
daughters showed significantly higher levels of selflessness and
asceticism compared to control daughters. Depressive tendencies
in anorexics’ mothers were associated positively and significantly
with their daughters’ selflessness. The results support the clinical
literature that depicts the anorexic daughters’ readiness to sacrifice
themselves for the family’s needs. Clinical implications are drawn.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders
Association.
Keywords: selflessness; depression; self-denial; asceticism; anorexia
INTRODUCTION
The presumed relationships between parent’s
(especially mother) characteristics and offspring’s
later psychopathology attract theoreticians and
researchers.Theformeraremuchquickerinpointing
at causal relationships between mother’s behaviour
European Eating Disorders Review
Eur. Eat. Disorders Rev. 16, 184–190 (2008)
*Correspondence to: Eytan Bachar, Ph.D., Hadassah Univer-
sity Medical Center, PO Box 12000, Jerusalem 91120, Israel.
Tel: þ972-50-7874493. Fax: þ972-2-6777938.
E-mail: bachare@mscc.huji.ac.il
yEytan Bachar and Kyra Kanyas contributed equally to this
paper.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Published online 25 October 2007 in Wiley InterScience (www.interscience.wiley.com) DOI: 10.1002/erv.840

Page 2

or characterological features, and their offspring
developing psychopathology or behavioural pro-
blems. Thus, in the field of Eating Disorders (ED),
self-psychology (Barth, 1991; Geist, 1989; Goodsitt,
1997) portrays the future Anorexia Nervosa (AN)
patient as one who did not experience enough
maternalreadiness for
her1self-object needs, that is, the mother was
experiencedasonewhodoesnotconveytothechild
the message that there are enough circumstances
and situations where mother can give up (even
temporarily)herownneedsandinterestsinorderto
fulfil the child’s needs. The future AN patient is
growing up to believe that instead, she should
ignore her own interests and view-points, for the
sake of fulfilling her mother’s needs.
The aim of the present study is to provide an
empirical setting for investigating self-psychological
assumption that parent’s (especially mother’s)
inability to serve self-object needs of the child
facilitates later development of AN in the child.
Longitudinalstudiesareobviouslythebestwayto
investigate causal relationships between parent’s
characteristics and later development of psycho-
pathology in offspring. Lacking the means for such
a research, we suggested the comparison of triads
of mother, father and anorexic daughter to control
triads. We further suggest logical hypothetical
relationships between the variables mentioned
below in order to add our contribution to the
debateabout the causalrelationship betweenparent
characteristics and child psychopathology.
Few studies investigated the relationship between
psychopathological characteristics ofthe parentand
psychological features of the anorexic daughter
using control group design of parents and daugh-
ters. Steiger, Stotland, Trottier, and Ghadirian
(1996) studied psychopathological traits such as
affectiveinstability, narcissism,
restrictive expression, as well as eating and body
image attitudes in mother, father and daughter in
Eating Disorders, other psychiatric disorders and
normals. They found correspondences between
daughter and parents’ psychopathological traits
and between mother’s and daughter’s eating con-
cerns. Despite this, parents of ED patients were not
discriminable on any of the variables studied from
relatives of the two control groups. The authors
conclude that transgenerational effect exists on
eating attitudes and psychopathological traits but
fulfillingsufficiently
compulsivity,
does not uniquely identify families in which clinical
ED syndromes occur.
In another study (Garcia de Amusquibar & De
Simone, 2003) 50 mothers of ED patients were
compared to 30 mothers of ‘non-consulting’ ado-
lescents. The mothers of the ED patients had more
frequent depression episodes and higher scores on
the bulimic factor of the EAT-26, and more bingeing
episodesthancontrols. Woodsideetal. (2002)found
that parents of ED patients (anorexia restrictive
type and anorexia bingeing-purging type) showed
elevated levels of perfectionism and more concerns
about weight and shape compared with controls.
But these two papers did not include the daughters
in their studies.
Fassino et al. (2002) did include in their design
mothers, fathers and anorexic daughters and found
that amongst the variables of temperament and
character structure that they had investigated, low
self-directedness was the common factor shared
by the daughter and her parents. But in contrast
with Steiger et al. (1996) who found that transge-
nerational similarities between parents and daugh-
ters did not uniquely discriminate anorexics from
control families, Fassino et al. did not have control
family groups to compare to.
The theoretical literature (mainly self-psycho-
logical) (Barth, 1991; Geist, 1989; Goodsitt, 1997),
and recently the empirical literature as well (Bachar,
Latzer, Kreitler, & Berry, 1999), depicts the ED
patient, primarily anorexic, as one who relinquishes
herown interests and ignores herown needs in order
toservetheinterestsandwellbeingofothers(Bachar,
1998; Bachar, Latzer, Kreitler, & Berry, 1999; Barth,
1991; Geist, 1989; Goodsitt, 1997; Sands, 1991). The
AN patient feels and behaves like a selfless person
serving others’ needs. She is liable to feel self-guilt
whenever she finds herself promoting her own
interests (Goodsitt, 1997). As a result, she often finds
herself living life in its narrowest parameters,
compromising her own development, denying even
her most basic needs, including nourishment.
Minuchin, Rosman, and Baker (1978) stress the
self-sacrificing behaviour and protectiveness of the
anorexic patient towards her parents, especially her
sensitivitytomother’swellbeingandherreadinessto
devote herself to alleviate mother’s depression.
Similarly Selvini-Palazzoli and Viaro (1988) describe
a pre-morbid family system in which the demands of
the family from its members to be self-sacrificing are
responded by total devotion of the to-be anorectic
daughter specifically to her mother.
We studied levels of selflessness—the readiness
to ignore one’s own needs and serve others’
1We used female form as the vast majority of ED patients are
females.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Eur. Eat. Disorders Rev. 16, 184–190 (2008)
DOI: 10.1002/erv
Mother Depression and Daughter’s Selflessness
185

Page 3

needs—using the Selflessness Scale (Bachar et al.,
2002). In addition to the Selflessness Scale, we
administered the asceticism sub-scale of the EDI
(Garner, 1991). We used this scale as it is a part of
one of the most used questionnaires in the field of
eating disorders, a sub-scale which taps content
area close to selflessness mainly in its self-denying
dimension.
We hypothesised that (1) level of selflessness and
asceticism of anorexic’s mother will be lower than
normal control mother’s, and that level of self-
lessness and asceticism of anorexic’s daughter will
be higher than normal control daughter’s. (2) That
mother’s depression will be associated positively
with her anorexic daughter’s selflessness.
METHOD
Participants
Twenty-eight AN daughters (mean age 21?4.1,
mean years of schooling 12.3?1.9) and their 28
mothers (mean age 50.5?5.8, mean years of
schooling 14.8?3) and 23 fathers (mean age
52.8?5.7, mean years of schooling 15.1?2.5).
Twenty-nine control girls (mean age 21.6?4,
mean years of schooling 13.1?2.1) and their
29 mothers (mean age 50.3?4.9, mean years of
schooling 15.3?3.9) and 28 fathers (mean age
52.9?6.2, mean years of schooling 14.6?3.1).
There were no significant differences between
anorexic and control family members in age and
years of education.
Five fathers of AN daughters and one control
father did not participate in the study claiming that
their schedules are busy.
AN daughters’ mean BMI was 16.2?1.2 and
control daughters’ was 21.1?3.4. All AN daughters
fulfilled DSM-IV criteria for anorexia, 21 of them
were diagnosed as restricting type and 7 as purging
type. Five of the AN daughters were also diagnosed
with co-morbid major depressive disorder. None of
the control families (parents and daughters) nor the
parents of the AN daughters fulfilled criteria of
psychiatric disorders of the DSM-IV.
Instruments
Beck depression inventory (BDI) (Beck & Steer, 1984)
The BDI is a 21-item self-report measure giving a
global score of depressive symptomatology. Each
item is answered on a 4-point Likert Scale with
higher scores indicating more severe symptomatol-
ogy. The BDI has good psychometric properties
including concurrent validity (r¼.73) and internal
consistency (Cronbach’s alpha¼.86) (Beck, Steer,
& Garbin, 1988). In the present study Cronbach’s
alpha was .80.
Selflessness Scale (SS) (Bachar et al., 2002)
The SS is a 15-item questionnaire on a 4-point
LikertScale,quantifyingthetendencytoignoreone’s
own needs and interests and serve others’. In a
previous study on eating disorders it distinguished
successfully between AN patients and normal
controls (Bachar et al., 2002). The internal reliability
of the instrument was reported in that study: Alpha
Cronbach coefficient was .61 (Bachar et al., 2002). In
the present study alpha Cronbach was .71. Construct
validity was demonstrated through factor analysis
on more than 1000 participants where four factors
were identified in an exploratory factor analysis:
Sacrifice for family, sacrifice for others, self-denial
and lack of self-interest. (Bachner-Melman, Zohar,
Ebstein, & Bachar, 2007).
Higher scores meant a higher tendency of the
participants to ignore their own needs and to serve
others’.
The asceticism sub-scale of the eating
disorder inventory-2 (EDI-2) (Garner, 1991)
The EDI-2 is a 91-item on a 6-point Likert Scale
self-report questionnaire consisting of 11 sub-scales
that assess specific cognitive and behavioural
dimensions of eating disorders; higher scores
indicate a higher tendency of the participant to
endorse attitudes in the specific sub-scale. The
author of this inventory (Garner, 1991) instructs
against the use of the total score of the entire
inventory and encourages the use of separate
sub-scales according to the target of the specific
study. For the purpose of the present study, the
asceticism sub-scale was administered due to its
aim to tap self-denying attitudes. Good internal
consistency(alphaCronbachbetween.80and.93for
the different sub-scales), convergent and discrimi-
nant validity were demonstrated (Garner, 1991).
The alpha Cronbach coefficient for the asceticism
sub-scale was .80 (Garner, 1991), and in the present
study was .78.
Structured clinical interview for DSM-IV (SCID)
(Spitzer, Williams, & Gibbon, 1994)
The SCID is a structured clinical interview for
assessing the frequency and severity of psychiatric
symptoms according to the DSM-IV criteria.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Eur. Eat. Disorders Rev. 16, 184–190 (2008)
DOI: 10.1002/erv
186
E. Bachar et al.

Page 4

Procedure
Anorexic participants were recruited by system-
atically reviewing all patients newly admitted to
either the psychiatric departments of Hadassah
Medical Center in Jerusalem, or Rambam Medical
Center in Haifa during a 4-year period. Control
daughters and parents were recruited directly
from the general population by the research staff
fulfillingcharacteristicsofage,ethnicity,occupation
and education equivalent to those of the AN
families.
AN daughters were administered the question-
naires and interviewed in the hospital clinics; all
other participants were interviewed and answered
questionnaires in their homes. All participants (AN
families and controls) signed an informed consent.
Data Analysis
We used t-test for independent samples to compare
anorexictocontrolfamilymembers.Giventhesmall
size of the two groups we performed, in addition,
Mann–Whitney non-parametric tests. Pearson cor-
relation coefficients were calculated to measure
associations between the variables in the study.
Measures of effect sizes were calculated: Cohen’s d
for the comparison of two independent groups. All
statistical calculations were performed using SPSS
13.0 for Windows.
RESULTS
Depressive tendencies in anorexics’ mothers were
associated significantly and positively with their
daughters’ selflessness (r¼.60, p¼.01). No corre-
lation was found between mothers’ depression in
the control group and their daughters’ selflessness
(see Table 1). The uniqueness of the association to
the anorexic daughter’s selflessness is outstanding
as there is no significant relation between parent’s
depressive tendencies and daughter’s depressive
tendencies, neither in the anorexics’ families nor
in the controls’ (see Table 1). In other words,
depressive tendencies in parents do not evoke
depressive tendencies in daughters or vice versa.
But depressive tendencies in anorexics’ mothers are
associated with selflessness in their daughters and
not in the control group.
Anorexics’ mothers tend to be less selfless than
control mothers (significant one tailed, t(52)¼1.81,
p¼.038, effectsize¼0.49,Mann–WhitneyZ¼?1.66,
p¼.049, see Table 2) that is, anorexics’ mothers tend,
less than controls, to serve others and tend to less
ignore their own needs. In contrast to mothers,
anorexic daughters tend to be significantly more
selfless than control daughters (t(54)¼2.79, p¼.007,
effectsize¼0.74, Mann–WhitneyZ¼?2.35,p¼.019,
see Table 2). Lastly, mothers of anorexics were
significantly less ascetic than control mothers (t(51)¼
2.15, p¼.035, effect size¼0.50, Mann–Whitney
Z¼?2.15, p¼.032), a finding that coincides with
their lower selflessness
mothers. Namely, anorexics’ mothers are showing
less self-denying behaviour than control mothers.
Anorexic daughters, similar to the findings of
their higher selflessness tendency compared to
control daughters, are also significantly more ascetic
compared to the control daughters (t(48)¼5.72,
p<.001,effectsize¼1.37,
?4.48, p<.001).
compared tocontrol
Mann–Whitney
Z¼
Table 1. All calculated correlations between parents and daughters in the anorexic and control families
DaughterAnorexic familiesControl families
MotherMother
BDI SS AsceticBDI SSAsceticism
BDI
SS
Asceticism
Daughter
.23 (n¼27)
.60?(n¼28)
.14 (n¼25)
?.22 (n¼27)
?.17 (n¼27)
?.01 (n¼25)
Father
SS
?.001 (n¼20)
?.12 (n¼20)
.08 (n¼20)
.29 (n¼25)
.06 (n¼25)
.27 (n¼25)
.13 (n¼28)
.12 (n¼28)
?.06 (n¼25)
?.19 (n¼27)
.24 (n¼27)
?.33 (n¼25)
Father
SS
?.24 (n¼26)
.16 (n¼26)
?.40 (n¼25)
.17 (n¼28)
.23 (n¼28)
.19 (n¼25)
BDI Ascetic
?.15 (n¼23)
?.11 (n¼23)
.14 (n¼23)
BDIAsceticism
.17 (n¼25)
?.09 (n¼25)
?.01 (n¼25)
BDI
SS
Asceticism
.22 (n¼23)
.23 (n¼23)
.19 (n¼23)
.26 (n¼28)
?.04 (n¼28)
?.10 (n¼25)
?p<.01 (two tailed).
BDI, Beck Depression Inventory; SS, Selflessness Scale.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Eur. Eat. Disorders Rev. 16, 184–190 (2008)
DOI: 10.1002/erv
Mother Depression and Daughter’s Selflessness
187

Page 5

DISCUSSION
The clinical literature (Barth, 1991; Goodsitt, 1997;
Minuchin et al., 1978) emphasizes future-AN-
patients’ tendency to be supersensitive to their
mothers’ psychological well being and their will-
ingness to devote themselves to alleviating their
mothers’depression.Itseemsthenthattheresultsof
the present study tend to support these clinical
descriptions. Depressive tendencies in mothers are
uniquely associated positively and significantly
with their anorexic daughters’ selflessness. These
resultsfromourtransgenerationalstudymayjustify
the efforts needed for conducting prospective
longitudinalstudiesforthepurposeofinvestigating
causal relationships. Correlational studies such as
the present one do not allow of course for causal
conclusions. It seems that only prospective studies
can determine whether the depression of anorexics’
mothers precedes the daughters’ selflessness or vice
versa. Having said this, we might still suggest that
the present study supplies a hint for the direction of
causality as it is not very plausible that a daughter’s
willingness to serve others’ needs, including (and
especially–atleastaccording totheclinical literature
e.g. Minuchin et al., 1978; Selvini-Palazzoli & Viaro,
1988) her mother’s, will evoke depression in the
mother. The results of the present study may
suggest that when the anorexic daughters sense
negative well being in parents, especially mothers,
they do not react with depression but with self-
lessnesswhichisamoreactiveandspecificreaction;
activating greater attunement and readiness to
serve mother’s needs.
Lower tendency of anorexics’ mothers for self-
denial (lower asceticism score compared to normal
mothers) and lower tendency to ignore their own
needs and serve others (lower selflessness score
compared to control mothers) might convey to their
anorexic daughters that mother cannot give, but
instead, should be given.
It is suggested that parent counselors will explore
whether this reading that ‘mother cannot give’
ensues from mothers’ exhaustion dealing with the
severe illness of their daughters (e.g. that mother
has become less selfless as a protective or defence
mechanism to cope with the difficulties associated
with caring for an anorexic daughter). If this is the
case it is suggested that the counselors will help the
parents to seek support from family and social
networks as well as from their own therapy and not
to refer to their daughters with their emotional
hardships due to the risk of intensifying the already
high levels of selflessness in their daughters.
Depression of anorexic’s father, in contrast to
mother’s depression, was not associated signifi-
cantly with his daughter’s selflessness. This may
suggest that the anorexic daughter is not attuned
anxiously to father’s affective state as she does with
mother. It is then suggested that parent counsellors
may encourage fathers to increase their involve-
ment in their daughters’ lives. Father’s ability to
give does not lag behind control father’s (they do
not differ in selflessness and asceticism levels from
controls) contrary to their wives vis-a-vis control
mothers. The finding that anorexic daughters are
not asalert to their father’s affective state and donot
tendtosacrificetheirinterestsforhissakeastheydo
with their mothers, may protect against the danger
of role reversal between parent and child.
High levels of selflessness such as those observed
in the anorexic daughters warrant therapeutic
intervention. Therapists working with the anorexic
daughters should help them distinguish between
Table 2.Means (and standard deviations) of BDI, SS and asceticism in anorexic and control families
Anorexic family Control family
MotherFatherDaughterMother Father Daughter
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
M (SD)
BDI7.32 (7.03)
(n¼28)
41.85 (5.67)
(n¼27)
2.46 (1.90)
(n¼25)
4.91 (3.45)
(n¼23)
41.40 (4.62)
(n¼20)
2.82 (2.54)
(n¼23)
19.70 (11.99)
(n¼27)
45.28 (4.63)
(n¼28)
6.46 (4.32)
(n¼25)
6.03 (4.21)
(n¼29)
44.52 (5.13)
(n¼27)
3.42 (1.93)
(n¼28)
5.46 (5.59)
(n¼28)
41.19 (5.28)
(n¼26)
2.66 (1.92)
(n¼25)
4.57 (4.02)
(n¼28)
41.79 (4.83)
(n¼28)
2.06 (1.43)
(n¼25)
SS
Asceticism
BDI, Beck Depression Inventory; SS, Selflessness Scale.
Copyright # 2007 John Wiley & Sons, Ltd and Eating Disorders Association.
Eur. Eat. Disorders Rev. 16, 184–190 (2008)
DOI: 10.1002/erv
188
E. Bachar et al.