Trauma,Dissociation,and Posttraumatic Stress Disorder in Female Borderline Patients With and Without Substance Abuse Problems

Abstract

Objective: To examine the associations of childhood traumatic experiences and childhood neglect with dissociative experiences and posttraumatic stress disorder (PTSD) in a population of female borderline personality disorder (BPD) patients with and without substance abuse.

Method: The sample included 64 female patients with BPD. Childhood traumatic experiences and childhood neglect were measured using the Structured Trauma Interview, dissociative experiences with the Dissociative Experiences Scale, and PTSD with the Structured Clinical Interview for DSM-IV.

Results: In general, dissociation scores were higher among those with a history of childhood trauma and neglect, in particular among those who reported both sexual and physical abuse before age 16, more than one perpetrator and severe maternal dysfunction. The prevalence of PTSD was clearly associated with the severity of childhood sexual abuse (CSA) in terms of the occurrence of penetration during CSA, intrafamilial CSA, a duration of CSA longer than 1 year and more than one perpetrator. Comorbid substance abuse problems modified the observed associations such that the associations mentioned above were also present or even more pronounced among those without substance abuse, whereas no associations were found in those with substance abuse.

Conclusions: The results suggest a moderately strong association between childhood trauma and neglect with dissociation and PTSD. However, trauma-dissociation and trauma-PTSD links were only observed among BPD patients without addictive problems. The findings are largely consistent with the literature. Potential explanations for the lack of a trauma-dissociation and trauma-PTSD link in the addicted subgroup are discussed.

There is consistent empirical evidence for a strong association between traumatic events and dissociative symptoms and posttraumatic stress symptoms [1, 2], phenomena that are frequently reported in borderline personality disorder (BPD). Borderline patients are known to have high rates of childhood abuse [3–15], high levels of dissociation [16] and frequently a comorbid diagnosis of posttraumatic stress disorder (PTSD) [17, 18]. Dissociation is commonly conceptualized as a psychological defence mechanism or a primary coping style to block out intolerable cognitions and emotions (i.e. to master childhood physical and sexual abuse [19, 20]). Some suggestive evidence indicates that successful psychotherapeutic treatment of BPD depends partly on the extent to which integration (as opposed to dissociation) is increased by, for example strengthening awareness, and hence control, of the dissociative processes maintaining fragmentation [21]. Knowledge about the specificity of the relationship between childhood trauma and dissociation might therefore be of substantial clinical interest.

Borderline patients often have a comorbid diagnosis of substance use disorder, and the results regarding the association of traumatic events and dissociative symptoms and PTSD among substance abusers are highly inconsistent. Most of the studies on this topic show only weak or no significant associations at all [19]. Several explanations have been provided for this discrepancy. In a recent study by Langeland et al. [19] it was shown that methodological issues (underreporting of childhood abuse; psychometric weaknesses of the dissociation measure) are not very likely to be responsible for the observed lack of associations. Rather, it was argued that traumatized individuals with limited capacities to psychologically dissociate may attempt to produce similar soothing or numbing effects by using psychoactive substances. Dissociative symptoms resemble intoxication or withdrawal symptoms. The authors showed that substance abusers might have learned to cope with childhood abuse through a mechanism of ‘chemical dissociation’ that developed instead of psychological dissociation or that may have replaced originally coping mechanisms through psychological dissociation.

In the current study we examine the association between childhood trauma and dissociation and PTSD in BPD patients. We are specially interested in possible differences in the association of childhood trauma and dissociation and PTSD between those borderline patients with a comorbid substance use disorder (SUD) (BPD + SUD) and those without (BPD–SUD). Based on the findings of Langeland et al. [19], we expect that the association between trauma (childhood sexual abuse and childhood physical abuse) and dissociation will be different among BPD–SUD patients, compared to BPD +SUD patients.

Method

Data were collected as part of a larger research project in which the efficacy of dialectical behaviour therapy (DBT) for the treatment of patients with a BPD was examined, for example the impact of baseline severity on the efficacy of DBT [22].

Subjects and procedures

The research subjects were 64 females participating in a randomized controlled trial comparing Dialectical Behaviour Therapy with Treatment as Usual. They were recruited from a pool of 92 clinical referrals from mental health and addiction treatment services within the greater Amsterdam area. Only those female patients who met more than six diagnostic criteria for DSM-IV borderline personality disorder [23] according to both the DSM-IV version of the Personality Diagnostic Questionnaire (PDQ4+) [24]; and the DSM-IV version of the Structural Clinical Interview for DSM-IV Axis II Personality Disorders (SCID-II) [25]; were eligible. In addition, all research subjects were required to have sufficient command of the Dutch language and reside within a 25-mile radius of Amsterdam. Subjects less than 18 years, who met DSM-IV criteria for bipolar disorder, a chronic psychotic disorder, or suffered from severe cognitive impairments, were excluded. Of the 92 referred patients, 28 patients were excluded, mostly because they could not be diagnosed with BPD. Of the remaining 64 patients, 20 (31%) were referred by addiction treatment services and 44 (69%) were referred by mental health services.

All patients were evaluated with a comprehensive assessment battery, administered by clinically experienced and trained diagnosticians. Borderline pathology and symptomatology was measured using the PDQ4+ and the SCID-II. Substance abuse pathology was measured with the EuropASI. The existence of PTSD was examined using the SCID-I, dissociation was examined with the Dissociative Experiences Schedule (DES) and to assess childhood sexual and physical abuse histories the Structured Trauma Interview (STI) was used.

PDQ4+

The Personality Diagnostic Questionnaire, fourth revised edition, is an updated version of the PDQR [24, 26]. The PDQ4+ is a selfadministered, forced-choice, true/false questionnaire. It measures all DSM-IV personality disorders. In this study it was used as a screening device for DSM-IV borderline personality disorder.

SCID-II

The Structured Clinical Interview for DSM-IV Axis II Personality Disorders [25, 27], is a semistructured interview of 108 questions, arranged according to diagnosis, yielding both categorical diagnoses and dimensional scores for each of the 10 DSM-IV personality disorders.

EuropASI

The EuropASI [28] is an European adaptation of the Addiction Severity Index (ASI) [29], a multidimensional instrument for assessment of substance abusers. This widely used semistructured clinical interview is designed to assess the severity of and problems caused by substance abuse. The instrument assesses problems in six different areas: drug and alcohol abuse, medical, psychiatric, legal, family/social and employment/support. For each area, the interview obtains objective information and the subject's judgements of severity and allows the interviewer to produce severity ratings. It has demonstrated both high reliability and validity. Subjects who have an alcohol or drug severity score of five or higher were considered to have clinically relevant substance abuse problems.

SCID-I

The Structured Clinical Interview for DSM-III-R [30, 31], is a semistructured interview, yielding both Axis I lifetime diagnoses (i.e. ever in life) and Axis I current diagnoses (i.e. past month), including PTSD. Posttraumatic stress disorder data were missing in one patient, and therefore the association between trauma and PTSD will be reported for only 63 patients.

DES

The Dissociative Experiences Schedule [16, 32, 33], is a 28-item selfreport questionnaire that measures the frequency of dissociative and associated experiences (e.g. depersonalization/derealization, absorption, amnesia and identity fragmentation). The test yields item and total scores that range from 0 to 100. Scores over 30 often indicate pathological levels of dissociation. The DES has been found to have good test–retest and spilt-half reliability, as well as good construct and criterion validity. In the current study, the internal consistency of the DES total score was excellent (Cronbachs α = 0.92). A subset of eight pathological dissociation items is often used to assess the presence of pathological dissociation (DEST) [34]. Dissocciative experiences schedule data were missing in eight patients, and therefore the association between trauma and dissociation will be reported for only 56 patients.

STI

The Structured Trauma Interview [35], is a semistructured interview with questions about traumatic experiences in childhood (i.e. before age 16) and in adulthood (at or after age 16) known to be risk factors in the development of psychopathology: sexual abuse, physical abuse, early loss or separation from one or both parents, witnessing domestic violence, neglect or other shocking experiences.

Statistical analysis

Differences between groups with regard to DEST scores were examined using general factorial ANOVA, and F-values for the main effects of trauma on dissociation. Differences between groups with the regard to the prevalence of PTSD were examined using χ 2 statistics. All analyses were conducted using SPSS 8.0 for Windows [36]. Due to the exploratory nature of the study the significance level was set at p < 0.10, and no Bonferroni corrections were applied.

Results

Patient characteristics

In this sample of 64 female BPD patients with a mean age of 34.9 years (SD = 7.7; range = 20–49 years), all but two patients had a Dutch nationality. Most patients had never been married, and approximately one-third was living alone (Table 1). The average number of years of education was 13.1 (SD = 3.6). The majority of patients was either unemployed or received a disability pension (Table 1). The average number of DSM-IV criteria for BPD was 7.3 (SD = 1.3). The majority of patients reported a history of at least one suicide attempt (71%) and/or a history of at least one self-mutilating act (93%).

Table 1.

Sample characteristics

Table 1 also shows that 57% of the patients reported childhood physical abuse (CPA), 71% childhood sexual abuse (CSA), and 43% a combination of CPA and CSA. In addition, 77% reported adult sexual victimization, and 72% adult physical victimization.

The mean DES score was 19.2 (SD = 14.2; range = 1–55), and the mean score on the DEST was 15.7 (SD = 14.7; range = 0–56). The lifetime prevalence of PTSD was 43% in the total sample, and 54% of the subjects reported clinically significant substance abuse problems.

Association between childhood trauma and neglect and dissociation and PTSD

Table 2 shows that in the total group (n = 56), elevated DES levels are found in female BPD patients reporting a combination of CPA and CSA (F = 3.9, df = 1,56; p = 0.05), sexual abuse by more than one perpetrator (F = 4.7, df = 1,56; p = 0.03), and witnessing violence (F = 3.0; df = 1,56; p = 0.09). Furthermore, severe maternal dysfunction is significantly associated with dissociation (F = 5.6, df = 1,56; p = 0.02). Analyses with the DEST revealed that only maternal dysfunction was significantly associated with the DEST total score (F = 6.6, df = 1,56; p = 0.01).

Table 2.

Relationship between childhood traumatization (Structured Trauma Interview, STI), and dissociation (Dissociative Experiences Schedule, DES), in non-addicted and addicted borderline patients

Table 3 shows that in the total group (n = 63), the presence of PTSD is associated with reports of CSA (χ 2 = 4.4, df = 1; p = 0.04), as well as with the severity of CSA, indicated by penetration (χ 2 = 4.8, df = 1; p = 0.03), intrafamilial perpetrators (χ 2 = 4.7, df = 1; p = 0.03), duration more than 1 year (χ 2 = 9.8, df = 1; p = 0.002), and more than one perpetrator (χ 2 = 5.2, df = 1; p = 0.02). Childhood neglect was not associated with the presence of PTSD.

Table 3.

Relationship between childhood traumatization (Structured Trauma Interview, STI), and posttraumatic stress disorder (SCID-PTSD) in nonaddicted and addicted borderline patients

Effect of comorbid substance abuse problems on the associations between childhood trauma, neglect, dissociation, and PTSD

In a stratified analysis, differentiating the total group of BPD patients into two subgroups according to the absence or presence of substance abuse problems, the observed associations between trauma, neglect and the level of dissociation, and PTSD only remain significant in the nonaddicted subgroup (Tables 2 and 3). The only exception is a negative association between CPA and the presence of PTSD, and between CPA + CSA and the presence of PTSD.

Discussion

In summary, the present findings indicate (i) high prevalences of childhood trauma and neglect as well as high levels of dissociation and frequent occurrence of adult PTSD in this sample of female BPD patients; (ii) a moderately strong association between several childhood traumatic experiences and childhood neglect with the level of dissociative experiences; (iii) a fairly strong association between the severity of childhood sexual abuse with the prevalence of PTSD; (iv) a modifying role for addictive problems such that the associations were only present (and more pronounced) among the non-substance abusing subsample, whereas no associations were found in the substance-abusing subsample; and (v) that levels of dissociation and PTSD among substance-abusing borderline patients were, irrespective of the presence of childhood trauma, highly similar to the levels reported by traumatized non-substance abusing patients.

These findings are largely consistent with the literature. As in previous studies [37–39], in the total sample the level of dissociation was found to be associated with a history of both childhood physical abuse and sexual abuse, and childhood sexual abuse by more than one perpetrator. Like in other studies, severe maternal dysfunction was found to be associated with the level of dissociation [40]. An association between trauma and PTSD was also supported.

The most remarkable finding of this study was the modifying role of addictive problems with respect to the associations found. To the best of our knowledge, this study is the first investigating trauma-dissociation and trauma-PTSD links in a population including both substance abusers and non-substance abusers. Our results are consistent with the general finding in literature that psychiatric samples do show associations between childhood trauma and neglect with dissociation, while addiction samples do not [19]. A number of explanations might account for this phenomenon.

First, the modifying role of addictive problems might be accounted for by differential reliability and validity of the instruments applied. More specifically, is it not unlikely that the self-reported DES does not only measure dissociative experiences, but also similar experiences induced by intoxication with psychoactive substances. From the literature, it is well known that self-report questionnaires in general are sensitive to contamination by state-like effects [41]. A similar effect of substance abuse on the PTSD diagnoses that were derived from scores on the SCID-I [30] is less likely because semistructured interviews tend to be more resistant against contamination by state effects [41]. The same would be true for the STI. Thus, the differential findings with respect to dissociation might partly be accounted for by over-reporting in the substance-abusing participants, but this explanation is less likely in relation to PTSD.

Second, the differential findings could be accounted for by the presence of symptomatology resulting from substance-related traumatization. In such a model, substance use disorder is already extant when the person is exposed to the traumatic event. This explanation is especially likely for the unexpectedly high prevalence of PTSD among substance-abusing subjects without a history of CPA [42]. Such scenarios are frequently encountered – for example, a heroin addict who, while attempting to procure drugs and ignoring or not registering subtle cues of imminent danger, is assaulted [43]. Another relatively common situation involves an alcoholic who, after drinking at a bar with friends, leaves with an acquaintance and is raped while intoxicated. On a psychological level, in altering the capacity for judgement and directed activity, substance intoxication may result in powerlessness or decreased self-efficacy, thereby increasing the risk for victimhood and actionless witnessing of traumatic events [43]. In addition, PTSD symptoms resulting from substance-related traumatization might be exacerbated by substance use [44].

Third, some authors have suggested that a traumadissociation link may not exist in substance abusers, particularly in male alcoholics, because these individuals may abuse substances to achieve dissociative-like states as a substitute for ‘true’ dissociation. This explanation is referred to as the ‘chemical dissociation’ hypothesis [19]. Our findings do not support this hypothesis, since levels of dissociation and PTSD among traumatized, substance-abusing subjects were as high as those reported by their traumatized, non-substance abusing counterparts.

In summary, the absence of a trauma-dissociation relation in subjects with addictive problems seems to be best explained by over-reporting of dissociative experiences through confusing of intoxication-related phenomena with dissociation, whereas the absence of a trauma-PTSD relation in subjects with addictive problems seems to be best explained by PTSD resulting from both substancerelated traumatization and childhood trauma.

Some study limitations should be noted. This is a cross-sectional study based on retrospective self-report, implying temporal ambiguity of cause and effect. The correlational nature of findings precludes drawing any firm conclusions. In addition, due to the limited sample size, the statistical analyses should be considered exploratory. Replication of this study in a larger sample is recommended.

Conclusion

In this explorative study of a group of borderline patients with and without a comorbid substance use disorder we found a moderately strong association between trauma and dissociation. This association was observed among BPD patients without addictive problems only. The absence of a trauma-dissociation relation in BPD subjects with addictive problems seems to be best explained by over-reporting of dissociative experiences, whereas the absence of a trauma-PTSD relation in subjects with addictive problems seems to be best explained by PTSD resulting from substance-related traumatization.