TRAUMA AND
DISSOCIATIVE
EXPERIENCES IN
EATING DISORDERS
Riccardo Daile Grave, M.D.
Manuela Oliosi
Patrizia Todisco
Claudia Bartocci
Riccardo Daile Grave, M.D., is Director of Reparto di Riabil­
itazione Nutrizionale - Casa di Cura Villa Garda. Via Monte
Baldo, 89 - 37016, Garda (VI') - Italy. Manuela Oliosi is a psy­
chologist at Reparto di Riabilitazione Nutrizionale - Casa di
Cura Villa Garda. Patrizia Todisco, M.D., is a physician at
Reparto di Riabilitazione Nuu'izionale - Casa di Cura Villa
Garda, Claudia Bartocci is a psychologist at Reparto di Riabil­
itazione Nuu'izionale - Casa di Cura Villa Garda,
For reprints write Riccardo Daile Grave, M.D., Reparto di
Riabilitazione Nutrizionale - Casa di Cura Villa Garda. Via
Monte Baldo, 83 - 37016, Garda (Vr) - Italy.
ABSTRACT
This study investigates the relationship between trauma, dissocia­
tive experiences, and eating psycho-pathology in a group of eating
disorderpatients. The Dissociation Questionnaire (DTS-Q) and a semi­
stmctured interview were used to assess 106 eating disorderpatients
at the start of an inpatient treatment program. DTS-Q scores were
evaluated for the eating disorder patients and compared with the
scores of20 schizophrenic patients and 112 high school graduating
students (controls). Of the eating disorder patients, 45.2 % report­
ed traumatic experiences. The highest trauma rate was reported &y
patients with bulimic symptoms. Significant differences werefound
in the prevalence of traumatic experiences between eating disorder
patients and contml subjects, but not between eating disoTder and
schizophrenic patients. The highest total DTS-Q scores were detected
in bulimia nervosa and anorexia neroosa binge eating/purging type
patients; the lowest DTS-Q scores were found in patients with binge
eating dis01'd"'; schizophrenia, and controls. Eating disorder
patients, in comparison with schizophrenic patients, ·rep01ted sig­
nificantry higher scores in identity confusion, loss of control, and
absorption. However, the only dissociative features which se",1IS to
link trauma, dissociation, and eating disorders are identity con­
fusion and loss of contml, since absorption is not sensitive to the
presence/absence of trauma.
INTRODUCTION
Recently several studies have reported a high i!,cidence
of past traumatic experiences (such as sexual and/or phys­
ical abuse) during the childhood and adolescence ofpatients
274
Witll eating disorders (Chandara & MalIa, 1989; Goldfarb,
1987; Oppenheimer, Howells, Palmer, & ChaloneI', 1985;
Palmer, Oppenheimer, Diagnon, Charloner, & Howells,
1990; Vanderlinden, Vandereychken, Van Dyck, & Vert­
omner, 1993; Waller, 1991; Waller, Ruddock, & Pitts, 1993).
Some authors even assumed that there could be a more spe­
cific relationship between traumatic experiences and the
developmen t of eating disorder (Bulik, Sullwar, & Rorty,
1989; Hall, Tice, Beresford, Wooley, & Hall, 1989; Vander­
linden, 1993). Other researchers, instead, sternly criticized
these studies for their metllOdology and concluded tI,at evi­
dence was lacking to support the theory that childhood sex­
ual abuse is a risk factor in bulimia nervosa (Hope & Hud­
son, 1992). Hope and Hudson (1992) argued that ti,e
differences in sexual abuse rates were partly due to the def­
inition of trauma employed in the various studies.
In 1994, Welch and Fairburn compared the rates of
reported sexual abuse in a community study among clinical
bulimic subjects, a psychiatric control population, and con­
trols. They discovered that ti,e rates of sexual abuse report­
ed by bulimic subjects were similar to those witll psychiatric
disorders and concluded that the association between child­
hood sexual experiences and the development of eating dis­
order need not imply a causal and specific relationship
between the two. Sexual abuse during childhood tI,en can
be best interpreted asjust one among many factors that may
increase the risk in the developmentofeating disorder (Gar­
ner, 1993). To better understand the possible link between
childhood trauma and the development of an eating disor­
der, several authors have considered the possible mediating
role of different factors including self-denigration, border­
line personality disorder, disclosure experiences, and dis­
sociation (Everill & Waller, 1995a).
The study of dissociation in the realm of eating disor­
der was overlooked until 1979, when Russell unveiled the
presence of minor dissociative "hysterical" mechanisms in
bulimic patients. However, only in the last decade has ti,e
relationship between dissociation and traumatic experi­
ences in eating disorder been extensively studied. Torem
(1986, 1987) underlined ti,e possible influence of an ego­
dissociation mechanism in the etiology and development of
eating disorder. Vanderlinden et al. (1993) demonstrated
that eating disorder patients (especially ones with bulimic
DISSOClATIO~,Vol. IX, No, 4, December 1996
DALLE GRAVE/OUOSI/TODISCO/BARTOCCI
and <l.l}piclll c"ling disorders) repon significantly higher le\',
eI~ ofdissociative experiences than Ilormal control sllbj(;CL~
and h}1)Olhesized that binge eating could fUllnion as a way
ofcoping"ilh leelings and memories associated with l.h~ u'au­
mao Rosen and Petl)' (1994) discovered that di~soci;llioll of
feelings and loss of control wcre significantl}' correlated to
eating rchth:d subscalcs of lhc Bulimia Tcst and lhc Eating
Disorder Im'CnlOI)' in a sample of college SlUdcrib.
Some preliminarvdata indicate Ih:1I dissociation lila)' act
as a mediating factor between history ofabu5e and bulimic
symplOmato]ogy ([vcrill & Waller, 1995a: l\kCarly, Gorr,
Baer. CeolTc, & I-Icn:og. 1994: Vanderlinden ('I aL. 1993). A
relationship between dissociation and other impulsi\'('-like
beh,Hiors h,IS also been reponed {Dernilrack. PUlrIam,
Bre\\'erton. Brand, &Gold, 1990; Van del" Kolk. Pcrl}'. & I-Ier­
mall, 1991). It has been suggested Ihat dissociation may be
the prim,u} defense mechanism lherd)} rcsulting in a
Mst.'nse ofdeadness- (Van der Kolk clal.. 1991). l,ollscquclU.l}'.
there ma)' be a need 10 block this negati\'c efTect through <I
~pecHic stimulus. Binge eating and other impulsi\'e behav­
iors (i.c.. CUlling) could provide this stimulation (Heat.er­
Ion & Baumeister, 1991: Van der Kolk et al.. 1991).
Another hrJ>othesis is that dissociation precedes binge
eating" (E\"erill & Walkr. 1995h): Ihis interprctation ad,,\)­
G.ltes that the defense mechanism ofdissociation. consequent
It) ,Ill earl}' traumatic experience. would be used to deal with
the ~LLhscqllcntstressful c\·cnL~.although it ma)' become iru.;f­
fecth'e and maladapli\'e oyer rime. Ilowe\"er. dissociation
\,ould pro\'ide a tempor-;:Iry refocusing ofanenlion. This may
cnablc bulimic subjecL" to initiale binge eating \,'ilhoUI deal­
ing \\ilh the long-tenll neg.uin· consequences oftheir behav­
iur (e.g., weighl gain, guilt.. and self-deprecation) (Healher­
tOil & BaulIlcistcr. 1991).
All of the above smdies inspired us to in\"esli~ate:
I ) The prescncc ofdi'SSOCiali\"c cxpericllccS ill cal­
ing disorder paticnts compared \\ith schizo­
phrenic subjects ~lI1d normal subjects;
2) The frequency and the lypes of tr.lllllla in eat­
ing disordcr paol'nlS and thcir relationship with
dissociation;
:\) The frequcncy and Sl'verit)' ofdissociative expe­
riences in eating disorder patients.
The stud)' included a group ofobese patients with binge eat­
ingdisordcr since the relationship between trdullla and dis­
~)ciati\'e experiences in binge eating disorder had nm been
~tudied !lefore.
A1.£T'HODS
Sllbjeds
The subjects were 106 female patients conS<.'"cuti\"e1y hos­
pitalized at the Eating DisorderCentcr in Carda.lt:al),. Theil'
diagnostic characlelistics \,'ere as foIlO\,'s: 52 anorexia ner­
\'osa palients (30 resu'ining type and 22 binge ealing/ pmg­
ing Iype) accordi ng to DS.\I-II/-U (Americ,1T1 Ps)'chiau'ic Ass0­
ciation. 1987).24 normal \,'eight bulimia nen'osa patients
according (0 DSJJ.ITl-R. and 30 binge eating disorder patients.
according to SpiLLcr et a1. (1993). Binge eating diMJrdn is
a recetlll)' defined eating disorder 111;\1 describes an eating
dislllrbance of a persoll SUffering from recurrenl episodes
ufbingc eating \\ilhoLlI characteristic compen.s:.tlory beha\' ­
iors ofhulirnia ncrvosa (i.l~.. \'omiting and purging) (Spit/cr
elal.. 1993). Recent studies ha\'C fonnd that obese subjects
witll binge eating disorder report more psychological dislress.
depressive symptoms, and psychiatric disorders lhan obcst=
non-bingers (~'larcuset<l!.. 1990; Yallo\1ik)'. Nelson. Dubbert,
& Spitzer, 1993). Table I shows the age. body mass index
(B;"II = weight in kilogrdllls divided by height s<luared in
melers), length of illness, educarion le\'el, Ihing siruation.
and social class or all eating disorder patients.
The DI5-Q daw of lhe palients were compared with a
group of 20 female schizophrenics parients hospitalized in
the ps),chiau'ic hospital of Ponton (Verona) diagnosed
according 10 lJSM-II1-U (Amcrican Psychiatric Association.
1987) criteria (age 33.9 ± 7.1 )'ears) and 112 female high
school gr-;:Iduating sllIdents as controls (age 18.1 :::: 0.8 years).
Measures
At lhe stan of the inpalient u'eauncnt, patients' clinical
histories \,'cre g<IIJlcred with sc\'cl"al qucstionnaires and a
semi-sIJ1.lclUred interview for eating disorder patients. The
Dissocialion Questionnaire (DIS-Q; Vanderlinden et aI.,
1992, 1993a) \\~dSadministered to ,lSSCSS dissociation. TIle DIS-­
Q is a 53-ilem self-reporting scale to mcasure dissociativc exlx.....
riences and has. besides a tOlal score. four subscales:
I) Idemity confusioll and frab'lnentation (den_....
alization and depersonalizalion);
2) Loss of comral over bchavior. thoughts and
l'motions;
3) Amnesia (memory lacunas);
4) Absorption (ellhanced cOllcenu,nioll. which
is sUPl>oscd 10 pia)' an imponaTll role in h)'p­
nosis).
The DI5-Q has been demonstrated to h,H"C good construct
validity and it is able to discriminate patients wilh djssocia­
li\'e pathology from all other ps)'chiau'ic catcgOlies (Van-
275
Dli'iOU\TIO~. \ol IX. \0 t Drmnbrr 1!l96
I
TRAUMA AND DISSOCIATIVE EXPERIENCES
RESULTS
1) Total DIS-Q between eating
disorder patients and con­
trols (p< .0001);
Dissociative Experiences in Eating Disorder
Subgroups Compared with Controls and
Schizophrenic Subjects
Table 2 shows the DIS-Q scores of eating
disorder patients compared with the con­
trol group and the schizophrenic subjects.
Analysis of variance (ANaVA) and
Scheffe's t-test were applied in order to
compare the DIS-Q scores of eating dis­
order patients with the scores of the 112
female high school students and the 20
schizophrenic subjects.
ANaVA found significant differences
between eating disorder patients, controls
and schizophrenic subjects in the total
DIS-Q (F = 17.05, d.£. = 5,232, p< .0001)
and on the four subscales: identity con­
fusion (F = 29.62, d.£. = 5,232, p< .0001);
loss of control (F= 7.43, d.£. = 5,232, p<
.0005); amnesia (F =4.96, d.f. =5,232, p<
.01); and absorption (F = 9.3, d.£. = 5,232,
p< .0001).
Post hoc testing showed significant
differences on the following subscales:
2) Identity confusion between
eating disorder patients and
conu'ols (p< .0001), eating
disorder patien ts and
schizophrenic subjects (p<
.05);
TABLE I
Patients' Clinical Data
ANR ANB BN BED
(n =30) (n =22) (n =24) (n =30)
Age (years) 21.9 ± 6.0 24.9 ± 4.2 22.3 ± 2.7 36.4 ± 13.2
BMI 14.1 ± 2.6 15.3 ± 2.5 19.7 ± 3.3 33.0 ± 6.6
Length of illness
(years) 3.9 ± 3.5 6.5 ± 4.3 5.9 ± 2.8 14.6 ± 10.2
Education
Elementary school 0(0%) 0(0%) 0(0%) 5 (17%)
Junior high school 12 (40%) 4 (18%) 9 (38%) 12 (40%)
High school 15 (50%) 15 (68%) 15 (62%) 12 (40%)
University 3 (10%) 3 (14%) 0(0%) I (3%)
Living Situation
Alone 2 (7%) 2 (9%) 0(0%) 7 (23%)
With parents 26 (87%) 16 (73%) 23 (96%) 8 (27%)
With a partner I (3%) 3 (14%) I (4%) 13 (43%)
With a child 0(0%) I (4%) 0(0%) 2 (7%)
Other situations 1 (3%) 0(0%) 0(0%) 0(0%)
Social Class
Low II (37%) 12 (55%) II (46%) 19 (63%)
Middle 17 (57%) 8 (36%) 10 (42%) I 0(33%)
High 2 (6%) 2 (9%) 3 (12%) 1 (4%)
ANR = anorexia nervosa rest'ricting type; ANB = anorexia nervosa binge eat-
ing/purging type; BN = bulimia nervosa; BED = binge eating disorder
derlinden, 1993b).
The presence of traumatic experiences was examined
by means of a clinical in terview and a self-report question­
naire for the eating disorders and schizophrenic patients and
with a self-reporrquestionnaire for the controls. In our study,
only severe forms of trauma prior to the onset of the eating
disorder were considered: incest (i.e., anything from fond­
ling to genital penetration between a child and a family
member); rape; physical abuse (i.e., repeated beating); and
severe psychological abuses (i.e., harsh and repeated humil­
iations or complete neglect or abandonment in cllildhood).
3) Loss ofconu'ol between eating disorder patients
and controls (p< .005), eating disorder patien ts
and schizophrenic subjects (p< .05);
4) Amnesia between eating disorder patients and
controls (p< .05);
5) Absorption between eating disorder patients
and controls (p< .005), eating disorder patients
and schizophrenic subjects (p< .005).
Statistical Analysis
Statview 4.5 software was used. One way ANaVA and chi­
squared (with continuity correction when appropriate) were
performed.
Dissociative Experiences in Eating Disorder Subgroups
Table 3 shows the DIS-Q scores of eating disorder sub­
groups. Analysis ofvariance (ANaVA) and Scheffe's t-testwere
applied in order to compare the DlS-Q score of eating dis­
order subgroups. ANaVA found notable differences between
-
276
DlSSOCIATlO:'\. 1'01. IX :'\0. 4. Dec<mbec 1996
Dl nlro) . .and arin Di rde P. ti n
OT DlS-Q2 DIS-Q4
D D D
, nu' I 112 1.9 I. 1.:- ± 2.6 _0.':-
hiz phr ni 2 ~. ± .6 2.0 0.7 2.1 ±0.7 2. .3 ±.
"
106 c ±O.? 2. d 0.9 2.Sbd ± 0.7 I.Sa ±O.S 2.9b O.
Df. QI .. idtmlit confusion; Dl Q2 - lo oj antral; Df' Q .. n'l1mf.~in; nr. Q4 = absorption; ED = eating di order. .One VIti
ANDY. : fl. p < .0- compared with flmtro{; ; b. P< .00 - compared wi/h control; c. p< .0001 (omparPrJ with ((mlmL; d. P< . (Om-
PMI'd willi ~(hizophTeni ;e. p < .OU- compared with chiwpl",mi .
ore for I'd Pari n
T :AI...
D
D 1 DIS-Q2
D
D
D
D 4
D
R
B
BN
FPD
3
2
2
2. ± 2.5 1. 2.3e ±0.9 I.
2.5 .5 2.7b ± 2.5 ±0.5 1.
2.7 _0.7 2.7 0.. 2.9a ±0.7 2.1
2. ± 0.' 1.9 0.7 2. ±D.7 1.5
1.0
O.
.7
3.0a ±0.8
2.3 _.7
01.' QI = ldenlil.y conJusi(m; DI.· Q2 -La ojcontroL; DI. = anm ia; Df. Q4 = ab orplion; At R - (111M' ia neroo a re tricl ­
mgt 'I) : t = anorexia nervo a bin r alin /purgingtJp . B . "" fiUlimifule7 0 a' BED =binge ratingtr ordR . Onewa.vA.t\o\' :(1.
p <.0 compar: d wilh B' . b. p< .01 mmpar: d wilh BED; r. P<. - compcm d with BED; d. P<. I compared with BED;
e. p.O· mmpar"d with B ".
r nee on th
ld nri
inrv
3
I) c.aIDi Qb twe n.u1~ \ ithbulimian I'-
II lsa ar d bin e aLing di 'ord I' (p<.O 5);
277
-
UlSSOCL-I.T10:\. \'01. IX. :\0..J. [}erelllher 1996
TRAUMA AND DISSOCIATIVE EXPERIENCES
TABLE 4
Frequency of Trauma in Controls, Schizophrenics, and Eating Disorder Subgroups
Controls Schizoph. ANR ANB BN BED
Type oJtmuma**
Incest 1 (0.8%) 0(0%) 0(0%) 3 (14%) 1 (4%) (0%)
Sexual abuse+ 2 (1.8%) 2 (10%) 2 (7%) 3 (14%) 3 (12%) 1 (3%)
Physical abuse 7 (6.2%) 2(10%) 2 (7%) 2 (9%) 2 ( %) 5 (17%)
Loss of family member 3 (2.6%) 0(0%) 2 (7%) 2 (9%) 1 (4%) 5 (17%)
Psychological abu e 3 (2.6%) 4 (20%) 0(0%) 4 (18%) 7 (29%) 3 (10%)
ANR = anorexia neroosa Testricting t)'jJe; ANB = anorexia neroosa binge eatillg/puTging type; BN = bulimia Ileroosa; BED = binge
eating disoTder.
'. ~ the expected values aTe too low to rio a statistical analysis. + &y others than Jarnily members
TABLE 5
DIS-Q Scores and Types of Trauma for the Eating Disorder Group
TOTAL DIS-QI DIS-Q2 DIS-Q3 DIS-Q4
N X SD X SD X SD X SD X SD
Non-trauma 58 2.2 0.7 2.3 0.9 2.3 0.7 1.7 0.8 2.9 0.7
Trautna 48 2.6a 0.7 2.6b 1.0 2.7" 0.7 2.la 0.9 2.9 0.9
Incest 4 2.8 0.6 3.1 0.9 3.0 0.5 2.1 0.3 3.2 0.3
Se>.Llal abuse+ 9 2.8 0.9 3.1 0.9 2.9 0.8 2.3 1.1 3.0 0.5
Physical abuse 11 2.2 0.6 2.2 0.8 2.3 0.7 1.7 0.6 2.9 0.6
Loss of family member 10 2.3 1.0 2.1 0.9 2.5 0.8 2.0 1.2 2.4 0.9
Psychological abuse 14 2.8 0.7 2.9 0.9 3.0 0.7 2.3 0.8 3.0 0.8
DfS-Q1 ~ Identity confusion; DIS-Q2 = loss ojcontrol; DfS·Q3 = amnesia; DIS-Q4 = absOlption. Olle way ANOVA: a. p < .01 com-
!JaTed with non-tTauma; b. p< .05 compared with non-tmwlla; + &y others than Jamily memben.
278
DISSOCIATIOX. Vol. IX. Xo. 4. December 1996
DALLE GRAVE/OLIOSI/TODISCO/BARTOCCI
-I) Ah!>urptil)l1 l.k:t\\t:t:1\ 'lll~t:CbWilli bulimia ncr­
\'OS;I :lnd bin~(·I·,lIillJ.:di-.orckr (p< .001). and
lx-tween subjecLS .... ilh auorc'xia Ilcnosa binge
t:alil'S I>urging I\ve .md bingcC'dlingdisordcr
(l'dlO I ).
Tmuma (md Dissociation
labk .. (''Ir(dlllinc, lilt: l>n'\'3lt'llct· of "'pes of U'allma in
cUlllrUls. schilophrenic.s :md e:lLlng disorder subgroup sub­
)«l••. 16.2ct: (n; l~) ofc.uing dhonlcrpaliclltsrcpont.-d trau­
malic expt·rienn',. Signillrdlll ditTen'nco emerged bel""eCII
eating disordcrwollp :,,,d conlrols.sll~in the prn-..lence
of tr.utnratic c),.perit'nc~ (chi 'o(llLared :: 26.5. d.r. :: -I, p
<.0001). \'0 hill- 110 difTt'rt'lIcl'" ",'n' found bclwecn ealing dis­
order group and ~hilOl)hr...mn (chi squared =.26. OF=1.
JF11.,..). 'Ill(" highest r.ut' ,,~ dcmonSlfated m muse ""ilh
bulimic bch;l\iof~ .mel ill 'lChilol>!Jrcnie subjc.·(.'"I5: anorexia
nerY()§;1 binge c;llmg pUf),'11lJ; I\'PC 63.6% (11'=14). bulimia
nenosa j8f( (n", II), billgccaung doorder 50% (n=15).and
;"chiIUI)hrcniOl -IOtl- (11- ): \\hcrC'..b. lJdUlIla "''<b ICS'> pn."\<l­
lenl :llllonganor('xi.IIII'nO".I re..lricting"1X'20% (n=6) and
controls 11't (n 16).
\'0\ A ,lIld Schcffe'.. H..:\l \\cre tl'>l:d to t-"'\<lhl,uc Ihe dif­
f(·rt"nc~~ IX'lwI'('n "ul~l'rl~ \\'ilh and \\;l!Hllll a rr:mmil histl>­
n '1 able .~ reporu IlISQscores according lodifTerenl forms
uf lr.Ultlla. Trauma p;lIiCllb '<01 cd "ignilicantl~higher than
thl' IUIIHralllll.t p;ltil'lIl~ iTl lcnmufthc 101:'11 D1SQ (F :6.15.
d.r. - 1.10-1. 1>< .(1) and on the lollo\\'ing subscales: idclllil)
wllfusion (F '" 4.19. d.f.• 1.1 0·1, 1>< .05): loss of conlrol (F
- nAn. d.f. '" 1,10-1, p< .01) and ;unnc..i" (F = 6.9, d.r. = I, I 0·1.
p<.0 I). ANOV,\ rnarkc'd !\i~nilicOlntdifTerences bel"wcen lrpe
of lr-aullia on lll..: t\lt.11 D1S-Q (F '" 2.46, eLf. '" 4,43, p< .05)
and 011 I hI' id"rHity fOll fll~ioll ~llbst'ak (F '" 2.99, el.f. = 4,4~,
p< .05).
Etail/I; D;su,.dliTS mllf Di,mx:;a/;fll/: f)"I!(fue"C)' olld Sigllifialllce
1II :lccnl'd:tllCl' wil h Varu!<,rlinden's sllggeslion (I \:193 Il).
a Clll-olfM:OfC ur~.~ 1'01 IIIC IOlal OI5-Q was used to study tile
frl'qllclIC)' 0\1111 ~1'\'l'ril}' ofdisSllriatin' experiences. 011r (lin:t
jilt hld('d ~N p:llienL~ (22.6%) with se\'ere di.ssoci:uh·e s)'1np­
tlll1l~. of\,'hom 19 (52.7' ) \\erl' subjected to st:'\'ere U,lllrllalic
c"'pcriCllccS (4 !W.'x\l:tl ahll~. I in('c~l,!l sc\ere ps)'choIOj.{i­
c.11.!hll"t-·, 4 ph),ical abuSt'. I loss orfarnily member), while
('iglu pOlticnb in lh..: llllld, dOlimcd 10 have llt:ver bccll
I"~d tu ll<lUlna, Di"ll(Kiali\'c "~llIplUms \\cre more com­
mOil in hlllimiOl ncnOS:t (37.5ll:) and in anorexia nenosa
binge ~adllg purging ('PC (31.8Cf) pauents Lllan in anorex­
ia neno~ rcsll;cling lypC (20<t:) and binge COlling disorder
J.3%) paticnts.Onl\ 3.6'1 of the conlrol group and 10%
flf -.c11l10phrenics paticllts I't'pont."() 'it-"\ere diS!iOCiau\e exIX-"
rit"nces (scon~ >2,9).
DISCUSSION
Ollr re"ult.:. MIPIXlrL otlll::r ..lttdies suggesung Ihal l.I"'Olll­
maliC' (·xpc·ri('nc('...tn,' more I)rl"\':tlenl in patients witll bulim­
ic beh<l\;or5 (Bushnell, \\'ells. & OalJc}'-SI'O\me. 1992: Pri­
bor & Din\\iddie. 1992; VOllldnlindcn ('I al.. 1993; Waller.
1991). \cC'ordillg 11l00Irdal;I.lrdllm:texperienc ",el'C lIlore.
frcquenth reponed'" anorexia nen'0S3 binge caling purg.
ing t\-pt.' paucn15 (66t{,) ,md bulimia l1en'QSOl paLienlS (59%)
than 1)\ alulft':\ia l\t.'no:..l rt.'..tricting l\pe pati~nlS (21 %). It
i~ inu.·rt.·..ling 10 nOle Ihal an e1e\otted nue of trauma eX'pe­
rienct'S "'~ also deteclt-"(I in obese patients ",;!h bingr eat­
ing di'>l'.mlcr (-I7ff). Thc-.e daGt ,*-em to suggrsl thaltrau­
rna could he a "':K'('lfic nsk factor in the de\'dopmelll orbinge
eating, Tht, d.at.. abo confinll thaI sexual abuse: ~ more fr~
quelll" repont."(1 in anorexia nenusa binge eating purging
1\1M: (28({) limn 111 anoreX'iOl !1cryosa restricling "pe (7,{).
In COlltr.lSt lO some slUdies (Root & Fallon. 19 : VandC"r­
Iindt'II, 199!S) , a lo\\cr l"ale uf 'oCXual abusc (16%) and 3 high­
er roue or "t-"\-ere 1)S\chol~';C:11 abust:' (29%) were found in
bullmiOl nen·OQ. II i possible. hO\\·c\'Cr. thai subjc."ClS ~;th
bulimi.1 lIen( I.are Ic~ like" lo remember tlleirabusi\'r sex·
tllli ('nCIlUnl(· lh.m olher lr,mmalic experiences due 10 their
dis."OC'ialion trndenC\'. OlS demonstrated by the high scon.."S
on the I>I~.
E:lling di..ordl'r p;Hil'nt~ reponed significantl}' higher
pR"\'alcnce Oftr.UIIl\Ollic experiences in comparison with con­
u'ols; hUl\"c\cr.the 10\\ pre\~alellce ofu,ullnil found in con­
Irub. e"llid hi' l'xplailll'd h) the facl IhOll in this grOllp lfau­
malic l'xpcl'icn,es. \\'ere eV;l.IUOlled only \\'ilh a self-report
qu,,:sliUllllairc and 1I0t wilh a direcl interview.
Nll ~ig"lifif·:l111 difft"'('ncl'~llll Ir:11lmatic experiences were
round bl'!ween l':t1ing disonler Olnd schizophrenic palients,
This re~lIh is ill aCCllI'rl with the finding of Welch and FOli\"­
hlll"ll (1!)!)tI) which ~Ial{'d IIl:t1 till' pn;v:t1cncc of lntlirnilllc
exp":l"it:ncl' i~ ~imibl' among p;ttients wilh ps)'chblfic di.sor­
deI'S.
Till' nl~ ~forn ohtaillcd h)' nul' cOlting diS(jrtlt'I' amI
'Whi/ophl'l'nic ~llbjecL~arc simibr lO Ihe scores reported by
VOllldcrlind":lI Cl oIl. (I~J3). Our findings re\'callhat eOlting
di"oUrdcr pal il'll L~ (t'~pI'rial1)' IhOM' wilh bulimia ncn'Osa and
anflrl'xi,t nl'''ll--:l bill~t, COlling/purging type) report SiJ,,'11ir­
icantl\' highl'" h.'\'ds ofdi~i;auvecxperiences than controls
and M:hi/ophft'llics 'iubjt'C1$. Thc!IC data could seem to sup­
IXlrt Iht.' II\1Xllhc~i~ Ih;11 :I clost:' associiluon exists bel\\·et."n
trauma. di~i<IUOIl.•md bulimic beha\;ors. HO\\'ever, the
hight'r It.',d ufdi~i<Hi\l'wmplOlIlS found in t-"3ung disor­
der p;ttienb ill comparison \\;th control subjec:lS could be
expl..inc.-d ll\ the facl that all tile eaLing disorders patienlS
\\en' ho.'>pit.lli/CtJ. and it i.. \\ell L.nown thaI tile Stress ofhos­
pil.lIi7';lllflll GUI GIII~' dll IIlcrease in dissociative s\lnptoms.
We ha\e tried lO Q\creomc this in comparing hospil3l­
iled ealing dbunler p;uicll1.... \\;Ih hospitalized schizophrentc
"UbJ«lS. Emmg di-.order palif':nts reponed significant high-
279
DhSOCl%nO\. \. LX. \ .. 4. Dm*t-19!li
TRAUMA AND DISSOCIATIVE EXPERIENCES
er scores in the three DI5-Qsubscales: identity confusion and
fragmentation, loss of control, and absorption.
In regard to a possible link between trauma, dissocia­
tion, and eating disorders, the dissociative features which
seem to differentiate eating disorder patients from
schizophrenic patients are identity confusion and fragmen­
tation and loss of control. Absorption is not sensitive to the
presence/absence of trauma. It is noteworthy that binge eat­
ing disorder patients have a high rate of traumatic experi­
ences but show low levels of dissociative symptoms (similar
to the DI5-Q cores for control subjects). A possible expla­
nation is that binge eating disorder patients tend to have a
less severe trauma history (i.e., sexual abuse is present in only
3% of the patients). Another reasoning could be that binge
eating disorder patients resort to overeating and binging:
the consequent weight gain may function as a physical bar­
rier for sexual triggers, hence, they don't need to use a "dis­
sociative coping style." Consequently, by becoming obese and
sexually unattractive, they build a psychological and physi­
cal barrier to distance themselves from others.
In conclusion, our results show that in certain subgroups
ofeating disorder patients (subjects with bulimia nervosa and
anorexia nervosa binge eating/purging type), but not in
binge eating disorder subjects, a high prevalence of trauma
is associated with a high level ofdissociative experiences, espe­
cially identity confusion and fragmentation and los of con­
trol. Our study, however, does not permit us to infer any
causality between trauma, dissociation, and eating disorder
symptomatology, and at this time, any possible causal link
would be speculative. Further research is necessary to under­
stand the role of trauma and dissociation in the etiology and
development of eating disorder, which can only be under­
stood by considering the role ofsuch experiences within the
context of the multifactorial nature ofeating disorders (Gar­
ner,1993).•
REFERENCES
American Psychiatric Association (1987). Diagnostic and Statistical
Manual ofMentalDisorders (3rd ed.-revised). Wa'hington, DC: Amer­
ican Psychiatric Association.
Bulik, C.M., Sullivan, P.F., & Rorty, M. (1989). Childhood sexual
abuse in women with bulimia. Journal ofClinical Psychiatry, 50, 460­
464.
Bushnell,j.A., Wells,j.E., & Oakley-Browne, M.A. (1992). Long­
term effects of intrafamilial sexual abuse in childhood. Acta Psy­
chiatrica Scandinavica, 85, 13&-142
Chandara, P., & Malia, A. (1989). Bulimia and dissociative'~tates:
A case report. umaditLnJournal ofPsychiatry, 34, 137-139.
280
Connors, M.E., & Morse, W. (1993). Sexual abuse and eating dis­
orders: A review. InternationalJournal ofEating Disorders, 13,1-11.
Demitrack, M.A., Putnam, r.w., Brewerton, T.D., Brand, H.A., &
Gold, P.W. (1990). Relation ofclinical variables to dissociative phe­
nomena in eating disorders. AmericanJournalofPsychiat,y, 147, 1184­
1188.
Everill,j.T., & Waller, G. (1995). Reported sexual abuse and eat­
ing psychopathology: A review of the evidence for a causal link.
InternationalJournal ofEatingDisorders, 18, 1-11.
Garner, D.M. (1993). Pathogenesis ofanorexia nervosa. Lllnce~ 341,
1631-1635.
Goldfarb, L. (1987). Sexual abuse antecedent ofanorexia nervosa,
bulimia and compulsive overeating: Three case reports. International
Journal ofEatingDisorders, 6, 675-680.
Hall, R., Tice, 1.., Beresford, T.P., Wooley, B., & Hall, A.K (1989).
Sexual abuse in patients with anorexia nervosa and bulimia. Psy­
chosomatics, 30, 73-79.
Heatherton,T.F., & Baumeister, R.F. (1991). Binge eating as escape
from self-awareness. Psych"'giwl Bulletin, 110, 8&-108.
Hechter, l.D., Schwartz, H.P., & Greenfield, D.G. (1987). Sexual
assault and anorexia nervosa. InternationalJournal ojEating Disor­
ders, 6,313-316.
Marcus, M.D., Wing, R.R., E,ving, 1.., Kem, E., Gooding, W., &
McDennott, M. (1990). Psychiatric disorders among obese binge
eaters. InternationalJournal ofEating Disorders, 9, 69-77.
McCarthy, M.H., Goff, D.C., Baer, 1.., Cioffi, j., & Herzog, D.B.
(1994). Dissociation, childhood trauma, and the response to flu­
oxetine in bulimia patien ts. InternationalJournal ofEating Disorders,
15,219-226.
McFarlane,A.C., McFarlane, c., & Gilchrist, P. . (1988) Post-trau­
matic bulimia, and anorexia nervosa. InternationalJournalofEating
Disorders, 7, 705-708
Oppenheimer, R., Howells, K., Palmer, L., & Chaloner, D. (1983).
Adverse sexual experiences in childhood & clinical eating disor­
ders: A preliminary description. Journal ofPsychiat,ic Researcl~ 19,
157-161.
Palmer, R.L., Oppenheimer, R., Dignon, A., Chaloner, D.A., & How­
ells, K (1990). Childhood sexual experiences with adults report­
ed bywomen with eating disorders: An extended series. BritishJour­
nalofPsychiatry, 156,699-703.
Pope, RG., & Hudson,j.!. (1992). Is childhood sexual abuse a risk
factor for bulimia nervosa? AmericanJournal ofPsychiatry, 149, 455­
463.
Pribor, E.F., & Dinwiddie, S.H. (1992). Psychiatric con-elates of
incest in childhood. AmericanJournal ofPsychiat,y, 146,52-56.
DISSOCLmo~, 1'01. IX. ~o. 4. December 1996
Root. M.P.• &: F;tllon, P. (1988).11lC~ incid~nc~of\;ctiI:nization cspc­
ri~ncc in a bulimic: ~"unplc.juumal ofInlnpt"rWnal l'iolma. J. I(i 1­
173.
RO:ICIl. t:.F.. &: I'clIy.I_G. (1!)9·I). DissodathcM;llcunddisordercd
c.tting. A"'I'nrlmjoumal ofQlnufll II)pnQsu. 16, 26&275.
R~II.G. (1979). 8ulimia n~!"'OQ:An ominous \";lrUm of;mort,,-­
U nc!"OQ... PJTr~aJi J'td,nlU'. 9. 429-+19.
">kwl. G.• &: l.C'i~ltlle1'". P. (1986). Is IhCl"~'" rd.u.ionship ~"ttn
"iCStl:t1 ...!lust or inroc :.tnd c:.tung dtsomCl"l? Csrurdinll jov",a/ rJj
ITirIfUllrr. JI. 65&660.
':tptl7.cr. R. L. Y;tIlO\~I:i.S.. W.lddell. T .. Wing. R.., Marcus. M.O..
Stunla.rd. A... Ot.... lill. M.. Mitchell.J.. 1I,!Sin.D.. 8: Home. R.L
(1993). Bing~ ~atitlg disorder: Its funhcr \-alid;ttion in a Illuhisitc
\IIKh./nvmatlOnQl.JoumalofJ-...DIIN~~11. 137-153.
ro~. M.s.« 1986). Dissoci:&tn~SUle'l pr~ting as an eating db­
<lHSer. ..t.".canjo,mu.J ofo,nu41 Ihf1nO$U. 29. 137-142_
Torc:m, ),1.5. (1987). EglH.... tC' ther.tfl\ for (,,".oiling disorder. A.".­
,anJovnwI t/o,n,ml 1I)'JJntnU., 10.94-103.
\'oUl del' Koll:.Ii.A..I'clT)'. C.• &: Ilcnn.m.J.L (1991). Childhood
III igins of !>Clf-c:lc~lructi\'c l>ch.nior. "'nvrirllli jtlllnuJI Qf Psychltltf)·.
148.1665-1671.
\andcrlindcn.j.. Nom:.J .. &: \'andcrercl:~n. W. (1992). A prodl­
ralguwwliu tn:ot."., tJjlJul;IIlUl 'U'f'l.'OMl. Nt....· '1'01'1:.: Brunn~1 Mud.
\.mdcrtindcn.j. (1993:1). DruonatlVl'~m.. lroUJPll'l and Ity/Jn~
Ul Rrs«Jrchfindrnp mid tlrnua/ appluut'QIU /II MUngdisortUrs. Oem:
Hluron
\ ,tIldcrlindcll,J.. V;mdc'rc}'ckcn. \\7.. V:m D)'cl:. R.. & Venom men.
J I. (I ~m!lh). Oissocialil'c cxperi!:nces and tl~11l1l1:1 ill calillg disor'­
(kr~. 11I1",ltIt;tJ/ltllJour/wi of1\"/;/11{ J)inmirol, 1J, IH7·194.
\\,Jllcr. G. (1991). Sexual :Ibusc :I.'l:l risk f;tclor in e;uing disordcr5.
flntuh jQU.mal of l'sJrhlfltTJ. 1'9. 6&1-671.
\\.llIer. G.. Ruddocl. A_. &: Pitu.. C. (1993). Wh~n is SoCxual abl.lSe
rdC"'o-anl 10 buhllltc di50rd~n?Ti,e \";llidit\ ofdiniC2ljudg~m~nt5.
I'uropmn J-Atmg J),J.Ordn &1!lm', J, 1"~151.
\\l'kh, S.L. & t·airbum. G.G. (l9'-J·I). Se"ual abuse and bulimia
ne!""Usa: Three inlcgr:lted C".t.SC control comp.lri'l()ll~.AMDi{lJlljlJllr ­
nlll of l'J:yclIlUtT1. 151. 402-407.
Y.lnO\~li. S.Z.. N~bon.J.E.. Oubhen, 8K. & Spil7.~r. R.L (1993).
\uoo:uion or bmge ~Iing disordn and )lS'"'Chialrlc comorbidin
III oboe ~bJ«ts. ,b",ua"javneal DfI'YJdliat". "0, H72-14i9.
281