American Journal of Orthopsychiatry, 70(3), July 2000

© 2000 American Orthopsychiatric Association, Inc.

PsychiatricDisordersin Adolescents Exposed to Domestic Violence and Physical Abuse David Pelcovitz, Ph.D., Sandra J. Kaplan, M.D., Ruth R. DeRosa, Ph.D., Frances S. Mandel, Ph.D., Suzanne Salzinger, Ph.D. The relationshipbetween abuse and psychiatric diagnoses was investigated in two groups of physically abused adolescents, 57 living in homes with interparentalviolence and 32 in homes without such violence, and in 96 nonabusedadolescents living in nonviolent homes. Adolescents in the first group werefound to be at greaterriskfor depression, separation anxiety disorder, post-traumaticstress disorder, and oppositional defiant disorder than

were those in the second group. Adolescents in the first group also appearedmore vulnerable to anxiety and depression.

espite growing recognition of the deleterious effect of interparental violence on children, few empirical studies have systematically investigated the psychological impact of such violence on adolescents. Recent surveys have found that physical abuse of adolescents constitutes 31.5% of documented maltreatment cases (U.S. Department of Health and Human Services, 1998). Estimates suggest a 30%060% percent greater chance of a child becoming a victim of abuse in families in which spouse abuse is present (Hughes, Parkinson, & Vargo, 1989; Jaffe, Wolfe, & Wilson, 1990). The most common trigger of marital violence is conflict over child-rearing, suggesting that children often get caught up in the violence (Pelcovitz & Kaplan, 1994). In their review of the literature on the impact of interparental violence, Kolbo, Blakely, and Engleman (1996) reported a consensus in the empirical literature that children who witness interparental violence are at increased risk for a number of emotional and behavioral problems. Children exposed to marital violence often experience significant D

academic difficulties (Pelcovitz, Bester, & Kaplan, 2000) while struggling with symptoms of anxiety, depression, post-traumatic stress, and aggression (Kilpatrick, Litt, & Williams, 1997; Meller & Borchardt, 1996). While researchers have begun to survey the negative impact of interparental violence, few have systematically examined the "double whammy" of being both a witness to and a direct victim of violence (Hughes et al., 1989). Estimates suggest that as the frequency of marital violence in the family increases, the probability of child abuse increases as well, from 5% with one incident of marital violence, to near certainty after 50 or more such incidents (Ross, 1996). It follows that identification of witnesses of interparental violence would be well advised to include an evaluation of child abuse. O'Keefe (1995) examined this double-exposure variable and found that, based on maternal reports, children who were both direct victims and witnesses of abuse had significantly more externalizing behavior problems than did nonabused witnesses. In one of the few studies that systematically

A revised version of a paper submitted to the Journal in October 1999. Work was funded by NIMH Grant #b5RO-MH-43772. Authors are at: Division of Child and Adolescent Psychiatny (Pelcovitz, Kaplan, DeRosa) and Division of Biostatistics, Department of Research (Mandel), Nortih Shore Universihj Hospital, Manhassett, N.Y.; and New York State Psychiatric Institute/Department of Psychiatny, Columbia Universihj, Nezw York (Salzinger).

360

Pelcovitz et al evaluated physically abused adolescent witnesses, O'Keefe (1996) obtained self-reports from abused and nonabused adolescents living in homes with marital violence. She found that the doubly exposed adolescents reported not only that their behavior was more aggressive, but that they also experienced significant levels of internalizing behavior, including depression and anxiety. A few of the retrospective studies examining the current functioning of young adults who had been child witnesses of interparental violence, have included child abuse variables. Henning et al. (1997) found that, compared to controls, college students who had witnessed interparental violence reported significantly more psychological distress, even after controlling for direct physical abuse, parental divorce, alcoholism, socioeconomic status (SES), and parental conflict. Silvern et al. (1995) noted similar findings for undergraduates with a history of witnessing interparental violence: even after physical and sexual abuse were controlled for, participants who had witnessed parental violence reported significantly more symptoms of depression and traumatic stress, as well as lower self-esteem. The literature on interparental violence and adolescent witnesses is sparse. Kolbo's extensive review of the effects of marital violence (Kolbo et al., 1996) found that only 6 of 29 empirical studies included children over the age of 13, and that none of the 29 studies focused exclusively on adolescents. This is unfortunate because, as adolescent trauma survivors begin to form intimate relationships, they may be particularly vulnerable to dating abusive partners, initiating violent conflict, and engaging in high-risk, impulsive behavior such as unprotected sex, pregnancy, running away, and delinquency (Pelcovitz & Kaplan, 1994). Understanding the developmental issues and psychological sequelae unique to adolescent witnesses of interparental violence is an essential step towards outlining effective prevention programs and psychotherapy interventions. None of the interparental violence research to date has utilized structured diagnostic interviews to assess psychological distress. Assessments of the children have been conducted using self-report instruments, often solely from the mother's perspective, despite the likelihood that the parents may be clinically depressed. No previous studies have determined whether the participants' parents met criteria for a psychiatric disorder. Increased

361 prevalence of depression in mothers involved in violent relationships is well documented (Griest, Wells, & Forehand, 1979; Jaffe, Wolfe, Wilson, & Zak, 1986). Fantuzzo and Lindquist (1989) have pointed out that parents at risk for depression are more likely to overestimate the level of their children's difficulties. Use of multiple sources of information in conjunction with standardized clinical interviews might yield more valid and reliable data. The study reported here examined the impact of different types of family violence on adolescents' psychological functioning by comparing adolescents who were direct victims of violence with adolescents who were both direct victims and exposed to interparental violence. (Survivors of sexual abuse by a family member were excluded.) The study sought to improve on previous methodologies by using an exclusively adolescent sample that was recruited directly from the New York State Department of Social Services Abuse and Maltreatment Register after confirmation of physical abuse (most prior studies have relied on shelter samples or children referred for treatment); relying on structured interviews and questionnaires for abuse history and diagnosis (administered blind with respect to subjects' abuse status); and using multiple informants, including parents, in the assessment protocol to determine whether the adolescents met criteria for a psychiatric disorder. It was hypothesized that physically abused adolescents who were exposed to marital violence would meet criteria for significantly more psychiatric diagnoses than their nonabused counterparts. The relative impact of such exposure was examined via a working assessment model using variables likely to affect adolescent functioning; the variables included adolescent physical abuse, psychiatric diagnosis of parents, parental care and protection, and family adaptability and cohesion, all drawn from the family functioning and family violence literature. As already noted, an extensive literature has documented the impact of physical maltreatment, which puts adolescents at risk for such psychiatric disorders and problematic behavior as difficulties in relationships and emotional dysregulation (Cicchetti & Lynch, 1995; Pynoos, 1994). It was also hypothesized that the psychological distress of parents is likely to affect their children's functioning because it taxes caretakers' ability to cope effectively with stress, in both marital and

362

PsychiatricDisordersand Domestic Violence

parental roles (Holden, Stein, Ritchie, Harris, & Jouiles, 1998). In addition, several studies (Davies & Windle, 1997; Kaplan, Pelcovitz, Salzinger, Mandel, & Weiner, 1997; Williamson, Borduin, & Howe, 1991) have found that children are more vulnerable to poor psychological outcomes when their families are characterized by low levels of family. intimacy, care, and cohesiveness; by rigidity of family relations; and by overprotection.

sure group based on mothers' reports that they themselves were battered by their adult partners (defined as hitting, slapping, pushing, or threats with a weapon on more than one occasion). Physical assault perpetrated by an adolescent participant was not assessed. Reliance on the mothers' reports of violence was decided after a comparative analysis of mothers' and fathers' answers showed virtually unanimous agreement. A detailed description of the subject group, and the overall subject pool, can be found elsewhere (Kaplan et al., 1997).

METHOD Subjects The study participants were 89 physically Measures abused adolescents, 96 nonabused adolescents, Group Categorization Exposure to violence. The Conflict Tactics Scale and their parents. Families of adolescents with documented physical abuse were recruited from (CTS) (Straus & Gelles, 1990) is a 19-item questhe New York State Central Register for Child tionnaire designed to assess styles of coping with Abuse between August 31, 1989 and August 31, conflict, and has been widely used in research on 1993. The 89 abused adolescents were divided into family violence (Straus, 1995). Families in the two groups: those in families where interparental comparison group completed the CTS exclude violence was also occurring (double exposure, those who may have been exposed to family vioN=32), and those in homes where interparental vi- lence. Internal consistency, as measured by Cronolence was not reported (single exposure, N=57). bach's alpha, ranges from 0.19 to .80. The validity The 96 nonabused adolescents and their parents re- coefficients, also measured by Cronbach's alpha, ported no familial violence. They were recruited range from .42 to .88. Exposure to sexual abuse. The Sexual Behavior by random digit dialing from the same communities and were matched to the two adolescent abuse Screen, a 12-item questionnaire administered to groups for gender, ethnicity, and age. All the fami- adolescents, was used to screen out those who had lies received up to a maximum of $250 for their had intrafamilial sexual abuse experiences. Interparentalviolence. The Family Disagreeparticipation. Inclusion criteria were: absence of any intrafa- ments Interview (Salzinger, Feldman, Hammer, & milial sexual abuse of the adolescent, the ability to Rosario, 1992) is structured interview designed to read English, and white race. An ethnically homo- elicit behavioral descriptions of abusive behavior geneous sample was recruited because the popula- among family members. The spouse abuse section tion of the geographic area of research was pre- of the interview consists of questions that assess dominantly Caucasian; consequently, the number the frequency of self-reported violence between of minority adolescents available for recruitment parents, the severity of the violence, the source of was relatively small and comparisons among racial disagreement, and the age of the adolescent at the onset of interparental violence. groups would have been statistically unreliable. Adolescents were included in the abuse sample if they were: 1) 12-18 years old, plus or minus six Adolescent Psychopathology Major depression. The Kiddie-Schedule for Afmonths; 2) the victim of physical abuse in the past two years at the hands of a mother, father, or step- fective Disorders and Schizophrenia (epidemioparent; and 3) part of an official investigation by logic version) (K-SADS-E), modified for DSMthe New York State Child Protective Services, 111-R (Orvaschel, Puig-Antich, Chambers, Abrizi, with documented evidence of physical assault or & Johnson, 1982), is a semi-structured diagnostic injury (including excessive corporal punishment, interview for children aged 6-17. Acceptable inbruises/welts, fractures, internal injuries, or bums). terrater reliability for this interview was reported Subjects were assigned to the single-exposure by Gammon, John & Weisman (1984). Adolescent diagnoses. These were made indegroup based on parents' reports. The physically by the first two authors, using a Best Espendently the double-expoto assigned abused subjects were

Pelcovitz et al timate Procedure (Leckman, Sholkomskas, Thompson, Belanger, & Weissman, 1982). In doing so, the two raters were blind to the abuse status and psychopathology of the adolescents' family members. The diagnoses were based on reviews of the K-SADS-E (Orvaschel et al., 1982), the Conners Teacher Rating Scale (Conners, 1969), the PTSD module from the Structured Clinical Interview for the DSM-III-R nonpatient edition (SCID-NP) (Spitzer, Williams, Gibbon, & First, 1990), and available school and medical information that was relevant to diagnosis but would not reveal the subject's group. PredictorVariableMeasures The following measures were used in the logistic regression analyses: Family cohesion andadaptability.These two dimensions of family functioning were measured on the Family Adaptability and Cohesion Evaluation Scale (FACES III) (Olson, Russell, & Sprenkle, 1985). A 20-item self-report measure completed by the adolescent, FACES III has alpha reliability ranging from .75-.90. Test-retest reliability is reported as .84. Parent-childbond. The Parental Bonding Instrument (PBI) is a 25-item self-report measure developed by Parker, Tupling, and Brown (1979) to assess the primary attitudes and behavior that comprise the parental contribution to the parent-child bond. Test-retest reliability is .76 for the care dimension, and .628 for the protection dimension. Pearson correlation coefficients of .76 and .63 (for both, p=0.001) were obtained for the care scale and protection scales, respectively. Parents'psychiatric diagnoses. The Structured Clinical Interview for DSM-III-R (SCID) (Spitzer, et al., 1990) is a semi-structured interview that has been widely used to assess diagnostic criteria for psychological disorders in a reliable fashion (Skre, Onstad, Torgersen, & Kringlen, 1991). Parents of each adolescent completed the interview to determine whether or not they had ever met criteria for a psychiatric diagnosis. Analyses Chi-square analyses were conducted to evaluate differences in adolescents' rates of psychiatric diagnoses across the double-exposure, single-exposure, and comparison groups. Current and lifetime prevalence were assessed for major depression,

363 dysthymia, attention-deficit/hyperactivity disorder (ADHD), conduct disorder, oppositional defiant disorder (ODD), post-traumatic stress disorder (PTSD), overanxious disorder, separation anxiety disorder (SAD), and abuse of cigarettes, drugs, and alcohol. Next, logistic regression models were computed to determine whether adolescents in the doubleexposure group were at greater risk for psychopathology and for using drugs and alcohol than were adolescents in the single-exposure group. The following demographic, parenting, and diagnostic variables were entered into the stepwise multiple regression model: gender, age at time of interview, marital status of parents, father involvement in the study, mother and father care and protection, family adaptability and cohesion, and psychiatric diagnosis of mother and father. The two variables, adolescent physical abuse and interparental violence, were then added to the model to evaluate the relative impact of exposure to domestic violence on psychiatric outcome. RESULTS Demographic data regarding the gender, age, and SES of participants were examined. Each group included an approximately equal proportion of male and female participants (single exposure, N=57, 46% male, 54% female; double exposure, N=32, 47% male, 53% female; control group, N= 96, 50% male, 50% female). Mean age at time of interview was 15.3 years (SD=1.7) for the singleexposure group, 15.4 years (SD=.9) for the double-exposure group, and 15.0 years (SD=1.7) for the control group. SES for each family was estimated by the median income level for the village of family residence. Statistical analysis thus revealed no significant differences across the three groups on any of the three variables. Chi Square Analyses The hypothesis that physically abused adolescents living in violent families would be at greater risk for psychiatric disorder than would physically abused adolescents living without interparental violence was supported. Interestingly, chi square analyses for each diagnostic category revealed that significantly more adolescents in the double-exposure group met criteria for lifetime diagnosis of SAD (22%) and PTSD (19%), than in the singleexposure group (7% and 2%, respectively, p<.05).

PsychiatricDisordersand Domestic Violence

364

Adolescents in the double-exposure group were also diagnosed with current dysthymia and with a history of ADHD at a higher rate (16% and 41%, respectively) than were adolescents in the singleexposure category (4% and 23%, respectively); however, this finding was not significant at the .05 level (p<.09). Compared to those in the nonabused comparison group, adolescents in both the double and singleexposure groups had significantly higher rates of diagnosis for a number of disorders, including major depression, conduct disorder, and ODD. No significant differences were found for overanxious disorder. Percentages of significant DSM-III-R diagnoses across the three groups are presented in TABLE 1.

Logistic Regression Models The vulnerability of the adolescents with double exposure to violence in their homes was also supported in the multivariate analyses, where they were shown to be at greater risk for developing depression, ODD, PTSD, and SAD. The significant predictor variables for these diagnoses are de-

picted in TABLE 2. Compared to adolescents in the single-exposure group, those with double exposure were five times more likely to be currently depressed and have a history of SAD, four times more likely to receive a current diagnosis of ODD, and 375.9 times more Table 1 PERCENTAGE OF ADOLESCENTS WITH DSM-III-R DIAGNOSTIC CATEGORIES

DISORDER Major Depression-L Dysthymia-L Dsythymia-C ADHD-L Conduct-L Conduct-C Opposition Defiant-L

DE' (N=32) 34 C c

22 16c

410 c 25 25 c c 25

GROUP SEb (N=57) c 42 C 14c 4 .d C 2 3 c .d 2 6 .d

21cd 19C

CONTROL (N=96) 15d 2d 1d 12 d 3d 2d 6d

12c d

5d 4d 6d 14 d

Opposition Defiant-C Post-Traumatic Stress-L Separation Anxiety-L Cigarette Abuse-L

25C 19c 22cc 31

2 7d 46c

Cigarette Abuse-C

25c.d

39C

14d

6 c, d

19

Id

Alcohol Abuse-L Drug Abuse-L

16c .d

21Cc

6d

Note. L=lifetime, C=current. Percentages in each row with different superscripts (c or d) differ significantly, p<.05, according to X' analyses. aDouble-exposure group. bSingle-exposure group.

Table 2 SIGNIFICANT PREDICTOR VARIABLES FOR PSYCHIATRIC DIAGNOSES INADOLESCENTS DIAGNOSISNARIABLE DepressiiC &Dysthymia-C Interparental violence ODD-C Interparental violence PTSD-L Mother care Father protection Family adaptability Interparental violence Separation Anxiety-L Interparental violence Cigarette Use-L Gender Age Adolescent abused Interparental violence Drug Use-L Age Adolescent abused Interparental violence Note. L=lifetime, C=current. aOdds ratio.

x

p

OR'

4.98

0.026

4.65

3.84

0.050

3.77

6.89 5.12 3.81 6.47

0.009 0.024 0.051 0.011

0.78 0.73 0.77 375.90

4.05

0.044

5.19

4.91 14.01 6.49 5.72

0.027 0.000 0.011 0.017

0.32 1.88 4.81 0.22

4.74 5.32 3.60

0.029 0.021 0.058

2.39 26.89 0.15

likely to receive a lifetime diagnosis of PTSD. The model revealed several contributors to the PTSD diagnosis, including mother care (X 6.89, p=.009), father protection (X =5.12, p=.02 4 ), family adaptability ( =3.81, p=.051), and interparental violence (X =6.47, p=.011). The PTSD assessments also revealed a variety of traumatic stressors, including discovery of a suicide, family violence, physical assault, extrafamilial sexual abuse and witnessing a car accident or serious injury. However, the number of adolescents with PTSD was small (N=9; double exposure=5, single exposure=l, and comparison group=3); therefore, the analysis may be unreliable and must be interpreted with caution. Nonetheless, the variables identified by the model seem to make intuitive sense as factors that may contribute to an adolescent's vulnerability to trauma after a stressful event (e.g., family chaos and lack of family support). Exposure to interparental violence was not a significant predictor for dysthymia, ADHD, conduct disorder, or overanxious disorder when the experience of direct physical abuse was included in the model. In addition to mood and conduct disorders, cigarette, alcohol, and drug abuse were also examined. While the experience of physical abuse accounted for most of the risk for cigarette and drug use (odds ratio for cigarettes=4.8 1, for drugs = 26.89), spouse abuse in the home did slightly in-

365

Pelcovitz et al crease the likelihood that the adolescent would abuse these substances (odds ratio for cigarettes= .22, for drugs=.15). This was not true for alcohol abuse, however, inasmuch as adolescents in the double-exposure group were at no greater risk for alcohol abuse than were those in the singleexposure group. DISCUSSION In violent family settings, there are numerous risk factors for adolescent survivors, ranging from repeated physical abuse and severe parental psychopathology to poor parenting skills and low family cohesion. These factors are often confounded in research methodologies. However, by including a group that was both physically abused and witness to interparental violence, and by utilizing thorough assessment interviews for both adolescents and their parents, the current study was able to draw more definitive conclusions about the impact of domestic violence on the adolescents. The use of a semi-structured diagnostic interview to assess adolescents exposed to marital violence is also novel in the literature. The hypothesis that adolescents with double exposure to violence in their homes may be more vulnerable to psychiatric disorder than those exposed to physical abuse without interparental violence was supported by the findings. Exposure to interparental violence, controlling for direct physical abuse, contributed significantly to models predicting PTSD, SAD, depression, and ODD. Overanxious disorder, dysthymia, ADHD, and conduct disorder were not significantly different in the two abuse groups, and their rates of diagnosis were consistent with theoretical predictions about the psychological impact unique to interparental violence. Anxiety disorders in the doubly exposed group were more likely to take the trauma-specific forms of PTSD and separation anxiety, rather than more generalized anxiety (since overanxious disorder was among those factors not significantly different in the two abuse groups). This is consistent with the view that when the protective shield of family cohesion is damaged by domestic violence, the adolescent is left without support or a safety net to buffer traumatic stressors. While both abuse groups were at risk for dysthymia (a mild depression), the double-exposure group was more likely to develop a more severe form and meet criteria for a major depressive episode. Similar rates

of externalizing disorders among trauma survivors are supported by the findings of Ford and colleagues (1999), who reported that exposure to a traumatic event was not uniquely predictive of a diagnosis of ADHD. The literature on the psychiatric sequelae of witnessing community violence reveals similar findings. In a review of this literature, Warner and Weist (1996) reported that inner-city children and adolescents exposed to violence in their neighborhoods were most frequently diagnosed with PTSD, SAD, and depression. While both retrospective (Raskin, Peeke, Dickman, & Pinsker, 1982) and correlational studies (Dykman et al., 1997) have

documented the high incidence of anxiety disorders in victims of child abuse, the current study is the first to demonstrate that physical abused adolescents exposed to interparental violence are at particular risk for developing an anxiety disorder. As one would expect, physically abused children are at risk for becoming violent themselves (Gabel, 1997a; Ross, 1996). In the current study,

abused adolescents who were exposed to interparental violence were at risk for ODD, but were less likely to meet criteria for conduct disorder than were abused adolescents not living with marital violence. This finding is in keeping with that by Ford et al. (1999), in whose sample of children a history of traumatic victimization predicted ODD rather than other disruptive behavior disorders. The adolescents in the single-exposure (physical abuse only) category were found to be at greater risk for engaging in aggressive behavior towards others, and "enacting" the impact of the trauma in more destructive terms (Pelcovitz & Kaplan, 1994).

It seems probable that the ODD seen in this sample may be a manifestation of depression. Kashani and Henigan (1997) have pointed out that the irritable mood of adolescent depression may at times be difficult to distinguish from the anger typical of disorders of conduct. Other investigators (Wenning, Nathan, & King, 1993) have found that chil-

dren with ODD are indeed at significant risk for depression.

For children who do not go on to develop conduct disorder, it has been hypothesized that ODD may instead be a precursor to a mood disorder or to a passive-aggressive personality style (Gabel, 1997b). One might hypothesize that adolescents in the double-exposure category, in contrast to those in the single-exposure category, may fear dealing

366

PsychiatricDisorders and Domestic Violence

with conflict overtly and rely instead on passive problem-solving strategies. This possibility is consistent with Gabel's (1997b) discussion of the overlap between ODD and passive-aggressive coping styles. Depression has long been associated with abuse (Kazdin, Moser, Colbus, & Bell, 1985). Elsewhere (Kaplan et al., 1998), the abused adolescents in the present sample have been reported as seven times more likely than nonabused controls to develop a major depressive disorder. While learned helplessness has been reported to be a major contributor to depression in the mothers of children exposed to interparental violence (Walker, 1984), it is probably an even more salient dynamic in children who are powerless in the face of the indiscriminate violence characterizing their households. In light of the high prevalence of depression in parents of abused children (Kaplan, Pelcovitz, Salzinger, & Ganeles, 1983), it is possible that there is also a generic component to the increased prevalence of depression in the present study sample. Although no empirical studies have examined the diagnosis of PTSD among adolescents exposed to interparental violence, certainly one would anticipate a high prevalence. However, the odds ratio of an over 300-fold increased risk for PTSD in the doubly exposed adolescents must be interpreted with caution, given the relatively small number of participants (N=9) who had ever met criteria for PTSD. It is of note, however, that previous research has documented highly increased risk in children exposed to domestic violence. For example, McCloskey and Walker (2000) found that among the 15% of their sample of children who met criteria for PTSD, 83% came from violent homes. The authors pointed out that the children exposed to domestic violence were at great risk for PTSD, whether or not they had experienced additional traumatic stressors. Straus, Gelles, and Steinmetz (1980) also reported that boys who witness paternal violence are at a 1,000% increased risk for assaulting their own partners as adults; interestingly, not only was interparental domestic violence a significant predictor of PTSD, so also were variables measuring family support and family adaptability. Unexpected cycles of interparental violence may serve to intensify family chaos and weaken parent behavior associated with safety and protection. Therefore, adolescents in the double-exposure category may be at greater risk

for feelings of helplessness and subsequent maladaptive beliefs and cognitive schema about themselves and the world around them, over which they feel they have little control. In contrast, adolescents in the abuse-only group have been characterized as actively involved in triggering the violence, and often perceive the abuse to be routine and "ordinary" (Pelcovitz et al. 1994). Feeling relatively empowered, i.e., feeling involved in the triggering, may serve as a resilience or protective factor in the development of some post-trauma sequelae (Roth, Lebowitz, & DeRosa, 1997). However, the adolescent's perception of being involved in the triggering may at the same time pave the way for a more aggressive problem solving approach to conflict. Understanding the developmental and psychological issues unique to adolescent witnesses of family violence is an essential part of prevention and intervention. As adolescent trauma survivors begin to forge new attachments and negotiate intimate relationships with others for the first time, they may be especially vulnerable to dating abusive partners, initiating violent conflict, and engaging in high-risk, impulsive behavior (Pelcovitz & Kaplan, 1994). For example, in the present sample, abused adolescents were 27 times more likely to abuse drugs than were adolescents in the nonabused control group. O'Keefe (1998) found, among other variables, that young adults who had witnessed interparental violence were more likely to be victims in a violent relationship if they accepted dating violence as the norm and had been victims of child abuse. Ross (1996) reported that fathers and mothers who had been physically abused as adolescents were, respectively, 1.77 and 2.64 times more likely to abuse their own children than were their peers who had not been thus abused as adolescents. While these types of maladaptive behavior among distressed adolescents may not be unique to family violence, if exposure to trauma exists, it is necessary that interventions address these issues in the context of coping with an abusive history. An intriguing question is whether a relationship exists between abuse and date of onset of psychiatric diagnosis. The work of McLeer, Deblinger, Atkins, Foa, and Rolphe (1988) has suggested that psychiatric illness can develop considerably later than date of onset of abuse. As reported elsewhere (Kaplan, Labruna, Pelcovitz, & Long, 2000), in

367

Pelcovitz et al the current study's total abuse sample, physical abuse typically occurred before the onset of psychiatric symptoms. However, the limitations of retrospective measures of psychiatric diagnosis did not permit careful assessment of the relationship

parameters for the assessment and treatment of children and adolescents with depressive disorders. Journalof the American Academy of Child and Adolescent Psychiatry, 37,

1234-1238. Bernstein, G.A., &Kinlan, J. (1997). Summary of the practice parameters for the assessment and treatment of children and adolescents with anxiety disorders. Journalof the American

between onset of abuse and onset of psychiatric Academy of Child &Adolescent Psychiatry, 36, 1639-1641. disorder. Consequently-a limitation of this Cicchetti, D., & Lynch, M. (1995). Failures in the expectable environment and their impact on individual development: study-it is possible that over time, a different patThe case of child maltreatment. In D. Cicchetti & D.J. Cotern of psychiatric disturbance may emerge. hen (Eds.), Developmental psychopathology, Vol. 2: Risk, Current findings are also limited by the ethnic disorder, and adaptation (pp. 32-71). New York: John Wihomogeneity of the (Caucasian) sample. As disley. cussed previously, this selection process was ne- Conners, K. (1969). A teacher rating scale for use in drug studies with children. American Journal of Psychiatry, 126, cessitated because the small number of recruitable 884-888. minority subjects would have rendered results of Dadds, M.R., Holland, D.E., Laurens, K.R., Mullins, M., Barracial comparisons unreliable. The generalizability rett, P.M., & Spence, S.H. (1999). Early intervention and prevention of anxiety disorders in children: Results at 2of the findings may also have been effected by use year follow-up. Journalof Consulting & Clinical Psycholof a sample of adolescents for which the abuse was ogy, 67, 145-150. reported and substantiated by child protective ser- Davies, P.T., &Windle, M. (1997). Gender-specific pathways vices. In some ways this is an improvement over between maternal depressive symptoms, family discord, and adolescent adjustment developmental psychology. Deprevious methodologies, which have relied on less velopmental Psychology, 33, 657-668. representative samples (e.g., shelter residents or R.A., McPherson, B., Ackerman, P.T., Newton, J.E., children referred for mental health treatment). It is Dykman, Mooney, D.M., Wherry, J., & Chaffin, M. (1997). Internalnot clear, however, whether a sample drawn from izing and externalizing characteristics of sexually and/or physically abused children. Integrative Physiological&Bea pool of adolescents whose abuse has been inveshavioral Science, 32, 62-83. tigated and documented represents a more severely Fantuzzo, J.W., & Lindquist, C.U. (1989). The effects of obeffected population. serving conjugal violence on children: A review and analyIn conclusion, physically abused adolescents sis of research methodology. Journalof Family Violence, 4, 77-94. who are also faced with interparental violence are more vulnerable to a number of psychiatric disor- Ford, J.D., Racusin, R., Daviss, W.B., Ellis, C.G., Thomas, J., Rogers, K., Reiser, J., Schiffman, J., &Sengupta, A. (1999). ders, including anxiety and depression. These findTrauma exposure among children with oppositional defiant ings should alert clinicians and child protective disorder and attention deficit-hyperactivity disorder. Journal of Consultingand ClinicalPsychology, 67, 786-789. services alike to the importance of thorough assessments of marital and family functioning when Gabel, S. (1997a). Oppositional defiant disorder. In J.D. Noshpitz, P.F. Kemberg, &J.R. Bemporad (Eds.), Handbook of children and adolescents are living in violent child and adolescent psychiatry. The grade school child: homes. Unfortunately, most adolescent witnesses Development and syndrome (pp. 351-359). New York: John Wiley. fail to receive thorough mental health evaluation and treatment (Kolko, Selelyo, & Brown, 1999). Gabel, S. (1997b). Conduct disorder in grade-school children. In J.D. Noshpitz, P.F. Kernberg, & J.R. Bemporad (Eds.), This is particularly troubling in light of the inHandbook of child and adolescentpsychiatry. The grade creased risk for a lifetime history of recurring afschool child: Development and syndrome (pp. 359-401). New York: John Wiley &Sons. fective disorder when initial episodes of depression or anxiety are not treated (American Acad- Gammon, G., John, K., &Weisman, M. (1984). Structured assessment of psychiatric diagnosis and psychosocial function emy, 1998; Bernstein & Kinlan, 1997; Dadds et and supports inadolescents: A role in the secondary prevenal., 1999). Expanding our understanding of the tion of suicide. In H.Sudak, A. Ford, &N. Rushford (Eds.), Suicide in children and adolescents (pp. 183-208). Littlelchallenges facing adolescent trauma survivors will ton, MA: John Wright. permit interventions designed to reduce the likeliGriest, D., Wells, K.C., & Forehand, R. (1979). Examination hood of the continued transmission of the intergenof predictors of maternal perceptions of maladjustment in erational cycle of domestic violence. clinic children. Journal of Abnormal Psychology, 88, References American Academy of Child & Adolescent Psychiatry Work Group on Quality Issues. (1998). Summary of the practice

277-281. Henning, K., Leitenberg, H., Coffey, P., Bennet, T., & Jankowski, M.K. (1997). Long-term psychological adjustment to witnessing interparental physical conflict during childhood. ChildAbuse and Neglect, 21, 501-515.

368

Psychiatric Disorders and Domestic Violence

Holden, G.W., Stein, J.D., Ritchie, K.L., Harris, S.D., & Jouriles, E.N. (1998). Parenting behaviors and beliefs of battered women. In G.W. Holden, R. Geffner, & E.N. Jouriles (Eds.), Children exposed to maritalviolence: Theory, research and applied issues (pp. 289-334). Washington DC: American Psychological Association. Hughes, H., Parkinson, D., & Vargo, M. (1989). Witnessing spouse abuse and experiencing physical abuse: A "double whammy "? Journal of Family Violence, 4, 197-209. Jaffe, P., Wolfe, D., &Wilson, S. (1990). Children of battered women. Newbury Park, CA: Sage Publications. Jaffe P., Wolfe D., Wilson, S.K., Zak L. (1986). Family violence and child adjustment: A comparative analysis of girls' and boys' behavioral symptoms. American Journalof Psychiatry, 143, 74-77. Kaplan, S., Labruna, V., Pelcovitz, D., & Long, J. (2000). Physical abuse, psychiatric disorders, and parental mental illness: A sequence analysis. Unpublished manuscript. Kaplan, S., Pelcovitz, D., Salzinger, S., & Ganeles, D. (1983). Psychopathology of parents of abused and neglected children and adolescents. Journal of the American Academy of Child andAdolescent Psychiatry, 22, 238-244. Kaplan, S., Pelcovitz, D., Salzinger, S., Mandel, F., & Weiner, M. (1997). Adolescent physical abuse and suicide attempts. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 799-808. Kaplan, S., Pelcovitz, D., Salzinger, S., Weiner, M., Mandel, F., Lesser, M., & LaBruna, V. (1998). Adolescent physical abuse: Risk for adolescent psychiatric disorder. American Journalof Psychiatry, 155, 954-959. Kashani, J., & Henigan, M. (1997). Dysthymic disorder in latency-age children. In J.D. Noshpitz, P.F. Kernberg, & J.R. Bemporad (Eds.), Handbook of child and adolescent psychiatry. The grade school child: Development and syndrome (pp. 327-342). New York: John Wiley. Kazdin, A., Moser, J., Colbus, D., & Bell, R. (1985). Depressive symptoms among physically abused and psychiatrically disturbed children. Journalof Abnormal Psychology, 94, 298-307. Kilpatrick, K.L., Litt, M., & Williams, L.M. (1997). Posttraumatic stress disorder in child witnesses to domestic violence. American Journalof Orthopsychiatry, 67, 639-644. Kolbo, J., Blakely, E., & Engleman, D. (1996). Children who witness violence: A review of empirical literature. Journal of InterpersonalViolence, 11, 281-293. Kolko, D.J., Selelyo, J., & Brown, E.J. (1999). The treatment histories and service involvement of physically and sexually abusive families: Description, correspondence, and clinical correlates. ChildAbuse and Neglect, 23, 459-476. Leckman, J., Sholkomskas, O., Thompson, W., Belanger, A., & Weissman, M. (1982). Best estimate of lifetime psychiatric diagnosis. Archives of General Psychiatry, 39, 879-883. McLeer, S.V., Deblinger, E., Atkins, M.S., Foa, E.B., & Rolphe, O.L. (1988). Post-traumatic stress disorder in sexually abused children. Journalof the American Academy of Child and Adolescent Psychiatry, 27, 650-654. McCloskey, L.A., & Walker, M. (2000). Posttraumatic stress in children exposed to family violence and single-event trauma. Journal of the American Academy of Child and Adolescent Psychiatry, 39, 108-115. Meller, W.H., &Borchardt, C.M. (1996). Comorbidity of major depression and conduct disorder. Journal of Affective Disorders, 39, 123-126.

O'Keefe, M. (1995). Predictors of child abuse in maritally violent families. Journalof InterpersonalViolence, 10, 3-25. O'Keefe, M. (1996). The differential effect of family violence on adolescent adjustment. Child and Adolescent Social Work Journal,13, 51-67. O'Keefe, M. (1998). Factors mediating the link between witnessing interparental violence and dating violence. Journal of Family Violence, 13, 39-57. Olson, D.H., Russell, C.S., & Sprenkle, D.H. (1985). Family and marital systems: A circumplex model. Terapia Familiare, 18, 39-58. Orvaschel, H., Puig-Antich, P., Chambers, W., Abrizi, M.A., & Johnson, R. (1982). Retrospective assessment of prepubertal major depression with Kiddie SADS. Journal of the American Academy of Child Adolescent Psychiatry, 21, 392-397. Parker, G., Tupling, H., & Brown, L.B. (1979). A parental bonding instrument. BritishJournalof Medical Psychology, 52, 1-10. Pelcovitz, D., Bester, J., & Kaplan, S. (2000). Academicfunctioning in adolescents exposed to maritalviolence. Unpublished manuscript. Pelcovitz, D., & Kaplan, S. (1994). Child witnesses of violence between parents: Psychosocial correlates and implications for treatment. Child and Adolescent Psychiatric Clinics of North America, 3, 745-757. Pelcovitz, D., Kaplan, S., Goldenberg, B., Mandel, F., Lehane, J., & Guarrera, J. (1994). Post-traumatic stress disorder in physically abused adolescents. Journal of the American Academy of Childand Adolescent Psychiatry, 33, 305-312. Pynoos, R.S. (Ed.). (1994). Posttraumaticstress disorder: A clinicalreview. Lutherville, MD: Sidran Press. Raskin, M., Peeke, H.V., Dickman, W., &Pinsker, M. (1982). Panic and generalized anxiety disorders: Developmental antecedents and precipitants. Archives of GeneralPsychiatry, 39, 687-689. Ross, S.M. (1996). Risk of physical abuse to children of spouse abusing parents. Child Abuse and Neglect, 20, 589-598. Roth, S., Lebowitz, L., & DeRosa R.R. (1997). Thematic assessments of posttraumatic stress reactions. In J.P. Wilson & T. Keane (Eds.), Assessing psychological trauma and PTSD (pp. 512-528). New York: Guilford Press. Salzinger, S., Feldman, R., Hammer, M., & Rosario, M. (1992). Constellations of family violence and their differential effects on children's behavioral disturbance. Child Behavior Therapy, 4, 23-40. Silvem, L,, Karl, J., Waelde, L., Hodges, W.F., Starek, J., Heidt, E., & Min, K. (1995). Retrospective reports of parental partner abuse: Relationships to depression, trauma symptoms and self-esteem among college students. Journal of Family Violence, 10, 177-202. Skre, I., Onstad, S., Torgersen, S., &Kringlen, E. (1991). High interrater reliability for the structured clinical interview for DSM-III-R axis I. Acta Psychiatrica Scandinavia, 84, 167-173. Spitzer, R.L., Williams, J.B., Gibbon, M., & First, M. (1990). Structured Clinical Interview for DSM-II-R NP [nonpatient]Edition (SCID-NP Version 1.0). Washington, DC: American Psychiatric Press. Straus, M. (1995). Manualfor the Conflict Tactics Scales (CTS). Family Research Laboratory, University of New Hampshire, Durham. Straus, M., &Gelles, R. (1990). Physical violence in American

369

Pelcovitz et al families: Riskfactors and adaptationsto violence in 8,154

Warner, R.S., &Weist, M.D. (1996). Urban youth as witnesses

families. New Brunswick, NJ: Transaction Books. Straus, M., Gelles, R., & Steinmetz, S.K. (1980). Behind

to violence: Beginning assessment and treatment efforts.

closed doors: Violence in the Americanfamily. New York:

Wenning, K., Nathan, P., & King, S. (1993). Mood disorders in children with oppositional defiant disorder: A

Doubleday/Anchor.

Journalof Youth and Adolescence, 25, 361-377.

U.S. Department of Health and Human Services, National

pilot study. American Journal of Orthopsychiatry, 63,

Center on Child Abuse and Neglect, Child Maltreatment

295-299. Williamson, J.M., Borduin, C.M., & Howe, B.A. (1991). The ecology of adolescent maltreatment: A multilevel examination of adolescent physical abuse, sexual abuse, and ne-

(1998). Reports from the states to the national child abuse and neglect data system. Washington DC: US Government Printing Office. Walker, L.E. (1984). The battered woman's syndrome. New

York: Springer.

glect Journal of Consulting & Clinical Psychology, 59,

449-457.

For reprints: David Pelcovitz, Ph.D., Division of Child and Adolescent Psychiatry, North Shore University Hospital, 300 Community Drive, Manhasset, NY 11030 [E-mail: [email protected])