INPATIENT COGNITIVE­ BEHAVIORAL TREATMENT OF EATING DISORDER PATIENTS WITH DISSOCIATIVE DISORDERS

Andrew P. Levin, M,D. Edward Spaustcr, Ph.D.

Andrew P. Levin, M.D. is Unit Chief of the Adult Partial lier by Torem (1986). Hospital Program,Holliswood Hospital andAssistant Clinical Despite these findings, little has been written address­ Professor of Psychiatry, Columbia University. ing treatment techniques for these ~dually diagnosed" patients. Typical treatment for eating disorders has focused Edward Spauster, Ph.D. is Clinical Director, Holliswood on a cognitive-behavioral therapy (CBT) model (Fairburn, Hospital. eeal., 1991). Although CBT approaches to dissociative patients (DDs) are well-known, (Andreason & Seidel, 1991; For reprints write Andrew P. Levin, M.D., 87-37 Palermo Caddy, 1985; Ross, 1989; Fine 1991a) it is unclear if these Street, Holliswood, N.Y., 11423. techniques are applicable when the patient suflers both dis­ orders. Torem hasdescribedan ego-state therapy (1987; 1989a) ABSTRACT which combines hypnoanalytic and behavioral techniques to reduce disordered eating. Although not specifically Although several investigations have noted an association between focused on the CBT modd, Torem's treatment paradigm eating disorders and dissociative disorders, little work has addressed includesseveral ofthe basic features ofcognitive-behavioral tlu: treatment ofpatients with both conditions. As an inpatient ser­ treatment in its overall scheme. vice focused on severely-ill eating disorder patients, il became nec­ As an inpatient service initially focused on severely-ill' essary to diagnose and treat roncomitant dissociative disorders. We EDs, our treatment program was forced to respond to the describe a cognitive-behavioral inpatient program developed and dissociative pathol0t.'"Y uncovered in this patientpopulation. specifically adapted to treat eating disorder patients with dissocia­ Tn this paper we describe an inpatient CBT program devel­ tive disorders. Patients were identified by self-report measures (the oped to treat the ~dually diagnosed" patient with an eating Eating Disorder Inventory and the Dissociative Experiences Scale) disorder and dissociative pathology. and clinical interoiews. Specific eatingsymptoms linked to post-trau­ Tn their descriptions ofCBT techniques with dissociative matic stress or conversion disorders were approached with stimulus and multiple personality disorder (MPD) patients, Ross control and hierarchical desensitization. individualandgroup cog­ (1989) and later Fine (1991a) highlighted the uncovering nitive thn-apy as well as diary techniques addressed irrational beliefs and correction ofcognitivedistortions. Thesecognitive restruc­ {()mrrwn to both disorders. Skill enhancement through relaxation turing strategies typically focused on issues such as respon­ training, anxiety management, and anger management, initially sibility for past abuse but neither author has described the tailored for the eating disorder patient, required adaptationJor dis­ use of specific techniques such as logging or rehearsal. sociativepatlwlogy. Addressingboth disorders reduced eatingpathol­ Caddy (1985) provideda case report ofCBT with an MPD ogy whereas inattention to dissociative symptoms led to continued including goal-setting, desensitization, self-management, disturbed eating and purging. Identification and treatment ojdis­ and relaxation exercises. Caddy appears to have employed sociative pathology may improve treatment outcome in the treatment­ unmodified CBT techniques developed foranxiety disorders resistant eating disorder patient. because panicandphobicsymptomsdominated in this patient and dissociative episodes did not disrupt the treaunent. It INTRODUCTION is unclear if these techniques would be manageable in a more severelydissociative patient. A briefreportbyAndreason Several recent'reports have focused on the association and Seidel (1991) described the usc of CBT, goal-setting, between eating disorders and sexual abuse (Hall, Tice, and behavionll limits in three patients with MPD, but was Beresford, Wooley, & Hall, 1989; Palmer, Oppenheimer, not specific about techniques that were employed. Dignon, Chaloner, & Howells, 1990) and, more specifical­ Tn addition to the small CBT literature for patients diag­ ly, between dissociative disorders (DDs) and eating disor­ nosed with dissociative disorders, a much larger literature ders (EDs) (Torem, 1990; Demitrack, Putnam, Brewerton, has described CBT techniques in post-traumatic stress dis­ Brandt, & Gold, 1990; Goodwin, Cheeves, & Connell, 1992). order (PTSD) , e.g., nlpe victims, warvetenlns, ordisaster vic­ McCallum, Lock, Kulla, Rorty, and Wetzel (1992) found a tims (see Saigh, 1992), and borderline personality disorder 29% incidence of dissociative disorders in a sample of 38 patients who were abused as children (Linehan, 1993). women with eatingdisorders. Further, McCallum eta\. (1992), Although PTSD is designated an anxiely disorder in DSM-//J­ reported a temporal relationship between dissociative phe­ R, several authors suggest it represents a form of dissocia­ nomena and eating behaviors, an association described ear- tive pathology (Braun, 1984). Conversely, dissociative dis-

178 I)]SSOClAlI0~. Vol. \ II, 11:0. 3, September 1994 LEVINjSPAUSTER ,•

on:l.erssuch asMPO have beenconceptualizedas PTSD (Spiegel, ation by an attending psychiatrist who develops diagnoses 1984), and PTSD symptoms are usually prominent in the in collaboration with otherteam members. In addition,each MPO patient (Lowenstein. 1991). Borderline perwnality patient receives a medical evaluation, the Eating Disorders pathology hasalsobeen recastasPTSD (Herman. 1992). CBT Inventory (Gamer,Olmsted,& Polivy, 1983),the Dissociative techniques utilized for PTSO patienrssuch as relaxation, desen­ Experiences Scale (Bernstein & PuUlam, 1986), and when siti7.ation, and as,.<;ertiveness trainingshould be applicable to appropriate, hypnotic evaluation ofdissociative symptoms. the complex dissociative patient. Our experience suggests CBT modalities are administered in group and individual­ that these techniques can be effective but require modifi­ ized formats. Specific groups include problem solving strate­ cation for the complex dissociative patient with an eating gies with daily behavioral goal-setting, relaxation training, disorder. hierarchical desensitization, cognitive restructuring, impulse Commonalities in cognitive and behavioral features of control strategies, and skill development for assertiveness eating disorder and dissociative disorder patients also pre­ and anger-management. In addition to CBT, treatment dict applicability of similar techniques. Both groups expe­ modalities include pharmacothentpy, psychosocial evalua­ rience affective dysregulation including anger, depression, tion and briefintervention,individual psychotherapy3-4/wk, and anxiery, as well as the impulsive behaviors which follow, conventionalgroup therapy, hypnotherapy, andTwclveStep e.g., self-mutilation, substance abuse, and binge/purge Alcoholics Anonymous, Narcotic Anonymous, and/or cycles. McCallum et al., (1992) noted that self-mutilation OvereatersAnonymous as appropriate andas tolerated. During was temporally related to dissociative experiences in their the period 1990 • 1993, the average length of stay on the sample of 38 eating-disordered women. Affective dysregu­ BDU was 23 days, but the "dually diagnosed" Ells tended to lation precipitating emotional flooding in both groups stay 2842 days. requires an overall focus on pacing, as described by Fine (l992b) in work with MPDs. Cognitive distortions of over· GOAL SETTING AND THE BECK LOG responsibility, perfectionism, and catastrophi:ting are well described in EDs (See Carner & Garfinkel, 1985) and DDs The techniques of goal-setting in a CBT milieu as (Fine, 199Ia). Skill deficits in time management, self-regu­ described by Levendusky and Berglas (1985) form the orga­ lation, andinterpersonalfunctioningfrequently preventprop­ nizing framework of the BDU program. Patients distill pre· er rolc functioning in these two groups and exacerbate eat­ senting symptoms into a a Problem/Goal/Method frame­ ing and dissociative pathology. work during the Grstthrec to five days ofhospitalization in Recognition ofthese common themes in cognitive and daily group therapy and individual sessions with their ther­ behavioral function enables bom the patientand treatment apists and primary nurses. Problems and goals are stated team to focus on specific goals for inpatient treatment and behaviorally; methodsentailshort-term techniques to be uti­ todeveloplong-range targetsforoutpatientcare.At thesame lized during inpatient treatment and long-tenn strategies time, response patterns and cognitive distortions differ in for aftercare. Examples ofgoals for bulimic and dissociative sevenll ways between uncomplicated EDs and "dually diag­ symptoms are included in Appendix). nosed" eatingdisorder patients. Failure to identifythesedif­ Thegoal-setting technique providesa specificdocument ferences may seriously undermine the treatment of the ED for the patient and therapist to utilize and amend through­ patient with dissociation. outlhe therapy. This represents a concretizationofthe prob­ lem idemification and paLiemeducation described inTorem THE INPATIENT SETIING (198gb) and Fine (1991a) thal arc the foundation for all CBTwork (Beck, Rush, Shaw, & Emery, 1979). By serving as The Behavioral Disorder Unit (fiDU) is a 25-bed locked a roadmap for the treatment, the Problem/Goal/Method inpatient unit u'eating adults ages 18-65 in a lOG-bed pri­ document pennits successiveapproachesto the targetbehav­ vate, for-profit psychiatric hospital. Patien IS are approximately iors, a process of behavioral "shaping" regarded by Linehan 4:1 female, 80% with a history ofchild abuse, 60% with dis­ (1993) as critical for the patient with poor self-regulation S(>ciativedisorder (10% MPD, 20%dissociativedisorderNOS, and impuLsh.ity. Theshapingapproach blendswell with Fine's 30% post-traumaticstressdisorder), 60%with substance abuse (1991a) overall emphasis on "pacing" in the treatmenl of (40% active, 20% in recovery for at least 3 months). and the dissociative patient. 50% with eating disorders. Approxjmately 50% of EDs meet Most EDs with DDs are fumiliarwith the behavioral man­ currentcriteria for bulimia nen'osa, 10% anorexia nelVosa, ifestations oftheir ED but only partially aware ofdissociative and 40% eating disorder NOS with most fulfilling the pro­ symptoms. Behavioral descriptionsofdissociative symptoms posed binge eating disorder criteria (Spit,er et aI., 1993). such as flashbacks or amnesias de-mystify these experiences ApprOXimately 50% of EDs are concomitantly diagnosed and lay the foundation for treatment. Patients are taught with a dissociative disorder. Taken togeLher, 75% ofpatients Beck's logging format (Beck et aI., 1979) to track admitLcd to the unit meet criteria for major depressive dis­ SiLUation/Thoughl/Feeling/Behaviorfor key symptoms. We ~rder, wi th approximately halfofthosewith co-existing depres­ have found the log to be useful in specifically elucidating SIon presenting suicidal ideation or behavior as well as self­ . experiencesidentiGed in an ongoing diary (Torem, 1989a). mutilation. AU patients receive a comprehensive psychiatric evalu- ...... J_L- 179_ -- CRT WITH DISSOCIATIVE EATING DISORDERS

RELAXATION TECHNIQUES Once this framework is established, patientsare encour­ aged to record dissociative experiences in diaries and apply All patients on the BDU learn relaxation techniques to the Beck log format to details ofeach experience. The ther­ enhance mastery over disturbing physiological arousal gen­ apist may employ hypnotic reconstruction to clarify events erated by anxiety provoking thoughts. These techniquesare precedingdissociative episodes. Patients then learn ground­ then available for usc in anxiety reduction, impulse control, ing techniques to employwith earlysignsofdissociation such and anger managemenL Instruction in progressive muscle as visual field constriction, fogginess, ordifficulty attending relaxation and calm scene visualization initially occurs in to a currentstimulus. Grounding techniques include main­ groupwilh individualized reinforcementbynursingstaITand tenance ofvisual attention, heightening of body kinesthet­ relaxation tapes. Selected patients arc referred for individ­ ic input (See Torem, 1989b), and focus on mental or writ­ ual biofeedback training. ten puzzles. When successful, these strategies significantly In ourexperience, relaxation techniquesare helpful for enhance the patient'ssenseofmastery and promotegreater eating disorder patients but appear to be problematic for participation in treatment. those with dissociative pathology. Closed-eye visualization exercises often precipitate intrusive recollections, flash­ HIERARCHICAL DESENSITIZATION backs, and frank dissociative episodes in patients with com­ FOR FEARED FOODS plex dissociative disorders. Instructions to "relax~ can trig­ ger memories of similar injunctions from abusers. Beach As part ofour usual treatment for EDs, patients identi­ scenes may precipitate fears ofsexualized exposure or pre­ fy feared foods and relate these to cognitive distortionsabout vious trauma occurring during summer months. their bodiesand nutrition,e.g., bread is highly fattening and As a resull., group participation is carefully monitored to be avoided. Patients musteata prescribeddietafter three and modified. Eligibilityfor relaxation group is assessed dur­ days oforientation and thereby areexposed to feared foods. ing the opening diagnostic evaluation, with severely disso­ Failure to maintainadequateintakeand completion ofmeals ciative patients being excluded ahogether. For those who leads to progressive sanctions involving withdrawal from the participate in group, modified instruction sets permit main­ community, e.g., restriction to room during free socializa­ tenance ofvisual contact and construction ofpersonalized tion time. These restrictions, not imfKlsed during the first scenes. Biofeedback with individual supervision is also more few days of hospitalization, prove more powerful once acceptable to these patients becauseitutilizes open-eye tech­ patients have formed bonds with peers. Most [Os are able niques. to overcome irrational avoidance of foods (e.g., bread or Patients may enhance relaxation scenes by producing pasta) with milieu support. repeated eXfKlsure, education drawings or paintings of their "safe place" in art therapy in nutrition, and the corrective experience ofweightstabi1­ which can be used to practice rela.xation individually. As ity. noted by Torem (1989b), self.hypnotic relaxation exercises Eatingdisordered patienrswith abuse historiesoften avoid are another method ofproviding mastery over arousal. BDU specific foods for reasons unrelated to weight concerns. therapists develop a personalized self-hypnotic exercise to Exploration in psychotherapy, often assisted by hypnoana­ replace progressive muscle relaxation. Audiotaping the lytic techniques (Torem, 1987) in individual therapy, reveals exercise, like the drawing ofthe safe place, gives the patient these foods trigger abuse memories and related responses. easy access to the experience. Patients afe encounlged to For instance, one patient'saversion to vegetables (an unusu­ practice relaxation three to five times per day and review al rood for EDs to avoid) was the result of classical condi­ progress regularly with the therapist and nursing staff. The tioning. During her childhood, conflicts with her parcllts dissociative patient often requires several weeks to develop over properfood consumption often escalated into violcnce a comfortable exercise and employ it on a regular basis. against her. Similarly, classical conditioning with general­ ization is responsible for the aversion to certain white foods MANAGEMENT OF DISSOCIATIVE EPISODES (association: semen). Patients come to recognize this first by cataloguing their aversions through the developmentof Inaddition todissociation precipitated byrelaxation tech­ a "dislike" lisL The dietician and therapist then formulate a niques, apparent "spontaneous~dissociation is common in plan to explore the dislikes with the patienL One approach this patient population and usually disrupts the patient's is to have patients prepare a feared or disliked food as part function. The treatment milieu promotes the formulation of a group activity, sharing feelings with peers and staff. that dissociation is a defensive maneuver triggered by an Following the meal patients participate in a "mood moni­ internal orexternalsignal ofdanger,and not"sfKlntaneousM toring" group which serves as a debriefing. These groups or mysterious. Dissociation is further reframed as a coping are a regular part of the daily routine and receive particu­ skill which was appropriate during abuse when escape was lar emphasis after the weekly meal preparation group. impossible butiscurrent1y no longeradaptive. r-dtients respond Similarly, purgingoften requires reconceplUalization for to these formulations with a sense of relief and increased the dissociative surviVQrofabuse. Whereas for the ED patient alliance with the treatment team. Management ofdissocia­ purging is an anxiety-reduction strategy in response to fears tive experiences thus becomes a clearly articulated goal in of weight gain, in the DO patient purging may represent a the treatment plan. re-cxperiencingofsensationsassociated with the abuse. Many

180 DlSSOCl~TlO:'i, Vol VII. :';0. 3. Scplember 19Q4 I survivors report nausea during or following forced oral sex. learns to recognize these cognitive distortions in family inter­ Vomiting then produced a reduction in the uncomfortable actions. In fact, EOs in general, and those with concomitant arousal (revulsion, fear, gag reflex) and nausea associated histories of abuse, often repeatedly choose to interact with with the abuse. This response is triggered anew whenever family despite resultantfeelings ofworthlessness, anger, and memories of abuse produce similar arousal cues. Patients guiltwhich then initiate disordered eating. To interruptthis who purgein this manneroftendeny specific concernsabout behavioral cycle, we first ask the patient to suspend comaet weight, insteaddescribing purgingas cleansingorrelieffrom with family for a several-day period to uncover the thoughts discomfort. Onceequipped with !.he behavioralformulation and feelings involved in this behavior. The patient usually that forced oral sex produced nausea and revulsion and that requires significant staff support to achieVE only a two- to memories ofabuse revive !.hese feelings, !.he patient is extri­ three-day hiatus in contact. The social worker may also inter­ cated from the sexual implications and connotations ofthe vene with the family (if the patient permits) to explain the abuse. The result is a decrease in shame and powerlessness. temporary decrease in contact, although the rationale given If the specific feelings ofnausea or revulsion exist in a sep­ is minimal, such as "this is routine procedure." Each time arate ego-state or alter then the reframing is accomplished the patient has an urge to talk with a family member, she during individual sessions with that alter. records the situation using the Beck format and utilizes this Following this behavioral reframing, the patient devel­ material for individual and group psychotherapy. Typical opsa hierarchyofthe foods with the therapistand dietician. sequences mightbeasfollows: (1) Situation- hearinganoth­ Exposure to these foods is first done through imagination, er patient talk about her family; (2) Thought-I must keep with relaxation and hypnosis techniques to counter the anx­ in touch with my family or they will be angry at me; (3) iety response. Exposure in vivo begins with very small Feeling-guiltand urges to gorge on desserts; (4) Behavior amounts of the feared food followed by post-exposure - bingeand/orpurge (hopefullyblocked by decreased avail­ debriefing and recording of the experience in an appro­ ability offood and monitoring ofbathrooms in the milieu). priate log or diary. Subsequent exposure proceeds stepwise Other typical cognitions include a need to reaffirm the fam­ up the hierarchy as tolerated, usually taking 10-14 days to ily's affection for her despite their abuse, fear of injuring desensitize the patientto three to five foods. This behavioral the family by withdrawing her attention and thereby pro­ progression provides a specific scheme to implement the voking further abuse, or an idealized perception that the process of"ratification"described in Torem's ego-state hyp­ family is in fact loving and supportive so that she is selfish notherapy (l989a). In our experience an ego-state or alter and unjustified in any anger toward them. The strength of will cooperate with a stepwi~e approach because it permits these cognitions and the resultant anxiety override negative gradual exposure to experiences, thoughts, and feelings which experiencesand preventthe patiemfrommodifying the base­ may trigger memories. line interaction patternwith the family despite revulsion toward them. Prior to instituting this treatment strategy we found COGNITIVE RESTRUCTURING that abused EDs were unable to interrupt disordered eating due to ongoingcognitions and feelings precipitated by fam­ Like other CRT programs for EDs (e.g., Garner & ily comacts. Garfinkel, 1985), the Behavioral Disorders Unit focuses on cognitive distortions about food, body image, perfectionism IMPULSE CONTROL and over-responsibility. Distortioned cognitions of food (e.g., starches cause weight gain) are addressed in nutrition Currentdatasuggest that patients with histories ofabuse education groups and exposure to these foods. Movement are at increased risk for self-destructive and suicidal behav­ groups enhance the patient's ability to experience feelings ior (van der Kolk, Perry, & Herman, 1991) comparedto non­ expressed in body postures and to challenge typical body abused patients. Further, both dissociative symptoms image cognitions. As noted earlier, enhancement of kines­ (Demitrack etal., 1990), anddissociative disorders (McCallum thetic awareness through movement strengthens ground­ eta!., 1992) aresignificantpredictorsofself-destructive behav­ ing skills for patients who dissociate. ior in EDs. As in any milieu treating dissociative patients, In addition, for the ED with a history of abuse, inter­ control ofself-destructive behavior on the BDU assumes top personal interaction with former or current abusers often priorityamong treatmentgoals. Toward this end, the patient triggers self:deprecation and feelings of over-responsibility utilizes the diary and Beck log to identifY cognitive distor­ which, in turn, trigger binge/purgebehaviortosootheaccom­ tions underlying self-destructive behavior, bringing this panying dysphoria. The patient is often unaware ofthe basis material to individual and group psychotherapy. for the cognitive distortions which initiate the disordered Whenself-destructive feelings dominate the patient'sdaily eating pattern. Once ED patients are identified as victims of function, the patient utilizes more intensive self-monitoring abuse on the BDU, they enroll in a psychoeducational group techniques. Self-destructive urges are registered on a 0-10 focused on the cognitive sequelaoftrauma, e.g., over-respon­ Subjective Units ofDistress Scale (SUDS) at frequent inter­ sibility, perfectionism, and catastrophizing (Fine 1991a). vals (15', 30', or 60' as appropriate), with staff reviewing Patients then employ the Beck log to analyze behaviors on these ratings at least twice per shift. Patient failure to per­ the unit that may involve these thought patterns. form monitoring result~ in restriction ofmovementandsocial­ After identifYing these patterns on the unit, the patient ization (e.g., full-time room restriction withoutsocialization,

lSI n - I CBT WITH DISSOCIATIVE EATING DISORDERS

oropenareas reSlriction with nogroup participation). Once ness with family are often limited to boundary strengthen­ the log is in place, a response plan is developed for the spe<:­ ing, such as controlling frequency of contact and content !rum of urge intensities_ Low SUDS scores are countered ofconversationswith abusive £ami Iy members.Similarly,con­ with relaxation exercises. whereas higher scores signal use frontation of the abuser is usually discouraged during the of the unit "safe room," contact with staff, or medication. hospital stay and patients are educated that this may not be The "safe room" is a room with minimal, comfonable fur­ helpful in the future. This contrasts with the general goal nishings where patients can write, draw, or use relaxation offacilitating family expression ofconflict in the treatment techniques. PatienLS must alert staff when they enter and ofuncomplicalCdeatingdisorders (Gamer& Garfinkel, 1985). depart. The room is mhcrwise off-limits, making it always Lastly, because virtually all survivors maintain diSlorted aV'o:lilable for patients in crisis. When self-destructive urges cognitions around responsibility and blame (Fine. 1991 a), arecomplicatedby dissociation the patientcmpioysground­ assertiveness training must occur in a therapeutic context ing techniques in !.he response plan. Intensive self-moni­ which challenges these dislOrtions. OthclWise, patientsexpe­ toring can also be applied to high frequency urges to purge rience skills-training as an indictment of previous respons­ which mayoccur in the dissociative ED patientduring a peri­ es to abusive situations. Victims must learn that passivity dur­ od offrequent and severe flashbacks. ing childhood abuse was an appropriate response and did not reflect complicity or responsibility. Similarly, current SKILLS DEVELOPMENT aggressive responses can be reframed as appropriate given distorted cognitive appraisal of threats from others. Patients with [Os are usually deficient in adult manage­ Reassessment of perceived danger can then fonn the basis ment skills and victims ofabuse are doubly impaired. Both for anger-management. individual andgroup modalities in ourse tting focus on inter­ personal response strategies with specific emphasison iden­ ANGER-MANAGEMENT tification of maladaptive passive and aggressive behaviors. Patients first utilize the Beck log to identify these patterns Anger-management involves the identification of trig­ in interactions on the unit with other patients and staff. For ger situations, the thoughts and feelings they precipitate. example. patientsoften reportdifficulty asking nursingstaff and resultantbehaviors. Patientsare laught typical sequences for assistance when they are feeling anxious or depressed. and common cognitive distortions in group and through Patients will log me situation (e.g., adifficultgroupsession), instructional materials. When anger is a frequent problem and resultant thoughts ('" will never get better"), feelings the patient keepsan anger log to record detailsofthe response (sadness orfear). and behavioral response (withdrawal from and its intensity. The ED patientoftenselectsaneatingbehav­ staffor rejection ofhelp). The patient will then discuss this ior in response to anger whereas the patient with a disso­ sequencewimthe therapistor in AssertivenessTraining Group ciative disorder may utilize withdrawal, frank dissociation or to understand how the response tostaffwas passive oraggres­ self-destructive behavior. In both [Osand OOs anger is often sive, rather than assertive (asking for help appropriately). denied or minimized. Anticipation ofangergenerating sit­ Patients are concurrently taught to constructassertive state­ uations and early identification ofthe somatic components ments using the phrases ~I think/I fcelll want" and then ofanger facilitate proper labeling offeelingsandenable the rehearse these prior 10 applying them on the unit. Forexam­ patient to avert negalive responses. Modeling of appropri­ pie. an assertive statement might be, ~I have been thinking ate anger exprcssion by peers and therapists is an integral I will never get belter. I feel sad. I want someone to listen to part of the learning process. my fears and sadness." Anger responses in patients who dissociate differ from A unit-derived understanding of these behavioral pat­ those in the non-dissociative group in two important wa~. terns is then expanded to comprehend typical interactions first, the dissociative patientmay develop angeron the basis with family or friends. But before attempting newstrategies ora traumatic or ~flashback" transference in which a per­ with the family the abuse survivor must learn how previous son orsiwatiol1 in the environmenttriggersa re-experiencing efforts to get empathy and suppan were rebuITed. This is oftrauma. Cognitive appraisal ofperceived threat is not con­ critical to avoid unrealistic expectations of family respons­ sonant with the situation and the patient is unaware ofthe es to assertive behaviors. The perfectionistic thinking com­ link to traumatic materiaL Beforestaffunderstood the nature mon 10 both [Os and abuse survivors is especially counter­ of these distortions, angry dissociative patients were often productive in seu.ing assertiveness goals with past orcurrent shunnedas "narcissistic"orintractable. As with feared foods, abusers. be they family, spouses, or employers. When the the hypnotic "affect bridge techniquc" (Watkins, 1971; see patient is rebuffed anew despite more assertive behavior, aJso Horevitz, 1991) may assist in tracing the feelings offear she will become frustrated and hopeless. and anger back to their antecedents permiuing elucidation In order to make assertiveness training more effective and decompression ofthe response. and reinforcing. the abuse survivor must learn to identify In addition La these specific triggers for anger, the dis­ receptive people beforeapplying assertive skills. Rather than sociative patientmay presentwith a "victim identityM in which attempt new, assertive behavior with family. survivors may all situations are interpreted as victimization experiences. need to mourn the wish for caringand protection from fam· Therapeuticinterventionsare r.tpidly undone by thisslance, i1y,and lowertheirexpecrations.ApproprialCgoalsforassertive- making early recognition essential to preserve Ireatmenl.

182 DlSSOCl\T10~.\01 \11. ~o 3. ~pmber 1m ______s1 - LEVIN/SPAUSTER

The referring therapist may have magnified the victim iden­ methods developed during brief inpatient stays can form tity as part of a need to identify or rescue victims ofabuse. the framev.·ork for outpatient care, The Problem/Goal/ Theinabilitytoform a U'catmcntallianceseenin these patients Method approach of Levcndusky and Berglas (1985) not is not typical of the purely cating disordered patient. who onlyfocuses patientandclinician butiswell-suited for report­ usually wishes to please others and perform compliantly. ing Prob'TCSS to monitoring agencies. A collabordtive treat­ TIlcrdpistempathycoupledwith rnatter-of-factcognitiveexplo­ memdirccted toward measurable,achievablegoals provides rdlion have been successful in correcting the distortion of patientand clinician Wilh a senseofmastery, gradually replac­ victimization and maintaining the treatment. Logworkfocused ing lhe passivity and powerlessness which cripple the disso­ on specific anger incidents can proceed after these initial ciative patient. SLCpS. A typical anger-management sequence runs as follows: APPENDIX 1 (I) Situation - patient asked to stop smoking during non­ smoking hour; (2) Thought - MstafT is singling me out, ~ I. Sample Treatment Plan for Bulimic Symptoms and "nooneever recognizes myneeds"; (3) Feeling-anger; (4) Behavior - verbally abuses slafffollowed by shame and A. Problem - I binge and purge whenever r feel anxious urges to hurt self. These behaviors may occur in a dissoci­ and overwhelmed. ated state. Once a treaunent alliance has been developed, the patient can attempt elucidation ofthe anger sequence. B. Cool-To identify specific triggers for binge/purge This may require staffsupport overa scveral-day period and behavior and develop alternate coping strategies. hypnotic explordtion in the dissociated patient. A preprint­ ed writing assignment to assist in this process is given to the C. Methods patient as soon after the incident as she appears receptive. During individual review of the assignment, and in Anger­ 1. Keep Situation/Thought/Feeling log ofurges to Management group, the patient learns to identify the inap­ binge/purge to identify three most frequent propriately personalized nature ofher cognitions. The sur­ triggers. vivor is often able to identify the origin ofthese feelings and cognitions in the family of origin. Stan' and therapist then 2. Discuss triggers in Eating Disorders Group and validate for the patient that cognitions ofbeing singled out individually. or neglected wcre accurate appraisals in the pasL Although the patiem may not be able to relinquish these responses 3. Learn relaxation technique in group and practice immediately, sheoften canlearn to identifythe physical symp­ four times a day. toms of rising anger (vocal tension, shaking, and/or palpi­ tations) and employ a delay strategy (focus on breathing, 4. Apply relaxation at start of binge/purge urges simplecounting,orwithdrawal from thesiluation) oraground­ on uniL ingstrategy to thcn permitprocessing. Delay averts the dam­ aging behavioral response and grounding blocks the disso­ 5. Practice relaxation on passes to home. ciative response, enhancing the patient's sense of mastery over an internal state which has previously made her feel II. Sample Treaunent Plan for Dissociation helpless. A. Problem-I dissociate and lose touch with things around CONCLUSIONS me when I am reminded of my brother's abuse.

Eating-disordered patientswith dissociative disorders reI' B. GoaI- To decrease episodes ofdissociation. resent a special group of "dually diagnosed" patients who require specifically tailored treatments. A modified cogni­ C. Metlwds tive-behavior therapy model which adapts to the unique cog­ nitive distortions and re-enacunent behaviors seen in this I. KeepSituation/Ibought/Feelinglogofdissociative population has been successful in our hands. Prior to insti­ episodes (ask staffand peers to help me remember tuting these techniques, patients frequently left the hospi­ what happened) to identify three most frequent tal premalUrely and did not develop appropriate follow-up rriggers. ~are ..More rigorous investigation of this population must Idenlify treatment elements most effective in reducing dys­ 2. Learn about early signs ofdissociation in Survivors functional eating, poor impulse and affect regulation, and Groups and with therapist, s~lr-destructivebehaviors, and in enhancing adaptive func­ lions Such as stabilization ofinterpersonal relationships. 3. Learn self-hypnosis with physician and how to use With increasing demand for mental health profession­ it to control memoriesand nashbacks. rrdctice four ~Is to provide effective, time-limited care, cognitive-behav­ times a day. Ioral treaunents are receiving greater emphasis. Goals and

183 D1550ClHlO\, \01. \ [[ \0. 3,SejlUrnbcr 1!l94 4. Learn grounding techniques in Survivors Group, Horevitz, R. HJfl7WSiJ in the tnalmenl of mll./Upk ~ disorder. from nursing staff, and with therapisL Practice four Paper presenled at the Eighth International Conference on times a day. Multiple Personality/Dissociative States, Chicago, November 15­ 17,1991. 5. Explore family interdctions that trigger dissociation in Family Roles Croup. Levendusky, P.G., & Berglas, S. (1985). The therapeutic contract program: An individual--oriented psychological treatmentcommunity. Psychothemf1J, 22. 3&-45. 6. Develop safety strategies for family interactions. Linehan, M.M. (1993). Cognitive-behavioral treatmmt ofborderline per­ 7. Practice self-hypnosis, grounding techniques, and sonality disorder. New York: Guilford Pres.,. safety strategies on pass with family.• Lowenstein. RJ. (1991). An office mental status examination for complex chronic dissociative symptoms and multiple pcrsonalily disorder. PsychialricClinio ofNorlhAmerita. J4, 567-604. REFERENCFS McCallum, K.E., Lock,J., Kulla. M.• Rorty. M.,& Wetzel, RD. (1992). Andreason, PJ.. & Seidel.joA (1991). Behaviorallechniques in Dissociative symptoms and disorders in patienti with eating disor­ the rreaunentofpatienuwith multiple personalitydisorder. Annals ders. DISSOCJATION, 5, 227-235. ofClinical Psychiatry. 4, 29-32. P-dlmer, RL, Oppenheimer, R., Dignon. A., Chaloner. D.A., & Beck,A.T., Rush,AJ.,Shaw, B.Y., & Emery,G. (1979) Cogn~I/uT­ Howells, K. (1990). Childhood sexual experienceswith aduhsreport­ an ofdqrressilm. New York: Guilford Press. ed by women with eatingdisorders: An extended series. BritishJournai ofPsychiatry, 56, 699-703. Bernstein, E., & Putnam, F.W. (1986). Development. reliability, and validily of a dissociation scale. jaumal ofNervous and Mental Ross, C.A. (1989). Multiple personality disorder: Diagnosis, clinicalfea­ Di.sea.se, 174, 727-735. tures, and treatment. New York: John Wiley & Sons.

Braun. B.G. (1984). Towards a lheory ofmultiple personalily and Saigh, P.A. (Ed.) (1992). Posl-traumatic stress disorder: A bthauibraL Olher dissociative phenomena. Psyr:hiatTU Clinia ofNorth Amniro, aptJroat:h Ioassemnml and tnalmenl. New York.: Macmillan Publishing 7,171-194. Company.

Caddy, G.R. (1985). Cognitive behavior lherapy in the treatmem Spiegel, D. (1984). Multiple personalily as a post-traumatic stress of multiple personality. &haviorModifiallUm, 9, 267-292. disorder. Psychiatric Clinics ofNorli& Ammw, 7, 101-110.

F~irbum, C.G.,jones, R, Peveler, R.C., Carr, SJ., Solomon, RA... Spitzer, RL., Yan()\'Ski, S., Wadden, T., Wing, R., Marcus, M.D., o <:<)Onor, M.E., Burton,j., & Hope, R.A. (1991).Three psycho­ Stunkard, A., Devlin, M.. Mitchell,j.. Hasin, D., & Horne, R.L. logical treaunent5 for bulimia neTV0S3.: A comparati\'e uial. Archim (1993). Binge eating disorder: 1t5 further validation in a multisite ofGenn-aJ Psychiatry, 48,463-469. study. IntemahonaJJoumalof&hngDis~,13, 137-153.

Fine, C.G. (1991a). The cogniti\'e sequelae of incest. In R.I'. Kluft Torem, M.S. (1986). Eatingdisorderanddissociative states. In F.E.F. (Ed.): Incest-rew.ted syndromes ofadult psychopathology, (pp. 161-182). Larocca (Ed.), Eating dis(ffders: t.JJective care and treatment, (pp. 141­ Washmgton, DC: American Psychiatric Press. 150). Sl. Louis: Ishiyaku Euroamerica, Inc.

Fin~, C.C. (199Ib). Treatment stabilization and crisis prevention: Torem, M.S. (1987). Ego-state therapyforeatingdisordcrs.Amtritan Pacln.g ~e ~e.rapy of the multiple personality disorder patient. Journal ofCliniali Hypnmis, 30, 94-103. PsydUmrn: Clinus ofNcmh Amerna. 14, 661-675. Torem, M.S. (1989a). Ego-state hypnotherapy for dissociative eat­ Carner, D.M., & Carfinkel. P.E. (1985). Handbook ofps,dwtlu:raf1J ing disorde~. H'!/J1WS, 16,52-63. for anorexia nnvosa and bulimia. New York.: Guilford PreM. Torem, M.S. (l989b). Recognition ;,md management of dissocia­ Ga.me~, D.M., Ol~s~ed, M.P., & Polivy.j. (1983). Dcvdopmentand tive regressions. Hypnos, 16, 197-213. vahdaoonofa mulodimensionalealingdisorderinventoryforanorex­ ia nervosa and bulimia. Intnnalionaljoumal of&ting Dis~ 2, Torem, M.S. (1990). Covert multiple personality underlying eat­ 15-34. ing disorders. America-n journal ofPsye!lolherafrJ, 44,357-368.

Goodwin,j.M., Wilson, N., & Connell, V (1992). NaLUral history of van der Kolk, B.A., Perry.j.C., & Herman,j.L. (1991).Childhood severe symptoms in borderline women treated in an incest group. origins of self-destructi\'e behavior. American jouT'lU11 ofPs)'chiatry, DISSDaA-TION, 5, 221·226. 148,1665-1671.

Hall, KC.W., Tice, L, Beresford, T.P., Wooley, B., & Hall, AX Watkins,j.G. (1971).TheaIfcctbridge: A hypnoanalytic tcchnique. (1989). Sexual abuse in patients with anorexia nervosa and bulim­ Intemaliooaljournal ofainical and Expmmmlal Hypnosis. 19, 21-27. ia. I's]chmomalics, 30, 75-79.

Herman.J. (1992). Trauma and m::m.oery. Basic Books.

184 DlSSOClHIO\". \'01 \'II. \"0. 3. Sl'pttmbcr l~~