Child Abuse & Neglect, Vol. 21, No. 10, pp.

941-952, 1997
Copyright © 1997 Elsevier Science Ltd
Pergamon Printed in the USA. All rights reserved
0145-2134/97 $17.00 + .00
PII S0145-2134(97)00055-0

ANXIETY, DEPRESSION, AND DISSOCIATION IN

WOMEN HEALTH CARE PROVIDERS REPORTING A
HISTORY OF CHILDHOOD PSYCHOLOGICAL ABUSE

KATHLEEN S. FERGUSON AND CHRISTINE M. DACEY

Department of Psychology, Xavier University, Cincinnati, OH, USA

ABSTRACT

Objective: The purpose of this study was to explore the relationship between childhood psychological maltreatment and
adult manifestations of depression, anxiety, and dissociation.
Method: Women health care professionals reporting a history of childhood psychological maltreatment (n = 55) were
compared to a nonabused control group (n = 55) on the three dimensions of anxiety, depression, and dissociation. The
Childhood Experiences Questionnaire, a measure constructed specifically for this study to assess abuse history, was used
to determine group membership. Participants were administered the State-Trait Anxiety Inventory (STAI), the Beck
Depression Inventory (BDI), and the Dissociative Experiences Scale (DES).
Results: A significant discriminant function analysis using the STAI, BDI, and DES as predictor variables was able to
correctly classify 74.5% of the psychologically abused participants and 89.1% of the nonabused group, with an overall hit
rate of 81.8%. Statistically significant differences were obtained between the abused and nonabused groups on the STAI,
BDI, and DES.
Conclusions: Interpretation of these results suggests that participants who reported a history of childhood psychological
abuse suffer significantly higher levels of depression and anxiety, and more frequent dissociative experiences, than the
nonabused women. © 1997 Elsevier Science Ltd

Key Words--Psychological abuse, Anxiety, Depression, Dissociation, Nurses.

INTRODUCTION

THE PROBLEM OF childhood psychological maltreatment has been attracting much needed
professional and public attention in recent years (Brassard, Germain, & Hart, 1987; Garbarino,
Guttman, & Seeley, 1986; Garrison, 1987). Hart and Brassard (1987) suggested that psychological
abuse is inherent in all other forms of child maltreatment, and may be just as damaging as other
types of abuse. Psychological maltreatment often coexists with physical and/or sexual abuse
(Claussen & Crittenden, 1991; Egeland, Sroufe, & Erikson, 1983; Garbarino et al., 1986), and most
of the persisting negative effects of abuse are psychological in nature (Brassard et al., 1987).
Although there is a general consensus in the professional community that psychological abuse is
a serious problem that warrants investigation, little empirical research has been reported which has
focused on this issue. Clinicians have long been aware of the potential psychological damage from
childhood experiences of gross emotional abuse or neglect, yet their interventions have been based

Submitted for publication December 27, 1996: final revision received December 27, 1996; accepted April 9, 1997.
Requests for reprints should be sent to Kathleen Ferguson, M.A., Department of Clinical Child Psychology, University of
Kansas, Lawrence, KS 66045.
© 1997 Elsevier Science Ltd. All rights reserved.
941
942 K.S. Fergusonand C. M. Dacey
more on intuition than on empirical findings (Rosenberg, 1987, p. 166). Difficulty in operational-
izing psychological maltreatment may contribute to the lack of investigation of this important issue
(McGee & Wolfe, 1991; Shaver, Goodman, Rosenberg, & Orcutt, 1991).
In 1983, during the International Conference on Psychological Abuse of Children and Youth, an
attempt was made to establish an adequate working definition of psychological maltreatment which
would enhance research efforts in this area (Brassard et al., 1987). This definition was refined and
empirically tested by Hart and Brassard (1991), and they ultimately identified five distinct subtypes
of psychologically abusive parental behaviors. The five subtypes of psychological abuse include
spurning (verbal battering, hostile degradation, rejection), terrorizing (threatening to hurt or kill a
child, child witnesses family violence), isolating (preventing child from interacting with peers or
adults outside of the family, confining child in a closet or closed room for an extended period of
time), exploiting/corrupting (modeling or encouraging child to engage in antisocial or criminal
behavior, encouraging child to use drugs or alcohol, keeping child in role of surrogate parent to
other children or as surrogate spouse), and denying emotional responsiveness (emotionally ne-
glecting child, being psychologically unavailable to child, ignoring child s attempts to interact with
parents).
A few recent studies have indicated that psychological maltreatment is associated with dysfunc-
tional social and emotional development in children (Claussen & Crittenden, 1991; Egeland,
Sroufe, & Erikson, 1983; Ney, 1987; Ney, Fung, & Wickett, 1994; Vissing, Straus, Gelles, &
Harrop, 1991), and that the negative effects may persist into adulthood (Braver, Bumberry, Green,
& Rawson, 1992; Briere & Runtz, 1988; Gross & Keller, 1992; Moeller, Bachman, & Moeller,
1993). In several retrospective studies involving university students, childhood psychological
maltreatment has been associated with symptoms of anxiety, depression, interpersonal sensitivity,
low self-esteem, dissociation, and borderline personality traits (Braver et al., 1992; Briere & Runtz,
1990; Gross & Keller, 1992). Because psychological maltreatment often occurs in conjunction with
other types of abuse, the correlates of psychological abuse are often difficult to disentangle from
the consequences of abuse in general. Therefore, this study attempted to identify a group of
psychologically abused adults with no reported history of physical and/or sexual abuse in order to
examine psychological abuse correlates without the influence of other types of childhood abuse.
Specifically, the relationship between childhood psychological abuse (as defined by Hart &
Brassard, 1991) and symptoms of anxiety, depression, and dissociation in a sample of adult women
health care providers was investigated. Limited research has been reported which has studied the
relationship between each of these variables and a history of psychological abuse. Research by
Gross and Keller (1992) supported higher levels of depression (as measured by the BDI) in
university students who reported a history of childhood psychological abuse. Briere and Runtz
(1988) supported higher levels of anxiety (as measured by the Hopkins Symptom Checklist) in
university women who reported a history of childhood psychological abuse. Studies have demon-
strated a link between childhood physical and sexual abuse and the development of dissociative
disorders (Chu & Dill, 1990). Because dissociative defenses seem to be used in response to
psychologically overwhelming experiences such as trauma and abuse, it would seem likely that the
painful experience of parental psychological abuse might precipitate dissociative experiences as
well. Briere and Runtz (1988) reported a correlation between childhood psychological maltreat-
ment and symptoms of dissociation in adulthood; however, the dissociation measure used was not
well-standardized. Consequently, the present study explored the possible relationship between
psychological maltreatment and dissociation using a well-standardized measure of dissociation.
Although prior research has examined the relationship between psychological abuse and each of
these variables separately, the research is limited and anxiety, depression and dissociation have not
been considered in combination. In order to better appreciate the broader clinical impact of
psychological maltreatment, this study looked at the three variables in combination in a multivar-
iate design.
 Childhood psychological abuse 943
Table 1. Summary of Demographic Characteristics of the Abused and Non-Abused Groups
Abused Non-Abused t-Value
Variable (n = 55) M (SD) (n = 55) M (SD) (df = 108) p

Age 36.7 (9.0) 34.6 (9.0) - 1.27 .206

Income $60,182 (23,758) $66,454 (24,938) 1.35 .180
Education 14.9 (1.3) 15. I (1.2) 0.47. .641
Abused Non-Abused

n P n P X2 value (df 3) p

Marital Status
Single 12 23.6% 7 12.7%
Married 37 67.3% 41 74.5%
Divorced 4 7.3% 7 12.7%
Widowed I 1.8% 0 0%
3.82 .281
Occupation
Registered nurse 49 89.1% 53 96.4%
Secretary 4 7.3% 2 3.6%
Respiratory
therapist 1 1.8% (I 0%
Pastoral counselor 1 1.8% 0 1)%
2.82 .420

A review of the literature failed to identify any standardized instruments which could be used to
assess psychological maltreatment. Previous studies of psychological abuse correlates have utilized
measures constructed by the researchers themselves to identify participants with psychological
maltreatment histories (Briere & Runtz, 1988; Gross & Keller, 1992). For the purposes of this
study, the empirically-supported definition of psychological maltreatment proposed by Hart and
Brassard (1991) was used to identify psychologically abused participants. More specifically, the
five subtypes of psychologically abusive parental behaviors identified by Hart and Brassard (1991)
were used to construct the Childhood Experiences Questionnaire (CEQ), the psychological abuse
measure used in this study.

METHOD

Participants

Participants included 110 women health care professionals employed at an urban hospital in the
Midwest. A summary of the demographic characteristics of the participants is presented in Table 1.
The experimental group consisted of 55 women ranging in age from 20 to 61 years (M = 36.7;
SD = 9.0) who reported a history of childhood psychological maltreatment on the Childhood
Experiences Questionnaire, a measure constructed specifically for this study. The control group
included 55 women ranging in age from 22 to 58 years (M = 34.6; SD = 9.0) who reported no
history of childhood physical, sexual, or psychological abuse on the questionnaire. Both groups
were comprised of predominantly White, middle-class, college-educated women with no signifi-
cant differences found between the groups on the demographic variables of age, yearly family
income level, years of education, marital status, or occupation (see Table 1).

Materials
The Childhood Experiences Questionnaire (CEQ) is a 3 0 - i t e m self-report measure which was
specifically constructed for this study to assess abuse history. The questionnaire elicited informa-
944 K. S. Ferguson and C. M. Dacey

tion regarding childhood experiences of physical, sexual, and psychological trauma. Eleven items
in the questionnaire assessing psychological maltreatment were based on Hart and Brassard's
(1991) definition of psychologically abusive parental behaviors. Two items included spurning
(made fun of you or intentionally humiliated you in public; belittled your abilities or accomplish-
ments, insulted you, called you negative names, ridiculed you or degraded you), two items included
terrorizing (threatened to physically harm you, threatened to kill you, threatened to abandon you;
witness one of parents hit or beat up other parent or other family member), two items included
isolating (locked you in a room or confined you in a closet, basement or some~'other area for an
extended period of time; forbid you to have friends or kept you from interacting-with people outside
of the family), three items included exploiting/corrupting (kept you home from school to help with
the housework, or to take care of your brothers and sisters, or to take care of one or both parents;
allowed you or encouraged you to use drugs or alcohol, or encouraged you to engage in criminal
behavior; treated you as if you were the parent in your family), and two items included denying
emotional responsiveness (neglected you emotionally, were psychologically unavailable, or
showed disinterest in your thoughts, feelings, activities or ideas; ignored your attempts to cuddle
with them as a child or pushed you away when you tried to hug or kiss them). Participants were
asked to indicate the frequency of such behaviors in childhood by circling a response of 0-NEVER
(behavior never occurred), 1-RARELY (occurred once or twice during childhood), 2-SOME-
TIMES (happened occasionally, one to five times a year), or 3-OFTEN (occurred on a regular
basis, once a month or more). Using this 4-point Likert scale, a total score for psychological
maltreatment was obtained by summing the 11 items related specifically to a history of childhood
psychological maltreatment, with a possible range of scores from 0 to 33.
In order to examine the correlates of psychological abuse independently of the consequences of
other types of abuse, participants who reported experiences of childhood physical and/or sexual
trauma on the questionnaire were excluded from the primary analysis. For the purposes of this
study, the definitions of physical and sexual abuse provided by The Child Abuse Prevention and
Treatment Act (Public Law 100-294) were used to identify participants who may have suffered
these types of abuse. Physical abuse is defined as the physical injury of a child by a person who
is responsible for the child's welfare, and is characterized by punching, beating, kicking, biting,
burning, hitting with an object, or otherwise harming the child. The CEQ included three questions
to screen for physical abuse, based upon this definition. Sexual abuse is defined as rape, incest,
molestation, prostitution, pornography, or other forms of sexual exploitation of a child by a person
responsible for the care of a child (National Center on Child Abuse and Neglect; NCCAN, 1992),
and was assessed by two questions on the CEQ.
Anxiety was measured with the State-Trait Anxiety Inventory, Form Y (STAI; Spielberger,
1983) which consists of two sets of 20 statements that the respondent rates according to how he or
she feels at the present time (state) and how the respondent generally feels (trait). State anxiety is
described as a situational experience and trait anxiety is thought to be a general disposition to
respond with anxiety across situations. Items are given a weighted score of 1 to 4, with 1 indicating
the absence of anxiety and 4 indicating the presence of a high level of anxiety. The items are then
summed to obtain scores for state anxiety and for trait anxiety, with a range of possible scores from
20 to 80 for each measure of anxiety. Because this study investigated anxiety as an enduring
symptom rather than as a reaction to a specific situation, only the Trait-Anxiety score was used in
data analysis. Adequate reliability data are reported for the STAI including test-retest reliabilities
ranging from .73 to .86, and a median alpha coefficient of .90. Validity coefficients of .75 and .80,
respectively, were obtained by correlating the STAI with the IPAT Anxiety Scale and the Taylor
Manifest Anxiety Scale (Spielberger, 1983).
Depression was measured with the Beck Depression Inventory (BDI; Beck & Steer, 1987),
which is a 21-item self-report inventory designed to measure depressive symptoms and the severity
of those symptoms. Each item consists of four statements which are scored from 0 (normal mood)
 Childhood psychological abuse 945

to 3 (depressed mood) and then summed for a total score which can range from 0 to 84. A total
score ranging from 1-9 would be interpreted as normal mood or asymptomatic, 10-18 as mild to
moderate depression, 19-29 as moderate to severe depression, and 3 0 - 6 8 as extremely severe
depression (Beck & Steer, 1987). Beck and Steer (1987) reported a mean correlation of .60 between
clinical ratings of depression and BDI scores for a nonpsychiatric population. Split-half reliability
of .93 has been reported for this instrument and internal consistency has been estimated at .85.
Test-retest reliabilities range from .48 to .86, and acceptable levels of validity have been reported
with coefficients ranging from .62 to .77 (Beck & Steer, 1987).
Dissociation was measured with the Dissociative Experiences Scale (DES; Bemstein & Putnam,
1986) which is a self-report measure of the frequency of dissociative experiences. There are 28
questions to which the individual responds by marking a 100 millimeter line with a slash to indicate
how often, from 0% to 100% of the time, a particular experience occurs. The instrument is scored
by measuring the length of the line to the slash mark in millimeters for each question and then by
computing the mean across all 28 items, with a range of possible scores from 0 to 100. Bernstein
and Putnam (1986) reported test-retest reliability of .84 and split-half reliability coefficients ranging
from .50 to .79. Criterion-related validity was demonstrated in that the DES was able to adequately
differentiate between participants with a dissociative disorder and participants without that diag-
nosis (Bernstein & Putnam, 1986).

Procedure

Women health care professionals attending a mandatory hospital inservice program were
recruited by the investigator to participate in the study. A standardized script which explained that
participation in the study would involve completing several questionnaires related to both adult and
childhood experiences was read to the group by the investigator. Volunteers who agreed to
participate were asked to complete a consent form. The consent forms were collected by the
investigator and maintained separately in order to provide anonymity. Participants then completed
a demographic questionnaire, followed by the STAI, DES, BDI, and the CEQ. The responses to the
CEQ were used to determine group membership. Based upon a 4-point Likert scale (0 to 3), those
participants who responded 3-OFTEN to one or more items related specifically to a history of
childhood psychological maltreatment, and 0-NEVER to all of the physical and sexual abuse
items, were assigned to the abused group (n = 55). Claussen and Crittenden (1991) maintained that
most people have experienced single acts of psychological maltreatment as children, but it is the
cumulative effect of recurrent episodes of psychological maltreatment that is harmful. Therefore,
inclusion in the abused group was contingent upon participant reports of chronic experiences of
psychological abuse. A total score for the questionnaire was obtained by summing the scores for
the psychological maltreatment items for the abused group. Participants who reported no history of
physical, sexual, or psychological abuse were assigned to the nonabused group. Specifically, those
participants who responded 0-NEVER or 1-RARELY to all of the psychological abuse items, and
0-NEVER to all of the physical and sexual abuse items, were assigned to the nonabused group
(n = 55). The total score on the questionnaire for the control group was also obtained by summing
the scores for the psychological maltreatment items.
In order to differentiate the long-term effects of psychological abuse from other types of abuse,
participants who reported childhood physical and/or sexual abuse on the questionnaire were
excluded from the primary analysis. The physical and sexual abuse items were based on the
definitions provided for such types of maltreatment by the NCCAN. Specifically, participants who
indicated having been hit, punched, kicked, beaten with an object, or receiving physical punishment
that resulted in cuts, bruises, broken bones, or burns either 2-SOMETIMES or 3-OFTEN were
considered physically abused (n = 30). Participants who responded either 1-RARELY, 2-SOME-
TIMES, or 3-OFFEN to the sexual abuse questions were placed in the sexual abuse group (n =
946 K.S. Ferguson and C. M. Dacey

Table 2. Means, Standard Deviations, and Univariate F-Ratios for the Abused and Non-Abnsed Groups on the STAI,
BDI, and DES
Predictor Variable Abused (n = 55) M (SD) Non-Abused (n = 55) M (SD) F-value (df = 108)

STAI 44.6 (10.2) 31.2 (6.3) 69.06**

BDI 11.8 (5.8) 4.2 (3.8) 66.61"*
DES 10.0 (8.3) 5.0 (3.9) 16.11"

Note. STAI = State-Trait Anxiety Inventory Trait-Anxiety score, BDI = Beck Depression Inventory, DES = Dissociative
Experiences Scale.
*p < .0001; **p < .00001.

19). Those participants who met criteria for more than one form of abuse were placed in the
combined abuse group (n = 20).
Data collection was attempted at 13 different inservice meetings with a total of 418 individuals
in attendance. Out of the 418 women who were approached, 211 (50.48%) agreed to participate.
Because the inservice program required those in attendance to complete various proficiency tests
and to demonstrate competency in certain medical procedures, it is possible that women who
declined to participate in the present study were reluctant to engage in further test-like activities,
and/or did not want to spend any more time in the inservice than what was required. Although no
specific demographic data are available for the nonparticipants, it is assumed that their demo-
graphic composition did not differ significantly from the participants as all of those in attendance
were employed as either nurses, respiratory therapists or medical secretaries in the hospital. Of the
211 questionnaires obtained, 110 (52.1%) were suitable for inclusion in the primary analysis, and
101 (47.9%) were excluded from the analysis due to either reported physical and/or sexual abuse,
or incomplete surveys. Of the participants excluded from data analysis, 30 (14.2%) reported a
history of childhood physical abuse, 19 (9.0%) reported a history of childhood sexual abuse, and
20 (9.5%) reported a combination of physical, sexual, and psychological abuse. Twenty-seven
(12.8%) participants reported levels of abuse that did not meet the criteria for inclusion in any of
the groups. Five (2.4%) surveys were returned to the investigator incomplete, and were, therefore,
ineligible for use in the study.

RESULTS

The psychological abuse group obtained a mean score of 9.82 (SD = 4.2) on the 11 items of
psychological maltreatment on the Childhood Experiences Questionnaire, and the nonabused
control group obtained a mean of .254 (SD = .44). There was a statistically significant difference
between the groups on their report of childhood psychological maltreatment, t(108) = - 16.83,
p < .0001, with the abused group reporting a significantly higher frequency of experiences of
childhood psychological abuse than the control group.
Means and standard deviations for the abused and nonabused groups on the STAI Trait-Anxiety
score, BDI, and DES are presented in Table 2. Univariate F-tests were conducted between the
control and experimental groups on the three predictor variables (see Table 2), and significant
differences were found between the two groups on all three variables (p < .0001). Abused women
reported higher levels of anxiety and depression and greater frequency of dissociative experiences
than nonabused women.
A discriminant function analysis was conducted to determine a statistically significant function
which maximized the differences between the abused (n = 55) and nonabused (n = 55) groups
using the STAI Trait-Anxiety score, BDI, and DES as predictor variables. The analysis produced
a significant function, X2 (3, N = 110) = 73.01, p < .00001, which was able to correctly classify
 Childhood psychological abuse 947

Table 3. Discriminant Function Structure Coeffi-

cients for the STAI, BDI, and DES
Predictor Variable Structure Coefficient

STAI .806
BDI .791
DES .389

Note. Variables listed in order of magnitude. STAI =

State-Trait Anxiety Inventory Trait-Anxiety score,
BDI = Beck Depression Inventory, DES Dissociative
Experiences Scale.

74.5% of the abused group and 89.1% of the nonabused group, with an overall hit rate of 81.8%.
Structure coefficients, which are used to determine the unique contribution of each predictor
variable to the function, indicated that all three predictor variables reached the suggested .30
elimination level and, therefore, could be used in the interpretation of the significant function (see
Table 3). Consequently, abused women had higher anxiety, depression, and dissociative experi-
ences scores than did nonabused women.
The primary intent of this study was to examine the correlates of psychological abuse with abuse
history. However, since a large number of women (n = 69) were necessarily excluded from the
analysis due to their reports of other forms of abuse, it seemed appropriate to look at the
relationship between the different forms of abuse and emotional correlates. Consequently, a
multivariate analysis of variance was conducted with abuse group serving as the independent
variable and the STAI, BDI, and DES scores as the dependent measures. The means and standard
deviations for each abuse group on the STAI Trait-Anxiety score, BDI and DES are presented in
Table 4. Using the Wilk's criterion, results of the MANOVA indicate an overall statistically
significant difference among abuse groups on the dependent measures, approximate F(4, 174) =
5.82, p < .0001 (see Table 4). Follow-up univariate analyses were conducted to compare
differences among abuse groups on each of the three dependent measures (see Table 4).
A one-way analysis of variance to test for differences among the abuse groups on the STAI
Trait-Anxiety score indicated a statistically significant difference, F(4, 174) = 14.31, p < .0001.
Using the Scheffe procedure to probe the significant findings, a statistically significant difference
was established between the psychologically abused group (M = 44.64) and the control group

Table 4. MANOVA, Means, Standard Deviations, and Follow-Up Univariate Analyses Comparing Abuse Groups on
the STAI, BDI, and DES
Wilks Value Hypoth. df Error df Approx F
MANOVA .6487 12.0 455.36 6.74*

Abuse Groups

Control a Psych b Physical c Sexual d Combined e

M (SD) M (SD) M (SD) M (SD) M (SD) F (4, 174)

STAI 31.2 (6.3) 44.6 (10.2) 39.7 (11.2) 38.9 (11.5) 40.4 (9.6) 14.31"
BD1 4.2 (3.8) 11.8 (5.8) 8.1 (7.7) 10.7 (9.6) 8.5 (5.6) 11.33"
DES 5.0 (3.9) 10.0 (8.3) 12.5 ( 12.5 ) 14.0 (12.6) 14. l (9.2) 6.98*

Note. STA1 = State-Trait Anxiety Inventory Trait-Anxiety score, BD1 = Beck Depression Inventory, DES = Dissociative
Experiences Scale.
~'n = 55; bn = 55; Cn = 30; dn -- 19; en -- 20.
*p < .0001.
 948 K. S. Ferguson and C.M. Dacey

(M = 37.33) at the .01 level. No differences among other types of abuse were indicated with regard
to anxiety.
To test for differences among the abuse groups on the BDI, a oneway analysis of variance was
conducted and a statistically significant difference was found F(4, 174) = 11.33, p < .0001.
Probing using the Scheffe procedure indicated significant differences between the sexual abuse
group (M = 10.68) and the control group (M = 6.0), and between the psychological abuse group
(M -- 11.78) and the control group (M =- 6.0) at the .01 level.
A oneway ANOVA was conducted to test for between-group effects on the DES. A statistically
",significant difference was established among abuse groups, F(4, 174) = 5.76, p < .0001. Probing
these results using the Scheffe procedure indicated significant differences between the sexual abuse
group (M = 14.0) and the control group (M = 5.0), and between the combined abuse group (M =
14.1) and the control group at the .01 level.

DISCUSSION

The purpose of this study was to explore the relationship between childhood psychological
maltreatment and adult manifestations of depression, anxiety, and dissociation. Based upon the
predictor variables of the STAI, BDI, and DES, a statistically significant discriminant function was
derived which accurately differentiated abuse victims from nonabused women. The psychologi-
cally abused group reported greater trait anxiety and depression, and more frequent dissociative
episodes. With an overall hit rate of 81.8%, the utility of the discriminant function in identifying
victims of psychological abuse was clearly demonstrated. These findings add empirical support to
the limited available literature which suggests that psychological abuse in childhood is related to
symptoms of psychological distress in adulthood (Braver et al., 1992; Briere & Runtz, 1988; Briere
& Runtz, 1990; Gross & Keller, 1992). The current study operationalized psychological abuse
based upon the empirical work of Hart and Brassard (1991), and thus differentiated it from other
types of childhood maltreatment, which enabled the investigation to focus more specifically upon
the relationship between psychological abuse and adult functioning.
In addition to contributing to the empirical research, the study contributes to the clinical
understanding of the effects of psychological trauma. Only 38.2% of the participants in the
experimental group reported on the CEQ that they considered themselves to have been psycho-
logically maltreated as children. Consequently, abused womens' perceptions significantly differed
from the classification of abuse status derived from their responses to items using the established
operational definition. Because the selection criteria for group membership were stringent, these
results suggest that significant underreporting of psychological abuse occurs, and, consequently, the
clinician cannot simply rely on client self-report in the assessment of psychological trauma. In light
of this information, the clinical significance of using a combination of measures such as the STAI,
BDI, and DES for the identification of clients who may have been psychologically abused is
supported.
The mean Trait-Anxiety score for the psychologically abused group was 44.6 (SD = 10.2) which
is approximately one standard deviation above the normative sample mean of 34.79 (SD = 9.22)
reported by Spielberger (1983). The mean Trait-Anxiety score for the control group was 30.6
(SD = 6.2), which falls within the normative range. Not only did participants in the abused group
report higher levels of Trait-Anxiety, but they reported significantly higher levels of State-Anxiety
(M = 41.67, SD = 10.52) than participants in the control group (M = 31.95, SD = 7.82), t(108)
= - 5.50, p < .0001. Participants in the abused group reported higher levels of anxiety as a
situational response (state) at the time of the survey, and they also reported more of a general
disposition to respond with anxiety across situations (trait) than participants in the control group.
The abused group also reported higher levels of depression than the control group. The mean
 Childhood psychological abuse 949

BDI score for the abused group was 11.8 (SD = 5.8), which falls in the range of mild to moderate
depression (Beck & Steer, 1987). Thirty-five participants (54.5%) in the experimental group
obtained BDI scores in the range of 10 to 18, which indicates mild to moderate depression, and six
participants (11%) scored in the range of 19 to 30 which indicates moderate to severe depression.
Only four participants (7.2%) in the control group reported BDI scores in the range of mild to
moderate depression, and no participants in the control group scored in the moderate to severe
range. These results are clinically significant in that the majority of women who reported
psychological abuse as children reported symptoms of mild to moderate depression, while the
majority of nonabused participants reported a normal mood with no depression.
These results, as well as the results concerning anxiety, are congruent with, and lend support to,
the limited existing research reporting higher levels of depression and anxiety in adults who were
psychologically abused as children (Briere & Runtz, 1990; Gross & Keller, 1992). The results also
suggest that the psychologically abused women suffer concurrent symptoms of depression and
anxiety. These depression and anxiety score elevations are not indicative of serious pathology, but
extremely elevated scores were not expected for this nonclinical sample of high-functioning
professional women. However, the mildly elevated scores are still of clinical concern. In addition
to the generally unpleasant feeling states that result from anxiety and depression, such compro-
mised emotional functioning may further impact the interpersonal relationships, physical health,
work productivity, and general quality of life of these women.
Participants in the abused group reported a higher frequency of dissociative experiences than the
control group. However, the mean DES scores for both the experimental group (M = 10.0; S D =
8.3) and the control group (M = 5.0; S D = 3.7) fell within the normal range of dissociative
experiences (M = 10.8; S D = 10.2) as defined by Ross, Joshie, and Currie (1990). In clinical
studies, DES scores above 20 indicate that the person is having more dissociative experiences than
the average person, and DES scores above 30 are usually associated with post-traumatic stress
disorder or dissociative identity disorder (Bernstein & Putnam, 1986). No participants in the control
group obtained DES scores above 20. Although the DES scores of the experimental group were
significantly higher than those of the control group, only three participants obtained scores in the
20 to 30 range, and two participants obtained scores greater than 30. Although the dissociative
scores of the abused women do not demonstrate clinical diagnostic value, they do suggest that those
women who suffered psychological abuse as children reported significantly more dissociative
experiences than nonabused women. Consequently, abused women may have a higher propensity
towards the use of dissociation as a defense in the face of psychological distress than do nonabused
women. These findings lend support to Briere and Runtz's (1988) research which indicated that
psychological maltreatment was associated with dissociation. Although Briere and Runtz (1988)
reported a significant relationship between psychological abuse and dissociation, the results of their
study remain questionable due to the lack of standardization of the measures used. This study
utilized a dissociation measure with reported acceptable psychometric properties (Bernstein &
Putnam, 1986) and supported the relationship between dissociative experiences and childhood
psychological abuse, especially when the DES is studied in combination with the STAI and BDI.
Results of the post-hoc analyses are also consistent with previous research. Gross and Keller
(1992) reported that psychological abuse proved to be a more powerful predictor of depression,
self-esteem, and attributional style than physical abuse. Egeland and colleagues (1983) reported
that nonphysically abused children of psychologically unavailable caregivers demonstrated more
serious problems than physically abused children. Ney (1987) reported that verbal abuse by parents
had a greater negative impact on the child than did parental physical abuse. However, children
subjected to combinations of verbal, physical and/or sexual abuse suffered more severe conse-
quences than children exposed to only one type of abuse. Further investigation revealed that
children who experienced a combination of physical neglect, physical abuse, and verbal abuse
suffered the most serious psychological consequences (Ney, Fung, & Wickett, 1994). In the present
950 K. S. Ferguson and C. M. Dacey

study, the psychological abuse group reported significantly higher levels of depression and anxiety
than the control group. The sexual abuse group reported significantly higher levels of depression
and dissociation than the control group, and the combined abuse group reported higher levels of
dissociation than the control group. The physical abuse group did not report higher levels of
anxiety, depression, or dissociation than the control group. These findings suggest that the impact
of psychological abuse alone is significant, and the degree of psychological distress reported by the
psychologically abused women on the three dependent measures indicates that psychological
abuse, though difficult to define, can be just as damaging as other forms of child abuse.
The present study did not control for the influence of significant people in the participants' lives
who may have mediated experiences of parental abusiveness. Participants who suffered psycho-
logical abuse in childhood may have received support and nurturance from other adults such as
teachers, coaches, relatives, or friends, and this may have had a buffering effect on the negative
experiences of abuse. Also, this study did not take into account any psychotherapy that the
participants may have received. Treatment would be expected to decrease the negative symptoms
experienced by victims of abuse. In future investigations of psychological abuse, it would be
interesting to explore mediating variables that may help to ameliorate the negative effects of abuse,
and to identify variables that enhance resiliency.
It would also be useful to replicate this study with different occupational samples. The majority
of participants in this study were nurses, and, consequently, the level of reported abuse and the
results obtained may not be generalizable to other populations. Of the 211 participants in the
present study, 151 (72%) reported having experienced either psychological, physical, or sexual
abuse in childhood. The high rate of reported abuse in this study is congruent with the reported
abuse of 33 health care providers in a pain treatment center of which 55% reported a history of
emotional abuse, 36% reported a history of physical abuse, and 33% reported being victims of
sexual abuse (Karol, Micka, & Kuskowski, 1992). Karol and colleagues (1992) suggested that the
providers' own abuse histories might impact on their style of care giving. An abuse history may
positively enhance therapeutic understanding and empathy or may be detrimental to care giving
(Karol et al., 1992, p. 484). Given the high rates of abuse reported among health care workers in
the Karol and colleagues (1992) study and in the present investigation, further empirical research
would be helpful to determine whether individuals who have suffered childhood psychological
maltreatment, or any other type of abuse, tend to gravitate toward the helping professions, including
nursing, medicine, psychology, social work, and so on. It would also be valuable to explore how
a provider's own abuse history might impact on the nature of interactions with patients.
Although certain methodological limitations exist, the results of this study lend clarification and
support to the existing research demonstrating an association between childhood psychological
abuse and psychological difficulties in adulthood (Briere & Runtz, 1988; Gross & Keller, 1992).

REFERENCES

Beck, A. T., & Steer, R. A. (1987). Beck Depression Inventor manual. New York: The Psychological Corporation,
Harcourt Brace Jovanovich.
Bernstein, E. M., & Putnam, F. W. (1986). Development,reliability and validity of a dissociation scale. The Journal of
Nervous and Mental Disease, 174, 727-735.
Brassard, M., Germain, R., & Hart, S. (1987). Psychological maltreatment of children and youth. New York: Pergamon
Press.
Braver, M., Bumberry,J., Green, K., & Rawson, R. (1992). Childhood abuse and current psychologicalfunctioning in a
university counseling center population. Journal of Counseling Psychology, 39, 252-257.
Briere, J., & Runtz, M. (1988). Multivariate correlates of childhood psychological and physical maltreatment among
university women. Child Abuse & Neglect, 12, 331-340.
Briere, J., & Runtz, M. (1990). Differentialadult symptomatologyassociatedwith three types of child abuse histories. Child
Abuse & Neglect, 14, 357-363.
Chu, J. A., & Dill, D. L. (1990). Dissociative symptomsin relation to childhood physical and sexual abuse. American
Journal of Psychiatry, 147, 887-892.
 Childhood psychological abuse 951

Claussen, A. H., & Crittenden, P. M. (1991). Physical and psychological maltreatment: Relations among types of
maltreatment. Child Abuse & Neglect, 15, 5-18.
Egeland, B., Sroufe, L. A., & Erikson, M. (1983). The developmental consequences of different patterns of maltreatment.
Child Abuse & Neglect, 7, 459-469.
Garbarino, J., Guttman, E., & Seeley, J. (1986). The psychologically battered child: Strategies for identification, assessment.
and intervention. San Francisco, CA: Jossey-Bass.
Garrison, E. G. (1987). Psychological maltreatment of children: An emerging focus for inquiry and concern. American
Psychologist, 42, 157-159.
Gross, A. B., & Keller, H. R. (1992). Long-term consequences of childhood physical and psychological maltreatment.
Aggressive Behavior, 18, 171-185.
Hart, S., & Brassard, M. (1987). A major threat to children's mental health: Psychological maltreatment. American
Psychologist, 42, 160-165.
Hart, S., & Brassard, M. (1991). Psychological maltreatment: Progress achieved. Development and Psychopathology, 3.
61-70.
Karol, R. L., Micka, R., & Kuskowski, M. (1992). Physical, emotional, and sexual abuse among pain patients and health
care providers: Implications for psychologists in multidisciplinary pain treatment centers. Progressional Psychology;
Research and Practice, 23, 480-485.
McGee, R. A., & Wolfe, D. A. (1991). Psychological maltreatment: Toward an operational definition. Development and
Psychopathology, 3, 3-18.
Moeller, T. P., Bachman, G. A., & Moeller, J. R. (1993). The combined effects of physical, sexual, and emotional abuse
during childhood: Long-term health consequences for women. Child Abuse & Neglect, 17, 623-640.
National Center on Child Abuse and Neglect (1992). Child abuse and neglect: A shared community concern (DHHS
Publication No. 92-30531). Washington, DC: U.S. Department of Health and Human Services.
Ney, P. G. (1987). Does verbal abuse leave deeper scars: A study of children and parents. Canadian Jaurnal of Psychiatr)',
32, 371-378.
Ney, P. G., Fung, T., & Wickett, A. R. (1994). The worst combinations of child abuse and neglect. Child Abuse & Neglect,
18, 705-714.
Rosenberg, M. S. (1987). New directions for research on the psychological maltreatment of children. American Psychol-
ogist, 42, 166-17l.
Ross, C. A., Joshi, S., & Currie, R. (1990). Dissociative experiences in the general population. American Journal of
Psychiatry, 147, 1547-1552.
Shaver, P. R., Goodman, G. S., Rosenberg, M. S., & Orcutt, H. (1991). The search for a definition of psychological
maltreatment. Development and Psychopathology, 3, 79-86.
Spielberger, C. D. (1983), Manual for the State-Trait Anxie~ Inventor. Palo Alto, CA: Consulting Psychologists Press.
Vissing, Y. M., Straus, M. A., Gelles, R. J., & Harrop, J. W. (1991). Verbal aggression by parents and psychosocial
problems of children. Child Abuse & Neglect, 15, 223-238.

RgSUME

Objeetif: Cette 6tude avait pour but d'explorer la relation entre les mauvais traitements psychologiques en enfance et les
manifestations de drpression, d'angoisse et de dissociation en ~ge adulte.
Mrthode: On a compar6 55 femmes travaillant ~t tltre de professionnelles de la sant6 qui avaient 6t6 maltraitres
psychologiquement durant leur enfance, h u n groupe contr61e de 55 personnes qui n'avaient jamais 6t6 abusres. Trois
facteurs ont 6t6 mesurrs: l'angoisse, la drpression et la dissociation. Le questionnaire Childhood Experiences (exprriences
d'enfance), qui a 6t6 conqu expressrment pour cette 6tude, 6valuait les antrcrdants abusifs e t a servi ~t choisir les personnes
qui feraient partie de l'rtude. On a administr6 le State-Trait Anxiety Inventory, le Beck Depression Inventory et le
Dissociative Experiences Scale.
Rrsultats: On a men6 une analyse discriminante en se servant de ces trois instruments comme variables, lesquels ont su
prrdire avec succrs les exprriences abusives dans 74,5 p.c. des cas pour le groupe de femmes abusres et dans 89,1 p.c. des
cas pour le groupe contrrle. Dans l'ensemble, le taux de succ~s &ait de 81,8 p.c. Des diffrrences d'une importance
statistique considrrable ont 6t6 notres entre les deux groupes.
Conclusions: En interprrtant ces rrsultats, on est port6 ~ conclure que les participantes qui rapportent avoir subi des mauvais
traitements psychologiques vivent des niveaus de drpression et d'angoisse plus marqurs et des exprriences dissociatives
plus frrquentes.

RESUMEN

Objetivo: E1 prop6sito de este estudio fue explorar la relaci6n entre el maltrato psicologico en la nifiez y manifestaciones
adultas de depresi6n, ansiedad y disociacion.
Mrtodo: Se compararon mujeres profesionales en salud que reportaban una historia de maltrato psicolrgico en la nifiez (n =
55) con un grupo control no abusado (n = 55) en las tres dimensiones de ansiedad, depresi6n, y disociaci6n. Para determinar
los miembros de cada grupo se utiliz6 el Cuestionario de Experiencias Infantiles un medida construida especfficamente para
952 K. S. Ferguson and C. M. Dacey

este estudio para evaluar la historia de abuso. Alas participantes se les administraba el "State-Trait Anxiety Inventory," el
Inventario de Depresi6n de Beck, y la Escala de Experiencias Disociativas.
Resultados: Un an~ilisis de la funci6n significativa discriminatoria utilizando et STAI, BDI, y DES como predictor de
variables fue capaz de clasificar correctamente el 74.5% de las participantes abusadas psicol6gicamente y el 89.1% de|
grupo no abusado, con una tasa general de certeza de 81.8%. Se obtuvieron diferencias estadisticamente significativas entre
los grupos abusados y no abusados en el STAI, BDI y DES.
Conclusiones: La interpretaci6n de estos resultados parece sugirir que las participantes que reportan una historia de abuso
psicol6gico en la nifiez sufren significativamente m~is altos niveles de depresi6n y ansiedad y experiencias disociativas con
mas frecuencia que las mujeres no abusadas.