Peer Support for Abuse Survivors
Guidelines for Treating Dissociative Identity Disorder in Adults (2005)
Corresponding author: James A. Chu, MD
115 Mill St. Belmont, MA 02478
Phone: 617 855-2761 Fax: 617 855-2674
Copyright 1994, 1997, and 2005 by the International Society for the Study of Dissociation. The Guidelines may be reproduced without the written permission of the International Society for the Study of Dissociation (ISSD) as long as this copyright notice is included and the address of the ISSD is included with the copy. Violations are subject to prosecution under federal copyright laws.
Additional copies of the guidelines (US $5 for members, $10 for nonmembers) can be obtained by writing to the ISSD at 60 Revere Dr., Suite 500, Northbrook, IL 60062 USA. The correct citation for this revision of the Guidelines is: International Society for the Study of Dissociation. (2005). [Chu, J.A., Loewenstein, R., Dell, P.F., Barach, P.M., Somer, E., Kluft, R.P., Gelinas, D.J., Van der Hart, O., Dalenberg, C.J., Nijenhuis, E.R.S., Bowman, E.S., Boon, S., Goodwin, J., Jacobson, M., Ross, C.A., Sar, V, Fine, C.G., Frankel, A.S., Coons, P.M., Courtois, C.A., Gold, S.N., & Howell, E.]. Guidelines for treating Dissociative Identity Disorder in adults. Journal of Trauma & Dissociation, 6(4), in press.
At its meeting in Vancouver , BC , Canada , in May 1994, the Executive Council of International Society for the Study of Dissociation (ISSD) adopted the Guidelines for Treating Dissociative Identity Disorder (Multiple Personality Disorder) in Adults (1994). The Guidelines presented a broad outline of what to date was considered to be effective treatment for Dissociative Identity Disorder (DID). However, Guidelines like these are never finished and require ongoing revisions. A first revision of the Guidelines was proposed by the ISSD’s Standards of Practice Committee and was adopted by the ISSD Executive Council in 1997 after substantial comment from the ISSD membership and several revisions. This current revision was requested and approved by the ISSD Executive Council, and utilized the expertise of a Task Force of expert clinicians and researchers.
These Guidelines are not intended to replace clinical judgment. However, they summarize expert consensus concerning safe and effective treatment for DID patients. Where a clear divergence of opinion exists in the field, the Guidelines attempt to present the different points of view about the issue. The Guidelines strive to be as free as possible of bias toward any theoretical approach to treatment.
These Guidelines focus specifically on the treatment of DID. They are a practical guide to the management of patients – primarily adults over the age of 18 – and represent a synthesis of current scientific knowledge and rational clinical practice. However, DID is only one of the dissociative disorders. There continues to be a need to explore the phenomenology and treatment of other forms of pathological dissociation (e.g., Depersonalization Disorder, Dissociative Amnesia, etc.) as well as non-pathological forms of dissociation (e.g., the relation of trance states to dissociation). However, principles of treatment of DID may also be applicable to some extent in the treatment of other dissociative disorders.
There are now separate Guidelines for the Evaluation and Treatment of Dissociative Symptoms in Children and Adolescents (International Society for the Study of Dissociation [ISSD], 2004), available through the ISSD and published in the Journal of Trauma & Dissociation, 5(3), 119-150. The American Psychiatric Association has publi shed Practice Guidelines for the Treatment of Patients with Acute Stress Disorder (ASD) and Posttraumatic Stress Disorder (PTSD) (American Psychiatric Association, 2004). Since DID patients almost universally suffer from co-morbid PTSD, the reader may wish to consult those documents in addition to these Guidelines in developing treatment plans for dissociative disorder patients.
Considerable progress has been made in the diagnosis, assessment, and treatment of dissociative disorders during the past decades, as reflected by increased clinical recognition of dissociative conditions, the publication of numerous research and scholarly works on the subject, and the development of specialized diagnostic instruments. Peer reviewed publications have appeared in the international literature from clinicians and investigators in the United States, Canada, Puerto Rico, the Netherlands, Norway, Switzerland, Great Britain, Germany, Italy, France, Sweden, Spain, Turkey, Israel, Australia, New Zealand, Japan and other countries. These have included case reports, clinical case series, investigations using standardized diagnostic instruments, open clinical trials, treatment outcome studies, studies of psychophysiology and neurobiology, neuroimaging, and studies of cognition, among others. The Guidelines present key findings and generally accepted principles that reflect current scientific knowledge and clinical experience specific to the diagnosis and treatment of Dissociative Identity Disorder (DID; American Psychiatric Association, 2000a). It should be understood that these Guidelines supplement, but do not replace generally accepted principles of psychotherapy and psychopharmacology. Undoubtedly, future research will add to our present understanding of treatment of DID and other dissociative disorders.
The Guidelines are not intended to dictate the treatment for a specific patient. Treatment should always be individualized. Therapists should always conform to the prevailing standards of care, mental health codes and related local laws, as well as to ethical principles of their professional disciplines. There is a consensus in the dissociative disorders field that treatment for DID is most effective when it adheres to the basic principles of psychotherapy and psychiatric medical management, utilizing specialized techniques as needed.
By themselves, the Guidelines are not intended to be construed or to serve as a standard of clinical care. These parameters of practice reflect the state of the art in this field at the present time. They are not designed to include all proper methods of care or to exclude other acceptable treatment interventions. Moreover, adhering to the Guidelines will not necessarily result in a successful treatment outcome in every case. Clinicians must use their judgment concerning the appropriateness for a particular patient of a specific method of care in light of the clinical data presented by the patient and options available at the time of treatment.
The Guidelines were developed by psychiatrists, psychologists, and other mental health practitioners in active clinical practice, research or other academic endeavors. The Guidelines were extensively reviewed by members of ISSD. Contributors and reviewers were asked to base their recommendations on an objective evaluation of available evidence.
DID and dissociative disorders are not rare conditions. In studies of the general population, a prevalence rate of DID of one to three percent of the population has been described (Murphy, 1994; Ross, 1991; Waller & Ross, 1997), although some researchers have criticized the methodology of these studies, and have suggested a somewhat lower prevalence. Clinical studies in North America, Europe, and Turkey have found that between one to 20 percent of patients on general inpatient psychiatric units, adolescent inpatient units, and in substance abuse, eating disorders, and obsessive compulsive disorder treatment may meet DSM-IV-TR (American Psychiatric Association, 2000a) diagnostic criteria for DID, particularly when evaluated with structured diagnostic instruments. Many of these patients had not been clinically diagnosed previously with a dissociative disorder (Bliss & Jeppsen, 1985; Goff, Olin, Jenike, Baer, & Buttolph, 1992; Latz, Kramer, & Highes, 1995; McCallum, Lock, Kulla, Rorty, & Wetzel, 1992; Ross, Anderson, Fleisher, & Norton, 1991; Modestin, Ebner, Junghan, & Erni, 1995; Ross et al., 1992; Saxe et al., 1993; Tutkun et al., 1998).
Accurate clinical diagnosis affords early and appropriate treatment for the dissociative disorders. Seven studies of 719 DID patients have shown that they spent five to 11.9 years in the mental health system before they were diagnosed as having DID (Boon & Draijer, 1993a; Coons, Bowman, & Milstein, 1988; Martínez-Taboas, 1991; Middleton & Butler, 1998; Putnam, Guroff, Silberman, Barban, & Post, 1986; Rivera, 1991; Ross, Norton, & Wozney, 1989). While progress has been made in educating the professional community about the prevalence and clinical presentation of dissociative disorders, these seven studies suggest that many cases of DID and related disorders are still being missed, misdiagnosed, and inappropriately treated.
The primary difficulties in diagnosing DID result from lack of education among clinicians about dissociation, dissociative disorders, and the effects of psychological trauma. This leads to limited clinical suspicion about dissociative disorders and misconceptions about their clinical presentation. Most clinicians have been taught that DID is a rare disorder with a florid, dramatic presentation. In fact, DID is a relatively common disorder which presents with subtle symptoms in a patient who may minimize or conceal symptoms. DID patients commonly present in a polysymptomatic fashion with dissociative and PTSD symptoms embedded in a matrix of symptoms such as depression, panic, somatoform symptoms, eating disorder symptoms, etc. that may lead only to diagnosis of these co-morbid conditions. This results in long and frequently unsuccessful treatment for these other conditions.
Further, almost all practitioners were taught standard diagnostic interviewing and mental status examinations that do not include questions about dissociation, PTSD symptoms, or a history of psychological trauma. Since DID patients rarely directly volunteer information about dissociative symptoms, absent questions about such symptoms, or recognition of them when they present, the clinician cannot diagnose DID. Accordingly, the sine qua non for the diagnosis of DID is the use of diagnostic interviews that inquire about dissociation, supplemented when necessary by screening instruments and structured interviews that assess the presence or absence of dissociative symptoms.
The Diagnostic and Statistical Manual, 4 th Edition, Text Revision (DSM-IV-TR; American Psychiatric Association, 2000a) defines the following diagnostic criteria for Dissociative Identity Disorder (300.14):
A. The presence of two or more distinct identities or personality states (each with its own relatively enduring pattern of perceiving, relating to, and thinking about the environment and self).
B. At least two of these identities or personality states recurrently take control of the person’s behavior.
C. Inability to recall important personal information that is too extensive to be explained by ordinary forgetfulness.
D. The disturbance is not due to the direct physiological effects of a substance (e.g., blackouts or chaotic behavior during Alcohol Intoxication) or a general medical condition (e.g., complex partial seizures). Note: In children, the symptoms are not attributable to imaginary playmates or other fantasy play.
In recent years, there has been debate about the diagnostic criteria for DID. Some have suggested that a set of polythetic criteria would more accurately portray the typical polysymptomatic presentations of DID patients (Dell, 2001) . Others have argued that the current criteria are sufficient (Spiegel, 2001). Still others have suggested that dissociative disorders should be reconceptualized as among “Trauma Spectrum Disorders,” emphasizing their intimate association with overwhelming and traumatic circumstances (Davidson & Foa, 1993).
Dissociation is defined as “A disruption in the usually integrated functions of consciousness, memory, identity, or perception” (American Psychiatric Association [APA], 2000a, p. 519). Dissociation is an ongoing process in which certain information (such as feelings, memories, and physical sensations) is kept apart from other information with which it would normally be logically associated. Dissociation can be a psychological defense mechanism that also has psychobiological components. Generally, it is thought to originate in “… a normal process that is initially used defensively by an individual to handle traumatic experiences [that] evolves over time into a maladaptive or pathological process…” (Putnam, 1989, p. 9).
DID patients have distinct identity or personality states, each with its own relatively enduring pattern of perceiving, relating to, and thinking about the environment and self (modified from DSM-IV-TR [APA, 2000a]). Alternate identities are dissociated parts of the mind that the patient experiences as separate from each other. Many terms have been developed to describe the DID patient’s subjective sense of self-states or identities. These include: personality, personality state, self-state, disaggregate self-state, alter, alter personality, alternate identity, part, part of the mind, part of the self, entity, among others. It should be noted that while many of these terms are useful and/or descriptive, some are associated with particular theoretical and conceptual points of view; all are somewhat metaphorical, and some are more useful in certain contexts than others . Some Task Force members advocated for the use of terms such as “dissociated parts of the personality” or “dissociative parts of the personality” to describe the degree of internal separateness and autonomy. The DSM-IV-TR (American Psychiatric Association, 2000a) uses the term “alternate identity” and the Guidelines use this term for consistency.
Clinicians should also attend to the unique, personal language with which DID patients characterize their subjective self-states or identities. Patients commonly refer to themselves as having, among others, “parts,” “parts inside,” “aspects,” “facets,” “ways of being,” “voices,” “multiples,” “selves,” “ages of me,” “people,” “persons,” “individuals,” “spirits,” “demons,” “lines,” and “others.” It can be helpful in working with DID patients to use their own unique descriptive terms for their alternate identities. However, some members of the Guidelines Task Force recommend that clinicians avoid using terms such as “people,” “persons” or other terms that might convey or reinforce a belief that the alternate identities are truly separate individuals. Other Task Force members are of the opinion that judicious use of these latter terms, in and of themselves, would not contribute to reinforcing such beliefs.
Alternate identities have been defined in a number of ways. For example, Putnam (1989) describes them as “ highly discrete states of consciousness organized around a prevailing affect, sense of self (including body image), with a limited repertoire of behaviors and a set of state dependent memories.” Kluft (1988) states that: “ A disaggregate self state (i.e., personality) is the mental address of a relatively stable and enduring particular pattern of selective mobilization of mental contents and functions, which may be behaviorally enacted with noteworthy role-taking and role-playing dimensions and sensitive to intrapsychic, interpersonal, and environmental stimuli. It is organized in and associated with a relatively stable (but order effect dependent) pattern of neuropsychophysiologic activation, and has crucial psychodynamic contents. It functions both as a recipient, processor, and storage center for perceptions, experiences, and the processing of such in connection with past events and thoughts, and/or present and anticipated ones as well. It has a sense of its own identity and ideation, and a capacity for initiating thought processes and action” (pp. 55).
In short, the alternate identities are intrapsychic entities that have a sense of self, have an emotional repertoire, and can process information. They have both the potential for “being-in-the world” behavioral enactments as well as subjective symbolic and metaphorical characteristics. They have aspects of both structure and process.
It is outside the scope of these Guidelines to provide a comprehensive discussion of current theories concerning the development of the alternate identities in DID (see Loewenstein & Putnam, 2004, and Putnam, 1997 for a more complete discussion). Briefly, however, many believe that alternate identities result from the inability of many traumatized children to develop a unified sense of self that is maintained across various behavioral states, particularly if the traumas occur before the age of five. DID develops during the course of childhood and rarely, if ever, derives from adult-onset trauma (unless it is superimposed on pre-existing childhood trauma) . Traumatic experiences, particularly severe, repetitive trauma, produce extreme states of experiences in the child. Simultaneously, development of discrete, personified “behavioral states” in the child are thought to encapsulate intolerable traumatic memories and affects to mitigate their effects on overall development. Also, disturbed caretaker-child attachments and parenting further disrupt the child’s ability to integrate experiences that occur in different contexts (e.g., with family, in traumatic situations, with friends, in school). Dissociation of traumatic experiences in this way may preserve relationships with caretakers and allow more normal maturation in other developmental areas such as intellectual, interpersonal, and artistic endeavors. Secondary structuring of these discrete behavioral states occurs over time through a variety of developmental and symbolic mechanisms resulting in the characteristics of the specific alternate identities. Once this process begins, it may generalize to help the patient manage and cope with a variety of life events and experiences. Further, alternate identities may diverge considerably in number, complexity, and sense of separateness as the child proceeds through latency, adolescence, and adulthood (Kluft, 1984; Putnam, 1997).
Another etiological model posits four factors that are required for DID to develop: 1) the capacity for dissociation, 2) experiences that overwhelm the non-dissociative coping of the child, 3) secondary structuring of the DID alternate identities with individualized characteristics such as names, ages, genders, and 4) lack of soothing and restorative experiences that leave the child to find ways of comforting him/herself after overwhelming experiences (Kluft, 1984). In particular, factor 3 phenomena may be very diverse and may differ significantly from patient to patient. For example, parameters that may lead to very elaborated alternate identity systems include multiple traumas, high levels of creativity and intelligence, and extreme withdrawal from the world with intense involvement in fantasy experiences, among others. Accordingly, most experienced therapists pay relatively limited attention to the overt presentational characteristics of the different alternate identities. They focus instead on the cognitive, affective, and psychodynamic information embodied by the alternate identity as well as that of the alternate identities as a “system” of representation, symbolization, and meaning in the mind.
The theory of “Structural Dissociation,” another etiological model, is based on the ideas of Janet and attempts to create a unified theory of dissociation that includes DID. This theory suggests that in dissociation there is a basic failure of integration of systems of ideas and functions of the personality. Following trauma, the personality divides into an “apparently normal part of the personality” (ANP) dedicated to daily functioning and an “emotional part of the personality” (EP) dedicated to defense. Defense in this context is related to psychobiological functions related to survival in response to life threat such as fight/flight, not to the psychodynamic notion of defense. It is hypothesized that chronic trauma and/or neglect leads to additional secondary structural dissociation of EPs. In this model, DID is produced by a tertiary level of structural dissociation among both ANPs and EPs related to long-standing trauma, neglect, and attachment pathology during early development (Nijenhuis & Van der Hart, 1999; Van der Hart, Nijenhuis, Steele, & Brown, 2004).
In short, these developmental models posit that DID does not arise from a previously mature, unified mind or “core personality” that becomes “shattered” or fractured. Rather, DID results from a failure of normal developmental integration caused by overwhelming experiences and disturbed caretaker-child interactions during critical early developmental periods leading to the development and elaboration of discrete, personified behavioral states. This view may help clinicians understand how patients may vary considerably in the number, type, and characteristics of the alternate identities. Further, it helps explain how patients could experience themselves as having very numerous alternate identities and/or seemingly bizarre alternate identities such as animals, deities, or mythical beings.
A careful clinical interview and a thoughtful differential diagnosis can usually lead to the correct diagnosis in persons who have DID (Coons, 1984). The patient should be asked about episodes of amnesia, fugue, depersonalization, derealization, identity confusion, and identity alteration (Steinberg, 1995), age regressions, autohypnotic experiences, hearing voices (Putnam, 1991a), passive-influence symptoms such as “made” thoughts, emotions, or behaviors (Dell, 2001; Kluft, 1987a), and somatoform symptoms such as bodily sensations related to past trauma (Nijenhuis, 1999). Loewenstein (1991a) has described an office mental status examination that inquires about many of the symptoms of DID, including evidence of alternate identities, amnesia, autohypnotic phenomena, PTSD, somatoform symptoms, and affective symptoms.
Clinicians should keep in mind that some persons with DID do not realize (or do not acknowledge to themselves) that their internal experience is different from that of others. In keeping with the view that dissociation is a defense against uncomfortable realities, alternate identities and other dissociative symptoms are commonly denied and disavowed by persons with DID.
The diagnosis of DID is nearly universally associated with an antecedent history of significant traumatization – most often first occurring in childhood (Putnam et al., 1986). Virtually every systematic study that has examined the trauma history of DID patients has found that DID patients have a higher rate of early childhood trauma than any other clinical group. Accordingly, the diagnostic process should include an effort to outline the patient’s trauma history. Clinicians should use careful clinical judgment about how much detail of traumatic experiences to pursue during initial interviews, especially when those experiences seem to be poorly or incompletely remembered. A premature trauma anamnesis may evoke a florid decompensation (i.e., severe posttraumatic and dissociative symptoms). Due to their dissociative amnesia, DID patients often provide a fragmented history during initial treatment; a more complete personal history typically emerges over time. Because most DID patients also have Posttraumatic Stress Disorder (PTSD), the diagnostic process should assess whether symptoms of PTSD are present. Assessment for PTSD may provide another avenue into the patient’s trauma history.
There are three classes of instruments that assess dissociation: clinician-administered structured interviews, clinician-administered measures, and self-report instruments.
There are two structured interviews for the dissociative disorders: the Structured Clinical Interview for DSM-IV Dissociative Disorders-Revised (Steinberg, 1994a, 1994b, 1995), and the Dissociative Disorder Interview Schedule (Ross, 1997; Ross, Heber, Norton, Anderson, Anderson, & Barchet, 1989).
The Structured Clinical Interview for DSM-IV Dissociative Disorders-Revised (SCID-D-R) is a 277-item interview that assesses five symptoms of dissociation: amnesia, depersonalization, derealization, identity confusion, and identity alteration. Most items have follow-up questions that request a description of the experience, specific examples, and an estimate of the experience’s frequency and impact on social functioning and work performance. The SCID-D-R diagnoses the five DSM-IV dissociative disorders; it also yields a score for each of the five dissociative symptoms and a total score. These scores are based on frequency and intensity of symptoms. The SCID-D-R takes 45 to 180 minutes or more to administer if subjects endorse many positive responses to interview questions and are allowed to elaborate on their answers extensively. The interviewer, whether a clinician or a trained technician, must have considerable familiarity with dissociative symptoms. The SCID-D-R has good-to-excellent reliability and discriminant validity. Total SCID-D-R scores have been shown to have a correlation of .78 with the Dissociative Experiences Scale (Boon & Draijer, 1993b) and .78 to .93 with the MID (Dell, 2004; Gast et al., 2003; Somer & Dell, 2005).
The Dissociative Disorder Interview Schedule (DDIS) is a 132-item structured interview with a yes/no format that assesses the symptoms of the five DSM-IV dissociative disorders, somatization disorder, borderline personality disorder, and major depressive disorder. The DDIS also assesses substance abuse, Schneiderian first-rank symptoms, trance, childhood abuse, secondary features of Dissociative Identity Disorder, and supernatural/paranormal experiences. The instrument usually takes 30 to 60 minutes to administer. The DDIS makes categorical diagnoses and yields an index of the number of items that were endorsed in each section of the interview. The DDIS does not assess frequency or severity of symptoms. The DDIS had an overall interrater reliability of 68,. an overall kappa of .96 for clinician-DDIS agreement on the diagnosis of DID, and a sensitivity of .95 for the diagnosis of DID (Ross, 1997). False-positive diagnoses of DID occurred in less than 1% of cases (Ross, 1997). The DDIS’ secondary features of DID and Schneiderian first-rank symptoms correlated with the DES at .78 and .67, respectively (Ross, 1997). Effective use of the DDIS requires less training than does the SCID-D-R.
There is one clinician-administered inventory that measures dissociative states, the Clinician Administered Dissociative States Scale (CADSS) (Bremner et al., 1998). It has 27 items with 19 subject-rated items and 8 observer-scored items, all rated on a 0-4 scale. It has three factors that assess symptoms of amnesia, depersonalization and derealization. The CADSS has an interrater reliability with an intraclass correlation coefficient of .99 for the subject-rated scale, .92 for the total score, but only .34 for the observer rated items. Cronbach’s alpha was 92. for the scale as a whole, but somewhat lower for the individual symptom subscales. Correlation was significant with the Dissociative Experiences Scale (DES; r = .48, df = 49, p < .001), and with the SCID-D (r = .42, df = 40, p = .005). The CADSS scores were significantly different in PTSD patients with high dissociation compared to PTSD patients with low dissociation, non-PTSD combat veterans, schizophrenics, patients with affective disorders, and healthy controls. CADSS scores increased significantly with exposure to traumatic memories in a subgroup of PTSD participants. The CADSS is viewed as a reliable and valid measure of present-state dissociative symptomatology and readily allows for repeated measures. It has been used primarily in research on psychophysiology and psychopharmacology of PTSD.Self-Report Instruments
There are six self-report measures of dissociation that have been used with some frequency. Five of the instruments (i.e., the Dissociative Experiences Scale [DES], the Questionnaire of Experiences of Dissociation [QED], the Dissociation Questionnaire [DIS-Q], Somatoform Dissociation Questionnaire [SDQ] and the Multiscale Dissociation Inventory [MDI] ) are brief screening inventories. The Multidimensional Inventory of Dissociation (MID) is a multiscale diagnostic instrument.
The Dissociative Experiences Scale (DES; Bernstein & Putnam, 1986) is the first and most successful of the self-report measures of dissociation. It has been translated into many languages from its original English version. As of 1997, the DES had been used in over 250 published studies (Carlson, 1997). The DES is a 28-item self-report instrument. Items are rated on a continuous scale (original version) or on an 11-point Likert scale (revised version) that ranges from 0 (“never”) to 100 (“always”). DES items primarily tap absorption, imaginative involvement, depersonalization, derealization, and amnesia. The DES has excellent internal consistency: Cronbach alphas of .95 (Frischholz, et al., 1990) and .96 (Boon & Draijer, 1993b), and split-half reliabilities of .83 (Bernstein & Putnam, 1986) and .93 (Pitblado & Sanders, 1991). The DES has a four-to-eight week temporal stability of .84 (Bernstein & Putnam, 1986) and a four-week temporal stability of .96 (Frischholz, et al., 1990). The DES correlated .78 with the SCID-D-R, .85 with the SDQ-20, and 90 with. the MID (Dell, 2004). Construct validity of the DES was supported by a steady progression of mean DES scores (from low to very high) across the following groups: nonclinical population, patients with a trauma history, patients with PTSD, patients with Dissociative Disorder Not Otherwise Specified (DDNOS), and patients with DID (Carlson & Putnam, 1993). Using a cutoff score of 30, DES scores predicted a diagnosis of DID with a sensitivity of .74 and a specificity of .80 (Carlson, Putnam, Ross, Torem, Coons, Dill, Loewenstein & Braun, 1993). Similarly, Draijer and Boon (1993b) used a cutoff score of 25 to predict the diagnosis of DID with a sensitivity of .93 and a specificity of .86. A shorter version of the DES, the DES-Taxon (DES-T), utilizes eight questions from the DES that are most closely identified with a taxon (class) of individuals who demonstrate “pathological dissociation” (Waller, Putnam, & Carlson, 1996).
The Questionnaire of Experiences of Dissociation (QED; Riley, 1988) is a 26-item, true/false self-report instrument. Its items were drawn from the literature about hysteria, multiple personality, non-DID dissociative patients, and temporal lobe epilepsy. The QED was shown to have a Cronbach alpha coefficient of .77 (Riley, 1988). The QED has discriminated groups of nonclinical individuals and alcoholics from groups of somatization disorder, PTSD, and DID patients (Dunn, Ryan, Paolo & Miller, 1993; Riley, 1988). A QED study of college students found six factors: depersonalization, process amnesia, dissociated body behavior, fantasy/daydreams, trance, and imaginary companions (Ray, June, Turaj & Lundy, 1992). Although there is a modest research literature on the QED, the instrument does not appear to be frequently used by clinicians.
The Dissociation Questionnaire (DIS-Q; Vanderlinden, Van Dyck, Vandereycken, Vertommen, & Verkes, 1993; Vanderlinden, 1993) is a 63-item, five-point Likert format, self-report instrument. The initial item-pool (N = 95) from which the DIS-Q was developed was comprised of the DES, Perceptual Alteration Scale (PAS; Sanders, 1986), QED, and additional items that were derived from dissociative patients. Data from four samples yielded a 63-item DIS-Q, with a stable four-factor solution that explained 77% of the variance: (1) identity confusion and fragmentation, (2) loss of control, (3) amnesia, and (4) absorption. The total DIS-Q had a Cronbach alpha coefficient of 96;. the subscales had alpha values of .94, .93, .88, and .67, respectively. The total DIS-Q had a three-to-four-week temporal stability of 94;. subscales had stabilities of .92, .92, .93, and .75. The DIS-Q discriminated among diagnostic groups: nonclinical individuals, general psychiatric patients, DDNOS patients, and DID patients. The DIS-Q correlated .85 with the DES (Vanderlinden et al., 1993; Vanderlinden, 1993). Developed in The Netherlands, the DIS-Q is more commonly used by European than North American clinicians and researchers.
The Somatoform Dissociation Questionnaire-20 (SDQ-20) is a 20-item self-report instrument using a five-point Likert scale (Nijenhuis, Spinhoven, Van Dyck, Van der Hart, & Vanderlinden, 1996). Based on the clinical/descriptive work of Janet (1889), the SDQ-20 is explicitly conceptualized as a measure of somatoform dissociation. The SDQ-20 items address tunnel vision, auditory distancing, getting stiff, psychogenic blindness, difficulty urinating, insensitivity to pain, psychogenic paralysis, pseudoseizures, and so on. The SDQ-20 was shown to have a Cronbach alpha coefficient of .95 and correlated .76 with the DIS-Q and .85 with the DES (Nijenhuis et al., 1996, 1999). Its construct validity was shown to be supported by its solid correlation with reported trauma (Nijenhuis, Spinhoven, Van Dyck, Van der Hart, & Vanderlinden, 1998). SDQ-20 scores discriminated among diagnostic groups even when controlling for general psychopathology (Nijenhuis, Van Dyck, Spinhoven, Van der Hart, Chatrou, Vanderlinden, & Moene, 1999). A shorter version of the SDQ-20, the SDQ-5, is comprised of five items from the SDQ-20 (Nijenhuis, 1999). The SDQ-5 was developed as a screening instrument for dissociative disorders and correlates well with findings of the longer inventory.
The Multidimensional Inventory of Dissociation (MID) is a 218-item self-report, multiscale measure of pathological dissociation that makes diagnoses and yields a comprehensive dissociative profile (Dell, 2004). The MID has 23 dissociation scales that vary in length from three to 12 items. The 23 scales achieved good-to-excellent Cronbach alpha values in a large clinical sample (range = .84 to .96; median alpha value = .91) and good-to-excellent temporal stabilities in a small clinical sample over a four- to eight-week test-retest interval (range = .82 to .97; median temporal stability coefficient = .92). The mean MID score was shown to have a four- to eight-week temporal stability coefficient of .97. The MID’s convergent validity was demonstrated by the instrument’s high correlations with five other measures of dissociation: DES (r = .90), DIS-Q (r = .83), SCID-D-R (r = .78-92), QED (r = .75), and SDQ-20 (r = 75). A factor. analysis of the MID’s items indicated that the MID has 12 first-order factors and one second-order factor: dissociation. The MID had a positive predictive power of .93, a negative predictive power of .84, and an overall predictive power of .89 for major dissociative disorder (i.e., DID or DDNOS-1). The MID is the only measure of dissociation that has validity scales: Defensiveness, Rare Symptoms, Attention-Seeking Behavior, Factitious Behavior, and Neurotic Suffering.
The Multiscale Dissociation Inventory (MDI; Briere, 2002) is a 30-item multiscale measure of dissociation with a 5-point Likert format. The MDI is fully standardized, allowing t score comparisons to a normative group of trauma-exposed men and women. It yields six subscales – Disengagement, Depersonalization, Derealization, Emotional Constriction/Numbing, Memory Disturbance, and Identity Dissociation – and a total dissociation scale. The scales have good to excellent coefficient alphas, and retain their strong psychometric properties in clinical and nonclinical populations (Briere, Weathers, & Runtz, 2005). The scales show their expected relationships to trauma history and to other known dissociation scales, including the DES (Briere, 2002; Dietrich, 2003). A cutoff score of 15 on the Identity Dissociation Scale yields high specificity (92%) and sensitivity (93%) for identifying individuals with Dissociative Identity Disorder. The MDI takes approximately 15 minutes to complete, and has been written for a sixth grade reading level.
Some measures that are frequently used in psychological testing (e.g., Rorschach, MMPI-2, WAIS-R, MCMI-III) can provide understanding of the patient’s personality structure and may suggest DID (Armstrong, 1991). In combination with more specific diagnostic testing (e.g., SCID-D-R, DES, etc.), standardized psychological tests may aid the clinician in differential diagnosis and prognosis, the identification of co-morbid disorders, and the evaluation of treatment options. It should be noted, however, that commonly used psychological tests were not designed to detect dissociative disorders, and may lead to misdiagnosis when the psychologist (1) is not familiar with the typical responses of dissociative patients on these tests, (2) does not administer additional dissociation-specific tests (such as structured clinical interviews), and (3) does not inquire specifically about dissociative symptoms during the clinical or testing interview.
There has been a heated debate in the professional literature concerning the so-called “iatrogenesis” of DID. Expert opinion in the dissociative disorders field has argued strongly against the notion that DID can be produced iatrogenically (cf. Gleaves, 1996). No study in any clinical or research population has yet demonstrated that the full clinical syndrome of DID can be produced in this fashion. On the other hand, false positive diagnoses of DID may arise from a variety of circumstances.
Clinicians should be alert to the possible false positive diagnosis of DID. Dissociative symptoms are central in other dissociative disorders, and may be present in PTSD as well as other disorders, such as somatization disorder. It should not be assumed that symptoms such as amnesia or even identity “fragmentation” automatically signal the presence of DID. The identity problems that occur in personality disordered patients may occasionally be misdiagnosed as a symptom of DID by clinicians who are unfamiliar with the clinical features of DID. Mood changes in bipolar patients, especially bipolar patients with co-morbid PTSD, have also been confused with clinical DID. Some psychotic patients with delusions of being inhabited by other people may also be misdiagnosed as DID. Of course, it should be noted that some DID patients have co-morbid bipolar or psychotic disorders. In these cases, a careful diagnostic evaluation, including diagnostic interviews and psychological testing, may be essential for an adequate differential diagnosis.
Patients who already have some dissociative features may be particularly at risk for the misdiagnosis of DID. When such patients are subjected to premature, intense exploration of trauma memories without adequate symptom stabilization, they may subjectively have a sense of personality fragmentation that can be misdiagnosed as DID. Naïve clinicians may also confuse the patient’s investment in a metaphorical “inner child” or similar phenomena with clinical DID. Further, clinicians who are poorly trained in hypnosis, and use this modality in treatment, may confuse hypnotic phenomena such as the production of “ego states” with clinical DID (Watkins & Watkins, 1997).
As with any psychiatric condition, a presentation of DID may be factitious or malingered. Clinicians should be alert to atypical presentations of apparent DID, especially in situations where there is strong motivation to simulate an illness (e.g., pending legal charges; Coons, 1991; Coons & Milstein, 1994; Draijer & Boon, 1999; Kluft, 1987c; Thomas, 2001). Research studies have shown that the SCID-D can be useful in differentiating “imitative” DID patients from those who actually meet DID diagnostic criteria. Instead, many of the imitative patients meet criteria for DSM-IV cluster B personality disorders and show a significantly different profile on the SCID-D (see also Boon & Draijer 1993c, Draijer & Boon, 1999 ). In addition to the clinical interview, comprehensive evaluation of possible factitious and/or malingered DID may include (a) measures of dissociation (e.g., DES, MID, SDQ-20, SCID-D, DDIS), (b) standardized measures of PTSD, (c) measures of malingering (e.g., Structured Interview of Reported Symptoms [SIRS]; Rogers Bagby, & Dickens, 1992), although recent studies have suggested that some DID subjects may be misdiagnosed as feigning on the SIRS, (d) standard psychological tests (e.g., Millon Clinical Multiaxial Inventory [MCMI-II; Millon, 1997] or the Rorschach), (e) a review of all available clinical documentation, and (f) use of collateral sources of information. In forensic settings, some or all of these assessments are likely to be required to supplement the clinical interview.
The DID patient is a single person who experiences himself/herself as having separate self-states or alternate identities, each of which has relative psychological autonomy from one another. These subjective identities may take executive control of the patient’s body and behavior, and/or influence the patient’s experience and behavior. It is important for clinicians to keep in mind that despite the DID patient’s subjective experience, the patient is not a collection of separate people sharing the same body. The DID patient should be seen as a whole adult person, with the alternate identities sharing responsibility for life as it is now. All the alternate identities together make up the identity or personality of the human being with DID. In the clinical setting, clinicians working with DID patients generally ought to hold the whole person to be responsible for the behavior of any or all of the alternate identities, even in the presence of amnesia for the behavior, or lack of a sense of control or agency over the behavior.
It is the consensus of expert opinion that wherever possible, treatment should move the patient toward better integrated functioning (see below). Although the therapist may, at times, address alternate identities as if they were separate, a fundamental tenet of the therapeutic work with DID is to bring about an increased degree of communication and coordination among them. The therapist should keep in mind that the “patient” is the entire spectrum of alternate identities, not just the “host” (defined as the alternate identity that has the most presence in the outside world at a given time), the alternate identity that bears the official name of the person, or any group of identities. Thus, it is counter-therapeutic to treat any alternate identity as if it were more “real” or more important than any other. It is also counter-therapeutic to suggest that the patient create additional alternate identities, to name alternate identities when they have no names (although the patient may choose names if he/she wishes), or to suggest that alternate identities function in a more elaborated and autonomous way than they already are functioning.
On the other hand, it is counterproductive to tell patients to ignore or “get rid” of alternate identities ( although it is acceptable to provide strategies for the patient to resist the influence of destructive or self-destructive alternate identities, or to help control chaotic emergence of certain alternate identities at inappropriate circumstances or times ). In addition, the therapist should not “play favorites” among the alternate identities or exclude apparently unlikable or disruptive identities from the therapy (although such steps may be necessary for a limited period of time at some stages in the treatment of some patients to provide for the safety and stability of the patient or the safety of others). The therapist should help foster the idea that all alternate identities represent adaptive attempts to cope or to master problems that the DID patient has faced. Accordingly, the alternate identities can be helped to find more adaptive ways to solve problems rather than using solutions that are dysfunctional, unsafe, or problematic.
Most experts in the dissociative disorders field agree that the most stable treatment outcome is fusion – complete integration, merger, and loss of separateness – of all identity states (e.g., Kluft, 1993a). However, a considerable number of DID patients will not be able to achieve full fusion and/or do not see fusion as desirable, even after undergoing considerable treatment. Accordingly, a more realistic long-term outcome may be a cooperative arrangement, sometimes termed “resolution” – that is, optimally integrated and coordinated functioning among alternate identities that allows optimal vocational, interpersonal, intrapsychic, and emotional functioning. Patients who achieve this kind of resolution frequently are more vulnerable than those who gain stable fusion to later decompensate into florid DID and/or PTSD when sufficiently stressed. Many factors can contribute to patients achieving this kind of resolution rather than fusion. These factors can include avoidance of unresolved, extremely painful life issues, lack of financial resources for treatment, co-morbid medical disorders, advanced age, significant unremitting DSM Axis I and/or Axis II co-morbidities, and/or significant narcissistic investment in the alternate identities and/or DID itself, among others (see below).
Terms such as “integration” and “fusion” are sometimes used in a confusing way. Integrationrefers to the work on all forms of dissociated mental processes throughout treatment. Kluft (1993a, p. 109) defines integration as, “[An] ongoing process of undoing all aspects of dissociative dividedness that begins long before there is any reduction in the number or distinctness of the identities, persists through their fusion, and continues at a deeper level even after the identities have blended into one. It denotes an ongoing process in the tradition of psychoanalytic perspectives on structural change” Fusion refers to a point in time when two or more alternate identities experience themselves as joining together with a complete loss of subjective separateness. Final fusion refers to the point in time when the patient shifts his/her sense of self from that of having multiple selves to that of a unified subjective self. This implies fusion of all alternate identities and the persistence of a sense of subjective unity. Kluft (1984) defined “stable fusion” as occurring after 27 months (two years after an initial three months) without evidence of alternate identities. Even after final fusion, additional work on “integration” of the patient’s dissociated ways of thinking and experiencing may continue. For instance, the therapist and patient might need to work on integrating an ability previously held by one alternate identity across all roles in which it would now be appropriate, or learn what the patient’s new pain threshold is, or how to integrate all the dissociated ages into one chronological age, and to re-gauge appropriate and healthy exercise or exertion levels for the patient’s age. Traumatic and stressful material also may need to be reworked from this new unified perspective.
S tudies of treatment outcome for DID and of cost efficacy for DID treatment have shown that DID patients can have a very successful treatment outcome. Single case descriptions of successful treatments for DID date back more than a century. Outcome data for groups of patients treated by a single clinician have demonstrated that many DID patients can achieve and sustain substantial improvement including complete final fusion and integration (Kluft, 1984, 1986a). Systematic outcome studies also have shown a positive outcome for DID when using a model including direct work with alternate identities and trauma material to help achieve greater integration for the DID patient. The first such study followed up 20 DID patients an average of three years after intake (Coons, 1986). The majority of patients were in treatment with therapists unfamiliar with DID, but supervised by experienced clinicians. Nonetheless, two-thirds of the clinicians reported moderate to great improvement in their patients.
In the Netherlands, a chart review study of 101 dissociative disorder patients in outpatient treatment for an average of six years found that clinical improvement was related to the intensity of the treatment with more comprehensive therapies having better outcomes (Groenendijk & Van der Hart, 1995). The largest and most systematic treatment outcome study reevaluated 54 DID inpatients two years after discharge to outpatient treatment (Ellason & Ross, 1997). As a group, there were significant overall decreases in psychopathology including number of Axis I and Axis II disorders, decreased DES scores, decreased depression on the Beck and Hamilton scales, and decreased dissociative symptoms on all of the DDIS subscales. Patients who achieved final fusion according to rigorous criteria were the most improved.
Two studies of outcomes and cost-efficacy of DID treatment have concordant findings suggesting that outcome depends on patients’ clinical characteristics (Loewenstein, 1994; Putnam & Loewenstein, 2000) . The more treatment responsive group of DID patients showed significant remission of symptoms within three to five years of beginning appropriate treatment. A second group with more alternate identities and more personality disorder features showed good outcome but required hospitalizations in addition to outpatient treatment. A third group, characterized by the longest period of treatment before DID diagnosis, largest number of alternate identities, and most personality disorder problems had a much longer, more costly, and more difficult course. Overall, however, treatment approaches specifically targeting DID showed reductions in overall psychiatric treatment cost after the first year compared with prior treatment for these patients.
These preliminary studies have notable limitations including the diverse and non-standardized nature of the therapy and lack of comparison groups. Nonetheless, in aggregate, they indicate that many DID patients improve with treatments focused on their dissociative symptoms and that overall treatment costs may be saved in the long-term by using the phased trauma treatment model for these patients (see below).
These studies also point to a major issue in treatment and treatment planning: the heterogeneous nature of the DID population. The literature identifies several subgroups of DID patients (cf. Kluft, 1994). At one extreme are relatively high functioning and highly motivated patients with relatively few co-morbidities and reasonable social supports who have a relatively rapid and straightforward treatment course. At the other extreme are those who function at the level of disabled, chronically and persistently ill psychiatric patients. They may have multiple co-morbidities, such as severe substance abuse, eating disorders, organic mental disorders, mood disorders, and/or multiple medical problems. Another severely impaired subgroup may be involved in violent or criminal subgroups and/or may be abusive to their own children or violent towards others. A third group is intermediate between these extremes with varying levels of co-morbidity, psychosocial dysfunction, and interpersonal pathology. Accordingly, treatment planning for DID patients should take account of the motivation and personal resources of the patient to engage in an intensive, demanding psychotherapy focused on major life change.
Over the past two decades, the consensus of experts is that complex trauma-related disorders – including DID – are most appropriately treated with a phase or stage oriented approach. The most common structure for this is a treatment consisting of three phases or stages:
(1) safety, stabilization and symptom reduction,
(2) working directly and in depth with traumatic memories, and
(3) identity integration and rehabilitation.
(cf. Brown, Scheflin, & Hammond, 1998; Chu, 1998; Courtois, 1999; Herman, 1992a; Kluft, 1993a; Steele, Van der Hart, & Nijenhuis, 2001, 2004; Van der Hart, Van der Kolk, & Boon, 1998). The writings of Kluft (1993a) and Steele, Van der Hart and Nijenhuis (2005), among others, address many of the specific issues in phase oriented DID treatment and other dissociative disorders. Phase oriented treatment of dissociative disorders is not a new idea. Pierre Janet advocated a similar phase oriented treatment for dissociative disorders beginning in the late 19 th century (cf. Brown, Scheflin, & Hammond, 1998; Van der Hart, Brown, & Van der Kolk, 1989).
Complex PTSD (Herman, 1992b, 1993; Van der Kolk, Roth, Pelcovitz, & Mandel, 1993) is a construct that fits many DID patients (Courtois, 2004). These patients usually have had repeated traumas typically beginning in childhood and spanning several developmental periods. In addition to PTSD symptoms, they have major difficulties with affect regulation, dissociation, and body image, the latter often manifesting itself as eating disorders, self-destructive attacks on the body, and somatization. They may have substantial relational pathologies including major problems with trust and enmeshment in violent or abusive relationships. They often view the world as dangerous and traumatizing, and see themselves as shameful, damaged, and responsible for their own traumatization. These patients commonly have significant problems with self-destructiveness and are often refractory to standard treatments (Herman, 1993; Van der Kolk et al., 1996). Accordingly, treatment for complex PTSD is a long-term, multi-modal, relatively eclectic psychotherapy designed to address the multitude of clinical difficulties with which these patients struggle.
A detailed description of phase oriented treatment for DID is beyond the scope of these Guidelines (consult references in the previous paragraphs). However, it is common for DID patients to initially present to treatment with suicidal, self-harming, and/or self-destructive behaviors that may lead to a continual and/or repetitive series of crises. These are often related to serious psychosocial stressors that have undermined the patient’s stability. They may have co-morbid conditions including PTSD, substance use disorders, eating disorders, somatoform disorders, anxiety and mood disorders (primarily in the depression spectrum), personality disorders (often a mixed personality disorder with avoidant, obsessive, borderline, and/or narcissistic traits), and, less commonly, organic mental disorders, and/or psychotic disorders.
The phases of treatment are to some extent a heuristic construction. The phases describe the dominant orientation of the therapeutic work during a particular stage to assist the DID patient develop greater overall life adaptation, safety and stability while, when indicated, working on traumatic material. For instance, in the stabilization phase, treatment may focus at times on traumatic experiences, but in a more distanced and cognitive way to help the patient more effectively separate past events from contemporary experiences. In the middle phase of treatment, stabilization and symptom management is often necessary to prevent patients from becoming overwhelmed by the nature of the work on trauma and to help stabilize them if they do. A focus on rehabilitation and better overall life adaptation is essential throughout any form of mental health treatment and should occur in each phase of treatment.
In the initial stages of treatment, emphasis should be placed on establishing a therapeutic alliance and educating patients about their difficulties and about the process of treatment. Maintaining a sound treatment frame in the context of a therapeutic holding environment is critical to establishing a stable therapy that potentially results in a successful outcome. As with other patients with complex PTSD, initial work with DID patients must address any major self-destructive behaviors and other issues that may potentially jeopardize patients’ physical or psychological safety.
Research studies of DID patients suggest that a history of suicide attempts and self-injurious behavior is found in a very high percentage of DID patients. No studies have been done to compare suicide risk between DID and other patient groups. However, since the overwhelming majority of DID patients meet DSM-IV-TR diagnostic criteria for current or past PTSD, data on suicide in PTSD are relevant to suicide risk in DID. The National Comorbidity Study (Kessler, 2000) found that a diagnosis of PTSD is associated with a six-fold increase in the likelihood of an initial suicide attempt, an odds ratio higher than that for any other anxiety disorder and about half that for mood disorders. Further, individuals with PTSD have an approximately equal odds ratio for making a suicide plan or impulsive suicide attempt compared with those with Major Depression. Of course, many DID patients also meet diagnostic DSM-IV-TR criteria for Major Depressive Disorder, which may increase the odds that DID patients will make suicide attempts.
Recent studies have also shown that childhood maltreatment in and of itself is associated with an increased risk of suicidal behavior (Arnow, 2004). In these studies, suicidal risk was increased in a dose dependent manner by multiple abuse events and by multiple forms of abuse: physical, sexual, emotional abuse and neglect.
Studies have also found that childhood sexual abuse is strongly associated with a history of self-injury, eating disorders, and other forms of parasuicidal behavior (Van der Kolk, Perry, & Herman, 1991). Higher rates of childhood sexual abuse and other forms of maltreatment also increase risk of suicide attempts substantially, with rates increasing with more forms of abuse experienced (Arnow, 2004). The high rates of severe, repetitive childhood sexual abuse and other forms of maltreatment in the histories of DID patients, strongly indicate that self-injurious and self-damaging behavior will be common as well in this population.
These data and other studies suggest that suicidal and/or self-destructive behaviors may be exceptionally common among DID patients (Putnam et al., 1986; Ross & Norton, 1989a). While studies have not addressed the prevalence of completed suicide in DID patients, many Guidelines Task Force members reported knowing of cases of DID patients who have ultimately committed suicide. Accordingly, clinicians treating DID must be aware and take seriously the potential for suicide and/or self-destructiveness in this patient population.
DID patients usually give a history of being abused or having their safety disregarded throughout their early lives. They tend to reenact these paradigms in their lives, venting their aggression, shame, fear, horror and other overwhelming affects onto themselves through self-destructive behaviors. Accordingly, one major cornerstone of the treatment of the DID patient is to help the patient achieve safety from behaviors that may make them dangerous to themselves, dangerous to others (especially to their minor children), and/or vulnerable to exploitation or violence by others. Many DID patients will exhibit these sorts of problems when they present for treatment or may reveal that they are occurring as treatment progresses. Most commonly, these problems include suicidal and/or parasuicidal behaviors, alcohol and/or substance abuse, enmeshment in violent or exploitative relationships, eating disorder symptoms, violent or aggressive behavior toward others, lack of food, clothing or shelter, and high-risk behaviors that may subject them to danger. The latter can include fugues or wandering in dangerous neighborhoods or environments, driving recklessly, engaging in unsafe sexual practices, and/or failure to attend to medical problems, among others. Safety issues may recur during treatment and need to be addressed vigorously and thoroughly.
Without successful resolution of the myriad safety problems that the DID patient brings to therapy, little will be accomplished overall in the treatment. The clinician should keep a very high index of suspicion for covert unsafe behaviors and address them and address them vigorously, being sure to bring their management into the forefront of treatment. A careful history often shows that the DID patient’s unsafe behaviors may have gone on for years. These behaviors often can best be understood as self-regulatory, even self-soothing, adaptive strategies that are logically related to the patient’s history of traumatic experiences and attempts to cope with these. Accordingly, they are usually best worked with as adaptations to be shaped in a different direction, rather than “bad” behaviors to be eliminated. At the same time, the therapeutic alliance and the sense of a “holding environment” for the DID patient may be facilitated if the therapist strongly takes a stand for “non-abusive values” to self or others (Loewenstein, 1993).
Safety issues should be addressed in a comprehensive and direct manner. Other treatment issues may need to be put on hold until safety is established. Interventions should include: 1) education about the necessity for safety for the treatment to be successful; 2) identifying alternate identities who act unsafely and/or control unsafe behaviors; 3) development of agreements with the latter and with all alternate identities to help the patient delay acting on unsafe impulses and/or to use alternative strategies for management of problems (see below); 4) cognitive therapy, cognitive-behavioral therapy, and/or dialectical behavior therapy to help the patient address the cognitive distortions and affect dysregulation that frequently underlie unsafe behaviors, and to develop alternative behavioral repertoires to remain safe; 5) use of symptom management strategies such as grounding mechanisms, crisis planning, self-hypnosis and/or medications (see below) to provide alternatives to unsafe behaviors; 6) development of specific treatment plans to manage eating disorder and/or substance use problems that may include referral to specialized treatment programs; 7) involvement of appropriate agencies if the clinician has a reasonable suspicion that the patient is abusive to children or to vulnerable adults or is in danger of acting violently towards another person (following the laws of the jurisdiction in which the clinician practices); 8) helping the patient with appropriate resources for self-protection from domestic violence (following the laws of the jurisdiction in which the clinician practices); and 9) insisting that the patient seek treatment at a more restrictive level of care, including hospitalization, to prevent the patient from harming self or others.
Commonly, as part of the management of the DID patient, the clinician will develop “safety contracts” or “safety agreements” with the patient’s alternate identity system to provide a structure for the patient to cease unsafe behaviors such as suicide attempts or self-mutilation. From both a clinical and medico-legal perspective, these agreements are not a substitute for the clinician’s judgment about the patient’s safety. They must be interpreted in the total context of the patient’s clinical situation and should be reviewed on a regular basis with the patient. The clinician should always insist on more restrictive treatment alternatives if, in the clinician’s clinical judgment, the patient is unsafe, despite the patient’s insistence on the validity of the “contract.”
Safety agreements may be best conceptualized as delaying or temporizing strategies, much as a substance abusing patient might use the AA structure as an alternative to drinking. In general, the clinician obtains assent from all alternate identities that “I will not hurt myself or kill myself, or anyone else external or internally, accidentally or on purpose at any time” (Braun, 1986, p. 12) or something similarly comprehensive. However, clinicians should recognize that no language is free of loopholes, and they should insist that patients comply with the spirit of the agreement. In addition, clinicians should not bear the burden of making a contract with each alternate identity. Instead, the patient should be helped to develop strategies (e.g., using ideomotor responses) to make sure that all alternate identities acknowledge that they are bound by the contract.
Other behaviors may be added into the agreement when necessary such as use of alcohol or drugs, driving recklessly, etc. These agreements are only really effective if the patient has a “safety plan” as an alternative. The safety plan may include a hierarchy of alternative behaviors including: contacting friends, leaving the setting where the patient feels unsafe, using symptom management strategies such as self-hypnosis, grounding and containment techniques, using medications as needed, and, finally, calling the therapist and waiting for a return call and/or going to the emergency department if the patient feels imminently unable to maintain safety.
Patients may more readily participate in time-limited safety agreements, especially early in treatment, as they may be reluctant to give up long-standing self-regulatory behaviors forever. Experienced clinicians generally try to negotiate agreements for several weeks or months, and will not renew these agreements more frequently than session to session (except during crises when they may be renewed during a phone contact to avoid immediate hospitalization). The need for safety agreement renewal more frequently than session to session suggests that the treatment needs to be modified, often by finding a more restrictive setting for the patient such as partial or inpatient hospitalization.
Some patients may experience these agreements as more concrete if they are written down and signed by some or all of the alternate identities. However, the clinician should not have any greater confidence in the reliability in these agreements if they are written and signed. The fundamental issue is the patient’s honest commitment to the agreement and taking personal responsibility for their safety, which may include asking for assistance instead of acting unsafely when overwhelmed. Working on the issue of honesty concerning the safety agreement also may focus on important issues in developing the therapeutic alliance (see below).
Use of safety agreements may help patients to eventually realize that they have greater control over safety issues than previously realized, to help them understand their ambivalence concerning personal safety, and to more effectively mobilize their efforts to control acting unsafely. Frequently, discussion of cessation of unsafe behaviors brings a wealth of crucial material into the therapy concerning the alternate identity system, the patient’s history, transference issue (especially traumatic transference themes), and dominant ideas and beliefs that shape the patient’s behavior.
Most models of phase oriented treatment begin with an initial period of symptom stabilization. The focus in this phase of treatment is the management and control of symptoms rather than exploration of traumatic memories. For example, if the patient has a spontaneous flashback or episode of intrusive recall of trauma during treatment, the therapist should help the patient modulate the intensity of the experience and not encourage detailed discussion of the material. In this phase the clinician would assist the patient with the development of personal and environmental safety, modulation of psychophysiological arousal levels, improved affect tolerance and impulse control, control of posttraumatic and dissociative symptomatology, diagnosis and stabilization of co-morbid disorders, better functioning in daily life, and improved capacity for engaging in mutually supportive relationships.
A variety of treatment interventions and strategies may be needed to assist the patient in stabilizing. These include psychoeducation about trauma-related difficulties; teaching the patient skills in “grounding” from dissociative and posttraumatic symptoms; teaching the patient skills for “containment” of dissociative, posttraumatic, and affective eruptions; helping the patient through cognitive-behavioral techniques to recognize problematic thought patterns and beliefs; and helping the patient with better adaptation to current life stresses and interpersonal problems. Psychopharmacological interventions such as medications for mood, anxiety, posttraumatic, sleep, and thought disorders may be a helpful adjunct to treatment (see section on psychopharmacology, below).
Chronically traumatized individuals, including DID patients, may benefit from periodic “ego-strengthening” interventions such as skill building, soothing images, reaffirming statements, and calming imagery such as “safe places.” When patients spontaneously experience intrusive traumatic imagery, they often benefit from learning strategies that help them delay, contain or control the level of intrusiveness of the traumatic material into their daily functioning. Specific early stabilizing interventions in DID are designed to help with spontaneous and uncontrolled dysfunctional switching of alternate identities, passive influence, amnesia, somatoform symptoms, and similar experiences that are highly disruptive and may be linked to intrusive PTSD symptoms.
Typical interventions include psychoeducation concerning the disorder, techniques to improve internal communication and co-consciousness among alternate identities, and strategies for them to have safe ways of communicating as well as containing their symptoms. Issues of accountability are usually discussed here, including how the patient as an individual is held accountable for the conduct of all alternate identities, in the external world, in therapy, and within the internal system. The strategies designed to improve internal communication may include techniques to encourage negotiation between the alternate identities, acknowledgement of the importance of all alternate identities, and commitments by all alternate identities for safety from self-harmful and/or suicidal behaviors (see above). In some cases the stabilization plan must include rehabilitation or crisis centers, specialized substance abuse or eating disorder programs, social services involvement such as child protection agencies, and residential, partial hospital and/or inpatient treatment.
As described in a prior section, DID patients have a broad range of ego-strength, commitment to treatment, social supports, life stresses, economic resources, and other factors that may make them more or less able to undertake a demanding, change-oriented treatment. Accordingly, many patients may continue in Phase 1 treatment for long periods of time – sometimes even for their entire treatment course. These patients may make considerable improvements in safety and overall functioning, but may not be able to participate in an extensive, emotionally intense, detailed exegesis of their trauma history. The focus of treatment for chronically low-functioning patients should remain on stabilization, crisis management and symptom reduction, rather than on the details of traumatic memories and fusion of alternate identities. Several factors have been described that may influence a decision towards maintaining such a focus including severe attachment problems, minimal ego strength and coping capacity, ongoing enmeshment with perpetrators, severe DSM-IV-TR Axis II pathology, significant medical problems, and ongoing substance abuse and dependency (cf. Boon, 1997, Kluft, 1997). In general, patients should only move into Phase 2 work if they are adequately stabilized in the initial phase, and provide adequate informed consent (see below) for entry into the next phase.
As part of learning about the nature of their disorder, DID patients must begin to understand, accept and access the alternate identities that play an active role in their current life. Clinicians must accept that successful treatment of DID almost always requires interacting and communicating with the alternate identities in some way. Early in the treatment, therapists and patients must establish safe and controlled ways of working with the alternate identities that will eventually lead to co-consciousness and greater integration. In order to work with alternate identities, they must be accessed directly or indirectly (see below). A complete discussion of this is beyond the scope of these Guidelines; see Putnam, 1989, Ross, 1997, Kluft, 2001, and Kluft and Fine, 1993 for more extensive discussions of treatment of DID.
Some alternate identities may insist that they do not inhabit “the body” of the host identity. In this form of “delusional separateness,” they may insist that suicide or self injury has no effect on themselves, only on other identities. Accordingly, severe safety problems can result from this issue. It is important to challenge this extreme form of dissociative denial directly. However, in some cases, it may take many sessions to erode delusional separateness, even with the alternate identities acknowledging the contradiction of finding themselves in the body that they deny inhabiting.
Alternate identities can be accessed directly, e.g., “I need to talk to the one(s) who went to Atlantic City last night and had unsafe sex.” On the other hand, experienced clinicians usually develop a repertoire of skills to access alters more indirectly. For example, the therapist may suggest that the alternate identities engage in inner conversations with one another, may use non-verbal responding such as ideomotor signals in response to questions (Hammond, 1990), or may insist that “everybody listen” when important matters are being discussed. Often the clinician can “talk through” or “talk over” the alternate identity presenting in the therapy session to communicate with other alternate identities relevant to the current clinical issues. The patient can be asked to “listen inwardly” to hear what the other alternate identities have to say. The latter techniques are helpful since they may allow rapid discussion of material among alters and may attenuate the disruptive effects of frequent switching that can occur when many different identities try to communicate with the therapist. Implicitly, these techniques break down dissociative barriers by encouraging sharing of thoughts, feelings, and perspectives that are subjectively sequestered within alternate identities. Hypnotic induction may facilitate the emergence of alternate identities (see section on hypnotic techniques) especially during safety crises, but, under ordinary circumstances, hypnosis is not necessary to allow alternate identities to emerge in the treatment – firm and persistent encouragement alone usually succeeds.
It is often helpful to generate an ongoing “map” or “roster” of the patient’s current view of the alternate identity system. This can help in understanding the subjective relationship among alternate identities as well as to indicate where hidden identities may exist that influence or control symptoms or safety issues. Some clinicians recurrently schedule a kind of “roll call,” calling the names of each known alternate identity and getting their assent that they are “present” and “listening” (Kluft, 1993a). In “mapping” a patient’s system, clinicians should not try to identify or elicit identities solely for the sake of mapping. It can be counter-therapeutic and potentially destabilizing to ask patients to reveal parts of themselves before they are psychologically prepared to do so. In general, alternate identities should be elicited as they appear naturalistically or if they have relevance to current clinical issues. On the other hand, in situations involving significant safety problems, repeated acting out by the patient, and/or at times of therapeutic impasse, it can be important to directly elicit alternate identities, previously known or not, that are experienced as causing these difficulties.
Many systems have been developed to attempt to make a typology of alternate identities. Discussion of this is beyond the scope of these Guidelines. All alternate identities should be conceptualized as attempts to solve life problems and as adaptations to disturbances in the patient’s early development and subsequent life. Accordingly, experienced clinicians working with DID commonly actively seek to access and safely engage alternate identities who embody intense mistrust, negative affects, or urges to harm the person’s body, and may be experienced as malevolent, rageful, violent, and/or internally abusive. With proper treatment, these alternate identities can be transformed and become the source of important strengths and resiliency during treatment (Watkins & Watkins, 1988).
Patients with a history of interpersonal trauma in childhood often have major difficulties with trust. This frequently manifests itself towards their therapists and can play out in a variety of complex transference manifestations (Brown, Scheflin, & Hammond, 1998; Davies & Frawley, 1997; Pearlman & Saakvitne, 1995). DID patients frequently report extensive childhood histories of traumatic experiences, usually involving maltreatment and/or neglect by family members, caretakers and others in positions of authority or trust. Accordingly, clinicians should never underestimate the difficulties these patients may experience in establishing a therapeutic alliance. In addition, treatment may begin to erode dissociative barriers and defenses leading to greater intrusion of traumatic memories. This may engender additional fears of loss of control due to increased awareness of extreme affects and disturbing cognitions. Feeling vulnerable, patients may manifest more difficulty with trust, fearing that they will be abused or manipulated as they were in childhood. Such “traumatic transference” reactivity may be intense among various alternate identities either overtly or covertly (e.g., the host appears to trust the therapist, while other identities feel vulnerable and sabotage the therapy). Finally, patients with a history of child abuse – especially incest – may be at particular risk of sexual exploitation by authority figures, including mental health professionals (Kluft, 1990). DID patients with a history of therapist abuse usually require an even longer time to develop a sense of safety in treatment, let alone, trust.
The clinician should attend to issues of the therapeutic alliance from the beginning of treatment. The clinician should be actively aware of the potential difficulties that building a therapeutic relationships can engender for the DID patient. It is helpful for the clinician to structure sessions to include education about the nature of DID and trauma treatment, the intense discomfort that can be engendered during treatment, and to anticipate and make explicit traumatic transference issues, particularly negative transferences. The clinician’s judicious use of containment and grounding techniques in sessions to avert crises also can help begin to build a therapeutic alliance with the DID patient. Patients may alternate from pressures to “open everything up to get out all the memories” to intense phobic avoidance of trauma issues. The clinician can foster a realistic therapeutic alliance by structuring the treatment to balance both of these pressures and to find a pragmatic balance between attenuation of trauma material and working more in depth on trauma material. One clinical study suggests that DID patients who are able to improve their therapeutic alliance in a real way may have a better and more rapid positive outcome than DID patients who are not able to do so (Kluft, 1994).
Effective therapy for DID usually requires a therapist who is engaged and who actively structures the treatment by anticipating difficulties and by having a clear plan to help the patient through the stages of treatment. A gradual fostering of a real therapeutic alliance with the DID patient will often occur as the clinician helps the patient pace the therapeutic work, learn skills for mastering symptoms and crises, separate the traumatic past from the present, and change PTSD and DID based cognitive distortions. The therapist’s active insistence on safety and recovery usually is a contrast for the patient with persons from the past who remained passive or unconcerned about the DID patients safety and well-being.
In this phase of treatment, the focus turns to working with the DID patient’s memories of traumatic experiences. It is generally accepted among experienced clinicians that effective work in this phase involves remembering, tolerating, and integrating overwhelming past events. Optimally in this stage, work on traumatic memories can be carefully planned out and scheduled: which memories will be the focus, at what level of intensity, which alters will participate, how to maintain safety during the work, and procedures to contain material if the work becomes too intense. Patients benefit when therapists help them use planning, information, exploration, and titration strategies (cf. Fine, 1991; Kluft, 2001) to develop a sense of control over the emergence of traumatic material. Specific interventions for DID patients involve working with alternate identities that experience themselves as holding the traumatic memories. These interventions help broaden the patient’s range of emotions and affects across alternate identities, and assist the patient as a whole with tolerating the affects associated with the trauma such as shame, horror, terror, rage, helplessness, confusion, anger and grief.
The patient and therapist may elect to work with memories spontaneously if they emerge in therapy, assuming that there is adequate time in session, and that the patient can work on memory material without significant life disruptions. Accordingly, as the various elements of a traumatic memory emerge, they are explored, rather than being re-dissociated or rapidly contained. Over time, and often with repeated iterations, the material in these memories is transformed from traumatic memory to what is generally termed narrative memory (cf. Brown, Scheflin, & Hammond, 1998 for a comprehensive review of trauma and memory in treatment). In addition to abreaction – the intensive discharge of emotions and tensions related to the trauma, the mechanism of change is one of repeatedly re-accessing and re-associating the fragmented and dissociated elements of traumatic memories (Van der Hart & Brown, 1992).
Active work on traumatic memories ultimately aims to bring together most dissociated aspects of traumatic experience: the affects associated with the trauma, the physiological and somatic representations of the experience, the sequence of events that occurred, to the extent that they can be remembered or reconstructed, with adult cognitive awareness and understanding of the role of self and others in the events (Braun, 1988; Brown, Scheflin, & Hammond, 1998; Chu, 1998).
As the patient re-experiences the events, attempts can be made to “detoxify” them by placing them in a more understandable context, and by finding alternative meanings for them. The patient’s unrealistic views of him/herself in the context of trauma (“It’s all my fault,” “I asked for it,” “I wanted it,” “I should have stopped it,” “I made it happen,” etc.) can be systematically challenged using the patient’s adult observing ego to understand what really occurred. Also, during this stage, the detailed recall of the past commonly leads to work on resolution of ambivalence and conflict concerning old or current relationships with family members and significant others and the latter’s remembered roles in traumatic experiences. In addition, during this stage, DID patients may grapple with their fear and/or ambivalence about change and recovery ( Chu , 1998).
The process of Phase 2 work allows the patient to gain a sense of control over the experiences and their reactions to them, and to build a better understanding of his/her personal history and sense of self. In addition, DID patients become able to recall the traumatic experiences across alternate identities, especially those who were amnestic or without emotional response to them. Some authors have used the term “synthesis” for this process (Van der Hart, Steele, Boon, & Brown, 1993). Synthesis can be described as a controlled therapeutic process designed to assist alternate identities that experience themselves as “holding” traumatic memories to share these with other alternates who experience themselves as being unaware of this material or do not regard it as part of their autobiographical memory .
However, even with careful therapeutic planning, destabilization can occur during this stage of treatment. This may require a return to focus on Phase 1 issues such as stabilization, internal communication, containment, and symptom management, as well as work on resistance and reluctance among alternate identities to continue trauma work. In addition trauma-based cognitive distortions and/or transference reactivity also may interfere with work on memories and will need to be systematically addressed. In some cases, destabilization may involve safety problems that require more restrictive levels of care such as partial hospital or inpatient treatment.
In this phase of treatment, work on traumatic material may need to occur repeatedly at many different levels of affective intensity to help the patient more fully integrate the totality of the recalled traumatic experiences. As this occurs, the alternate identities may experience themselves as less and less separate and distinct. Spontaneous and/or facilitated fusions among alternate identities may occur as well as this process unfolds. Facilitated fusions often involve “fusion rituals.” These therapeutic ceremonies usually involve imagery or hypnosis that “… are perceived by some … patients as crucial rites of passage from the subjective sense of dividedness to the subjective sense of unity …” (Kluft, 1986, quoted in Kluft, 1993a, p. 119). The patient’s experience is that alternate identities join together with an image of joining together or becoming unified “… [These rituals] merely formalize the subjective experience of the work that therapy has already accomplished …” (Kluft, 1993a, p. 120).
Fusion rituals are often misunderstood. Many clinicians attempt to press for fusion before it is appropriate to do so. Fusion rituals are useful when previous psychotherapeutic work has caused a separateness to no longer serve a meaningful function for the patient’s intrapsychic and environmental adaptation and when the patient is no longer narcissistically invested in maintaining the particular separateness. Clinicians should not attempt to press for fusion before the patient is clinically ready for this. Premature attempts at fusion may cause significant distress for the DID patient or, alternatively, a superficial compliance with the alternate identities attempting to please the therapist by seeming to disappear. Premature fusion attempts can also occur when the therapist and patient collude to avoid particularly difficult therapy material. On the other hand, ongoing, dispassionate education of the DID patient about the manifold issues that the question of unification brings up may be helpful during any stage of DID treatment.
In the third phase of DID treatment, patients make additional gains in internal coordination and integration, and usually begin to achieve a more solid and stable sense of who they are and how they relate to others and to the outside world. In this phase, DID patients may continue to fuse alternate identities and improve their functioning in a more and more unified manner. They may also need to revisit the trauma history from a more “unified” perspective. As patients become less fragmented, they usually develop a greater sense of calm, resilience, and internal peace. They may make more coherent sense of their past history and deal more effectively with current problems. The patient shifts focus from the traumatic past and towards living better in the present. Work on loss, grief, and mourning may be profound in this stage as the patient grapples with the realistic perception of the many losses that the traumatic past has engendered. On the other hand, the patient may begin to have less focus on the past traumas, directing his/her energy to living better in the present. With a greater level of integration, the patient may be more able to sort out traumatic “memories” and decide that some of these may be more related to subjective symbolism concerning the past, or to dissociative experiences that seemed “real” at the time but did not occur in objective reality.
Many of the tasks of late phase treatment of DID are similar to the those of non-traumatized patients who function well, but are experiencing emotional, social, or vocational problems. However, the unifying DID patient may need specific coaching about dealing with everyday life problems in a non-dissociative manner. Similarly, the patient may also need help in tolerating everyday stresses, petty emotions and disappointments as a routine part of human existence.
The primary treatment modality for DID is usually individual outpatient psychotherapy. Frequency of sessions and duration of treatment may depend on a number of variables including the patient’s characteristics, the abilities and preferences of the clinician, and external factors such as insurance reimbursement and the availability of skilled therapists. As described in the preceding sections, there is a broad spectrum of DID patients with respect to motivation for treatment, resources for treatment, and co-morbidities that may affect the course of treatment. As with other patients with complex posttraumatic disorders, treatment for DID patients generally is long-term, usually requiring years, not weeks or months of treatment.
Frequency of sessions provided may vary depending on a variety of factors including the goals of the treatment and the patient’s functional status and stability. The minimum frequency of sessions for most DID patients with a therapist of average skill and experience is once or twice a week, with many experts in the field recommending twice a week. Long-term supportive Phase 1 treatments usually occur once or twice per week depending on the patient’s ability to manage symptoms and maintain themselves at an outpatient level of care. Task Force members differ about frequency of sessions for change oriented treatments involving intensive Phase 2 and Phase 3 therapy work. Some opine that many patients will need to be seen at least two times per week and often more frequently during these phases to provide sufficient intensity for the trauma work and to keep a focus on everyday events in the patient’s life. Others believe that once to twice weekly therapy may be sufficient to accomplish the work of these phases in selected patients
Some contributors to these Guidelines believe there is a potential danger for dysfunctional dependence that can occur in some patients with more intensive therapy, especially in patients prone to regression. Other contributors suggest that the type of treatment, not its frequency, is the critical variable in the development of a treatment impasse due to regression. For example, some highly unstable, chronic patients can be helped to not regress by the stabilizing effect of a highly structured and supportive treatment that occurs several times per week over many years.
In certain circumstances, a greater frequency of sessions (up to three or more per week) can be scheduled on a time-limited basis to aid the patient in maintaining the highest possible level of adaptive behavior and/or (as an alternative to hospitalization) in containing self-destructive and/or severely dysfunctional behavior. For patients newly discharged from inpatient treatment, a period of sessions at a greater frequency may sometimes be necessary to help the patient make the adjustment from the high frequency of sessions and/or greater level of interpersonal support provided in many inpatient programs. Very frequent outpatient sessions should be generally limited to brief periods for re-stabilization, and regression should be minimized by emphasizing the need for the patient to be appropriately independent and to use the most mature coping skills possible.
While the usual 45-50 minute session remains the norm for most therapists, some therapists have found extended sessions useful. DID patients have sometimes benefited from two 75-90 minute sessions or one extended session and one 45-50 minute session per week. Specific clinical situations have led to the practice of occasionally extending sessions beyond their usual length, e.g., for preplanned trauma memory processing. In all clinical situations, therapists need to attempt to help patients to reorient themselves to external reality and cease processing of traumatic memories well before the scheduled end of therapy sessions, although they can only influence, but never control, the patient’s ability to reorient to the present. Repeated difficulties with grounding the patient in current reality at the end of sessions necessitates that the therapist make this issue the focus at the beginning of a subsequent session to help understand the causative factors and to suggest interventions (e.g., letting the patient know some minutes before the end of the session to initiate the process of re-orientation to leaving therapy).
There is a divergence of opinion concerning very lengthy sessions (e.g., sessions longer than 90 minutes), with some experienced clinicians doubting if they are ever required, and others finding them useful for specific purposes. If used, they should be scheduled, structured, and have a specific focus such as completion of processing of traumatic memories and imagery, or administration of a diagnostic battery. They may also be indicated when logistics force the patient to come to the therapist infrequently, but to work intensely while there.
Medication management sessions for the DID patient often require 25-30 minutes or, occasionally, a 45-50 minute session. This is necessary to both discuss medication management and to handle the often complex psychological responses of the DID patient to taking medication. The DID patient’s traumatic transference, posttraumatic cognitive distortions, and DID trance logic may all combine to require more in-depth explanations of the risks and benefits of medications than for other patients. DID patients are often both medication phobic and medication seeking, the latter often to avoid painful psychotherapeutic issues.
Initial medication evaluation sessions may require one or more 45-50 minute sessions to take an adequate history and to educate the patient and initiate a medication trial. Some psychiatrists will schedule a single initial 75-90 minute session to accomplish this purpose. Subsequent sessions should be handled with the same frequency as for non-DID patients. More frequent visits may be needed when medications are being initiated or significantly adjusted, especially if the patient is unstable. Once stabilized on medications, the patient usually can be seen once per month, or even less often for stable patients on long-term medication regimens. As discussed below, it is vital for the medication prescriber and the primary psychotherapist to have clearly defined roles and boundaries to assure that there is only one primary psychotherapist for the patient.
© Copyright 1998-2005. All rights reserved. Contact: Last edited: 01/02/03.