sign in sign up
<<
Home , Personality disorder

MPD, BORDERLINE DISORDER AND : ACOMPARATIVE STUDY OF CLINICAL FEATURES

David Fink, M.D. Michael Golinkoff, Ph.D.

David Fink, M.D., is Associate Director of the Dissociative personality disorder, borderline personality disorder, and Disorders Program atThe Institute ofPennsylvania Hospital schizophrenia. In order to identify and define similarities in Philadelphia, Pennsylvania. and differences, the traits of these groups were measured using an extensive battery of diagnostic interviews and psy­ Michael Golinkoff, Ph.D., is Associate Director ofthe Evalu­ chometric instruments. The hypothesis to be tested is that ation Unit at The Institute of Pennsylvania Hospital and MPD represents a unique clinical entity, with stable and Clinical Associate in , University of Pennsylvania measurable clinical characteristics, which differentiate it Medical School, both in Philadelphia, Pennsylvania. from both schizophrenia and borderline personality disorder. For reprints write David Fink, M.D., Associate Director, Dissociative Disorders Program, The Institute ofPennsylva­ MULTIPLE PERSONALITY DISORDER nia Hospital, Philadelphia, Pennsylvania 19139. AND SCHIZOPHRENIA

ABSTRACT The relatively low prevalence of MPD in relation to schizophrenia, as well as the power of the Bleulerian and Multiple personality disorder (MPD) has at times been confused with Schneiderian conceptualizations ofschizophrenia, have re­ both schizophrenia (SCHIZ) andborderlinepersonality disorder (BPD). sulted in the frequent misdiagnosis ofdissociative In this study, 38patients withDSM-III-R diagnosis ofMPD (N = 16), as schizophrenia. In a review of 100 cases of MPD, Putnam, SCHIZ (N = 11), or BPD (N = 11) were evaluated with a battery of Guroff, Silberman, Barban, and Post (1986) found that 49% structured interviews (SCID, DDIS) and psychometric tests (MMPI, had received a priordiagnosis ofschizophrenia. Rosenbaum MCMI, DES) in order to define distinguishing features among the (1980), in a review ofreported cases of MPD between 1903 three diagnostic groups. MPD was differentiated from SCHIZ on the and 1978, noted a dramatic decline in the reports of MPD great majority oftest measures. MPD was not differentiatedfrom BPD afterthe newdiagnosis of"schizophrenia"based on Bleuler's on MMPI or MCMI, but these groups differed in many clinical ideasbecame a mainstayofpsychiatric diagnosis. Rosenbaum features, particularly measures ofseverity ofabuse and dissociative contends that after the introduction of this new diagnostic symptoms. entity, the incidence ofMPD declined as these patients were now more often diagnosed as having schizophrenia. Within INTRODUCTION the Bleulerian , however, MPD would be miscon­ strued as schizophrenia primarily on the basis of auditory Multiple personalitydisorder (MPD) historicallyhasbeen and associated findings, rather than on the subsumed within the boundaries ofother m~or psychiatric core features of associative loosening, affective blunting, disturbances (Bliss, 1980; Greaves, 1980; Rosenbaum, 1980; , and ambivalence. Schneider's eleven First Rank Coons, 1984; Putman, Guroff, Silberman, Barban, & Post, Symptoms (FRS) were an effortto establish greaterdiagnostic 1986; KIuft, 1987; Horevitz & Braun, 1984; Clary, Burstin, & reliability for schizophrenia, and were to be considered Carpenter, 1984). In the nineteenth century it was both pathognomonic for the disorder in the absence oforganicity understood and disputed in terms ofits relation to hypnosis; (Schneider, 1959). Recently, however, the specificity of the subsequentlyitwas understood as a manifestation ofhysteria first rank symptoms has been refuted. Their occurrence has (Bliss, 1980; Greaves, 1980). More recently, MPD has been been demonstrated in a number of psychiatric conditions. included within the diagnosis ofschizophrenia (Bliss, 1980; Andreason and Akiskal (1983) have noted the high occur­ Rosenbaum, 1980) and most recently the borderline condi­ rence ofFRSs in patients with affective disorders. Carpenter, tions (Horevitz & Braun, 1984; Clary, et al., 1984). Until the Strauss, and Muleh (1973) have also noted the lack of recentdevelopment ofDSM-III (and DSM-III-R), with its strict diagnostic specificityin theSchneiderianFRSsand have found set of criteria for MPD, borderline personality disorder and that they are reported in 20 to 50% of cases of manic schizophrenia, it has been difficult to determine to what depressive illness. extent our efforts to establish diagnostic classifications for KIuft (1987b) has described the prevalence ofFirstRank these three diagnostic groupshave represented careful study Symptoms in patients with MPD. In his review ofinitial con­ of homogeneous patient groups. The present research tacts with thirty MPD patients, he found that they endorsed compares groups of patients with the diagnosis of multiple an average of 3.8 Schneiderian symptoms. Each patient

127 DISSOCIATIO:-l, Vol. III, :-10. 3: September 1990 endorsedfrom one to eightsymptomswith a total ofl08FRSs Etiologically, both MPD and BPD are considered to be endorsed in all. The eightSchneiderian symptomsendorsed primarily psychodynamically based forms of psychopathol­ were: (1) voices arguing, (2) voices commenting on one ogywhich reflectearlydevelopmental life experiences. Several action, (3) influences playing on the body, (4) thought investigators have consideredthe relationship ofMPD psychic withdrawal, (5) thought insertion, (6) made feelings, (7) organization to borderline personalityorganization. Claryet made impulses, and (8) made volitional acts. aI. (1984), in reviewing treatment of eleven patients with Other investigators, while specifically reviewing neither MPD, contrasted the defensive styles and personality orga­ Bleuler's criteria nor Schneiderian first rank symptoms in nizations ofthese cases to the psychoanalytic understanding MPD patients, have noted considerable symptom overlap of to borderline personality organization. They conclude between schizophrenia and MPD. Bliss (1980), in a report of that MPD represented a "special instance ofborderline per­ fourteen cases of MPD, found that 64percent heard voices, sonality disorder" (p. 98). Kernberg (in press) states that 36% saw visions, 73% felt that someone was trying to influ­ MPD is not to be associated with anyone level ofpersonality ence their minds, 55% experienced someone controlling organization but can be found in neurotic, borderline, and their mind, and 27% had thought broadcasting. Putnam et psychotic character structures. al. (1986), in a review of the clinical phenomenology of 100 Phenomenologically MPD and BPD show considerable cases, found auditoryhallucinationsin 30%, visual hallucina­ overlap. However, there remains considerable debate as to tions in 28%, apparent in 21 % and apparent whether MPD represents a variant ofborderline personality thought disorder in 19%. Given the high rate of endorse­ organization or whether it represents a distinct defensive ment by MPD patients of Schneiderian FRSs and other organization. symptoms frequently associated with schizophrenic illness, To better define and clarifY similarities and/or differ­ the potential for confusion between schizophreniaand MPD ences among the three DSM-III-R diagnostic groups of is readily apparent. schizophrenia, BPD and MPD, subjects with these respective diagnoses were tested with a broad battery of psychological MULTIPLE PERSONALITY DISORDER tests and a series ofstructured interviews. AND BORDERLINE PERSONALITY DISORDER METHOD Like schizophrenia, borderline personality disorder (BPD) has been a diagnostic group with an evolving degree Subjects: the three groups were selected from inpatient ofclinical specificity (Perry & KIerman, 1978; Gunderson & and outpatient psychiatric facilities in the Philadelphia area. Kolb, 1978; Spitzer, Endicott, & Gibbon, 1979; Liebowitz, Subjects were originally diagnosed by their psychotherapists 1979; Lerner, Sugarman, & Gaughran, 1981; Kroll, Martin, and referred to the studybased on these diagnoses. Subjects' Lari, Pyle, & Zander, 1981; Andrulonis, Glueck & Strobel, diagnoses were confirmed using the Structured Clinical 1982; Clarkin & Widiger, 1983; Frances & Clarkin, 1984; Interview for DSM-III-R - Patient Version (SCID) (Spitzer, Akiskal & Chen, 1985). Despite an abundant literature on Williams, & Gibbon, 1987).Thescm, however, does nothave "borderline" states, it was only with DSM-III that clear diag­ a module on dissociative disorders; therefore, the diagnosis nostic distinctions were made between the borderline and of MPD was made according to DSM-III-R criteria and con­ schizophrenic conditions. While the subsequent distinction firmed by data collected by another psychiatric interview, the between schizophrenia and borderline personality disorder Dissociative Disorder Interview Schedule (DDlS) (Ross, He­ has been effected relatively easily, the potential over-inclu­ ber, Norton, & Anderson, 1988). The diagnosis of BPD was siveness of the borderline diagnosis with relation to other confirmedbasedon datacollected throughoutthe interviews DSM-III conditions remains an area ofconsiderable concern and based on the patients, endorsement offive ofeightDSM­ (Perry & Lerkman, 1978; Lerner, Sugarman, & Gaughran, III-R (American Psychiatric Association, 1987) criteria for 1981). BPD. The DDIS does ask specifically about each of the eight Efforts to differentiate MPD and BPD have been compli­ symptomsfrom which the diagnosis ofBPD is made. Subjects cated by the ongoing consideration ofthe two at were then sorted into three groups based on whether they different levels of theoretical conceptualization. Phenome­ metcriteriafor MPD, forschizophrenia (SCHIZ) , andforBPD. nologically, MPD and BPD share many diagnostic features The MPD groupconsistedofsixteensubjects and the BPD and (Putnam et aI., 1986; KIuft, 1987a, 1987b; Horevitz & Braun, SCHIZ groups consisted of eleven subjects each. 1984; Clary et aI., 1984). In a study of twenty MPD patients, Procedure: The purpose ofthe study ('We are interested Ross, Heber, Norton and Anderson (in press) found that in better understanding factors, particularly relating to past twelve (60%) metDSM-III criteriafor BPD. Horevitz and Braun events and currentsymptoms, importantin assessing various (1984), in a review of the phenomenology of thirty-three psychiatric diagnoses.") was explained to each subject, and cases ofMPD, found that70% metDSM-III criteriafor BPD while informed consent to participate was obtained. The scm and the other 30% did not. The use of the Global Assessment DDlSwere then administered byone oftwo trained clinicians. Scale (GAS) further differentiated the groups. A significant When these were completed the subjects were given tlu-ee difference existed between those individuals with the high­ additional questionnaires to fill out: (1) the Minnesota est GAS measurement and those with the lowest GAS scores. Multiphasic Personality Inventory (MMPI); (2) the Millon In addition the GAS score differentiatedbetween MPD patients Clinical Multiaxial Inventory (MCMI); and (3) the Dissocia­ who met criteria for BPD and those who did not. tive Experiences Survey (DES).

128 • j ~~ rt. :~." I,,; ",,- .~ • ~ ..~ DISSOCLUION, Vol.lII, :-10. 3: September 1990 ~ FINK/GOLINKOFF

Materials: The fopowing is a briefdescription ofeach of dominantly met criteria for schizophrenia and rarely had the instruments utilized in the study. other diagnoses. The scm (Spitzer et aI., 1987) is a structured interview Differences among the groups with respect to the life­ that guides the clinician through a systematic and thorough time prevalence ofa according to assessment of symptoms needed to meet criteria for the the scm were also evaluated,¥ith a significantmain effectfor following currentand pastAxis I disorders: Mood Disorders, group (F(2,35) = 10.40, P < .01). Post hoc Tukey tests Psychotic Disorders, Substance Use Disorders, Dis­ revealed that the MPD and BPD groups did not differ from orders, Somatoform Disorders, Eating Disorders, and Ad­ each other with respect to their meeting lifetime criteria for justment Disorders. a Major , but both had a significantly higher TheDDlS (RossetaI., 1988) is a psychiatric interview that percentage than the SCHIZ group (MPD: F(I,35) = 14.48, P questions subjects about specific experiences that meet cri­ < .01; BPD: F(I,35) = 18.69, P < .01). teriafor dissociative disorders as well as symptomsfrequently Table 2 presents a summary ofthe information generat­ associated with dissociative disorders. The MPD criteria em­ ed by the two measures specifically designed to distinguish ployed by the DDIS are the NIMH Research Criteria which between MPD patients and other psychiatric patients. include the DSM-III criteria plus two additional criteria. The The DDIS clearly confirmed the presence ofMPD in the two are: (1) that two or more have been in MPD group and ruled it out in the other two subject groups. control ofthe subject'sbodyon three ormore occasions; and However, the interView also highlighted other differences (2) a form of exists among the different personal­ among the groups. A series ofANOVAs was completed look­ ities. This instrument also assesses borderline personality ing at types ofsymptoms among the three subject groups. disorder criteria as defined by DSM-III-R. In addition, there There was an overall group effect for the presence of are sections assessing somatic, depressive, Schneiderian first somatic symptoms (F(2,35) = 12.05, P < .01). Post hoc rank psychotic and substance symptoms. There are analyses revealed that both the BPD (F(1,35) = 9.04, P < .01) also questions about previous psychiatric treatment, as well and SCHIZ (F(I,35) = 22.75, P < .01) groups reported signif­ as any history of physical or . icantly fewer symptoms than did the MPD group. The DES (Bernstein & Putnam, 1986) is a questionnaire There was also a significant overall effect for the pres­ that asks respondents to mark the percentage of time that ence of Schneiderian first rank symptoms (FRS) (F(2,35) = each oftwenty-eightdissociative-like experiences happens to 10.29, P < .01). The BPD group reported fewer FRS than ei­ them. therthe MPD (F(1 ,35) = 13.88, p< .01) and the SCHIZgroups The MMPI (Hathaway & McKinley, 1983) and MCMI (F(I,35) = 17.63, p< .01). The MPD and SCHIZ groups did not (Millon, 1982) are self-report personality inventories. They differ significantlyfrom each other (F(I,35) = .72, p < .41) on present the subjects with 566 and 175 true/false questions, the mean number of FRS reported. respectively. Theyare then scoredforseveralscales. Reliability The groups did not differ at all with respect to history of measures for the MMPI scales in psychiatric populations (F(2,35) = 1.2, P < .31). (Hathaway & McKinley, 1983) vary between .36 and .93. Thus, a combination ofthese two semi-structured diag­ Millon (1982) reports reliability coefficients of between .61 nostic interviews helped to differentiate these three groups. and .91 for the MCMI scales. The scm was able to isolate schizophrenic subjects from Analyses ofvarious (ANOVAs) were conducted on scores generated by each of these measures for the main effect of subject group. When there is a signif­ icant main effect for subject group on any of the TABLE 1 scores, pain¥ise posthoc comparisonswere conducted. Frequency ofscm Derived DSM-III-R Diagnosis The Tukey honestly significant difference (HSD) for the Three Subject Groups method was selected because, according to Hays (1981) "... it is simple, widely used, and flexible in SCHIZ application" (p. 434). MPD BPD Diagnosis (n =16) (n =11) (n =11) RESULTS 12 10 1 The data were analyzed with respect to whether Schizophrenia 2 0 11 the diagnostic interviews and/or the personality in­ ventories were able to distinguish the MPD subjects Substance Abuse 0 3 1 from the other two groups of psychiatric subjects. 12 3 1 Table 1 summarizes the information generated by Somatoform Disorder 0 0 0 the scm. The MPD and BPD groups showed similar 2 1 1 diagnostic data according to the scm, with a pre­ dominance of patients being diagnosed with mood disorders. The m;;yor difference between the MPD Number of Diagnostic and BPD groups was the greater incidence ofanxiety Criteria met Lifetime 3.5 3.3 1.5 disorders in the MPD group. The SCHIZ groups pre-

129 DISSOCLUIOX, Vol. III. ~o. 3: September 1990 , MPD, BQ,RDE~I~E,PERSON~ITY OJSORDER &, SCHIZOPHRENIA , :- >' ~ -' .' • .. ., .'

comparative means) forthe three TABLE 2 groupswas significant (F(2,35) = 19.20, p< .01). Post hoc Tukey Summary of DDIS and DES Information tests showed the MPD group to have significantly higher DES MPD BPD SCHIZ RESULT scores than the BPD (F(I,35) = 16.96, P < .01) and the SCHIZ Percent MPD Criteria 100 0 0 a,b group (F(I,35) = 34.90, P < .01. Percent Major Dep. Criteria 75 91 18 b,c Therefore, combining the infor­ Percent Substance Abuse 47 55 27 mation obtained from the DES Percent Physically Abused 75 64 18 b,c and the DDIS allows for further differentiation ofthe MPD from Percent Sexually Abused 94 64 9 b,c the BPD group. # Somatic Symptoms 8.8 3.6 6 a,b Tables 5 and 6 present the # Schneiderian Symptoms 4.8 1.2 5.6 a,c means ofthe MMPI and MCMI-II DES Score 48.6 23.5 12.6 a,b scales for each of the three groups. A multiple regression a: BPD significantly different from MPD. analysis for each of these two b: SCHIZ significantly different from MPD. personality inventories was per­ c: SCHIZ significantly different from BPD. formed with the subscales as de­ pendent measures and subject group as the independent vari- able. In addition, post hoc (Tukey tests) comparisons between each pair ofgroups was TABLE 3 performed as well. The significant differences for each scale Frequency of Sexual Abuse for MPD and BPD Subjects between the pairsofgroupsare also summarizedin those two tables. The following discussion of results will focus on the Frequency MPD BPD most interesting comparisons. The most striking finding for both the MMPI and MCMI was 0 1 4 how clearly differentiated the SCHIZ group was from each of the other two groups on many of the scales, but how few of 1 - 10 1 5 the scales differentiated the MPD and BPD groups from each > 10 14 2 other. The comparison between the MPD and BPD groups on the personality measures is of most interest. The only scale that was significantly different between the MPD and BPD TABLE4 groups on theMMPlwastheHSscale (F(I,35) =4.77,p<.05), Age of Onset of Sexual Abuse for MPD and BPD Subjects with the MPD group yielding higher scores. The MCMI was even more striking in its lack ofdifferen­ Age (years) MPD BPD tiation between the MPD and BPD gr"oups. Of the 25 scales measured only the B (alcoholism) scale showed a significant difference (F(I,35) = 4.65, P < .05). However, as can be seen 1 - 5 13 1 in Table 6, the MCMI clearly differentiated the SCHIZ group 6-10 2 3 from each of the other two groups in that their scores on > 10 0 3 twelve subscales for the BPD group and seven subscales for the MPD group were significantly different. MPD andBPD subjects based on the presence ofsymptoms of schizophreniaand the absence ofothersymptoms. The DDIS DISCUSSION was then able to distinguish the MPD and BPD groups from each other based on the greater incidence of dissociative In the current study we have looked at a number of disorder symptoms and Schneiderian FRS in the MPD group. clinical validators in order to define areas ofdifference and The DDIS assesses history of sexual abuse which is con­ of similarity among schizophrenia, borderline personality sidered to be an important etiological factor in the develop­ disorder, and multiple personality disorder. While MPD has ment ofdissociative phenomena. A high incidence ofsexual frequently been subsumed within the other two diagnostic abuse was reported by both the MPD and BPD group. How­ categories, our results define a number of clinical features ever, as Tables 3 and 4 illustrate, the severity of abuse as which differentiate MPD from both schizophrenia and BPD . measured by the overall frequency of abuse and the age of In keeping with the major etiologic model of MPD abuse onset is much greater in the MPD group. (KIuft, 1987a), we found that early and severe childhood The differences in mean DES scores (see Table 2 for trauma was a hallmark of our MPD population. Seventy-five

130 DISSOCIATION, Vol. III, No.3: September 1990 FINK/GOLINKOFF

percent of the MPD patients reported , and 94% reported sex­ TABLE 5 ual abuse. In contrast, in the BPD group, Mean MMPI Scale Scores for the Three Subject Groups 64% reported physical abuse and 64% reported sexual abuse. Of the schizo­ phrenics, 18% reponed physical abuse Scale MPD BPD SCHIZ RESULT and 9% reponed sexual abuse. While both the MPD and BPD groups reported L 47.4 46.5 46.8 high rates ofchild abuse, the severity of F 89.8 84.5 70.2 b,c abusewas markedlygreaterfor the MPD K 49.5 44.8 51.4 c patients. In comparing the characteris­ HS 75.9 67.0 59.6 a,b tics ofthe sexual abuse reported by the BPD and MPD patients, the MPD group D 86.3 92.8 75.5 b,c reported a greaternumberofassailants, BY 74.2 72.7 66. c an earlier age of onset, a later age of PD 85.0 87.4 78.1 cessation ofabuse, and a greater num­ MF 49.0 50.1 54.1 ber ofkinds of childhood sexual expe­ PA 82.1 81.3 67.9 b,c rience. PT 81.6 85.2 72.1 b,c Our results also supportan associa­ tion between extreme childhood abuse SC 104.1 94.5 83.7 b and dissociative experiences. We found MA 66.6 58.9 55.2 b a high correlation between the DES SI 71.4 77.0 63.2 b,c scoresand all severityofabusemeasures. The MPD group had a mean DES score a: BPD significantly different from MPD. twice as high as the mean score for the b: SCHIZ significantly different from MPD. BPD group. This finding is consistent c: SCHIZ significantly different from BPD . with the work of Herman, Perry, and van der Kolk (1989), who found a cor­ relation between DES scores and severity of abuse in a con­ schizophrenic patients. trolled studyofagroup ofborderline subjects. They propose Schneiderian first rank symptoms were found to be a model for conceptualizing survivor syndromes extremely common in MPD patients. While only two of the with a range of adaptation to various degrees of childhood MPD patients met DSM-III-R criteria for schizophrenia, 94% trauma. In this scheme, dissociative disorders represent an reported at least one Schneiderian first rank symptom, with extreme form of adaptation, borderline personality an im­ a mean of4.8 FRSs per patient. This compared with a mean mediate form, and certain somatizationand anxietydisorders of 5.6 FRSs in the schizophrenic group. Ross et al. (1989) a less extreme form ofadaptation to abuse. In this light, it is found a higher rate ofFRSs in their MPD group than in their noteworthy thatourthreegroups also differedwith regard to schizophrenics, with a mean of6.6 per patient. A differential report of somatic and anxiety symptoms. While few of the item analysis ofthe Schneiderian signs in this study reveals a patients in our study met DSM-III criteria for somatization high rate of auditory hallucinations and passive influence disorder, the mean number ofsomatic complaints were 9, 4, experiences in the MPD group, with relatively little interfer­ and .6, respectively, for the MPD, BPD, and schizophrenic ence with thought process, i.e., thought withdrawal or groups. In addition, 75% ofthe MPD patients met DSM-III-R broadcasting. In lightofthe pronounceddifferencesbetween criteria for a current anxiety disorder in contrast to 27% of the MPD and schizophrenia groups on all other measures, the borderline patients and 9% of the schizophrenics. including the MMPI, MCMI, DES, and abuse history, the Our findings support a clear differentiation between specificity of Schneiderian signs for schizophrenia must be schizophrenic and multiple personality patients in terms of abandoned. Schneiderian signs are non-specific; in fact, they psychological profiles on the MMPI and MCMI inventories. should raise suspicion ofMPD or another dissociative disor­ MPD was distinguishable from schizophrenia on both the der rather than a thought disorder. MMPI and the MCMI. The predominant mean MMPI profile In contrast to tlle clear differentiation of MPD from for the schizophrenic group was 8/4/2, whereas it was 8/2/ schizophrenia across psychological test measures, the MPD 4 for the MPD group. The MPD patients showed significantly and BPD groups showed many areas of similarity. However, greater overall elevations than did the schizophrenics on all MPD patients were clearly differentiated from BPD patients scales except 5, the Masculinity/Feminity scale. It is note­ by differences in antecedent history, specifically the severity worthy that the MPD patients had markedly higher scores and extent ofchildhood physical and sexual abuse, and the than the schizophrenic patients on the F and Sc scales, which degree of dissociative symptoms. are considered to be the mostspecificfor schizophrenia. The Indeed, many areas of overlap were found in the phe­ MCMI also differentiated the MPD group from the schizo­ nomenology of MPD and BPD patients. Both patient groups phrenic group, with the MPD patients showing a much more were polysymptomatic and met DSM-III-R criteria for an av­ acute and polysymptomatic picture than that shown by the erage of at least two other concurrent major psychiatric

131 DISSOCl-\.TIO:\', Yo!. III. :\'0. 3: September 1990 1982; Solomon, 1983; TABLE 6 Coons &Sterne, ] 986; Bliss, Mean MMPI Scale Scores for the Three Subject Groups 1984; Kemp, Gilbertson, & Torem, 1988). Two avail­ able studieshave compared Scale MPD BPD SCHIZ RESULT MPD and BPD patients us­ ing the MMPI. Kemp et aI. Disclosure 77.9 81.9 64.3 c (1989) found no significant DesirabiIity 42.5 36.5 55.5 c differences among mean Debasement 85.6 90.8 57.7 b,c scales to differentiate the groups. Coons and Fine Schizoid 81.0 79.9 66.4 (1990) report a 68% accu­ Avoidant 102.9 103.5 81.3 b,c racy rate for the differenti­ Dependent 71.6 74.9 66.4 ation ofMPD MMPI profiles Histrionic 46.3 49.1 54.1 from profiles from a large Narcissistic 40.1 45.0 58.7 sample of patients with a range of diagnoses includ­ Antisocial 65.1 72.8 64.4 ing BPD and schizophre­ Aggressive/ Sadistic 57.5 68.0 58.4 nia, butfail to reportwhich Compulsive 50.6 51.2 60.3 specific factors supported Passive-aggressive 87.4 99.5 68.4 c their distinguishing the Self-defeating 97.3 104.5 70.5 b,c groups. Schizotypal 81.3 79.5 70.2 The discrepancy be­ tween the historical!de­ Borderline 91.9 98.1 66.4 b,c scriptive and psychometric Paranoid 59.8 62.7 63.5 findings for the MPD and Anxiety 71.4 85.0 50.6 b,c BPD groups raises a num­ Somatoform 56.8 60.8 52.8 bel'ofimportantquestions. Bipolar: Manic 50.7 44.6 39.0 The first is whether MPD represents a subset or vari­ Dysthymic 81.7 91.4 54.7 b,c ant of BPD (Bliss, 1980; Alcohol Dependence 59.4 74.8 46.4 a Horevitz & Braun, 1984; Drug Dependence 63.0 73.4 49.1 c Clary, et aI., 1984). "Vhile Thought Disorder 74.0 74.5 61.6 c there were "borderline" Major Depression 76.4 86.6 47.8 b,c tendencies in our MPD 59.4 56.5 55.1 group, few MPD patients wouldhavebeendiagnosed with the disorder. Many nosologists have debated disorders. Importantly, the groups differed on the extent to the current over-inclusiveness of the borderline personality which they endorsed DSM-III-R criteria for borderline per­ disorder diagnosis (Perry & KJerman, 1978; Gunderson & sonality disorder. All ofthe patients in the BPD group met at Kolb, ] 978; Spitzer, Endicott, & Gibbon, 1979; Liebowitz, least five of eight DSM-III-R BPD diagnostic criteria for in­ 1979; Lerner, Sugarman, & Gaughran, 1981; Kroll, Martin, clusion in the study, with an average of 6.6 per patient. In Lori, Pyle, & Zundel', 1981; Andrulonis, Glueck & Stroebel, contrast, only two ofthe MPD patients met DSM-III-R criteria 1982; Frances & Clarkin, 1983; Akiskal & Chen, 1985). for borderline personality disorder with a mean of 3.7 BPD With regard to the MMPI and MCMI profiles ofthe MPD criteria for the MPD group as a whole. group, the specificity of the personality inventories them­ BothMMPI andMCMI profilesfor theMPD andBPD groups selves mustbeaddressed. The MMPlwas designedbefore MPD were remarkably similar. In both cases, the profiles were was characterized and well defined as a clinical entity. Subse­ consistentwith the mostcommonlyreported mean borderline quently, no systematic consideration of dissociative symp­ personality disorder profiles for these instruments (Patrick, toms within one subscale exists. Instead, a host ofsymptoms 1984; Evans, Ruff, Braff, & Cos, 1986). A number of investi­ common to MPD are found dispersed across several scales. gators who have studied MPD patients with the MMPI have No consolidation of these indicators has yet been catego­ obtained the same results. In general, they were character­ rized which might increase the specificity ofthe instrument. ized by extremely elevated F and Sc scales, an 8/2/4 profile, Additionally, the heterogeneityofmean MMPI and MCMI technically invalid inventories due to too many extreme profiles has been described (Antoni, Tischer, Levine, Green, subscale elevations, and a polysymptomatic picture & Millon, 1985a; 1985b). Millon has noted that a look at (Brandsma &Ludwig, 1974; Wilbur, Brandfeldt, &Jameson, combined MMPI and MCMI high point profiles of the 2/8 1972; Larmore, Ludwig, & Cain, 1977; Solomon & Solomon, MMPI profile resulted in three stable and distinctclusters. He

132 DISSOCIATIO~,Vol. III, ~o. 3: September 1990 FINK/GOLINKOFF

suggests that mean MMPI scale scores alone lack specificity Bernstein,E.M., &Putnam,F.W. (1986). Thedevelopment, reliability, and at times represent several distinct clinical pictures. In andvalidityofa dissociation scale.jownalofNel1JousandMentalDisease, this study such a combined high point analysis was not 174,727-735. possible due to the limited sample size, but it should be Bliss, E.L. (1980). Multiple personality: A report of fourteen cases considered for future investigation. This possible lack of with implications for schizophrenia and hysteria. Archives ofGeneral specificity ofmean MMPI profile findings for distinguishing PsychiatlY, 37, 1388-1397. MPD from BPD is further supported by the similar MMPI and MCMI profiles of Vietnam Veterans with PTSD (Fairbank, Bliss, E.L. (1984). A symptom profile of patients with multiple Keane, & Malloy, 1983; Keane, Malloy, & Fairbank, 1984). personality disorder, including MMPI results. Joumal ofNel1Jous and Itmaybethatthe psychometricprofile onMMPI andMCMI MentalDisease, 172, 197-202. identified in PTSD, BPD, and MPD represents acommon final Brandsma,j.M., &Ludwig,AM. (1974) .Acaseofmultiplepersonality: pathway for three groups of patients which share overarch­ Diagnosis and therapy. InternationalJoumal ofClinical andExperimental ing features: an extremely high degree of internal disorga­ Hypnosis, 3, 216--233. nization, a high level of affective instability, and extreme distress. All three disorders are highlighted by marked af­ Carpenter, W.T., Strauss, j.S., & Muleh, S. (1973). Are there fective states and a failure to fully integrate certain experi­ pathognomonicsymptomsin schizophrenia:Anempilicinvestigation of Schneider's first rank symptoms. Archives ofGeneral Psychiatry, 28, ences. They are all partially defined by either splitting ofthe 847-852. ego, alternate personalityformation, and/or flashback phe­ nomena. Co-morbidityofthese disorders aside, the similarity Clarkin, j.F., & Widiger, TA (1983). Prototypic typology and the in psychological profile calls into question the instruments' borderline personalitydisorder.JournalofAbnormalPsychology, 92, 263­ reliability in differentiating severe character pathologyfrom 275. post-traumatic disturbances. Clary,W.F., Burstin, Kj., &Carpenter,j.S. (1984). Multiple personality The profile of these measures that is defined as the disorderand borderlinepersonalitydisorder. Psychiatric Clinics ofN01th characteristic profile for BPD may often actually represent a America, 7, 89-99. case of MPD and should raise clinical suspicions of a disso­ ciative disorder. Diagnosis can best be made and confirmed, Coons, P.M. (1984). Thedifferential diagnosisofmultiplepersonality: then, by evaluating the extent ofdissociative symptoms, and A comprehensive I'eview, Psychiatric Clinics ofNOlth America, 7, 51..u7. reviewing the patient's history for evidence of severe child Coons, P.M., &Fine, c.G. (1990) .Accuracyofthe MMPI in identifying abuse.• multipIe personality disorder. Psychological Reports, 66, 831-834.

Coons, P.M., &Sterne,AL. (1986). Initialandfollow-up psychological REFERENCES testing on a group of patients with multiple personality disorder. Psychological RepOlts, 58, 43-49. Akiskal, H.S., Chen, S.E., Davis, G.C., Puryatian, V.R., Kashgarian, M., &Bulugen,TM. (1985). Borderline:An adjective in searchofa noun. Danesino,A,Daniels, T, &McLaughlin,T J. (1979) .]ojo,Josephine, Journal ofClinical Psychiatry, 46, 41-48. andJamie: Astudy ofmultiple personality by means ofthe . Journal ofPersonality Assessment, 43, 300-313. Allison, R.B., &Wagner, C.F. (1983). Diagnosisofmultiple personality with the Rorschach: A confirmation.Journal ofPersonality Assessment, Evans, R.W., Ruff, R.M., Braff, D.L., & Cox, DR (1986). On the 47,143-149. consistency ofthe MMPI inborderlinepersonalitydisorder. Perceptual and Motor Skills, 62, 579-585. American Psychiatric Association. (1987). Diagnostic and Statistical Manual (3rd Edition - Revised). Washington, DC: Author. Fairbank, SA, Keane, TM., & Malloy, P.F. (1983). Some preliminary data on the psychological characteristics ofVietnam Veterans with Andreason, N.C., &Akiskal, H.S. (1983). The specificityofBleulerian post-traumatic stress disorders. Joumal of Consulting and Clinical and Schneiderian Symptoms:Acritical re-evaluation. Psychiatric Clinics Psychology, 6, 912-919. ofN01thAmerica, 6, 41-54. Frances,A., &Clarkin,j.F. (1984). Reliabilityofcriteliaforborderline Andrulonis, PA, Glueck, B.C., Stroebel, C.F., & Vogel, .G. (1982). personality. AmericanJoumal ofPsychiatry, 141, 1080-1084. Borderline personality subcategories. Journal ofNel1Jous and Mental Disease, 170, 670-679. Greaves, G.B. (1980). Multiple personality: 165 years after Mary Reynolds. Joumal ofNervous and Mental Disease, 168, 577-596. Antoni, M., Tischer, P., Levine,j., Green, C., & Millon, T (1985). Refining personality assessment by combining MCMI high point Gunderson, j.G., & Kolb, j.E. (1978). Discriminating features of profilesandMMPI codes part I.JournalofPersonalityAssessment, 49,392­ borderline patients. Amm1canJournai ofPsychiatry, 135,792-796. 398. Hathway, S.R., & McKinley,j.C. (1983). Manualfor the administration Antoni, M., Tischer, P., Levine,j., Green, c., & Millon, T (1985). andscoringoftheMMPI. Minneapolis, Minnesota: National Computer Refining personality assessment by combining MCMI high point Services. profiles and MMPI codes part Ill.Joumal ofPersonality Assessment, 49, 508-515. Hays, W.L. (1981). Statistics (3rd Edition). New York, New York: Holt Rinehart Winston.

133 DISSOCIATIO:-;, Vol. III, :-;0. 3: September 1990 MPD, BORDERLINE PERSONALITY DISORDER & SCHIZOPHRENIA

Herman,].L., Perry, Cj., & van der Kolk, BA (1989). Childhood Rosenbaum, M. (1980). The role of the term schizophrenia in the traumainborderlinepersonality disorder. AmericanJournalofPsychiatry, decline of the diagnosis of multiple personality. Archives of General 146, 490-495. Psychiatry, 37, 1383-1385.

Horevitz, R.P., & Braun, B.G. (1984). Are multiple personalities Ross, CA, Heber, S., Norton, G.R., & Anderson, G. (1988). The borderline: An analysis of33 cases. Psychiatric Clinics ofNorth America, dissociative disorders interview schedule. DISSOCIATION, 2, 169-189. 7,69-83. Ross, CA, Heber, S., Norton, G.R., & Anderson, G. (in press). Kernberg, O.F. (in preparation). Some clinical observations on Differences between multiple personality disorder and other multiplepersonalitydisorder. Multiplepersonality disorder and dissociation. diagnosticgroupsonstructuredinterview.JournalofNervous andMental Ed. B.G. Braun and RP. Kluft. Disease, 177,487-491.

Keane, T.M., Malloy, P.F., & Fairbank,].A (1984). Empirical Schneider, K (1959). Clinicalpsychopathology (5th Edition). NewYork: development of an MMPI subscale for the assessment of combat Greene & Stratton. related PTSD. Journal ofConsulting and , 52, 888-891. Solomon, RS., & Solomon, V. (1982). Differential diagnosis of Kemp, K, Gilbertson, AD., & Torem, M.S. (1988). The differential multiple personality. Psychological Reports, 51, 1187-1194. diagnosisofmultiple personality disorderfrom borderlinepersonality disorder. DISSOCIATION, 1(4), 41-46. Solomon,R (1983). Use oftheMMPlwith multiplepersonalitypatients. Psychological Reports, 53, 1004-1006. Kluft, R.P. (1987a). An update on multiple personality disorder. Hospital and Community Psychiatry, 38, 363-373. Spitzer, RL., Endicott,]., & Gibbon, M. (1979). Crossing the border into borderline personalityand borderline schizophrenia. Archives of Kluft, R.P. (1987b). First rank symptoms as diagnostic clues to GeneralPsychiatry, 36, 17-24. multiple personality disorder. AmericanJournal ofPsychiatry, 144, 293­ 298. Spitzer, R.L., Williams,]., & Gibbon, M. (1989). The structured clinical interview for DSM-III-R. New York: Biometics Research Kroll,]., Martin, K, Lari, S., Pyle, R, & Zander,]. (1981). Borderline Department, New York State Psychiatric Institute. personality disorder: Construct validity of the concept. Archives of General Psychiatry, 38,1021-1026. Wagner, E.E., & Heise, M.R. (1974). A comparison of Rorschach recordsofthree multiple personalities.JournalofPersonality Assessment, Larmore, K,Ludwig, AM., &Cain, RL. (1977). Multiple personality: 38,308-331. An objective case study. BritishJournalofPsychiatry, 1031, 35-40.

Lerner, H.D., Sugarman, A, & Gaughran,]. (1981). Borderline and schizophrenic patients: A comparative study of defensive structure. Journal ofNervous and Mental Disease, 169, 705-711.

Liebowitz, M.R (1979). Is borderline a distinct entity? Schizophrenia Bulletin, 23-28.

Ludwig,AM., Brandsma,].M.,Wilbur, C.B., Bendfeldt,F., &jameson, D.H. (1972). The objective study ofa multiple personality. Archives of GeneralPsychiatry, 26, 298-310.

Millon, T. (1982). TheMillon ClinicalMultiaxialInventoryManual (2nd Edition). Minneapolis, Minnesota: National Computer Services.

Osgood, c.E., &Luria, Z. (1954).Ablind analysis ofacase ofmultiple personality using the semantic differential. Joumal of Abnormal Psychology, 49, 579-591.

Osgood, C.E., Luria, Z., & Smith, S.W. (1976). The three faces of Evelyn: A case report. Joumal ofAbnormal Psychology, 85, 256-270.

Perry, ].c., & Klerman, G.L. (1978). The borderline patient: a comparativeanalysisoffoursets ~fdiagnosticcriteria. Archives ofGeneral Psychiatry, 38,1021-1026.

Pope, H.S., & Lipinski,].F. (1978). Diagnosis in schizophrenia and manic depressive illness. Archives ofGeneral Psychiatry, 35, 811-826.

Putnam, F.W., Guroff,jj., Silberman, E.K., Barban, L., & Post, RM. (1986). The clinical phenomenology of MPD: Review of 100 recent cases. Journal ofClinical Psychology, 47, 285-293.

134 DISSOCIATION, Vol. III, No.3: September 1990