In Review

Promising Psychotherapies for Personality Disorders

George Hadjipavlou, MD, MA ; John S Ogrodniczuk,

Objective: To provide a narrative review of recent research on the psychotherapeutictreatment of patients with personality disorders (PDs).

Method: We conducted PubMed and PsyclNFO searches of recently published articles thatreported on the treatment outcomes of psychotherapies for PDs. Our focus was on studiesthat used randomized controlled trial (RCT) methodology. The search period was fromJanuary 2006 to June 2009.

Results: The effectiveness of various psychotherapy treatment packages for PDs is wellsupported by favourable results from RCTs. Beneficial effects of psychotherapy includedreduced symptomatology, improved social and interpersonal functioning, reduced frequencyof maladaptive behaviours, and decreased hospitalization. Equivalent effects among theinterventions we compared were common. Many of the treatments studied required onlylimited training by therapists. Most studies were focused on treating patients with borderlinepersonality disorder (BPD). Some findings were suggestive of psychotherapy beingcost-effective; however, few studies actually included formal cost analyses. Only one studyincluded follow-up of treated patients beyond 1-year posttreatment.

Conclusions: There is strong support for the use of psychotherapy to treat patients withPDs. However, most of the evidence is limited to BPD. The findings of recent studies holdpromise for training and practice. Future research should attend to identification ofappropriate patient-treatment matches, elucidation of active treatment ingredients, andillumination of factors that are common among treatments that account for their equivalenteffects.

Can J Psychiatry. 2010;55(4):202-210.

Clinical Implications

• Patients with PDs can be treated effectively using various psychotherapy treatmentpackages, suggesting a better prognosis than is often assumed.

• From a best practices standpoint, psychotherapy appears to be the treatment ofchoice for PDs.

• Dissemination of effective treatment packages into training and practice realmsappears promising as several tested treatments demanded only modest trainingperiods.

Limitations

• There is limited research on treatments for disorders other than BPD.

• Long-term follow-up of treated patients remains scarce.

• Systematic economic evaluations of tested interventions are in short supply.

Key Words: psychotherapy, personality disorders, randomized controlled trials,narrative review

202 + La Revue canadienne de psychiatrie, vol 55, no 4, avril 2010

Promising Psychctherapies for Personaiity Disorders

Considerable progress has been made in the psychotherapyof PDs.'"** Once engendering a pervasive therapeutic

nihilism, PDs are starting to be viewed as treatable with amuch better prognosis than previously thought.'"''' Evidencefrom RCTs demonstrating the effectiveness of various formsof psychotherapy, coupled with findings from several longitu-dinal studies, suggests that such increased clinical optimism iswarranted. ""''' This is encouraging, given that PDs are com-monly encountered in nonclinical samples, with prevalencerates of 9% to 13%,'^ and are among the most frequentlytreated conditions by psychiatrists in outpatient clinics.'* PDsalso pose a serious public health problem, as they are associ-ated with significant functional impairment, suicide risk,extensive treatment use, and worse outcomes in the treatmentof depression."

Prospective longitudinal studies have found surprisingly highrates of remission for PDs.'"*'' ''* For instance, the McLeanStudy of Adult Development found that 88% of patients diag-nosed with BPD severe enough to require hospitalization nolonger met diagnostic criteria by the 10-year follow-upperiod, and of those whose symptoms remitted, only about 6%went on to have a recurrence."*"" Similarly, in the Collabora-tive Longitudinal Personality Disorders Study," less thanone-half of the patients followed still met criteria for a PDdiagnosis after 2 years. Further, it has been estimated thatpatients with PDs receiving psychotherapy experience recov-ery 7 times faster, compared with the natural course of the ill-ness.^" Taken together, these findings suggest that thetraditional view of PDs espoused by the DSM as "enduring"conditions that remain "stable"'*' '' ^*' in the long-term—a

Abbreviations used in this article

ASPD antisocial personality disorder

AVPD avoidant personaiity disorder

BPD borderline personaiity disorder

CBT cognitive-behaviourai therapy

DBT dialectical behaviourai therapy

DDP dynamic deconstructive psychotherapy

DSM Diagnostic and Statisticai Manuai of Mentai

Disorders

GPiVl general psychiatric management

ISTDP intensive short-term dynamic psychotherapy

MACT manuai-assisted cognitive therapy

iVIBT mentalization-based therapy

PD personality disorder

RCT randomized controiied triai

SFT schema-focused therapy

ST supportive therapySTEPPS Systems Training for Emotionai Predictability and

Probiem Soiving

TAU treatment as usual

TFP transference-focused psychotherapy

view which fosters pessimism about the potential for positivetherapeutic outcomes—has not been borne out by empiricalevidence.'*

This is especially true when we consider outcome data fromRCTs of psychotherapy. There have been several systematic

'"' ^ and meta-analyses,^"' ' '' including a Cochrane, ^ supporting the use of psychotherapy for PDs. This

contrasts with the evidence for phannacological interven-tions, which has been less encouraging.'•''•^'' Although all ofthese reviews considered psychotherapy to be a promisingtreatment with favourable outcomes, they also cautionedthat, given the small number of high-quality RCTs available(mainly before 2002), such a view was still preliminary. Nev-ertheless, the emergence of such an evidence base is a crucialadvance in solidifying the place of psychotherapy as a treat-ment of choice for PDs.

Interestingly, a 2007 review found that between 2003 and2006, twice as many RCTs of psychotherapy for PDs hadbeen published than in the preceding 2 decades.* Since then,several other, methodologically rigorous studies have beenpublished. The goal of our paper is to provide an overview ofthese new research developments. Our primary focus is ontherapies that have been tested by well-designed RCTs. Inparticular, we are interested in studies of psychotherapy thathave been published in peer-reviewed journals during thepast 3 years. To identify relevant studies, we conductedPubMed and PsycINFO searches dating from January 2006to June 2009. Although we searched for studies on every PD,most available data pertain to BPD, which is refiected herein.

It does not escape our attention that focusing solely on RCTshas limitations, the most obvious of which is overlookingother potentially relevant evaluative approaches, and privi-leging research that potentially may not generalize easily toclinical practice. The problem of generalizability has beenidentified as the bane of RCTs,^ as studies that include care-fully selected patients who are treated by experts in academiccentres providing services that are not generally available, donot seem directly relevant to everyday clinical care. From thisvantage, the Cochrane Review of BPD concluded by stating aneed for "real-world studies."^^' '' ^ These recent RCTs beginto address that need. Almost all of the studies reviewed hereshare features of both efficacy and effectiveness trials; forinstance, most are conducted in everyday clinical settings,include participants with comorbid conditions common toPDs, and many provide treatment by nonexpert therapists.

Our review is intended as an updated guide to new psycho-therapy treatment packages that may be relevant to cliniciansinvolved in the care of patients with PDs. Although we aremainly concerned with new treatment packages, we alsoinclude recent findings from rigorous trials lending addi-tional support to previously tested approaches such as DBT.All of these treatment packages derive from and are elabora-tions on established principles and practices; most have theirroots in either psychodynamic or cognitive-behaviouralapproaches. None offers an entirely unique or novel

The Canadian Journai of Psychiatry, Voi 55, No 4, Aprii 2010 203

in Review

approach. Although they differ in focus and technique, theyalso have much in comriion. We provide a brief description ofeach therapeutic approach and thé inäjbr findings from recentclinical trials that support their effectiveness. We then discussimplications for clinical practice and future research.

Evidence From Psychotherapy Research onPDs (2006-2009) .

Modified Psychodynamic Approaches ',Mentalization-Based Therapy. MBT is a complexpsychodynamic tréátmeht that is i"ooted in attachment theoryand draws on concepts' from cognitive psychology, Batemanand Fonagy^' describe MBT as "a focus for therapy ratherthan a specific therapy in itself," employing "a reiteration ofwell-knowh basic therapy practices such as support, empathy,exploration and challenge,"' ' ^ ' ^ The focus of MBT is onenhancing mentalization,

Mentalization is the capacity to understand behaviour, one'sown and that of others, regarding underlying mental states (forexample, thoughts and feelings). It is a mental process thatinvolves recognizing how one's perceptions of oneself andothers are not equivalent to reality but are representationsclosely related to one's thoughts, feelings, and desires,"'^^''"^MBT seeks to enhance this reflective capacity, which is pos-ited to be disrupted in patients with BPD—particularly in thecontext, of relationships that activate their attachmentsystem—and underlies their disturbed interpersonal related-ness. The integration of one's experience of one's own mindwith the view presented by the therapist is a key component ofMBT,"

MBT was initiallydescribed as a psychoanalytically orientedtreatment approach in the setting of partial hospitalization, atreatment package that consisted of both individual and grouppsychotherapy, and had proven effective in an 18-monthRCT, compared with routine psychiatric care.^' Bateman andFonagy" recently reported impressive findings from an8-year follow-up study of MBT demonstrating its long-termefficacy. This is the first long-term study following patientstreated with an evidence-based psychotherapy. Patients(/2 = 41 ) who participated in the.original trial wer.e followed up8 years after initial randomization (5 years after treatmentcompletion). Patients treated with MBT exhibited substantialclinically and statistically significant differences from thosetreated with usual psychiatric care. They had far fewer suicideattempts (23% and 74%, respectively)^ reduced emergencydepartments visits, hóspitalizatiohs; and use of mental healthservices; dramatically fewer rnèdiçation prescriptions; andincreased social, educational and vocational functioning. Per-haps inost striking was that only 13% of patients treated withMBT still met criteria for BPD, compared with 87% of thecontrol group. Despite these fairly remarkable improvements,patients in both groups continued to show impairment in gen-eral social function, albeit far less for those who receivedMBT; 46% of the MBT group had Global Assessment of

Functioning scores above 60, compared with 11 % of theTAU group,

Transference-Focused Psychotherapy. TFP is á structured,object relations approach, which emphasizes the integrationof affect-laden mental representations of self and others thatwere originally derived through the intemalization of attach-ment relationships with caregivers,''^"'''' Understanding howthese internal representations become activated in thehere-and-now relationship with the therapist is a key part oftherapy. In this way, negative affect states, particularlyaggression, are gradually controlled by understanding themas they unfold in the context of the transference,''^'^^ TFP aimsfor full recovery, which encompasses reducing suicidalityand self-injurious behaviour, improving behavioural controland affect regulation, and enhancing the ability to pursuegratifying relationships and meaningful life goals. It consistsof twice-weekly individual therapy provided in the context ofa clear contract and a consistent frame, TFP was the firstexperimental therapy for BPD to be tested under randomizedconditions against another established treatment for BPD,that is, DBT.'"

In an RCT (n = 90) comparing DBT, TFP, and ST, patients inall 3 treatment groups showed significant, and generallyequivalent improvement in social and global functioning,and significant decreases in depression and anxiety. Treat-ments were delivered in community mental health settingsand included a broad range of patients with BPD—not justthose exhibiting parasuicidal behaviour. While both TFP andDBT significantly improved suicidality, only patients in theTFP group demonstrated reductions in verbal assault, irrita-bility, and direct assault. Overall, TFP led to improvement ina broader range of outcome measures (10/12) than DBT(5/12) and ST (6/12), A subsequent analysis of the trial data *found that TFP resulted in significant improvements inattachment style (a change to more secure attachments), nar-rative coherence, and reflective functioning, which were notobserved with DBT or ST,

Preliminary Findings From Other PsychodynamicApproaches. ISTDP has recently shown promise in a smallRCT (n 7= 27) of mixed PDs, ^ Patients were randomly allo-cated to either ISTDP or a minimal contact control condition,ISTDP encourages the awareness and experience of previ-ously unconscious feelings associated with maintainingsyhiptonis and dysfunction, while clarifying and challengingdefences in collaboration with the patient. Patients and thera-pists meet on a weekly basis and mutually decide on when toterminate treatment (ah average of 28 sessions). Patientstreated with ISTDP not only experienced significant reduc-tions in their symptoms and interpersonal problems, they alsoshowed improvement in function, including doubling theiroverall work hours and rates of employment, and decreasingmedication use (69% stopped all medications). By the end ofthe follow-up period, 86% of participants no longer met crite-ria for any PD, and treatment costs were calculated to have

204 La Revue canadienne de psychiatrie, voi 55, no 4, avrii 2010

Promising Psychotherapies for Personaiity Disorders

been thrice offset by reduetions in medication and disabilitypayments."

Another manual ized psychodynamic approach, DDP, wastested in an RCT of patients with fairly severe BPD andcomorbid alcohol abuse.^^ A significant proportion of thèsepatients were also diagnosed with ASPD. DDP maintains anonjudgmental and rioridirective stance that focuses on help-ing patients identify and verbally express their emotions, con-struct coherent narratives of their interpersonal experiences,and integrate polarized attributions about self and others,without resorting to compensatory, maladaptive behaviours(for example, alcohol abuse) to alleviate painful affects. Par-ticipants were randomized to receive either 10 individualweekly sessions of DDP over 12 to 18 months (« = 15) or toTAU in the community (n = 15). Participants who receivedDDP showed significant improvement at 12 months inparasuicidal behaviour (decreasing from 73% of participants,pretreatment, to 30%), incidents of alcohol abuse (67% and30%) and need for institutional care (67% and 10%), as wellas reductions in overall medication prescriptions and mea-sures of BPD pathology, depression, and dissociation. DDPwas carried out by therapists who had very little prior psycho-therapy training and was actually less time-intensive than rou-tine community care.

Modified CBT ApproachesDialectical Behavioural Therapy. DBT conceptualizes thecore problem of BPD as a habitual breakdown of patients'cognitive, behavioural, and emotional regulation systemswhen they experience intense emotions.^' It is thought tofacilitate change through the learning of emotion regulationskills in the validating environment of the treatment. DBT is acomprehensive treatment package that involves 4 modes oftherapy: individual, in which the therapist oversees treatmentintegration and manages life-threatening behaviours and cri-ses; group skills training, including mindflilness, distress tol-erance, emotion regulation, and interpersonal effectiveness;skills generalization through telephone contact outside of nor-mal therapy hours; and a consultation team designed to sup-port therapists in working with difficult clients. AlthoughDBT also comprises 4 stages of treatment, evidence fromRCTs primarily pertains to the first stage, which involveseliminating dangerous and disruptive behaviours.^'

Since Linehan et al's'"' seminal study of DBT almost20 years ago, there have been several RCTs conducted by dif-ferent research groups demonstrating its efficacy.^' The larg-est (n = 180) and most recent trial"" compared DBT withGPM, a standardized set of treatment practices derived fromthe American Psychiatric Association practice guidelines forthe treatment of BPD. GPM consisted of weekly individualpsychodynamic psychotherapy, case management, andmedication management. Subjects were randomly assignedto 1 year of DBT or GPM. Both treatment groups showedimprovements on most clinical outcomes, including reduc-tions in frequency and severity of self-injurious behaviours;

reduced BDP symptoms, general symptom distress, depres-sion, and anger; lower rates of general health care use; andimproved interpersonal functioning. There were nobetween-group differences on any of the outcome variables.Treatment retention (62%) was the same in the 2 treatments.

There was one other well-designed controlled trial of DBTfor BPD published within our review period."* In this study,DBT (n = 52) was compared with nonbehavioural therapy bycommunity psychiatrists with expertise in the treatment ofBPD (n = 49), a condition intended to control for nonspecifictreatment effects of expert therapy. Participants in bothgroups showed substantial improvements, but those receiv-ing DBT were about one-half as likely to attempt suicide(23%) as those treated by community experts (46%), to visitthe emergency department for suicidal ideation (15.6% and33.3%, respectively), or to drop out of treatment (25% and59%, respectively). DBT-treated patients also had signifi-cantly lower rates of psychiatric hospitalization and use ofcrisis services. Although at 1 -year follow-up there were nosignificant differences between treatment groups in the fre-quency of nonsuicidal acts of self-harm, subjects receivingDBT had a lower medical risk associated with their self-injurious behaviour. Subjects in both groups showed similarimprovements on measures of depression, suicidal ideation,and reasons for living.

Schema-Focused Therapy. SFT is an integrative therapy thatbrings together elements of cognitive therapy, behaviouraltherapy, object relations, and gestalt therapy. It focuses onpatients' maladaptive schémas or pervasive patterns of think-ing, feeling, and behaving that are developed during child-hood and are associated with problems in one's identity andsense of self, interpersonal functioning, and affectregulation."*^^^ In this approach, BPD is thought to involveregression into early maladaptive modes of being that are tiedto specific schémas and associated intense emotional states.Therapy involves recognition of self-perpetuating processesthat maintain maladaptive schémas and render them resistantto change. Identifying and changing maladaptive schémas isthe main focus of treatment. Changing schémas involves bothcognitive and experiential work, including such approachesas limited adaptive reparenting (emphasizing acceptance andvalidation) and empathie confrontation. Maladaptive behav-iours outside of therapy are also addressed. Recovery is thegoal of treatment, and is achieved when maladaptive schémasno longer dominate patients' lives, allowing them to imple-ment more adaptive coping skills.

A multicentre RCT"^ comparing 3 years of outpatient,twice-weekly SFT to TFP in 86 patients with BPD found sig-nificant improvements in both groups, including reductionson all DSM-IV BPD symptoms, increased quality of life, andpositive changes in personality, which were apparent after1 year of treatment. Although after 3 years of treatmentresults generally favoured SFT, these findings have beencontested by the consultant hired to provide supervision to

The Canadian Journai of Psychiatry, Voi 55, No 4, Aprii 2010 205

In Review

the TFP therapists who felt that their adherence was less thanadequate.''^

SFT has been adapted into a group format, and its effective-ness was recently tested in a small RCT (« = 32) of female out-patients with fairly severe BPD."" Patients in this study eitherreceived TAU, which largely comprised of weekly individualpsychotherapy, or TAU plus SFT, SFT consisted of 30 weeklysessions during 8 months, and included components of emo-tional awareness training, psychoeducation, distress manage-ment, and schema change work. Patients receiving SFTimproved significantly on a broad range of clinical measuresof borderline psychopathology and showed increases in over-all global functioning, while the control group either stayedthe same or deteriorated. In addition, 94% of SFT patients nolonger met criteria for BPD, compared with 8% receivingTAU—a positive effect that was maintained or increased after6 months of follow-up. There was a remarkable 100% reten-tion in the SFT groups. SFT has been shown to be cost-effective,"' and its adaptation into a group format will likelymake it more so.

Other Trials of CBT. An RCT in a community setting investi-gated the value of combining up to 30 one-hour sessions ofCBT, which aimed to correct maladaptive core beliefs andoverdeveloped behavioural patterns that impair ñinctioning,with TAU in 106 patients with BPD,"*'"' Although bothgroups showed sustained and comparable improvement,results favoured the addition of CBT, Despite receiving anaverage of only 16 sessions of CBT by therapists with limitedtraining during the course of 2 years, patients randomized tothis group experienced a significant reduction in mean num-ber of suicidal acts, levels of distress, dysfunctional thinking,and state anxiety, compared with TAU. The 2 groups did notdiffer in the number of psychiatric hospitalizations, emer-gency department visits, or depression ratings,

MACT is a short-term (6 sessions) individual therapy thatincorporates elements of DBT (for example, emotion regula-tion strategies), CBT (for example, correcting negatively dis-torted thinking), and bibliotherapy. Weinberg et al " adaptedand tested this approach to address deliberate self-harm in30 patients with BPD who were randomized to either receiveMACT (n = 15) or TAU (n = 15). MACT patients attended all6 sessions and exhibited significantly fewer acts of deliberateself-harm, compared with control subjects, with decreasesboth in frequency (93% less) and in severity at 6-monthfollow-up. These results diverge from a previous, larger RCTin which MACT was not found to improve outcomes, com-pared with TAU,'' However, 38% of patients randomized toMACT in this earlier study did not attend any sessions, butwere included in the analyses. Also, in the study byWeinberg et al,'" MACT was provided as an adjunctive treat-ment, and deliberate acts of self-harm were distinguishedfrom suicide attempts.

The efficacy of brief CBT (12 weeks) for reducing self-injurious behaviour^^ was tested in an RCT of adolescents andyoung adults. Although participants in this study were not

evaluated for the presence of a PD, their extensivemaladaptive beliefs and repetitive acts of self-harm sug-gested high rates of personality pathology. CBT in additionto TAU not only significantly decreased repetitive acts ofself-harm, compared with TAU alone, but also resulted inimproved self-esteem and problem-solving ability, andreduced depression, anxiety, and suicidal ideation.

Connor et al" carried out an RCT of CBT in violent men withASPD. Participants were randomized to TAU plus CBT {n =25) for either 6 or 12 months, or just TAU {n = 27). The CBTintervention focused on addressing beliefs about self and oth-ers, and behaviours that impair social and adaptive function-ing. Decreased rates of verbal and physical aggression werereported by participants in both groups at 12-monthfollow-up, without a significant difference between the 2,However, this study was underpowered to detectbetween-group differences in clinically relevant outcomes, Anonsignificant trend was reported in the CBT interventionregarding reduced rates of harmful alcohol consumption,improved social functioning, and more positive beliefs aboutothers,

A small RCT compared the effectiveness of CBT and briefdynamic therapy in the treatment of AVPD,^" Patients wererandomized to either 20 weekly sessions of CBT (« = 21),brief dynamic therapy {n = 23), or a wait-list control. CBTwas found to be more effective than brief dynamic therapy inameliorating symptoms of AVPD. At follow-up, only 9% ofpatients in the CBT group and 36% of those receivingdynamic therapy still met diagnostic criteria for AVPD.Results favouring CBT in this trial differ from a previous40-week RCT, which found both modalities to be equallyeffective in treating Cluster C PDs." As the authors point out,these are encouraging findings as AVPD has been shown tobe persistent with a reported tendency to worsen over time.'"

Treatment Approaches That Emphasize PsychoeducationSystems Training for Emotional Predictability and ProblemSolving. STEPPS is an adjunctive, multifaceted, 20-weektreatment program designed to supplement patients' ongoingcare, be it psychotherapy or case management."''" STEPPScombines elements of CBT and skills training with a systemscomponent, which actively involves people with whom thepatient interacts regularly and has designated as their systemmembers (family, significant others, and health care profes-sionals). Systems members are provided education aboutBPD and how best to interact with the patient, while patientsare encouraged to share their treatment notebooks with them.This systems approach is an innovative addition that aims todirectly address some ofthe interpersonal problems that are amajor feature of the disorder. Participants attend 2-hourweekly group seminars organized around learning specificemotional, cognitive, and behavioural self-managementskills,

STEPPS was tested in a 20-week RCT with a 1-year follow-up in outpatients with BPD who either received STEPPS plus

206 La Revue canadienne de psychiatrie, vol 55, no 4, avril 2010

Promising Psychotherapies for Personality Disorders

TAU (« = 65) or just TAU (« = 59). Participants treated withSTEPPS showed significantly greater improvement than con-trol subjects on a broad range of clinical measures, includingdecreased levels of depression, negative affects and disturbedcognitions, reduced impulsivity, and better interpersonal andglobal functioning. Although there was a significant reduc-tion in emergency department visits, there were no significantbetween-group differences for overall use of crisis services,suicide attempts, and self-harm. Given that other treatmentprograms (for example, DBT) have proven efficacious inreducing self-harming behaviour while demonstrating a muchless impressive effect on such symptoms as depression,STEPPS, with its brief and easy to implement format, mayserve a valuable, complementary role.'^

Social Problem Solving Plus Psychoeducation. An importantfinding from prospective longitudinal studies of PDs is thatproblems in social relationships and social functioning arepersistent, often despite substantial improvements in symp-toms.'* Problem-solving therapy plus psychoeducation is apractical psychosocial treatment package intended to improvesocial competence by teaching patients with PDs how to iden-tify solutions to problems in l i v i ^ *

Huband et al * conducted an RCT of problem-solving therapyfor PDs. Patients in this study were randomly allocated toreceive either 3 individual sessions of psychoeducation with16 weeks of group-based problem-solving trainingcomplementing existing treatment {n = 87) or to a wait-listcontrol (n = 89) in which they could still receive TAU. Noneof the study therapists (psychologists and psychiatric nurses)had previous training in the treatment program before a 2-daycourse attended as part of the trial. Participants who receivedproblem-solving therapy plus psychoeducation showed sig-nificant improvement in social problem-solving ability,higher social functioning, and reduced anger expression,compared with control subjects. It is not known how much ofthe improvement can be attributed primarily to socialproblem-solving skills as even very brief psychoeducationhas been shown to be of some benefit.^' Unfortunately, therewere high dropout rates, with an overall completion rate of48%.

DiscussionPDs are complex conditions that manifest in myriad ways. It ishard to imagine that a single therapeutic modality would workwell for all patients, even if they carry the same diagnosis. Tlierecent emergence of various empirically, testedpsychotherapeutic treatments is thus an important advance.Unfortunately, psychotherapy research on PDs other thanBPD is lagging, and much work needs to be done to identifyeffective therapies for these patients.

Taken together, these studies speak to the importance ofwell-structured approaches within a coherent treatmentframework that is understandable both to practitioners and topatients.'" The effective treatments we reviewed here sharenumerous common features. For instance, they incorporate

psychoeducational, motivational, cognitive, and interper-sonal components, albeit with differing degrees of emphasis,while affective and behavioural techniques are more vari-able.^" Given substantial differences across studyparameters—outcome measures, duration of treatment, indi-vidual compared with group format, treatment setting, lengthof follow-up—comparisons about overall treatment effec-tiveness cannot easily be made. No psychotherapy treatmentpackage can be said to be clearly superior.

Instead of pitting different therapies against each other, itmay be more productive to consider how differentapproaches may work better for different patients, or howthey can complement each other or be provided in a rationalstepwise manner.'^'^" Psychotherapy treatments such asSTEPPS and social problem solving are intended to comple-ment patients' existing care, while others, such as SFT orTFP are designed as comprehensive treatments fostering theambitious goal of recovery. It seems that where some thera-pies do less well, others succeed. For instance, STEPPSyields robust antidepressant effects, whereas DBT tends tounderperform in this domain; and DBT significantly reducesself-injurious behaviour, while SJEPPS does not.'^

A stepwise or sequenced approach may consider differenttherapeutic interventions at different phases of patients' ill-ness.^°'^' Livesley*' has proposed a general framework forintegrating therapeutic approaches and techniques in asequenced manner that conceptualizes patients' personalitypathology in stages, balancing treatment priorities (for exam-ple, safety) with amenability to intervention (for example,behavioural symptoms, compared with core traits), andeipphasizes nonspecific components of therapy.

Except for its inclusion as a comparator in the RCT of TFP,there has been no systematic investigation of supportive psy-chotherapy in the treatment of PDs. This is unfortunate as it isarguably the most commonly practiced form of therapy.^^ AnRCT of supportive psychotherapy that is delivered in awell-defined framework aiming to maximize establishedcommon factors without adhering to any specific so-calledbrand techniques or elaborate theoretical orientation wouldbe a welcome addition to the evidence base.

Given that PDs run a chronic course, longer-term treatment isgenerally assumed to be necessary. The most recentlyupdated treatment guidelines for BPD from the NationalInstitute fpf Health and Clinical Excellence in the UnitedKiiigdpm even cautions agairis.tthe.use of brief psychologicalinterventions.^'' Ho\yèyer, thissêénis^^t'p^ds with recentfindings suggesting that brief ádjunctiye iritervéntipn^ (forexample, MACT, STEPPS, and problem-solving theráp>;)can have significant positive effects. Althpugh there' is ánimportant signal that short-term therapies pari haveameliorative effects, there is perhaps an even stronger signalthat longer treatments with higher doses are of greater bene-fit. In several studies, significant improvement was onlyobserved after 12 months of active treatment.

The Canadian Journal of Psychiatry, Vol 55, No 4, April 2010 207

in Review

Despite deriving considerable symptomatic improvementfrom these various forms of psychotherapy, patients with PDscontinue to show impairment in social functioning and dimin-ished quality of life.*"* This is consistent with prospective lon-gitudinal studies demonstrating that remission of symptomsdoes not correspond with a return to normal functioning.' ""'Further, data from these longitudinal studies suggest that PDscan be better conceptualized as hybrid conditions that encom-pass both more stable personality traits (for example, relatedto temperament) and intermittently expressed dysfunctionalsymptomatic behaviours (for example, self-harm).""'' Mosttherapies so far have primarily addressed these symptomaticbehaviours.^'' Psychotherapy research can make valuable con-tributions by developing strategies to enhance quality of life(rather than assuming it will improve as behavioural symp-toms resolve) and to more directly address those enduringtraits that persist even when the disorder remits.

RCTs of manualized therapies provide empirical support forthe so-called whole package. While these packages enablevaluable goals of training, replication, and dissemination,they hamper the identification of key treatment elements oractive ingredients.^^ The effective ingredients of these treat-ment packages have yet to be clearly identified. Could it bethat different ingredients are effective for different patients, orthat markedly different treatment approaches can improve thesame symptoms through different mechanisms?^ Forinstance, while patients treated with DBT may developincreased behavioural control or reduced impulsivity throughbehavioural skills training, MBT-treated patients may exhibitthe same improvement in behavioural control by focusing onmentalization, without receiving any specific behaviouralinterventions.'''*

Dismantling studies that isolate and test specific therapeuticinterventions within more elaborate treatment packages are animportant next step. Such an advance would likely lead toincreased treatment efficiency and wider dissemination andavailability of these treatments."* A common criticism of DBTis that it is resource-intensive and expensive. While much lesscostly, dismantled forms of DBT are available (for instance,outpatient DBT skills groups), it is not known whether theyshare some of the same positive outcomes as the whole DBTpackage.

Another important avenue for future research concerns theinvestigation of common elements of various types of psycho-therapy. Many authors have argued that the general finding ofequivalent effects of different types of therapy implies thatthere are particular features that are common across thesetherapies that account for their effectiveness.^^ Studies havesuggested that a contextual model, which emphasizes com-mon factors, is more consistent with the research evidencethan is a model that posits specific therapeutic ingredients.*^Three particular sources that are believed to explain variabil-ity of outcome in psychological treatments are therapists, thetherapeutic alliance, and expectations.

Despite the persistent view that psychotherapy treatment istoo expensive and not cost-effective, available data suggestotherwise.^ Recent studies have found that the costs of psy-chotherapy are not only offset by reductions in the use ofother health care services, including hospitalizations, butalso may result in savings. In addition to reduced direct coststo the medical system, there are decreases in indirect costsfrom increased productivity, and reduced absenteeism fromwork and imprisonment.^

Although the evidence base for psychotherapy in the treat-ment of PDs has grown substantially during the past fewyears, we echo previous reviews''^ in suggesting that our abil-ity to draw confident conclusions is still hampered both bythe small number of RCTs and by the methodological qualityof available studies. Replication in larger, adequately pow-ered studies is needed. Longer follow-up periods, such as inthe MBT trial, would further increase confidence that treat-ment gains are maintained, as would the inclusion of condi-tions that control for the effects of natural recovery. Althoughmuch work still needs to be done, in light of the promisingcurrent evidence base for psychotherapy, pessimism aboutthe treatment of PDs seems outdated.

AcknowledgementsThe Canadian Psychiatric Association proudly supports the InReview series by providing an honorarium to the authors.

References

1. Paris J. Clinical trials of treatment for personality disorders. Psychiatr ClinNorth Am. 2008;31(3):517-526, viii.

2. Gabbard GO. Do all roads lead to Rome? New findings on borderlinepersonality disorder. Am J Psychiatry. 2007;164(6):853-855.

3. Paris J. Recent advances in the treatment of borderline personality disorder.Can J Psychiatry. 2005;50(8);435-441.

4. Paris J. The treatment of borderline personality disorder: implieations ofresearch on diagnosis, etiology, and outcome. Annu Rev Clin Psychol.2009;5:277-290.

5. Fonagy P, Bateman A. Progress in the treatment of borderline personalitydisorder. Br J Psychiatry. 2006; 188:1-3.

6. Gabbard GO. Personality disorders come of age. Am J Psychiatry.2005;162(5):833-835.

7. Bartak A, Soeteman DI, Verheul R, et al. Strengthening the status ofpsychotherapy for personality disorders: an integrated perspective on effectsand costs. Can J Psychiatry. 2007;52(12):803-810.

8. McMain S, Pos AE. Advanees in psychotherapy of personality disorders:a research update. Curr Psychiatry Rep. 2007;9(l):46-52.

9. Gabbard GO. Psychotherapy of personality disorders. J Psychother Praet Res.2000;9(l):l-6.

10. Bateman AW, Fonagy P. Effectiveness of psyehotherapcutic treatment ofpersonality disorder. Br J Psychiatry. 2000;177:138-143.

11. Bateman A, Fonagy P. 8-year follow-up of patients treated for borderlinepersonality disorder: mentalization-based treatment versus treatment as usual.Am J Psychiatry. 2008; 165(5):631-638.

12. Silk KR. Augmenting psychotherapy for borderline personality disorder: theSTEPPS program. Am J Psychiatry. 2008;165(4):413-415.

13. Bateman AW, Tyrer P. Psyehologieal treatment for personality disorder. AdvPsychiatr Treat. 2004; 10:378-388.

14. Zanarini MC, Frankenburg FR, Hennen J, et al. Predietion of the 10-year courseof borderline personality disorder. Am J Psychiatry. 2006;163(5):827-832.

15. Lenzenweger MF. Epidemiology of personality disorders. Psyehiatr Clin NorthAm. 2008;31(3):395^03, vi.

16. Zimmerman M, Rothsehild L, Chelminski I. The prevalenee of DSM-IVpersonality disorders in psychiatric outpatients. Am J Psyehiatry.2005;162(10):1911-1918.

17. Skodol AE, Gunderson JG, Shea MT, et al. The Collaborative LongitudinalPersonality Disorders Study (CLPS): overview and implications. ] Pers Disord.2005;19(5):487-504.

208 '4' La Revue canadienne de psyciiiatrie, vol 55, no 4, avril 2010

Promising Psychotherapies for Personaiity Disorders

18. Skodol AE. Longitudinal course and outcome of personality disorders. PsychiatrClin North Am. 2008;31(3);495-503, viii.

19. Zanarini MC, Frankenburg FR, Reich DB, et al. The subsyndromalphenomenology of borderline personality disorder; a 10-ycar follow-up study.Am J Psyehiatry. 2007;164(6):929-935.

20. Perry JC, Banon E, lanni F. Effectiveness of psychotherapy for personalitydisorders. Am J Psyehiatry. 1999;156(9):1312-I321.

21. Verheul R, Herbrink M. The efficacy of various modalities of psychotherapy forpersonality disorders: a systetnatic review of the evidence and elinicalrecomtnendations. Int Rev Psychiatry. 2007;19(l):25-38.

22. Duggan C, Huband N, Smailagie N, et al. The use of psychological treatmentsfor people with personality disorder: a systematic review of randomizedcontrolled trials. Pers Mem Health. 2007; 1:95-125.

23. Leiehscnring F, Leibing E. The effectiveness of psychodynamie therapy andcognitive behavior therapy in the treatment of personality disorders:a meta-analysis. Atii J Psychiatry. 2003; 160(7): 1223-1232.

24. Leichsenring F, Rabung S. Effectiveness of long-term psyehodynamiepsychotherapy: a meta-analysis. JAMA. 2008;300(13):1551-1565.

25. Binks CA, Fcnton M, MeCarthy L, et al. Psyehologieal therapies for people withborderline personality disorder. Coehrane Database Syst Rev.2006;(l):CD005652.

26. Sehulz SC, Zanarini MC, Bateman A, et al. Olanzapine for the treatment ofborderline personality disorder: variable dose 12-week randomised double-blindplaeebo-eontrolled study. Br J Psychiatry. 2008;193(6):485-492.

27. Bateman A, Fonagy P. Mentalization-bascd treatment for borderline personalitydisorder. Oxford (GB): Oxford University Press; 2006.

28. Bateman A, Fonagy P. Treattnent of borderline personality disorder withpsyehoanalytieally oriented partial hospitalization: an 18-month follow-up. Am JPsychiatry. 200l;158(l):36-42.

29. Bateman AW, Ryle A, Fonagy P, et al. Psychotherapy for borderline personalitydisorder: mentalization based therapy and cognitive analytic therapy compared.Int Rev Psyehiatry. 2007;19(l):5l-62.

30. Choi-Kain LW, Gunderson JG. Mentalization: ontogeny, assessment, andapplication in the treatment of borderline personality disorder. Am J Psyehiatry.2008;165(9):l 127-1135.

31. Bateman A, Fonagy P. Effectiveness of partial hospitalization in the treatment ofborderline personality disorder: a randomized eontrolled trial. Am J Psyehiatry.1999;I56(1O):1563-1569.

32. Clarkin J, Yeomans FE, Kemberg OF. Psychotherapy for borderline personality.Arlington (VA): Ameriean Psyehiatric Publishing Inc; 2006.

33. Levy KN, Wassennan RH, Seott LN, et al. Empirical evidence fortransferenee-focused psyehotherapy and other psyehodynamic psychotherapy forborderline personality disorder. In: Levy RA, Albon JS, editors. Rosenbaum JF,series editor. Handbook of evidenee-based psychodynamic psyehotherapy:bridging the gap between science and practice. Totowa (NJ): Humana Press;2009. p 93-113.

34. Clarkin JF, Levy KN, Lenzenweger MF, et al. Evaluating three treatments forborderline personality disorder: a multiwave study. Am J Psychiatry.2007;164(6):922-928.

35. Yeomans FE, Clarkin JF, Kcmbcrg OF. A primer of transfercnee-foeusedpsychotherapy for the borderline patient. Oxford (GB): Rowman & LittlefieldPublishing Group, Inc; 2005.

36. Levy KN, Meehan KB, Kelly KM, ct al. Change in attachment patterns andreflective function in a randomized control trial of transferenee-foeusedpsychotherapy for borderline personality disorder. J Consult Clin Psychol.2006;74(6):1027-l040.

37. Abbass A, Sheldon A, Gyra J, ct al. Intensive short-term dynamic psychotherapyfor DSM-IV personality disorders: a randomized controlled trial. J Nerv MentDis. 2008;l96(3):211-216.

38. Gregory RJ, Remen AL, Soderberg M, et al. A eontrolled trial of psychodynamicpsyehotherapy for eo-occurring borderline personality disorder and alcohol usedisorder: six-month outeome. J Am Psychoanal Assoc. 2009;57(l): 199-205.

39. Lynch TR, Trost WT, Salsman N, et al. Dialeetieal behavior therapy forborderline personality disorder. Annu Rev Clin Psychol. 2OO7;3:181-52O5.

40. Linehan MM, Armstrong HE, Suarez A, et al. Cognitive-behavioral treatment ofchronically parasuicidal borderline.patients. Arch Gen Psychiatry.199l;48(12):1060-1064.

41. McMain SF, Links PS, Gnam WH, et al. A randomized trial of dialecticalbehavior therapy versus general psyehiatric management for borderline 'personality disorder. Atn J Psychialry. 2009;166(12):1365-j374. • '

42. Linehan MM, Comtois KA, Murray AM, ct al. Two-year randomized eontrolledtrial and follow-up of dialectical behavior therapy vs therapy by experts forsuieidal behaviors and borderline personality disorder. Areh Gen Psyehiatry.2006;63(7):757-766. • , , • ' r •. i - .

43. Gicsen-Bloo J, van Dyck R, Spinhovcn P, et al. Outpatient psjjehotherapy forborderline personality disorder: randomized trial of schema-focused therapy ystransferenee-focused psychotherapy. Arch Gen Psychiatry. 2006;63(6):649-658.

44. Farrcll JM, Shaw lA, Webber MA. A sehema-focused approach to grouppsychotherapy for outpatients wjth borderline personality disorder: a randomizedcontrolled trial. J Behav Ther Exp Psychia'try. 2009;40(2):317-328.'

45. Kellogg SH, Young JE. Schema therapy for borderline personality disorder.J Clin Psychol. 2006;62(4):445-458.

46. Yeomans F. Questions concerning the randotiiized trial of sehenia-focuscdtherapy vs transference-focused psyehotherapy. Areh Gen Psychiatry.2007;64(5):609-610; author reply 610-601.

47. yan Asselt AD, Dirksen CD, Amtz A, et al. Out-patient psyehotherapy forborderline personality disorder: eost-effectiveness of schctna-foeused therapy vtransference-focused psychotherapy. Br J Psyehiatry. 2008;l92(6):450-457.

48. Davidson K, Tyrer P, Gumley A, et al. A randomized eontrolled trial ofeognitive behavior therapy for borderline personality disorder: rationale fortrial, method, and deseription of sample. J Pers Disord. 2006;20(5):431-449.

49. Davidson K, Norrie J, Tyrer P, et al. The effectiveness of cognitive behaviortherapy for borderline personality disorder: results from the borderlinepersonality disorder study of eognitive therapy (BOSCOT) trial. J Pers Disord.2006;20(5):450-465.

50. Weinberg I, Gunderson JG, Hennen J, et al. Manual assisted cognitivetreatment for deliberate self-harm in borderline personality disorder patients.J Pers Disord. 2006;20(5):482-492.

51. Tyrer P, Tom B, Byford S, et al. Differential effects of tnanual assistedcognitive behavior therapy in the treatment of reeurrent deliberate self-harm andpersonality disturbance: the POPMACT study. J Pers Disord.2004;18(l):102-116.

52. Slee N, Gamefski N, van der Leeden R, ct al. Cognitive-behaviouralintervention for self-harm: randotiiised eontroilcd trial. Br J Psychiatry.2008;192(3):202-211.

53. Connor KM, Davidson JR, Hughes DC, et al. The iinpact of borderlinepersonality disorder on post-traumatie stress in the cbmtuunity: a study of healthstatus, health utilization, and funetioning. Compr Psychiatry. 2002;43(l):41-48.

54. Emmelkamp PM, Benncr A, Kuipers A, et al. Comparison of brief dynamic andcognitive-behavioural therapies in avoidant personality disorder. Br JPsychiatry. 2006; 189:60-64. '

55. Svartberg M, Stiles TC, Seltzer MH. Randotiiized, eontrolled trial of theeffeetiveness of short-term dynamic psychotherapy and eognitive therapy forcluster C personality disorders. Atn J Psyehiatry. 2004; 161 (5):810-817.

56. Blum N, St John D, Pfohl B, et al. Systetns Training for EmotionalPredictability and Problem Solving (STEPPS) for outpatients with borderlinepersonality disorder: a randomized controlled trial and I-year follow-up. Am JPsychiatry. 2008;165(4):468-478.

57. Blum N, Pfohl B, John DS, ct al. STEPPS: a cognitive-behavioral • ,systems-based group treatment for outpatients with borderlitie personalitydisorder—a preliminary report. Compr Psyehiatry. 2002;43(4):30l-310.

58. Huband N, McMurran M, Evans C, et al. Social problem-solving pluspsychoeducation for adults with personality disorder: pragtiiatic randotuisedcontrolled trial. Br J Psychiatry. 2007; 190:307-313.

59. Zanarini MC, Frankenburg FR. A preliminary, randomized trial ofpsyehoeducation for women with borderline personality disorder. J Pers Disord.2008;22(3):284-290. •'

60. de Groot ER, Verheul R, THjsburg RW. An integrativc perspective onpsychotherapeutic treatments for borderline personality disorder. J Pers Disord.2008;22(4):332-352.

61. Livesley WJ. Prineiples and strategies for treating personality disorder. Can JPsychiatry. 2005;50(8):442-450.

62. Hellerstein DJ, Markowitz JC. Developing supportive psyehotherapy asevidenee-based treatment. Am J Psychiatry. 2008;165(10):l355-r356; authorreply 1356. • . \ - , . • '

63. Tyrer P, Kendall T, Bateman A, et al. Borderlirie persotiality disorder: treatmentand management NICE clinical guideline 78 [Internet]. London (GB): NICE;2009 [cited 2009 Jun'15]. Available from: http//:www.nice.org.uk/CG78/Guidanee/pdf/English. . . . . . . .

64. Levy KN.' Psychotherapies and lasting change. Atu J Psyehiatry.2Ó08;l'65(5)':556-559.

65. Lebow J. Twenty-first century psyehotherapies: contetiiporary approaehes totheory arid practice.'Hobokcn (NJ): John Wiley & Sons, Inc; 2008: .'

66. Wampold B. The great psyehotherapy debate: models, methods, findings.Mahwah (NJ): Lawrenee Eribauin Associates; 200 i.

Manuscript received and accepted Septeitib'er 2009.' ciinicallnstructor, Department ofPsychiatry, University of BritishColumbia, Vancouvei", British'Coliimbia.: , , ' • '

^ Associate Professor, Deíjartrnent OfPsychiatry, University of BritishColumbia, Vancouver,.British Columbia:' ' ' ',

Addreskfor correspondence:- Dr JS Ogrodniczuk, Department ofPsychiatry, (jtiiversity of British Colutribia; Suite 420, 5950 UiiiversityBoulevard, Vancouver, BC VóT 1Z3; o"grbdnic@interchange.ubc.ca

The Canadian Journal of Psychiatry, Vol 55, No 4, April 2010 209

In Review

Résumé : Psychothérapies prometteuses pour les troubles de la personnalité

Objectif : Offrir une revue sommaire de la recherche récente sur le traitement psychothérapeutiquedes patients souffrant de troubles de la personnalité (TP).

Méthode : Nous avons recherché dans PubMed et PsycINFO des articles récemment publiés quifaisaient part des résultats de traitement de psychothérapie pour les TP. Nous avons mis l'accent surles études qui utilisaient la méthodologie des essais randomisés contrôlés (ERC). La période derecherche s'étalait de janvier 2006 à juin 2009.

Résultats : L'efficacité de divers ensembles de traitements de psychothérapie pour les TP est biensoutenue par les résultats favorables des ERC. Les effets bénéfiques de la psychothérapiecomprenaient une Symptomatologie réduite, un meilleur fonctionnement social et ¡nterpersonnel, unefréquence réduite des comportements mésadaptés, et une hospitalisation réduite. Des effetséquivalents parmi les interventions que nous avons comparées étaient communs. Nombre destraitements étudiés n'exigeaient qu'une formation limitée des thérapeutes. La plupart des étudesétaient axées sur le traitement des patients souffrant du trouble de la personnalité limite (TPL).Certains résultats suggéraient que la psychothérapie était rentable; cependant, peu d'étudescomprenaient vraiment des anaiyses formeiles de coûts. Une seule étude comprenait un suivi depatients traités au-delà d'un an après le traitement.

Conclusions : II y a un appui solide à l'utilisation de la psychothérapie pour traiter les patientssouffrant de TP. Cependant, la plupart des données probantes se limitent au TPL. Les résultats desrécentes études sont prometteurs pour la formation et la pratique. La future recherche devrait sepréoccuper d'identifier les correspondances patient-traitement appropriées, d'élucider les ingrédientsactifs du traitement, et de mettre en lumière les facteurs qui sont communs aux traitements et quisont responsables des effets équivalents.

210 • La Revue canadienne de psychiatrie, vol 55, no 4, avril 2010

Copyright of Canadian Journal of Psychiatry is the property of Canadian Psychiatric Association and its content

may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express

written permission. However, users may print, download, or email articles for individual use.