Parameters of Reported Childhood Sexual Abuse and Assault in Adolescents and Adults with Borderline Personality Disorder

Abstract

Objective:

Prior research has demonstrated a link between childhood sexual abuse and borderline personality disorder (BPD) in adolescents and adults and has indicated that more severe abuse is related to poorer psychosocial functioning. The present study describes the overall severity of sexual abuse/assault in adolescents and adults with BPD and compares both groups on specific parameters of abusive and assaultive experiences.

Methods:

Participants included 104 adolescent (aged 13–17 years) inpatients with BPD, and 290 adult inpatients with BPD. All participants completed two interviews that assessed the presence and severity of sexual abuse/assault.

Results:

Of the studied patients with BPD, 26.0% of adolescents and 62.4% of adults reported a childhood history of sexual abuse/assault before the age of 18. Adults had higher scores on an index of sexual abuse severity than adolescents, and a higher proportion of adults reported scores in the severe range. Adults with BPD were also more likely than adolescents to report having experienced sexual abuse/assault that occurred at multiple developmental stages, was frequent (i.e., weekly basis or more), was longer in duration (i.e., a year or more), and was perpetrated by a parent. The groups did not differ on other parameters.

Conclusions:

Taken together, these results suggest that adults with BPD are more likely to report childhood sexual abuse/assault than adolescents with BPD. Additionally, adults report histories of sexual abuse/assault that are more severe than adolescents with BPD, with specific differences observed in timing, frequency, duration, and perpetrator.

The relationship between experiences of sexual abuse in childhood and borderline personality disorder (BPD) in adulthood has been documented in several studies.^1–5 Although the link between abuse and BPD is consistently found, the strength of this relationship varies considerably by study, with one meta-analysis finding only a moderate effect size for the relationship between childhood sexual abuse and adult BPD.⁶ This observed variability suggests that the relationship between childhood sexual abuse and BPD symptomatology is likely complex and may be influenced by multiple biosocial factors.

Recently, there has been an increasing amount of research on BPD in adolescents, which has suggested that diagnosis of BPD in adolescents is valid, and that adolescent BPD shares a similar profile of risk factors and clinical correlates with adult BPD.⁷ With respect to sexual abuse, a meta-analysis of existing studies has found that a history of childhood sexual abuse is associated with a five-fold increase in odds of BPD diagnosis in youth.⁷ There has been less work comparing experiences of sexual abuse in adults and adolescents with BPD. One study⁸ found that borderline adults reported higher rates of sexual abuse (and other forms of childhood adversity) than borderline adolescents, who in turn reported significantly more abuse than psychiatrically healthy adolescents.

Additional studies have highlighted the importance of examining specific aspects of sexual abuse—rather than only its history—in patients with BPD. One study⁹ of 41 adult patients with BPD with a history of sexual abuse found that overall severity of sexual abuse (determined by scores on three dimensions related to perpetrator(s), duration, and type of abuse) was associated with severity of BPD symptoms. The most predictive severity dimension was duration; sexual abuse that was ongoing was associated with greater BPD symptoms and increased parasuicidality. Another study¹⁰ of 181 adult patients with BPD with a history of sexual abuse found that overall sexual abuse severity (determined by scores on several dimensions, including age, duration, frequency, relationship to perpetrator, and nature of abuse) was related to severity of BPD symptoms and level of psychosocial impairment. An additional study¹¹ found that among 61 adult patients with BPD, severity of childhood sexual abuse (determined by scores related to perception of abuse, relationship and number of perpetrators, disclosure events, duration, frequency, and overall life effect of abuse) was associated with increased risk of suicidal behavior. Comparatively less is known regarding these specific aspects of sexual abuse/assault in adolescents with BPD.

The current study examined the experience of specific parameters of childhood sexual abuse and sexual assault reported by adults with BPD in comparison to adolescents with BPD. To the best of our knowledge, it is the first study to compare specific aspects of abusive and assaultive experiences in these two developmental groups. This study also characterized the average level and distribution of sexual abuse severity in both the adolescent and adult borderline groups.

Method

The methodology of this study has been presented before in detail.¹² All study procedures were approved by the institutional review boards at the participating institutions. Briefly, all adults with BPD were inpatients at McLean Hospital in Belmont, Massachusetts who were admitted between June 1992 and December 1995. Each patient was initially screened to determine that he or she: 1) was between the ages of 18–35, 2) had normal or better intelligence, and 3) had no history or current symptomatology of a serious organic condition that could cause psychiatric symptoms, schizophrenia, or bipolar I disorder. Written informed consent was obtained from each patient. These adult participants were administered three diagnostic interviews by research staff blind to their clinical diagnoses. The instruments were: 1) the Structured Clinical Interview for DSM-III-R Axis I Disorders (SCID I)¹³, 2) the Revised Diagnostic Interview for Borderlines (DIB-R)¹⁴, and 3) the Diagnostic Interview for DSM-III-R Personality Disorders (DIPD-R)¹⁵. The inter-rater and test-retest reliability of these measures have been found to be good-excellent.^16,17

Adolescents (aged 13–17) with presumptive BPD were recruited from four units at McLean Hospital and one unit at the Icahn School of Medicine at Mount Sinai between the dates of August 2007 and September 2012. For adolescent participants, parents provided consent and adolescents provided assent. Adolescent participants, who all met the same exclusion criteria concerning intelligence and co-occurring disorders as adult participants, were then administered the following diagnostic assessments: 1) the Structured Clinical Interview for DSM-IV Childhood Diagnoses (KID-SCID);¹⁸ 2) the Revised Diagnostic Interview for Borderlines (DIB-R);¹⁴ and 3) the Childhood Interview for DSM-IV Borderline Personality Disorder (CI-BPD).¹⁹

To assess pathological experiences reported to have occurred before age 18, all participants were administered two interviews: 1) the Revised Childhood Experiences Questionnaire (CEQ-R);²⁰ and 2) the Abuse History Interview (AHI).²¹ The CEQ-R is a semi-structured interview which assesses a range of pathological and protective childhood experiences; its psychometric properties have been described elsewhere.²⁰ The AHI assesses seven parameters of childhood sexual abuse/assault that have been highlighted as having clinical importance.²² These parameters include: age of abuse, frequency of abuse, duration of abuse, relationship to perpetrator(s), nature of abuse, and use of force or violence.

A continuous measure of the severity of childhood sexual abuse/assault was calculated by adding the scores for each of the seven parameters of sexual abuse, resulting in a summary score with a possible range of 0–19. See Table 1 for a detailed summary of the scores assigned to each dimension of sexual abuse. For example, a patient who reported being sexually abused in early childhood/latency only (1 point), on a weekly basis (3 points), for more than a year (3 points), by one parent (3 points for relationship to patient and 1 point for number of perpetrators), with fondling (2 points), and with force or violence (1 point) would receive a score of 14 on this measure.

Table 1.

Distribution of Childhood Sexual Abuse and Assault Severity Scores for 104 Adolescents with BPD vs. 290 Adults with BPD

	BPD Adolescents (N=104)		BPD Adults (N=290)
	N	%	N	%
Sexual Abuse/Assault Severity
0–7 points	77	74.1	109	37.6
8–15 points	27	25.9	93	32.1
16–19 points	0	0.0	88	30.3

Note: Differences in level of severity were tested with the Wilcoxon rank-sum test; adults with BPD reported significantly higher scores than adolescents with BPD (z=−7.42, p<0.001).

Statistical Analyses

Logistic regression was used to compare the prevalence of each aspect of childhood sexual abuse/assault in adolescents with BPD vs. adults with BPD. However, Firth’s penalized logit method²³ was used in analyses of 11 variables due to sparseness of data in one or both groups (i.e., when cells contained data from less than five cases). Analyses pertaining to the severity of sexual abuse/assault were conducted using linear regression. The Wilcoxon rank-sum test was used in analyses comparing groups’ distribution of levels of severity scores. All analyses comparing parameters and levels of severity of sexual abuse/assault were conducted controlling for sex. Additionally, all between-group analyses of the individual abuse parameters used p-values that were adjusted for multiple comparisons using Hochberg’s step-up procedure.²⁴

Results

Participants

One hundred and four participants were adolescents between the ages of 13 and 17 who met DIB-R and DSM-IV criteria for BPD. Two hundred and ninety subjects were adults between the ages of 18–35 who met DIB-R and DSM-III-R criteria for BPD.

Demographic characteristics have also been described before.¹² Briefly, there were significantly more females in the adolescent sample than in the adult sample. In terms of GAF scores, adolescents with BPD had significantly lower scores (by about 4 points) than adults with BPD.

Severity of Sexual Abuse

Overall, 26.0% (27 of 104) of adolescent borderline patients and 62.4% (181 of 290) of adult borderline patients reported a childhood history of sexual abuse and/or assault that occurred before age 18. History of sexual abuse/assault was significantly more prevalent in borderline adults than adolescents, OR=7.01, p < 0.001.

Table 1 details the distribution in continuous sexual abuse/assault severity scores for the 104 adolescents with BPD and 290 adults with BPD. As can be observed in the table, for adults with BPD, severity scores were generally evenly divided in terms of low (0–7 points; 37.9%), moderate (7–15 points; 32.1%), and high (16–19 points; 30.3%) scores. By comparison, most (74.1%) adolescents with BPD scored in the low range, about a quarter (25.9%) scored in the moderate range, and none scored in the high range. The distribution of levels of severity scores differed significantly by group, with borderline adults reporting significantly greater sexual abuse/assault severity than adolescents with BPD (z=−7.42, p<0.001).

For patients reporting a history of childhood sexual abuse/assault, Table 2 shows the percentages from each group that reported each level of the seven parameters studied. With respect to the age at which abuse/assault occurred, there were no differences between the groups regarding whether abuse/assault occurred in early childhood/latency or adolescence only. However, a significantly higher number of borderline adults than adolescents reported experiencing abuse/assault at both developmental stages (i.e., childhood/latency and adolescence). In terms of frequency, borderline adults were significantly less likely than adolescents to report abuse/assault occurring on a monthly or less frequent basis but were significantly more likely to report abuse/assault that occurred on weekly or daily basis. The groups did not differ regarding rate of one-time abusive/assaultive experiences. In terms of duration, borderline adolescents were more likely than adults to report abuse of a month to less than a year in duration, whereas adults were more likely than adolescents to report abuse lasting one year or more. There were no between-group differences with respect to abusive experiences that lasted less than a month in duration. Borderline adults were significantly more likely than adolescents to report that a parent was a perpetrator of sexual abuse. The groups did not differ significantly regarding the rates of having a stranger or known person/sibling as perpetrators of abuse/assault. The groups also did not differ significantly regarding the number of perpetrators of abuse/assault, the nature of the most serious type of abuse/assault, or whether abusive/assaultive experiences occurred in the context of force or violence.

Table 2.

Parameters of Sexual Abuse and Sexual Assault among Adolescent and Adult Borderline Patients with Reported History of Childhood Sexual Abuse/Assault

	BPD Adolescents (N=27)		BPD Adults (N=181)
	N	%	N	%	Odds-ratio	p-valuea
Age of Abuse
Early childhood/latency (0–12) (1 point)	16	59.3	61	33.7	0.32	0.076
Adolescence (13–17) (2 points)	9	33.3	24	13.3	0.35	0.191
Both (3 points)*	2	7.4	96	53.0	11.75	0.006
Frequency
Once (1 point)*	0	00.0	33	18.2	10.75	0.495
Monthly/less (2 points)*	27	100.0	43	23.8	0.01	0.006
Weekly/daily (3 points)*	0	00.0	105	58.0	78.02	0.034
Duration
1 month or less (1 point)	9	33.3	28	15.5	0.33	0.165
> Month < 1 year (2 points)	18	66.7	20	11.0	0.03	<0.001
1 year or more*	0	00.0	133	73.5	167.85	0.006
Perpetrator (closest relationship)
Stranger (1 point)*	4	14.8	3	1.7	0.12	0.059
Known abuser/sibling (2 points)	22	81.5	99	54.7	0.24	0.076
Parent (3 points)*	1	3.7	79	43.6	15.34	0.021
Number of Perpetrators
One (1 point)	17	63.0	64	35.4	0.31	0.076
Two (2 points)*	5	18.5	45	24.9	1.43	0.889
Three or more (3 points)*	5	18.5	72	39.8	2.66	0.307
Nature of Abuse (most serious type)
Observation Without Contact (1 point)*	2	7.4	3	1.7	0.17	0.307
Fondling (2 points)*	5	18.5	36	19.9	1.08	0.889
Penetration (Oral/Vaginal/Anal) (3 points)	20	74.1	142	78.4	1.25	0.889
Force or Violence
Yes (1 point)	17	63.0	92	50.8	0.64	0.889

Note: All analyses controlled for sex.

^*Firth’s penalized logit method used in cases where cells contained n≤5.

^aP-values adjusted for multiple comparisons using Hochberg’s (1988) step-up procedure.

Discussion

This study yielded several findings pertaining to the experiences of sexual abuse and sexual assault in adults and adolescents with BPD. The first finding is that overall prevalence of sexual abuse/assault was higher in adults with BPD than adolescents with BPD, as was the overall severity of sexual abuse/assault. The distribution in severity scores also differed between these groups, with only adults with BPD reporting scores in the most severe range.

We also found that even among patients with a history of sexual abuse/assault, adolescents’ and adults’ experiences differed on several key parameters. Adults with BPD were more likely than adolescents with BPD to report having experienced sexual abuse that occurred at multiple developmental stages, occurred frequently (i.e., on a weekly basis or more), was longer in duration (i.e., a year or more), and was perpetrated by a parent. In addition to the aforementioned differences, the groups were similar on several important aspects of abusive/assaultive experiences. In particular, there were no differences between adults and adolescents in terms of age at which the abuse first started, the number of perpetrators, the nature of the abuse/assault (i.e., involving penetration or not), and whether abuse/assault occurred in the context of force or violence.

There are several possible explanations for the observed differences in parameters of childhood sexual abuse/assault in adolescents with BPD compared to adults with BPD. First, adults reported on experiences that occurred approximately 30 years prior to those reported on by the adolescent group; thus, the between-group differences may reflect larger cultural differences between these two age cohorts. Overall rates of sexual abuse/assault and other forms of childhood maltreatment have declined steeply since the early 1990s, as other indicators of child welfare have improved.²⁵ Finkelhor and Jones²⁵ describe several possible explanations for this overall pattern, noting that increases economic prosperity, agents of social intervention, and the availability of psychiatric interventions may contribute to preventing violence and fostering better welfare of children. Another possible explanation is that the two studied cohorts may have been exposed to different styles of parenting, depending on the era in which they were raised. More specifically, adults with BPD may have been more likely to have been raised with a “free range” parenting style with less parental supervision (thus increasing opportunity for abuse), whereas adolescents may have been raised with more supervision or “helicopter” parents. Additionally, there have been other relevant cultural shifts including increased collective awareness concerning the prevalence of childhood abuse and increased intervention when abuse is suspected or discovered. Furthermore, because they were older when assessed, the adult group had more years to accumulate experiences of sexual abuse, which may be reflected in the prevalence, duration, and/or extent of abuse. In particular, the adults had more time to begin dating, increasing the likelihood of abuse by intimate partners. Finally, it may be the case that individuals with extensive experiences of abuse are more likely to be stably symptomatic and therefore more likely to retain a diagnosis of BPD in adulthood.

It is also important to note that although the adults and adolescents differed with respect to the severity of sexual abuse/assault, they did not differ in terms of the severity of their borderline symptomatology. A previous study involving the same two samples¹² found that the two groups did not differ with respect to the severity of most (20/24) symptoms of BPD that were studied. Taken together, these findings suggest that although these two groups shared an outcome (i.e., similar levels of borderline psychopathology), they likely reached this outcome via different pathways of risk, consistent with an equifinality framework in developmental psychopathology²⁶. These findings also illustrate how history of sexual abuse is neither necessary nor sufficient to explain the development of BPD, particularly as many adult patients and an even larger proportion of adolescent patients did not report a history of this type of childhood adversity. These findings regarding the role of sexual abuse in the development of BPD are consistent with earlier work by Zanarini and colleagues.²⁷

These findings also highlight how experiences of sexual abuse/assault are themselves heterogeneous. The descriptive findings on specific facets of sexual abuse/assault present an advance in our understanding of the range of experiences reported by present-day adolescents with BPD. As other studies have highlighted,⁹ the specific characteristics of abusive experiences matter in terms of eventual effects of the abuse. Merely knowing that someone has a history of childhood sexual abuse/assault tells us little about the context in which these experiences occurred or their extent. It is therefore important for assessments of abusive/assaultive experiences to be multifaceted and for evaluations of abuse as a risk factor to be nuanced to better elucidate the role of these experiences in the pathogenesis of BPD.

These findings also have important clinical implications for the treatment of BPD. Patients with BPD and a history of sexual trauma, particularly of high severity, may benefit from treatments that are tailored to their specific needs. Fortunately, some treatments have been developed in recent years for patients with BPD who are also experiencing trauma-related sequelae. These include dialectical behavior therapy plus prolonged exposure (DBT-PE)²⁸, which combines elements of DBT and exposure-based trauma interventions, and DBT for post-traumatic stress disorder (DBT-PTSD)²⁹, which is designed explicitly for patients with a history of childhood sexual abuse (with or without co-occurring BPD or BPD features). In trials, these treatments have shown promise in improving outcomes for this group of patients and are also well-tolerated by patients.^28–29

Additionally, adolescents with BPD and a significant sexual abuse/assault history may constitute a particularly high-risk group for whom early intervention is warranted. Specifically, history of sexual abuse in adolescents with BPD has been found to predict stability of BPD from adolescence to adulthood.³⁰ Additional recent work has demonstrated a reciprocal relationship between victimization and BPD symptomatology over time, suggesting that symptoms elevate risk for further victimization and vice versa.³¹ Thus, treatment of BPD in these young patients may help prevent further exposure to risk, and attempts to reduce risk exposure and address the sequelae of abuse/assault can be beneficial for the trajectory of BPD symptoms over time. Fortunately, several treatments for adolescents with BPD have demonstrated efficacy, including adaptations of DBT³² and mentalization-based therapy (MBT),³³ and stepped-care treatment delivery systems have also been proposed for this population.³⁴

These findings also contribute to the growing body of literature supporting the validity of the diagnosis of BPD in adolescence. Other researchers⁷ have highlighted the importance of examining “aetiological validity” and other relevant clinical correlates in addition to symptomatology to fully evaluate the construct of BPD in adolescence. The present work and other recent studies using this sample^8,12 have highlighted convergence in BPD presentation (among multiple domains) between adolescents and adults with BPD and divergence in presentation between healthy adolescents and adolescents with BPD. Taken together, these findings and others in the literature suggest that BPD is a valid clinical diagnosis for adolescents and that symptoms of BPD are not indicative only of a particularly tumultuous developmental phase.

This study had several limitations. First, all participants were inpatients at the time of assessment; thus, findings may not generalize to less severely ill patients with BPD. Second, the size of the sample of adolescents with BPD and a history of abusive/assaultive experiences was relatively small, which may have limited power to detect some differences between the two groups. Third, information on pathological childhood experiences was collected via retrospective report. Although participants had to provide detailed vignettes of their experiences (with ambiguous accounts not counted), it is still possible for there to have been some inaccuracies in reporting. Additionally, we were limited in the number of variables we could examine; thus, there may have been other baseline differences between the groups that we could not describe. Furthermore, the present study’s design does not allow for the examination of the differential impact of early childhood sexual abuse vs. assaultive experiences that are more common in adolescence (e.g., “date rape”). Additional research is needed to elucidate the effects of the interplay between exposure to these different forms of adversity and other factors on the etiology of BPD. Finally, there were some issues with sparseness of data in some comparisons of parameters of sexual abuse/assault between adolescents and adults with BPD (i.e., some experiences were very rare or non-existent in the adolescent group). Although we accounted for this sparseness with appropriate statistical modeling techniques (i.e., Firth’s penalized logit), the significance of some between-group differences may be underestimated.

Taken together, these results suggest that adults with BPD are more likely to report sexual abuse/assault than adolescents with BPD. Additionally, adults report abuse/assault that is more severe than adolescents with BPD, with specific differences observed in timing, frequency, duration, and perpetrator.