PRELIMINARY
RESULTSOF THE FIELD
TRIAL OF PROPOSED
CRITERIA FOR
DISSOCIATIVE DISORDER
OF CHILDHOOD
Gary Peterson, M.D.
Frank W. Putnam, M.D.
Gary Peterson, M.D., is Clinical Associate Professor of
Psychiatry at the Program on Childhood Trauma at the
University of North Carolina School of Medicine, Chapel
Hill, North Carolina.
Frank W. Putnam, M.D., is Chief of the Dissociative Disorders
Program, Laboratory of Developmental Psychology, National
Institute of Mental Health, Bethesda, Maryland.
For reprints write Gary Peterson, M.D., Program on Childhood
Trauma, University of North Carolina School of Medicine,
Chapel Hill, NC 27599-7160.
ABSTRACT
Despite recent increased awareness of child and adolescent disso-
ciative disorders, the lack of an official "sanction" by the inclusion
of specific diagnostic entities intheDSM-III-R/W has seriously imped-
ed the acceptance of these conditions in youthful population. There
are also widespread concerns that the adult-oriented DSM criteria
are often inappropriate for child and adolescent cases. To correct
these deficiencies, the diagnosis ofDissociativeDisorder of Childhood
(DDoC) has been proposed. A preliminary, questionnaire-based, field
trial was initiated to test the proposed DDoC criteria. Results indi-
cate that the DDoC criteria identify a group of children who do not
meetDSM-III-R/ IV
nonetheless have significant levels of dissociative symptoms. These
criteria also serve to discriminate children with DDoC from another
group of children, frequently labeled with DissociativeDisorderNOS,
who have a different clinical picture. Based on these data, DDoC cri-
teria will be further refined.
INTRODUCTION
The last decade has seen a marked increase of interest in
the dissociative disorders, primarily multiple personality dis-
order (MPD). Initially, clinical attention and research stud-
ies focused almost exclusively on dissociative disorders in
adults. More recently attention gradually has begun to turn
towards the manifestations of pathological dissociation in
children and adolescents. This refocusing reflects the increas-
ing realization that MPD, in particular, appears to be a chron-
ic condition that first develops during childhood as a result
of severe, and usually repetitive trauma (Kluft, 1984; Bliss,
1986; Putnam, 1989). Kluft (1984, 1985) has enumerated
the reasons for the belated recognition of dissociative dis-
orders in children and adolescents including the lower index
of suspicion for these conditions among clinicians who treat
children and the fact that dissociative symptoms in children
are often attributed to other commonly diagnosed child-
hood conditions (such as conduct problems, attention-deficit
hyperactivity disorder and depression) . In the last five years,
a growing literature, currently totalling approximately 100
cases, has begun to detail the clinical phenomenology of dis-
sociative disorders in youth (e.g., Fagan McMahon, 1984;
Kluft,1984,1985; Bowman, Blix Coons, 1985; Weiss, Sutton,
Utecht, 1985; Riley Mead, 1988; Vincent Pickering,
1988; Dell Eisenhower, 1990; Hornstein Putnam, 1992).
Childhood dissociative disorders still have not received
much attention in standard texts nor, despite the vigorous
efforts of concerned colleagues, have they been included in
currently planned epidemiological surveys of childhood men-
tal illness. Frontline clinicians continue to be the major source
of information on dissociative disorders in youth. In many
instances, pathological dissociation in children and adoles-
cents is different from allied phenomena in adults (Putnam,
1993). Recognizing these differences, early pioneers often
qualified their descriptions of child cases with terms such as
"incipient multiple personality" (Fagan cMahon, 1984),
or "MPD in evolution" (Malenbaum Russel, 1987) . As addi-
tional cases have enlarged our perspectives, it has become
apparent that many children and adolescents have signifi-
cant levels of dissociation that fall short of clear-cut MPD and
are poorly characterized by the examples and criteria for
the diagnosis of dissociative disorder not otherwise speci-
fied (DDNOS).
Both authors have been repeatedly contacted by pro-
fessionals seeking advice on the applicability ofDSM disso-
ciative diagnoses in children and adolescents. Discussions
with child clinicians, particularly social workers and psychi-
atric nurses working with maltreated children, reveal that
many are afraid or unwilling to make the diagnosis of MPD
or some other dissociative disorder. A number of reasons
are commonly offered. It is frequently observed that, although
dissociative, the child in question does not fit theDSM-III-R
criteria for MPD. Many complain that the lack of any refer-
ences in the DSM to childhood dissociative disorders make
their supervisors reluctant or unwilling to consider these
diagnoses. They fear that third-party payers will not accept
these diagnoses in children without an official sanction.
Some consider the diagnosis of MPD too bizarre and stig-
matizing to give to a child.
There are many reasons for having a child-oriented dis-
sociative disorder diagnosis in the taxonomy: 1) to have
212
DISSOCIATION, Vol. VII. No, 4, December 1994
PETERSON/PUTNAM
a more accurate and developmentally appropriate diagnos-
tic description; 2) to alert clinicians to dissociative disorders
in children; 3) to avoid the stigma of an MPD diagnosis; 4)
to use as an intermediate diagnosis if an MPD diagnosis is
suspected but not established for some period of time; 5) to
administratively sanction and justify therapy directed toward
the resolution of dissociative symptoms in children and ado-
lescents; and 6) to encourage research on childhood disso-
ciative disorders.
History of the Proposed Diagnosis of "Dissociative Disorder
of Childhood"
In1989 the first author (G.P.) presented a paper describ-
ing a constellation of symptoms that seemed to capture the
nature of dissociative disorder in childhood (Peterson,
1989). He designated this constellation of symptoms as
"Dissociation Identity Disorder" (DID). The DID symptom
complex included three categories of symptoms and a rule-
out criterion: A) amnestic or trance-like experiences; B)
marked behavior fluctuations; C) behavioral and other
symptoms which are frequently seen in other disorders; and
D) the subject does not qualify for a DSM-III-R/IVdiagnosis
of multiple personality disorder.
During and after the 1989 meeting, many clinicianswork-
ing with children with dissociative disorders were encour-
aged and excited about the prospect of having a formally
recognized, child-oriented dissociative disorder diagnosis.
In February of 1990, the first author (G.P.) wrote to theDSM-
IV Task Force to petition for consideration of such a diag-
nosis. There are precedents in the system for diagnoses
which have one form in childhood and another in adult-
hood. For instance, the DSM-III-R includes childhood diag-
noses of conduct disorder, avoidant disorder, identity dis-
order, and overanxious disorder, which are thought to be
precursors of antisocial personality disorder, avoidant per-
sonality disorder, borderline personality disorder, and gen-
eralized anxiety disorder respectively (APA, 1987).
In 1990, the first author (G.P.) was appointed to the
American Psychiatric Association DSM-IV Task Force
Dissociative Disorders Work Group as a child psychiatry liai-
son advisor representing the American Academy of Child
and Adolescent Psychiatry. The second author (F.P.) was
already a member of the Dissociative Disorders Work Group.
The concept of DID was not received with great enthusiasm
by the Chair of the work group. He suggested that to be con-
sistent with his nomenclature the name should he changed
to include "dissociative" rather than" dissociation." In order
to conform with his terminology, DID was changed to
"Dissociative Disorder of Childhood " (DDoC). The first
author (G.P.) drafted the proposed text and criteria. In
November, 1990, he met with the second author (F.P.) and
Francis Waters, A.C.S.W., to finalize them. The DDoC pro-
posed criteria are listed in Table 1.
Over the next few years vigorous efforts were made to
interest the Dissociative Disorders Work Group in theDDoC
concept. However, in the DSM-IV Options Bookthere was no
mention of child, adolescent or youth in the dissociative dis-
orders section (American Psychiatric Association, 1993).
TABLE 1
Proposed Diagnostic Criteria for Dissociative
Disorder of Childhood
A) A disturbance of at least six months during which
either one or two of the following are present:
1) Recurrent amnestic periods or missing
blocks of time
2) Frequent trance-like states or appearing to
be in a daze or in another world
B) Perplexing, major fluctuations in behavior which
include at least two of the following:
1) Dramatic fluctuations in school or work
performance and behavior
2) Variations in apparent social, cognitive, or
physical abilities
3) Sudden, recurrent shifts in friendship
patterns
4) Changes in language, accent, and voice tone
5) Perplexing changes in preferences for
clothes, food, toys, games, etc.
C) At least three of the following:
1) Refers to self in third person or uses
another name to refer to self or parts
2) Has vivid imaginary companionship
3) Frequently disavows observed behavior
4) Exhibits frequent inappropriate sexual
behaviors or is sexually precocious
5) Has intermittent depression
6) Has auditory hallucinations from inside the
head
7) Has frequent sleep problems
8) Exhibits unprovoked explosive anger and
violent behavior
9) Exhibits other antisocial behaviors
D) Does not meet the criteria for Multiple Personality
Disorder.
213
DISSOCIATION. Vo1I1. No. I. December
DISSOCIATIVE DISORDER OF CHILDHOOD
TABLE 2
Mann-Whitney "U" Pairwise Comparisons of Factors and CDC Score
(Tie-Corrected Z score and p Value)
FACTORS MPD vs DDoC DDoC vs ODD
Attention Hyperactivity -.81NS -1.78 NS
Anxiety Phobias -.73 NS -1.60 NS
Depression -.04 NS -2.04p=.04
Eating Problems -.82 NS -.25 NS
Identity Disturbances -.81 NS -1.66 NS
Dissociative Symptoms -1.07 NS -1.66 NS
School Problems -.28 NS -2.O8p=.03
Thought Disorder Sx -.27 NS -2.05p=.04
Aggression -.44 NS -2.35p=.01
Amnesia -2.04p=.04 -1.34 NS
Conduct Problems -1.51 NS -1.52 NS
Hallucinations -.02 NS -1.53 NS
OCD Symptoms -.31NS -.59 NS
PTSD Symptoms -.17 NS -1.61 NS
Sleep Disturbances -.46 NS -2.27p=.02
Sexual Problems -.64 NS -3.04 p = .002
CDC Score -1.95 p = .05 -2.12p=.03
* Significant at p < .05 after Bonferroni correction for number of comparisons
Ironically, we were more successful in informing otherDSM-
IV work groups about the need to include dissociative dis-
orders in their rule-out differential diagnosis sections. Our
efforts did result in the inclusion of a mention of children
in the rule-out criterion for MPD (renamed "Dissociative
Identity Disorder" in the DSM-IV) and the inclusion of dis-
sociative disorders in the rule-out criterion for attention-
deficit/hyperactivity disorder. DDoC was not accepted into
DSM-IVasa diagnostic category, nor was it listed in the appen-
dices as a diagnosis for future consideration (APA, 1994).
Purpose of the Study
The purpose of this study was to provide a preliminary
test of the applicability of the DDoC diagnostic approach to
dissociative disorders in youth. In particular, we were inter-
ested in determining if the DDoC criteria would identify a
group of children and adolescents who had significant dis-
sociative symptoms but did not meet DSM-III-Rcriteria for
MPD. We were also interested in determining if there were
children and adolescents, thought to have significant dis-
sociative symptoms, who did not meet criteria for either MPD
or DDoC. Finally, we were interested in determining which
DDoC criteria appeared to be the most sensitive to non-MPD
dissociative disorders in youth.
There were serious time and funding constraints on this
study. We were interested in having data supporting the DDoC
concept to present to the DSM-IVDissociative Disorders Work
Group. No funds were available to support the project, nor
was there sufficient time to apply for funding. These con-
straints ruled out a clinical study involving structured inter-
views or similar approaches to the accrual and characteri-
zation of a clinical research sample. For example, it took
several years for Hornstein and Putnam to collect a com-
bined clinical sample of 64 child and adolescent dissocia-
tive disorder patients (Hornstein Putnam, 1992) . Drawing
upon the second author experience with clinician ques-
tionnaire studies (e.g. Putnam, Guroff, Silberman, Barban,
Post, 1986; Putnam Loewenstein, 1993), it was decided
to adopt a therapist questionnaire approach as a first order
approximation of the suitability of the proposed DDoC cri-
teria.
214
DISSOCIATION. Vol. Vii. No. 4. Decemher 1994
PETERSON/PUTNAM
TABLE 3
Contingency Table Comparison of Proposed DDoC
Criteria by Diagnostic Group
(Percent of cases positive for DDoC Criterion)
Criterion Percent MPD Percent DDoC Percent ODD Chi Sq. p Value
Al 75.4% 42.8%
A2 92.4% 94.2%
B1 88.6% 94.2%
B2 86.7% 91.4%
B3 35.8% 37.1%
B4 83.0% 60.0%
B5 47.1% 8.5%
Cl 71.7% 40.0%
C2 41.5% 28.5%
C3 71.5% 62.8%
C4 49.0% 51.4%
C5 75.4% 74.2%
C6 82.6% 60.0%
C7 67.9% 74.3%
C8 66.0% 62.9%
C9 56.6% 45.7%
23.0% 16.3, p = .0003
84.6% 1.2p=NS
53.8% 13.5, p - .001
7.6% 43.8, p = .0001
15.3% 2.2, NS
30.7% 14.8, p = .0006
0% 21.3,p=.0001
38.4% 10.6, p = .005
23.0% 2.4, p = NS
23.0% 10.4, p = .005
15.3% 5.5, p = NS
30.7% 10.4, p .005
23.0% 17.9, p = .0001
38.5% 5.5, p = NS
30.7% 5.5, p = NS
7.6% 10.0, p = .006
* Significant at p < .05 after Bonferroni correction far the number of comparisons.
METHODS
Questionnaire Design
Individual cases of children and adolescents with dis-
sociative symptoms were collected by means of a question-
naire completed by a clinician involved with the case. In
most instances the questionnaire was completed by the childs
primary therapist. The questionnaire was derived from the
data base developed by Putnam and Hornstein to facilitate
data collection in their two-site study of the clinical phe-
nomenology of child and adolescent dissociative disorders
(Hornstein Putnam, 1992).
The DDoC questionnaire consists of six parts. Part I
includes an introductory section explaining the purpose of
the study, confidentiality safeguards, information about the
clinician completing the questionnaire, demographic infor-
mation on the patient, and information about setting and
purpose of the clinician contact with the child.
Part II contains a 107-item symptom checklist inquiring
about behavioral problems, affective symptoms, dissociative
symptoms, cognitive and thought disorder symptoms, post-
traumatic symptoms, self-injurious and aggressive behaviors,
school problems, anxiety symptoms and miscellaneous symp-
toms. For each item on the checklist, the clinician checked
one of four possible options: "present," "past," "never," or
"unknown." Part II ends with a brief section inquiring about
full scale, verbal and performance IQ data, and the name of
the IQ measure the patient had been given.
Part III inquires about the child history of trauma and
maltreatment. The clinician is asked to indicate whether the
child has experienced physical abuse, sexual abuse, neglect
or abandonment, emotional abuse, death of a parent, domes-
tic violence, or other traumas that could be specified by the
respondent. For each trauma, the clinician was asked to indi-
cate his/her degree of certainty ("well documented," "sus-
pected," or "unknown" ) about the actual occurrence of the
alleged experience.
Part IV collects information on current and past DSM-
III-R diagnoses received by the child, and information on
substance abuse patterns. Part V collects information on the
215
DISSOCIATION. Vol. Cll. No. t. December 1994
DISSOCIATIVE DISORDER OF CHILDHOOD
TABLE 4
Mann-Whitney "U" Pairwise Comparisons of Proposed DDoC Criteria
(Tie-Corrected Z score and p Value)
Criterion MPD vs DDoC DDoC vs ODD MPD vs ODD
Al -3.07, p = .002 -1.08, p = NS -3.52, p =.0004
A2 -.33, p NS -1.16, p NS -.87, p = NS
B1 -.89,p = NS -3.48, p = .0005 -2.89, p = .003
B2 -.66, p = NS -5.37, p -.0001 -5.60, p = .0001
B3 -.12, p = NS -1.28, p = NS -1.40, p = NS
B4 -2.39, p = .01 -1.58, p = NS -3.76, p = .0002
B5 -3.78, p = .0002 -1.03, p = NS -3.11,p=.001
Cl x.94, p = .003 -.10, p = NS -2.23, p = .02
C2 -1.22, p = NS -.23,p=NS -1.21, p = NS
C3 -.86,p=NS -2.24, p = .02 -3.21,p=.001
C4 -.21,p=NS -2.59, p = .009 2.18,p=.02
C5 -.12, p = NS -2.52, p = .01 -3.04, p = .002
C6 -2.34, p = .01 -2.07, p = .03 -4.19, p = .0001
C7 -.63, p = NS -2.03, p = .04 -1.94, p = .05
C8 -.39, p = NS -1.75, p = NS -2.30, p = .02
C9 -.99, p = NS -2.30, p = .02 -3.14, p = .001
*Significant at p < .05after Bonferroni correction fornumber of comparisons
child family including dissociative and other mental dis-
orders in family members. Part VI consists of a checklist for
the proposed criteria for DDoC. The clinician is instructed
to check all of the items that applied to each case. DDoC
items were arranged in the A, B, C, and D groupings pro-
posed by Peterson (1991).
In addition, each clinician is asked to have a parent or
guardian complete a copy of the Child Dissociative Checklist
(CDC) and to return it with the questionnaire. The CDC is
a 20-item parent/adult observer report measure that quan-
tifies dissociative behaviors in children and adolescents. Prior
research has established the reliability and validity of the
CDC (Malinosky-Rummell Hoier, 1991; Hornstein 
Putnam, 1992; Putnam, Helmers, Trickett, 1993;Wherry,
Jolly, Feldman, Adam, Manjanatha, 1994; Putnam, Helmers,
Horowitz, Trickett, in press).
Sampling and Subjects
Copies of the questionnaire were distributed by the first
author (G.P.) under the auspices of the Department of
Psychiatry, of the University of North Carolina. Questionnaire
216
distribution took several forms. Copies were mailed to clin-
icians known to be working with dissociative child and ado-
lescent cases. Questionnaires were mailed to clinicians who
were members of the International Society for the Study of
Multiple Personality and Dissociation (ISSMP), a profes-
sional organization devoted to understanding and treating
patients with dissociative disorders. Copies of question-
naires were also distributed during professional workshops
and lectures on the subject of childhood dissociative disor-
ders given by the first author (G.P.) and his associates.
Distribution of questionnaires occurred over the period July,
1991 to July, 1993. Approximately 500 copies were distributed,
but the precise number is not available because some respon-
dents copied the questionnaire and redistributed it within
their clinical networks. One hundred and twenty-three com-
pleted questionnaires were returned by 76 clinicians. From
this sample, 102 questionnaires were selected for data anal-
ysis. Study sample questionnaires were selected such that no
clinician contributed more than three cases (2.9%) to the
sample. Whenever possible, a CDC completed by a parent,
guardian or other independent observer was also available
DISSOCIATION, vol. VII, \o. 4. December 094
PETERSON/PUTNAM
TABLE 5
Mann-Whitney "U" Pairwise Comparison of Summation of DDoC Criteria
(Tie-Corrected Z score and p Value)
Criteria Sum MPD vs DDoC
Sum of A Items -2.72, = .006
Sum of B Items -2.13, p =.03
Sum of C Items -1.98, p =.04
Total of All Items -2.89, p =.003
* Significant at p < .05 after Bonferroni correction for the number of comparisons
DDoC vs ODD MPD vs ODD
-1.47, p = NS -3.07, p = .002
-4.48, p = .0001 -4.60, p = .0001
-4.18, p = .0001 -4.77, p = .0001
-4.84, p = .0001 -5.11, p = .0001
(89 cases, 87.3%).
Data Analysis
Using the DDoC criteria in Part VI, cases were catego-
rized into three groups. Cases were given a diagnosis of Multiple
Personality Disorder (MPD) if the clinician indicated that
the childmetDSM-III-R criteria forMPD (DDoC Criterion D).
Cases were given a diagnosis of Dissociative Disorder of
Childhood (DDoC) if they did not meet criteria forMPD but
did satisfy DDoC criteria. Cases were given a diagnosis of
other dissociative disorder ( ODD) if they suffered a disso-
ciative disorder but did not satisfy criteria forMPD or DDoC.
Data were analyzed by these three diagnostic categories.
The large number of symptoms inquired after in Part
II necessitated grouping the many items into factors.
Following the model developed by Hornstein and Putnam
(1992) , behaviors and symptoms from Part II were grouped
into 16 factors. A table of the Part II items composing each
of these factors is available in Hornstein and Putnam (1992) .
The factors are: 1) attentional and hyperactivity problems;
2) amnesias and memory disturbances; 3) anxiety and pho-
bic symptoms; 4) conduct problems; 5) depression; 6) eat-
ing symptoms; 7) hallucinations; 8) identity disturbances;
9) obsessive and compulsive behaviors; 10) dissociative
symptoms; 11) post-traumatic stress symptoms; 12) school
problems; 13) sexual behaviors; 14) sleep problems; 15)
thought disorder and cognitive dysfunction symptoms; and
16) aggression toward self and others. Factor scores were
computed by adding up the number of items checked as
"present" or "past" for each factor. A trauma index, con-
sisting of the number of Part III trauma items checked as
"well documented," was computed for each case.
Statistical analysis was conducted using Statview II soft-
ware. All factor and CDC data were analyzed non-paramet-
rically, though group means are included to characterize
response patterns. Pairwise comparisons of factors andCDC
scores were analyzed with the Mann Whitney "U" test. Three-
way comparisons were analyzed with the Kruskal-Wallis H
statistic.Contingency table analyses with a Yates corrected
Chi Square statistic were used to compare proportions of
MPD, DDoC, and ODD cases with respect to a given variable.
Tables 2 - 5 contain Bonferroni corrections for the number
of variables compared in an analysis.
RESULTS
Demographic Comparisons
After imposition of the case selection rules described
above, the final sample of 102 cases consisted of 67 (65.6%)
females and 35 (34.3%) mates. The mean age was 11.2 4.6
(range 3.0-19.1) yearswith no significant difference between
males and females. The sample was contributed by a variety
of professional disciplines. Psychiatrists contributed 29.4%,
MSW social workers 28.4%, Ph.D/PsyD psychologists 18.6%,
master level therapists 10.8%. Miscellaneous other profes-
sionals contributed 12.8%. Using the proposed DDoC clas-
sification criteria, there were 54 MPD cases (52.9%) , 35DDoC
cases (34.3%) and 13 ODD cases (12.7%). There were no
significant differences in the gender or age distributions by
diagnosis.
The majority of cases were seen in outpatient settings
(73.5%) , followed by inpatient evaluations (15.7%), "other"
situations (8.8%), and clinics (1.9%). Most of the children
were seen for psychiatric or behavioral problems (80.4%),
followed by social service or child protection evaluations
(7.8%) , school-related evaluations (5.8%), custody and legal
evaluations (1.9%), and other reasons (5.8%). There were
no significant differences in the reasons for evaluation or
setting by age, gender or diagnostic group.
Clinical Factors
The focus of this study was to determine if theDDoC cri-
teria proposed by Peterson (1991) delineate and appropri-
ately characterize a group of children and adolescents who
suffer from dissociative disorders that are distinct from mul-
tiple personality disorder. In the past, such cases have been
lumped under the poorly defined diagnosis of Dissociative
Disorder Not Otherwise Specified (DDNOS). Table 2 pre-
217
DISSOCIATION, Vol. VII. No. 1, December 1994
DISSOCIATIVE DISORDER OF CHILDHOOD
sents the pairwise comparisons of the three diagnostic
groups on the sixteen factors and CDC scores.
In addition to the presence or absence of alter person-
alities (criterion D) , MPD and DDoC cases can be distinguished
from each other by differences in degree and types of amne-
sia and by CDC scores (MPD mean = 22.6 7.3; DDoC mean
= 19.25 7.3; Z -1.95, p = .05). There are no significant dif-
ferences between these two groups on any of the other fac-
tors. The DDoC cases had significantly less amnesia for gen-
eral material (Z = -2.83, p =.01) and fewer experiences of
lost time (Z = -2.05, p = .03) but did not differ for amnesias
related to trauma or amnesias related to specific events or
circumstances.
The DDoC cases were clinically distinguishable from the
ODD cases in that they had significantly more depression,
school problems, thought disorder and cognitive symptoms,
aggression, sleep difficulties, and sexual behavior problems.
Their CDC scores were also significantly higher (DDoC mean
= 19.25 7.3; ODD mean = 14.0 5.0; Z = -2.12, p = .03).
Interestingly, the DDoC cases were not significantly more
dissociative, amnesic, or post-traumatic than the ODD cases.
The MPD cases differed significantly from the ODD cases
on a large number of factors including: depression, identi-
ty disturbances, dissociative symptoms, school problems,
thought disorder and cognitive symptoms, aggression, amne-
sias, conduct problems, sleep disturbances, sexual behavior
problems, and CDC scores.
Trauma Histories
There were no statistical differences in the trauma his-
tories of the three groups. The mean number of "well doc-
umented" trauma items was essentially the same across all
groups (MPD 3.34 1.3, DDoC 3.07 1.5, ODD 3.09 1.3;
Kruskal-Wallis tie-corrected H = .77, p = NS) . Although MPD
subjects tended to have earlier ages of onset for the various
traumas, the age differences were not significant and many
clinicians indicated their uncertainty as to the exact age of
onset of the trauma.
Proposed DDoC Criteria
Table 3 presents the percentage of subjects in each diag-
nostic group who were positive for each of the proposed
DDoC criteria. Inspection of this table reveals that a number
of proposed DDoC criteria do not significantly discriminate
among the three groups: i.e., A2 (trance states), B3 (recur-
rent shifts in friendship patterns), B5 (perplexing changes
in preferences for clothing, food, toys, games etc), C2 (vivid
imaginary companionship) , C4 (inappropriate or precocious
sexual behavior), C7 (auditory hallucinations), C8 (frequent
unprovoked explosive anger and violent behavior). Table 4
presents the Mann-Whitney "U" pairwise comparisons of each
DDoC criterion for all combinations of diagnostic groups.
Again, criteria A2, B3, C2 and C8 do not significantly dif-
ferentiate between two or more diagnostic group combina-
tions. Only one criterion, C6 (sleep problems), is signifi-
cantly different across all three groups. The other criteria
are significantly different across two diagnostic combinations.
Another approach to this question is to simply sum the
number of DDoC items checked and see if these sums dif-
ferentiate the three groups. Thisworkswell in most instances.
The total sum of positive A, B and C criteria readily differ-
entiates the three groups (MPD mean = 10.7 2.8, DDoC
mean = 9.2 1.9, ODD mean - 4.46 2.1, Kruskal-Wallis tie-
corrected H = 34.77, p= .0001) . When considered separately,
the sums of A (MPD mean - 1.6, .6, DDoC mean = 1.3 .5,
ODD mean = 1.1 .6, Kruskal-Wallis tie-corrected H = 13.15,
p = .001), B (MPD mean = 3.3, 1.2, DDoC mean = 2.9 .7,
ODD mean =1.01.1, Kruskal-Wallis tie-corrected H = 27.10,
p = .0001), and C (MPD mean = 5.7, 1.7, DDoC mean = 5
1.6, ODD mean = 2.3 1.5, Kruskal-Wallis tie-corrected H
= 26.5, p = .0001) items each differentiate the three groups.
Table 5 presents the diagnostic pairwise comparison of the
sum of positive A, B, C and total of all DDoC items. This
approach was used in in attention deficit hyperac-
tivity disorder but was abandoned in favor of requir-
ing the core symptoms (inattention or hyperactivity-impul-
sivity).
Past and Present Diagnoses
Inspection of the non-dissociative symptom factor scores
shows high rates of affective and post-traumatic symptoms,
school and conduct problems and aggression relative to unpub-
lished data on non-dissociative disorder children and ado-
lescents (Putnam, unpublished data). An examination of
the past and current psychiatric diagnoses reported in Part
N of the questionnaire shows that 36% of the sample had
received a major affective disorder diagnosis, 33% had received
one or more of the disruptive behavioral disorder diagnoses
(conduct disorder, oppositional defiant disorder or atten-
tion deficit hyperactivity disorder), 47% of the sample had
received a diagnosis of PTSD, and 10% had received an anx-
iety disorder diagnosis. In addition, there was a wide array
of learning disabilities and adjustment disorder diagnoses.
Although MPD and DDoC cases tended to acquire more diag-
noses than ODD, these differences were not significant. This
pattern of multiple diagnoses is similar to that reported by
prior investigations (Hornstein Putnam, 1992; Putnam,
1993).
DISCUSSION
Limitations of the Study Design
By necessity, this study adopted a questionnaire format
to investigate the validity of the DDoC approach to dissocia-
tive disorders in children and adolescents. Although useful
in situations in which one wants to survey an issue or prob-
lem rapidly and inexpensively, questionnaire studies face a
number of methodological limitations (Putnam, Guroff,
Silberman, Barban, Post, 1986; Putnam Loewenstein,
1993) . Our sample is not random and represents the patients
of therapists who are interested in dissociative disorders in
children and adolescents. Questionnaires do not permit the
degree of standardization of assessment possible with struc-
tured interview-based clinical studies. We are dependent on
the judgment and understanding of the clinicians completing
the questionnaire. We do not know how precisely defini-
218
DISSOCIATION. Vol. VII. No. 4, December 1991
PETERSON/PUTNAM
tions of symptoms were applied in a given case. Nor did we
have an opportunity to independently verify the diagnoses
or establish interrater reliability on the use of the symptom
and DDoC criteria checklists. The lack of a non-dissociative
comparison group, although not uniquely the result of a
questionnaire design, also limits the larger generalizability
of our findings. These limitations should be kept in mind
by the reader as they constrain the conclusions that we can
draw from our data.
Validity of the DDoC Criteria Approach to Non-MPD
Dissociative Disorders
Nonetheless, our results suggest that a diagnostic
approach based on a DDoC model to non-MPD dissociative
disorders in children and adolescents is useful. The DDoC
criteria were able to identify a substantial number of cases
(34.3% of the sample), who, in the opinion of their thera-
pists, failed to fulfill DSM criteria for MPD. These cases had
clinically significant levels of dissociation based on their ques-
tionnaire dissociation factor scores and on their Child
Dissociative Checklist scores, completed by an independent
observer.
Many of the ODD cases also had elevated levels of dis-
sociative symptoms and did not statistically differ from the
DDoC cases on the questionnaire dissociative factor. This
is not surprising given that all subjects were selected by their
therapists because of their dissociative symptoms. ODD sub-
jects did have significantly lower CDC scores thanDDoC and
MPD subjects. A number of DDoC criterion items also sig-
nificantly differentiated DDoC and ODD subjects. In gener-
al MPD subjects were robustly different from ODD subjects
on virtually all measures of dissociation included in this study.
These results indicate that the DDoC criteria can partition
a highly dissociative sample into a three groups, which dif-
fer on an independent measure of dissociation (the CDC)
and on a number of non-dissociative clinical factors. The
clinical utility of these distinctions remains to be established.
Recommendations and Future Directions
It is apparent that the opportunity to influence the
IV has passed and future attempts to include a childhood
dissociative disorder diagnosis must look to the next revi-
sion of this influential document. DDoC criteria will be fur-
ther refined based on the results of this study and clinical
experience. For instance, it is becoming increasingly clear
from our unpublished data that for school age children, the
difference between the imaginary companions of children
with dissociative disorders and normal imaginary compan-
ionship is the degree to which the child cannot distinguish
that the imaginary companion is not "real." Future studies
will permit a better determination of the age/developmen-
tal stage "norms" for DDoC criteria such as imaginary com-
panionship.
During the last few years we have come a long way in
understanding childhood dissociative disorders. These con-
ditions are being increasingly recognized in a variety of clin-
ical settings by a range of primary clinicians and therapists.
This is good, but it is far from sufficient. Further acceptance
of dissociative disorders in children is going to require bet-
ter clinical descriptions and more appropriate diagnostic
criteria. TheDDoC field trial and our associated political lob-
bying with the DSM work groups made small but tangible
gains in informing the larger fields of psychiatry and clini-
cal psychology of the need to consider dissociative disorders
in youth. Future gains will require better information on
these conditions in youth and a more concerted and orga-
nized lobbying effort on the behalf of children afflicted with
dissociative disorders. 
REFERENCES
American Psychiatric Association. (1993).DSM-IV draft criteria.
Washington, DC: Author.
American Psychiatric Association. (1994).
manual of mental disorders,4th Edition. Washington, DC: Author.
Bliss, E.L. (1986).Multiple personality, allied disorders and hypnosis.
New York: Oxford University Press.
Bowman, E.S., Blix, S., Coons, P.M. (1985). Multiple personal-
ity in adolescence: Relationship to incestual experiences.Journal
of the American Academy of Child Psychiatry, 24,
Dell, P.F., Eisenhower, J.W. (1990). Adolescent multiple per-
sonality disorder.
Psychiatry, 29,359-366.
Fagan, J., McMahon, P.P. (1984). Incipent multiple personali-
ty in children: Four cases. and Mental Disease, 172,
26-36.
Hornstein, N., Putnam, F.W. (1992). Clinical phenomenology
of child and adolescent dissociative disorders.of theAmeri can
Academy of Child and Adolescent Psychiatry, 31,1077-1085.
Kluft, R.P. (1984). Multiple personality disorder in childhood.
Psychiatric Clinics of North America, 7,
Kluft, R.P. (I985). Childhood multiple personality disorder:
Predictors, clinical findings and treatment results. In R.P. Kluft
(Ed.),Childhoodantecedents of multiple personality (pp. 167-196).
Washington, DC: American Psychiatric Press.
Malenbaum, R., Russel, A.J. (1987). Multiple personality disor-
der in an 11-year-old boy and his mother.
Academy of Child and Adolescent Psychiatry, 26,
Malinosky-Rummell, R.R., 1-loier, T.S. (1991). Validating mea-
sures of dissociation in sexually abused and nonabused children.
Behavioral Assessment, 13,
Peterson, G. (1989). Assessment and diagnosis of child and ado-
lescent multiple personality disorder: Issues and procedures. In
B.G. Braun (Ed.), Disorders, 1989. Proceedings of the Sixth
International Conference on Multiple Personality/Dissociative States
176).Chicago: Rush University.
Peterson, G. (1991). Children coping with trauma: Diagnosis of
"Dissociation identity disorder." DISSOCIATION, 4,
219
DISSOCIATION. Vol. VII. No. 4. December 1994
DISSOCIATIVE DISORDER OF CHILDHOOD
Putnam, F.W. (1989). Diagnosis and treatment of multiple personality
disorder. New York: Guilford.
Putnam, F.W. (1993) . Dissociative disorders in children: Behavioral
profiles and problems. Child Abuse & Neglect, 17,39-45.
Putnam, F.W., Guroff, j jSilberman, E.K., Barban, L., Post,
R.M. (1986). The clinical phenomenology of multiple personality
disorder: Review of 100 recent cases. Journal of Clinical Psychiatry,
47,285-293.
Putnam, F.W., Helmers, K, Horowitz, L.A., Trickett, P.K. (In
press). Hypnotizability and dissociativity in sexually abused girls.
Child Abuse & Neglect.
Putnam, F.W., Helmers, K., Trickett, P.K. (1993). Development,
reliability and validity of a child dissociation scale. Child Abuse &
Neglect, 17,
Putnam, F.W., Loewenstein, R j. (1993) . Treatment of multiple
personality disorder: A survey of current practices.American journal
of Psychiatry, 150,1048-1052.
Riley, R.L.,  Mead, J. (1988). The development of symptoms of
multiple personality disorder in a child of three. DISSOCIATION,
1(4), 41-46.
Vincent, M., Pickering, M.R. (1988). Multiple personality disor-
der in childhood. Canadian Journal of Psychiatry, 33,524-529.
Weiss, M., Sutton, P jUtecht, A.J. (1985). Multiple personality
in a 10-year-old girl ofChild Psychiatry,
24,495-501.
Wherry, J.N., Jolly, J.B., Feldman, J., Adam, B., Manjanatha, S.
(1994). The Child Dissociative Checklist: Preliminary findings of
a screening instrument.journal of Child Sexual Abuse, 3,51-66.
220
DISSOCIATION. Vol.VII, No. 4. December 1991