Effectiveness of phase-oriented treatment for trauma-related dissociative disorders: a systematic review

ABSTRACT

Background: Dissociative disorders (DDs) are a group of mental health conditions characterised by disconnection from thoughts, feelings, memories, or sense of identity. It is believed that DDs, particularly dissociative identity disorder (DID) result from severe and chronic childhood trauma. Guidance from the International Society for the Study of Trauma & Dissociation (ISSTD) recommends trauma-informed phase-oriented treatment (PoT) to treat DDs. To date, there has been no systematic evaluation of treatment effectiveness of PoT in DDs.

Objective: This review aimed to systematically assess published studies and evaluate the evidence and effectiveness of PoT for DDs relative to other treatments.

Method: The search included four databases from 1980–2025 and reference lists of core papers. Studies were included if (1) participants have a primary diagnosis of DID or other specified dissociative disorder (OSDD), (2) it examined PoT and/or other treatments in these populations, and (3) it used at least one outcome measure pre- and post-treatment. Study quality was assessed using the Effective Public Health Practice Project (EPHPP) tool.

Results: 6088 articles were screened and 19 articles met the inclusion criteria. All treatment types were associated with improvements in symptoms of dissociation, general mental health symptoms, psychopathology, and general functioning. When compared to one another, each treatment type has its strengths depending on the outcome of interest.

Conclusions: This review demonstrates an emerging evidence base for using psychological interventions to treat DDs. High-powered controlled trials are needed to build upon these results to inform clinical practice.

HIGHLIGHTS

• Dissociative disorders such as dissociative identity disorder are considered to be severe forms of PTSD. Guidelines for treating these disorders are still in their infancy. The recommended treatment is trauma-based phase-oriented psychotherapy, but there has been no evaluation of its effectiveness.

• This systematic review summarised and evaluated 19 studies that examined the effectiveness of phase-oriented treatment for dissociative disorders as compared to other treatments.

• All types of treatment, including phase-oriented treatment, individual psychotherapy, and group therapy improved symptoms in dissociative disorders.

• High powered controlled trials are needed to inform clinical practice.

1. Introduction

1.1. Dissociation and dissociative disorders

Dissociation is the experience of disconnect from thoughts, feelings, memories, or sense of identity (American Psychiatric Association [APA], 2013; Loewenstein et al., 2024). When dissociation is severe and impairs a person’s ability to function normally it may be a symptom of a dissociative disorder (DD). There are several types of DDs and the most severe is dissociative identity disorder (DID; Reinders & Veltman, 2021). Research highlights a robust correlation between DDs and early childhood trauma (Chalavi et al., 2015; Rafiq et al., 2018). The Trauma model posits that DID is a severe post-traumatic stress disorder (PTSD) resulting from early/repeated childhood abuse (Dalenberg et al., 2012; Raison & Andrea, 2023). In this view, dissociation is a subconscious defence mechanism wherein the individual experiences disrupted integration of normal psychological functions in response to trauma, causing dissociation to protect against traumatic stress (Dimitrova et al., 2024). This coping mechanism persists beyond the offset of trauma and causes high degrees of psychiatric and psychosocial impairment. The opposite model is referred to as Fantasy model (Dalenberg et al., 2012) which states the dissociation is due to fantasy proneness and symptoms can be iatrogenically induced during treatment (Lilienfeld, 2007).

1.2. Treatments for dissociative disorders

Treatment guidelines for DDs recommended by the International Society for the Study of Trauma & Dissociation (ISSTD, 2011) include individually-focused psychotherapy as the primary treatment, with psychiatric medication as a helpful adjunct. Psychotherapy for DDs involves working through the trauma that triggers dissociation to improve understanding and coping with symptoms (Loewenstein et al., 2024). The recommended treatment modality is psychodynamic psychotherapy but can integrate other modalities too e.g. cognitive behavioural therapy (CBT), hypnosis, and group therapies. Alike treatments for other trauma-related disorders, treatment duration for DDs is generally long-term.

1.3. Phase-oriented treatment for dissociative disorders

ISSTD (2011) recommends that psychotherapy for DDs utilises a phase-oriented treatment (PoT) approach. This involves three phases: (1) stabilisation and symptom reduction; (2) treating trauma memories, and (3) identity reintegration and rehabilitation. PoT is based on structural dissociation theory (Steele et al., 2005); it suggests that individuals with trauma-related disorders and/or those experiencing pathological dissociation suffer from a division of identity into different parts, each with its own autobiographical underpinnings. The aim of PoT is to raise the integrative capacity of these identity states. Although PoT is the recommended treatment approach for DDs there are no published systematic reviews investigating its effectiveness. The Treatment of Patients with Dissociative Disorders (TOP DD) study (Brand et al., 2009) is currently the largest body of research investigating the most effective interventions for DDs. It analysed treatment outcomes of 280 DID/dissociative disorder not otherwise specified (DDNOS) patients over 30 months. They found that treatment was associated with decreased dissociative and PTSD symptomatology, depression, suicide attempts, self-harm, drug use, and hospitalisations, alongside improved social/emotional functioning.

1.4. Aims

Preliminary evidence points to the therapeutic efficacy of PoT for DDs (Subramanyam et al., 2020), but there has been no systematic evaluation of its effectiveness. Accordingly, the present systematic review aims to examine the effectiveness of PoT, as well as other treatments for DDs, on the four domains most affected in DDs: (1) dissociation, (2) general mental health symptoms, (3) psychopathology, and (4) general functioning in domains of cognitive, social, and occupational functioning.

2. Method

This systematic review follows Cochrane guidance on conducting review and Preferred Reporting Items for Systematic Review and Meta-analysis (PRISMA) guidelines (Page et al., 2021). The PICOS framework was used to determine eligibility criteria for study inclusion (Methley et al., 2014). This review protocol was pre-registered with PROSPERO (2021; CRD42021253478) with minor modifications made in February 2024.

2.1. Inclusion and exclusion criteria

Studies were included if they met the following criteria: (1) the sample included adult human participants; (2) published from 1980 onwards, the year that MPD was first mentioned in the DSM-III; (3) participants received a DSM primary diagnosis of DID, MPD, or other specified dissociative disorder (OSDD)/DDNOS; (4) examined the effectiveness of PoT for DDs, or compared PoT to other treatments, and (5) used at least one clear, measurable, and validated outcome from questionnaires, interviews, and/or tasks measuring symptoms, emotion regulation, global functioning, higher adaptive capacities, and quality of life pre- and post-treatment. No restrictions were placed on treatment type or length. Studies were excluded if they met the following criteria: (1) not published in the English language, and (2) non-peer-reviewed pieces of work.

2.2. Search strategy

A systematic search using four databases (PubMed, OVID, PsycINFO, Web of Science) was undertaken in January 2025. Grey literature was not included. The search terms used were based on a previous systematic review by one of the authors (Roydeva & Reinders, 2021), adding only keywords related to PoT. The search terms were: ‘dissociative disorder*’ OR ‘dissociative identity disorder*’ OR ‘DID disorder*’ OR ‘DID personalit*’ OR ‘DID psychiatr*’ OR ‘DID mental*’ OR ‘DID dissociat*’ OR ‘DID identit*’ OR ‘multiple personality disorder*’ OR ‘MPD’ OR ‘dissociative disorder not otherwise specified’ OR ‘DDNOS’ OR ‘other specified dissociative disorder’ OR ‘OSDD’ AND ‘treatment*’ OR ‘phase-oriented’ OR ‘outcome’ OR ‘therap*’ OR ‘longitudinal’. Databases were searched using title, abstracts, and keywords, and references were imported into Endnote. Reference lists of core papers were assessed for eligible literature.

2.3. Selection process

Searches from all databases were exported into Endnote 20 for duplicate removal by TG. The remaining references were then uploaded to Rayyan (www.rayyan.ai). Two reviewers (TG and LD) independently screened all titles and abstracts for inclusion according to the inclusion/exclusion criteria. Next, both coders independently screened all full-text studies for final inclusion. Authors were contacted to request relevant study data if not reported in-text. For both stages of screening, agreement between reviewers was calculated using percentages and Cohen’s Kappa (McHugh, 2012). Disagreements were resolved through discussion or with another member of the team.

2.4. Data extraction

Data extraction was undertaken using a prespecified spreadsheet created for the review. The following information was extracted from included studies: author(s); publication year; journal; study design; participant number; diagnosis; treatment interventions delivered; outcome measures; and study findings. Data extraction was independently conducted by the two reviewers with discrepancies resolved through discussion or consulting another member of the review team.

2.5. Quality assessment

Study quality and risk of bias was assessed using the Effective Public Health Practice Project (EPHPP; Figure 2; Armijo-Olivo et al., 2012). All papers were assessed for quality by both reviewers; disagreements were resolved through discussion or consulting another member of the team.

Figure 2.

Quality assessment ratings of included studies using the EPHPP tool (k = 19).

Notes: Study quality and risk of bias was assessed using the Effective Public Health Practice Project (EPHPP; Armijo-Olivo et al., 2012). This tool was selected as it provides a standardised means to assess quality of quantitative studies and develop recommendations from study findings. The EPHPP tool comprises six sections to be rated by reviewers: selection bias, design, confounders, blinding, data collection methods, withdrawal, and dropouts. The tool provides overall ratings of ‘strong’, ‘moderate’, or ‘weak’ in each domain. A global rating is determined for each study based on the number of sections rated as weak. EPHPP has good content and construct validity and adequate inter-rater reliability (Armijo-Olivo et al., 2012). Concerning global rating, 58% (k = 11) of studies were rated as weak, 32% (k = 6) as moderate, and 11% (k = 2) as strong. The domain ‘blinding’ produced the greatest risk of bias with 84% of studies being rated as weak. 58% of studies were rated as weak on the ‘confounders’ domain. EPHPP domains ‘study design’ and ‘selection bias’ received mostly moderate ratings. Concerning study design, most studies were rated moderate for their use of single case studies/series without a comparison/control group; only Baekkelund et al. (2022) was rated as strong for its use of an RCT. Concerning selection bias, 47% of studies were rated as strong with the remaining 53% rated as moderate. Data collection methods were an area of relative strength with all studies being rated as strong. All studies used valid/reliable outcome measures. Given the varying quality of studies and more than half receiving weak global ratings, initial conclusions can be drawn but findings need to be repeated with more methodologically robust designs.

3. Results

3.1. Study selection

A PRISMA flow diagram is shown in Figure 1.

Figure 1.

PRISMA flow diagram of steps of the systematic review (Moher et al., 2009).

Notes: 6088 articles were retrieved from the four databases, as well as 11 articles from the reference lists of core papers (see Supplementary materials). From a total of 6099 articles, 2521 duplicates were removed. 3578 articles were screened by title and abstract against inclusion/exclusion criteria. 3521 references were excluded due to failure to meet inclusion criteria, yielding 57 papers for full-text screening. 38 articles were excluded at this stage, leaving 19 articles which met eligibility criteria and included in the review. Two studies (Kellett, 2005; Ross & Haley, 2004) were included based on indirectly meeting inclusion criteria; they used DID/OSDD diagnostic instruments that were based on DSM criteria but did not specifically state that participants met DSM criteria for DID.

3.2. Study and sample characteristics

Study and sample characteristics are shown in Table 1. Studies were published between 1980–2022; 84% (k = 16) were published after 2000. Most studies (79%) took place in the U.S.A. (k = 15). All studies recruited participants from clinical settings. Concerning study design, six (32%) were uncontrolled single case studies and one was a single case experimental design. Eight (42%) were uncontrolled case series with sample sizes ranging from 3–72 participants. Two were cross-sectional studies; one was a cohort study, and one study was a randomised controlled trial (RCT).

Table 1.

Study and sample characteristics of studies (k = 19).

Study	Location	Study design	Setting	Sample (n)	Gender (% M/F)	Ethnicity	Age (M in years)	Diagnosis (comorbid diagnoses)	Diagnostic criteria	Medication
Choe and Kluft (1995)	USA	Case series	Inpatient	21	100% F	–	33.2	DID	DDIS	–
Boyd (1997)	USA	Case study	Outpatient	1	100% F	–	23	DID	SCID-D	–
Manchester et al. (1998)	USA	Case series	Outpatient	11	100% F	100% White	41.1	DID	DSM-III	–
Coons and Bowman (2001)	USA	Case series	Outpatient	12	100% F	100% White	39.6	DID	DSM-III	Unspecified anti-depressant, unspecified anti-anxiety, unspecified anti-antipsychotic
Gold et al. (2001)	USA	Case series	Outpatient	3	100% F	–	50	DID (2 PTSD, MDD, & PD; 1 BPD)	DSM-IV	–
Sar et al. (2002)	Turkey	Case study	Outpatient	1	100% F	–	18	DID	SCID-D	Paroxetine and Trazodone
Ross and Haley (2004)	USA	Case series	Inpatient	46	96% F	–	36.1	DID	DDIS	–
Humphreys et al. (2005)	USA	Case study	Outpatient	1	100% F	100% White	18	DID	DSM-IV	–
Kellett (2005)	UK	SCED	Outpatient	1	100% F	–	–	DID	NIMH RDoC	–
Gantt and Tinnin (2007)	USA	Case series	Outpatient	50	77% F	98% White; 2% not specified	38	37 DDNOS; 13 DID	DSM-IV	–
Brand et al. (2009)	USA	Cross-sectional	Outpatient	237	94% F	89% White; 2% African American; 2% Hispanic; 2% Asian; 5% Other	43.7	DID and DDNOSa	–	Unspecified anti-depressant, unspecified anti-anxiety
Myrick et al. (2012)	USA	Case series	Outpatient	29	91% F	86% White; 14% not specified	25.7	23 DID; 6 DDNOS	DSM-IV	Unspecified anti-depressant, unspecified anti-anxiety
Jepsen et al. (2014)	Norway	Case series	Inpatient	23	93% F	–	39.5	19 DDNOS; 4 DID	SCID-D	–
Brand and Loewenstein (2014)	USA	Cross-sectional	Outpatient	237	94% F	89% White; 2% African American; 2% Hispanic; 2% Asian; 5% Other	43.7	DID & DDNOS	–	Unspecified anti-depressant, unspecified anti-anxiety
Schiepek et al. (2016)	Austria	Case study	Outpatient	1	100% F	–	–	DID (BPD)	DSM-IV and SCID-D	–
Pollock et al. (2017)	USA	Case study	Outpatient	1	100% M	–	58	DID (MDD, SUD)	DSM-IV	–
Brand et al. (2019)	USA	Cohort study	Outpatient	111	88% F	86% White 4% Hispanic 2% Asian 2% Black 6% Other	43	DID & DDNOS	DSM-IV and DSM-V	–
Barbieri et al. (2022)	Italy	Case study	Outpatient	1	100% M	–	38	OSDD (PTSD & BPD)	SCID-V, DDIS	–
Baekkelund et al. (2022)	Norway	RCT	Outpatient	59	93% F	–	35.2	DID & OSDD	SCID-D	–

Notes: – = not reported. ^a = specific number of each diagnosis was not reported. BPD = borderline personality disorder; DDIS = dissociative disorders interview schedule; DDNOS = dissociative disorder not otherwise specified; DID = dissociative identity disorder; DSM = Diagnostic and Statistical Manual of Mental Disorders; MDD = major depressive disorder; NIMH RDoC = National Institute of Mental Health Research Domain Criteria; OSDD = other specified dissociative disorder; PD = panic disorder; PTSD = post-traumatic stress disorder; RCT = randomised controlled trial; SCED = single case experimental design; SCID-D = structured clinical interview for dissociative disorders; SUD = substance use disorder. Table summarising 19 included studies with columns for study, location, design, setting, sample size, gender, ethnicity, age, diagnosis and comorbidities, diagnostic criteria, and medication details.

619 adult participants were included, accounting for the fact that two studies analysed different outcomes from the same TOP DD study sample. Sample sizes ranged from 1–237 (median = 17). Most participants were female (91%) and ages ranged from 18–72 years. Ten studies did not report ethnicity; among those that did, most participants were White. All participants had primary diagnoses of either DID or OSDD. Three studies included participants with comorbid PTSD, three with depression, two with borderline personality disorder, and one each with panic disorder or substance use disorder. Most studies used diagnostic criteria from the DSM, including DSM-III (k = 2), DSM-IV (k = 7), DSM-IV SCID-D (k = 5), and DSM-V (k = 2). Three studies used the Dissociative Disorders Interview Schedule (DDIS) and one used the National Institute of Mental Health Research Domain Criteria (NIMH RDoC) for DID. 26% (k = 5) of studies included participants that were prescribed psychotropic medication, the most common being anti-depressants. 14 studies did not record or report medication use.

3.3. Quality assessment

EPHPP ratings are presented in Figure 2.

3.4. Outcome measures

Outcome measures and timepoints of administration are presented in Table 2. A total of 29 outcome measures were used across studies. These can be categorised into four domains of outcome measures assessing: dissociation, general mental health, psychopathology symptoms, and general functioning.

Table 2.

Outcome measures and timepoints of administration (k = 19).

Study	Outcome measure				Time-points
	Dissociation	General mental health symptoms	Psychopathology	General functioning
Choe and Kluft (1995)	DES	–	–	–	Pre, post
Boyd (1997)	DES	MMPI	–	–	Pre, post
Manchester et al. (1998)	DES	–	MCMI	GAF	Pre, post, 12–36mfu
Coons and Bowman (2001)	DES, DDFQ	–	BDI, PCL-5	–	Pre, 10-year-fu
Gold et al. (2001)	DES	MMPI	BDI, IES-R	–	Pre, post
Sar et al. (2002)	DES	–	–	CFIT	Pre, post
Ross and Haley (2004)	DES	SCL-90-R	BHS, BSSI	–	Pre, post, 3mfu
Humphreys et al. (2005)	DES	–	–	–	Pre, post
Kellett (2005)	DES, SSD	BSI	BDI	IIP-32, PSQ	Pre, post, 6mfu
Gantt and Tinnin (2007)	DES, DRS	SCL-45	IES-R, TAS	–	Pre, post, 3mfu, 6mfu
Brand et al. (2009)	DES	SCL-90-R	PCL-5	GAF, PITQ	5 time-points across therapy
Myrick et al. (2012)	DES	SCL-90-R	PCL-5	PITQ	Pre, 6mfu, 18mfu, 30mfu
Jepsen et al. (2014)	DES	SCL-90-R	BDI, IES-R	IIP-64	Pre, post, 12mfu
Brand and Loewenstein (2014)	DES	SCL-90-R	–	GAF	4 time-points across therapy
Schiepek et al. (2016)	Synergetic Model Feedback				Throughout treatment
Pollock et al. (2017)	DES	–	BAI, BDI, ERSQ	–	Pre, 3 timepoints across 14 m
Brand et al. (2019)	DES	–	DERS, PCL-5	PITQ	Pre, 12mfu, 24mfu
Barbieri et al. (2022)	DES	–	PCL-5	YSQ	Pre, mid-treatment, post, 6mfu
Baekkelund et al. (2022)	DES, MID	SCL-90-R	PTSD-SS	GAF, IIP-32,	Pre, post, 6mfu

Notes: BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; BHS = Beck Hopelessness Scale; BSI = Beck Suicide Inventory; BSSI = Beck Scale for Suicidal Ideation; CFIT = Culture Fair Intelligence Test; DDFQ = Dissociative Disorders Follow-up Questionnaire; DES = Dissociative Experiences Scale; DERS = Difficulties in Emotion Regulation Scale; DRS = Dissociation Regression Scale; ERSQ = Emotion Regulation Skills Questionnaire; FU = Follow-up; GAF = Global Assessment of Functioning; IES-R = Impact of Event Scale-Revised; IIP-32 = Inventory of Interpersonal Problems 32; IIP-64 = Inventory of Interpersonal Problems 64; M = Month; MCMI = Million Clinical Multiaxial Inventory; MFU = Month follow-up; MMPI = Minnesota Multiphasic Personality Inventory; MID = Multidimensional Inventory of Dissociation; PSQ = Personality Structure Questionnaire; PCL-5 = PTSD Checklist for DSM-V; PTSD-SS = PTSD Symptom Scale; PITQ = Progress in Treatment Questionnaire; SCL-45 = Symptom Checklist-45; SCL-90-R = Symptom Checklist-90-Revised; SSD = State Scale of Dissociation; TAS = Toronto Alexithymia Scale; YSQ = Young Schema Questionnaire. Table summarising outcome measures used in 19 studies, listing dissociation, mental health, psychopathology, general functioning measures, and timing of assessments such as pre, post, and follow-up.

Dissociation. Dissociation was assessed in 95% (k = 18) of studies. Dissociation was assessed using five outcome measures, including the Dissociative Experiences Scale-II (DES-II) which was used in all studies measuring dissociation. Single studies used the Dissociative Regression Scale, Multidimensional Inventory of Dissociation, State Scale of Dissociation, and one ‘dissociative disorders follow-up questionnaire’ created by Coons and Bowman (2001) for the study.

General mental health symptoms. General mental health symptoms were assessed in 53% (k = 10) of studies; seven used the Symptoms Checklist-Revised; two used the Minnesota Multiphasic Personality Inventory, and one used the Brief Symptoms Inventory.

Psychopathology. 13 studies assessed specific mental health symptoms; nine used PTSD-specific measures such as the Impact of Event Scale-Revised and PTSD Checklist. Six studies assessed depression using the Beck Depression Inventory and Beck Hopelessness Scale. A small number of studies assessed emotion regulation skills (k = 2), anxiety (k = 1), alexithymia (k = 1), personality disorder (k = 1), and suicidal ideation (k = 1).

General functioning. General functioning i.e. cognitive, social, and occupational functioning was assessed in 47% (k = 9) of studies. Four studies used the Global Assessment of Functioning and three used the Progress in Treatment Questionnaire. Single studies used the Personality Structure Questionnaire and Young Schema Questionnaire and the Inventory of Interpersonal Problems (IIP). One study measured cognitions, emotions, and behaviours using a personalised questionnaire based on an idiographic system model, and one measured IQ using Cattell’s (1949) Culture Fair Intelligence Test.

3.5. Overview of treatment interventions

Treatment interventions are presented in Table 3. Individual psychotherapy (k = 16) was the most common, followed by group therapies (k = 10).

Table 3.

Treatment approach, length, frequency, and facilitator (k = 19).

Study	Treatment	Treatment length	Treatment frequency	Treatment facilitator
Choe and Kluft (1995)	Psychodynamic psychotherapy Hypnosis Group (art, cognitive, movement, verbal)	23 days	Individual: 5 sessions a week Group: up to 12 times a week	–
Boyd (1997)	Three-stage phase-oriented psychotherapy: Phase 1: assessment, stabilisation, mapping the system Phase 2: reducing trauma and dissociation symptoms, integrating identities, hypnosis Phase 3: relationships, individuation, finances and career	Stage 1: 5 sessions over 17 days Stage 2: 54 sessions over 10 months Stage 3: 52 sessions over 12 months	–	Clinical psychologist
Manchester et al. (1998)	Neurofeedback training Neurofeedback group	13–27 months Training: 30 sessions Group: 10 sessions	–	–
Coons and Bowman (2001)	Psychotherapy, journal writing, art therapy, psychotropic medication, hypnosis, counselling, psychoeducation, group, taping, hospitalisation, day therapy, vocational rehabilitation, marital therapy, electric convulsive therapy, family therapy, dance therapy, exorcism	–	1–1.5 sessions a week	Therapists
Gold et al. (2001)	Contextual therapy (interpersonal relating, self-understanding, skills transmission)	8–30 months	–	Novice therapists
Sar et al. (2002)	Psychotherapy (integrative, target trauma cognitions, contextual systemic factors)	18 months	–	–
Ross and Haley (2004)	Trauma model therapy: Individual and group therapy (attachment patterns, self-blame, correct cognitive errors and maladaptive beliefs, coping skills)	Mean 18.2 days	Individual: 3 h a week Group: 35 h a week	–
Humphreys et al. (2005)	Humanistic therapy (assimilation model of identities)	200 sessions over 3 years	–	Student therapist
Kellett (2005)	Cognitive analytic therapy	28 sessions over 53 weeks	–	–
Gantt and Tinnin (2007)	Intensive trauma treatment programme: Psychotherapy (psychoeducation, trauma processing, reversing dissociation, video recording/written tasks, modification of victim mythology) Group (art therapy)	1–2 weeks	7 h a day, 5 days a week	–
Brand et al. (2009)	Psychotherapy, group, family therapy, couples therapy, art therapy, other expressive therapies	Mean 5 years	–	Therapists
Myrick et al. (2012)	Psychotherapy, group, family therapy, couples therapy, art therapy, other expressive therapies	30 months	–	Therapists
Jepsen et al. (2014)	Specialised inpatient treatment programme for trauma-related symptoms in childhood sexual abuse survivors: Psychotherapy, group, movement therapy, expressive art, occupational therapy, physical training	3 months	Individual: 1–2 sessions a week Group: two sessions daily	Psychiatrists, clinical psychologists, nurses, occupational and art Therapists, social workers
Brand and Loewenstein (2014)	Psychotherapy, group, family therapy, couples therapy, art therapy, other expressive therapies	Mean 5 years	–	Therapists
Schiepek et al. (2016)	Synergetic process management therapy (resource-focused interview, idiographic system model, individual psychotherapy, aftercare)	19 weeks	–	–
Pollock et al. (2017)	Three stage phase-oriented psychotherapy: Phase 1: psychoeducation, reduce self-destructive behaviour, self-care, increase safety Phase 2: processing trauma memories Phase 3: integration and maintaining identity, relearning relationship skills, tolerate stress, future goals	14 months Phase 1: 11 sessions Phase 2: 12 sessions Phase 3: 7 sessions	–	–
Brand et al. (2019)	Web-based psychoeducation programme (psychoeducation videos, video/written/behavioural exercises, symptom and emotion management techniques)	24 months	–	–
Barbieri et al. (2022)	Phase-oriented schema therapy (integrating CBT and NET techniques, skills training, expressive modalities)	85 sessions over 26 months	–	Therapists
Baekkelund et al. (2022)	Psychoeducation group	20 sessions	–	Therapists

Notes: – = not reported. CBT = cognitive-behavioural therapy; DDD = depersonalisation/derealisation disorder; NET = narrative exposure therapy. Table summarising 19 studies with details on treatment type, duration, session frequency, and professional facilitator roles such as therapists or psychologists.

Individual psychotherapy. 84% (k = 16) of studies used individual psychotherapy, with 8 types being used. PoT was most common (k = 3). Two studies delivered three-stage PoT in line with ISSTD guidelines (Boyd, 1997; Pollock et al., 2017); one used phase-oriented schema therapy (Barbieri et al., 2022). Three studies used trauma-focused psychotherapy: one based on attachment theory and CBT (Ross & Haley, 2004), one as part of an intensive treatment programme (Gantt & Tinnin, 2007), and the third as part of a specialised inpatient programme for childhood sexual abuse survivors (Jepsen et al., 2014). Single studies used other individual psychotherapies, including psychodynamic psychotherapy (Choe & Kluft, 1995), cognitive analytic therapy (CAT; Kellett, 2005), contextual therapy (Gold et al., 2001), psychoeducation (Brand et al., 2019), and humanistic therapy (Humphreys et al., 2005). Five studies used individual psychotherapy but did not specify modality.

Group therapy. 53% (k = 10) of studies used group therapies, with nine types. Five studies did not specify the type/focus of their intervention (Brand et al., 2009; Brand & Loewenstein, 2014; Coons & Bowman, 2001; Myrick et al., 2012; Ross & Haley, 2004). Four studies used group interventions for couples/families (Brand et al., 2009; Brand & Loewenstein, 2014; Coons & Bowman, 2001; Myrick et al., 2012), and two used art therapy groups (Choe & Kluft, 1995; Gantt & Tinnin, 2007). Single studies used other group therapies: cognitive and movement (Choe & Kluft, 1995), neurofeedback (Manchester et al., 1998), trauma-focused (Jepsen et al., 2014), psychoeducation (Baekkelund et al., 2022), and dance groups (Coons & Bowman, 2001). Nine studies used group therapies alongside other interventions, only one study used group therapy as the sole intervention (Baekkelund et al., 2022). Five studies used multiple group therapies concurrently (Brand et al., 2009; Brand & Loewenstein, 2014; Choe & Kluft, 1995; Coons & Bowman, 2001; Myrick et al., 2012).

Other miscellaneous treatments. 53% (k = 10) of studies used other treatments, comprising 19 types. One study contributed to a large proportion (12 treatments) of these (Coons & Bowman, 2001). Hypnosis was used as an adjunct in 15% (k = 3) of studies (Boyd, 1997; Choe & Kluft, 1995; Coons & Bowman, 2001). Single studies used treatments such as neurofeedback training (Manchester et al., 1998), web-based psychoeducation programme (Brand et al., 2019) synergetic process management therapy (Schiepek et al., 2016) and others. Three studies reported ‘other expressive therapies’ without providing specific modality information (Brand et al., 2009; Brand & Loewenstein, 2014; Myrick et al., 2012).

Treatment length and frequency. All but one study (Coons & Bowman, 2001) provided treatment length information, ranging from seven days to five years. The exception reported 1–1.5 treatment sessions per week. 26% (k = 5) of studies provided treatment frequency. Individual treatment sessions ranged from 1 to 5 weekly; group therapy session varied from two daily (12 times a week) up to 35 h a week.

3.6. Effectiveness of treatments

Clinical outcomes and effect sizes are presented in Table 4. Outcomes of interest used to infer treatment effectiveness were dissociation symptoms, mental health symptoms, and general functioning; outcome measures with the most available data from studies are presented in Figure 3.

Table 4.

Clinical outcomes of the treatment approaches (k = 19).

Treatment	Study	Dissociation outcomes	Mental health outcomes	General functioning outcomes
Phase-oriented treatment	Boyd (1997)	Decrease in DES from 48.2 at assessment to 2.0 at discharge. Self-reported identity unification. No longer met SCID-D DID diagnosis.	–	Improvements in quality of interpersonal relationships and subjective wellbeing/life functioning.
	Pollock et al. (2017)	No longer met DSM-IV DID diagnosis but still experiencing dissociation symptoms.	Improvements in depression and anxiety but still experiencing clinically significant symptoms.	Improvements in emotion regulation skills and reduced alcohol. Unchanged tiredness, low energy, sleep difficulties, and little interest in sex.
	Barbieri et al. (2022)	Improvements post-treatment.	Improvements in PTSD symptoms.	Improvements in schema domains disconnection/rejection, autonomy/performance, and vigilance/inhibition. Schema domain disconnection/rejection returned back to pre-treatment level at follow-up. Unchanged tiredness, low energy, and little interest in sex.
Individual psychotherapy	Gold et al. (2001)	Improvements post-treatment.	Improvements in PTSD and depression symptoms.	Self-reported improvements in self-sufficiency, employment, and relationships.
	Sar et al. (2002)	Successful integration of alters.	Decrease in traumatic content scores.	Increased control over affective responses to environment and increased IQ scores from 90 to 124.
	Kellett (2005)	Improvements that were maintained and augmented at 6-month follow-up. Self-reported decreased sense of fragmentation.	Improvements in depression and OCD symptoms that were maintained at 6-month follow-up.	Self-reported improved ability to self-manage and capacity to work.
	Humphreys et al. (2005)	Successful integration of alters.	–	–
Multiple interventionsx	Choe and Kluft (1995)	Improvements in absorption, changeability, and depersonalisation/derealisation scores, but increase in amnesia.	–	–
	Coons and Bowman (2001)	Increase in identity unification. 33% became fully integrated. Decrease in number of identity states among those who remained unintegrated.	Integrated participants showed improvements in PTSD, depression, and somatoform symptoms. No improvements among unintegrated participants.	–
	Ross and Haley (2004)	Improvements post-treatment (d = 0.63).	Improvements in overall psychological distress (d = 1.25), depression (d = 1.66), hopelessness (d = 1.17), and suicidal ideation (d = 1.12). Reductions in the number of suicide attempts and hospitalisations. Reductions were all sustained at 3-month follow-up except for hopelessness (d = −0.05).	–
	Gantt and Tinnin (2007)	Improvements post-treatment.	Improvements in overall psychological distress, PTSD, and alexithymia symptoms	–
	Brand et al. (2009)	Improvements post-treatment (d = 0.64).	Improvements in overall psychological distress (d = 0.71) and PTSD (d = 1.00) symptoms, but still experiencing clinically significant levels of PTSD. Reductions in self-harm, suicidal ideation/attempts, and hospitalisations.	Improved ability to manage affect, impulses, symptoms, and relationships (d = 1.61). Better social and work functioning.
	Myrick et al. (2012)	Improvements post-treatment.	Improvements in PTSD symptoms. Reductions in self-harm, suicidal ideation/attempts, and hospitalisations. Younger adults improved at a quicker pace than older adults.	–
	Jepsen et al. (2014)	No effect (d = 0.09).	Small improvements in overall psychological distress (d = 0.28) and depression (d = 0.19); medium improvements in PTSD symptoms (d = 0.59).	–
	Brand and Loewenstein (2014)	Improvements in voice hearing of dissociated self-states, including subjective self-division and hearing voices of self-states.	Improvements in overall psychological distress.	Participants became more functional over the course of treatment.
Other treatmentsy	Manchester et al. (1998)	100% reported identity unification. Dissociation scores within normal parameters at one-year follow-up.	All participants free from PTSD symptoms post-treatment and these improvements sustained at follow-up. Group process of treatment made them feel less alone, less isolated, and instilled sense of hope.	–
	Schiepek et al. (2016)	–	–	Positive effects on identity development and autonomy in life.
	Brand et al. (2019)	Medium improvements (d = 0.48). High dissociation group showed large improvements (d = 0.81); low dissociation group showed small improvements (d = 0.24).	Medium improvements in PTSD symptoms (d = 0.65) and reduced reports of self-injury. High dissociation group showed large improvements in PTSD symptoms (d = 0.93); low dissociation group showed small improvements (d = 0.32).	Large improvements in emotion regulation (d = 0.90) and adaptive capacities (d = 0.86). High dissociation group showed large improvements in emotion regulation (d = 1.04) and adaptive capacities (d = 0.94). Low dissociation group showed medium-approaching-large improvements in emotion regulation (d = 0.74) and adaptive capacities (d = 0.75). Most participants reported ongoing symptoms that required further treatment.
	Baekkelund et al. (2022)	Minimal effect post-treatment (d = −0.10) and at 6-month follow-up (d = −0.10).	Minimal effect on overall psychological distress (d = 0.10) and no effect on PTSD symptoms (d = 0.0) post-treatment. Minimal effect on overall psychological distress (d = 0.10) and PTSD symptoms (d = −0.10) at 6-month follow-up.	Small-approaching-medium effect on adaptive functioning post-treatment (d = 0.4) but large effect at 6-month follow-up (d = 1.10). Most participants reported ongoing symptoms that required further treatment.

Notes: – = not reported. DES = dissociative experiences scale; DID = dissociative identity disorder; DSM-IV = Diagnostic and Statistical Manual of Mental Disorders, 4th Edition; IQ = Intelligence quotient; OCD = obsessive compulsive disorder; PTSD = post-traumatic stress disorder; SCID-D = Structured Clinical Interview for Dissociative Disorders; ^x = Psychotherapy, journal writing, art therapy, mediation, hypnosis, counselling, psychoeducation, group, taping, hospitalisation, day therapy, vocational rehabilitation, marital therapy, electric convulsive therapy, family therapy, dance therapy, exorcism, other expressive therapies; ^y = neurofeedback training, web-based psychoeducation programme; psychoeducation-based group, synergetic process management therapy. Table summarising dissociation, mental health, and general functioning outcomes reported in 19 included studies, grouped by treatment type.

Figure 3.

Pre- and post-treatment outcomes for measures of dissociation, general mental health, psychopathology, and general functioning.

Notes: Outcome measures depicted represent one study with the most available data from each of the four outcome domains, namely dissociation (a), general mental health (b), psychopathology (c), and general functioning (d). Dissociation outcome is indexed using the Dissociation Experiences Scale (DES; k = 15); general mental health is indexed using the Symptom Checklist-90-Revised (SCL-90-R; k = 5); psychopathology (depression) is indexed using the Beck Depression Inventory (BDI; k = 6); and general functioning is indexed using the Global Assessment of Functioning (GAF; k = 4). Standard errors are not presented when they were not reported in the citing paper.

Phase oriented treatment. The three studies that delivered PoT all reported improvements in dissociation. Two found that participants self-reported personality unification and no longer met DID criteria, but were still experiencing dissociation post-treatment. The two studies that assessed mental health symptoms found improvements in depression, anxiety, and PTSD. One study found that participants were still experiencing depression and PTSD symptoms post-treatment. The three studies that assessed general functioning found improvements in quality of interpersonal relationships, subjective wellbeing, emotion regulation, and alcohol use. One study found improvements in schema domains of disconnection/rejection, autonomy/performance, and vigilance/inhibition. However, one study found unchanging symptoms of tiredness and little interest in sex, and another found that some cognitive domains, such as disconnection/rejection, returned back to pre-treatment levels at 6-month follow-up.

Individual psychotherapy. The four studies that delivered individual psychotherapies all found improvements in dissociation, and two studies reported personality unification. Concerning mental health symptoms, contextual therapy improved post-traumatic and depression symptoms, integrative psychotherapy decreased traumatic content scores, and CAT produced improvements in depression and obsessive-compulsive disorder (OCD) symptoms that were sustained at 6-month follow-up. For general functioning, contextual therapy improved self-sufficiency, relationships, and employment, CAT improved ability to self-manage and capacity to work, and integrative psychotherapy improved control over affective response and IQ. Contextual therapy produced improvements within timeframes that are considered brief compared to the three-to-five year course of treatment commonly projected for DID (Myrick et al., 2017).

Multiple interventions. Seven out of eight studies found positive effects of multiple interventions on dissociation. One study found reductions in absorption, changeability, and depersonalisation/derealisation scores, but an increase in amnesia. One study reported reduced voice hearing of dissociated self-states, and another found increased identity unification. One study found that 33% of participants became fully integrated, and for those who remained unintegrated, the number of identities decreased from a mean of 15–7. They also found that participants who became integrated showed improvements in depression, somatoform and PTSD symptoms whereas unintegrated participants did not. However, the specialised inpatient trauma programme found little effect on dissociation during the treatment period but larger improvements at follow-up. The six studies that assessed mental health symptoms all found significant improvements. PTSD was the most commonly improved symptom with five studies reporting improvements. Three studies found improvements in depression symptoms, and three studies found reductions in self-harm, suicidal ideation/attempts and hospitalisations. Single studies found improvements in somatoform and alexithymia symptoms. One study found large effect sizes on all domains except for hopelessness were sustained from post-treatment to follow-up. One study only found improvements in mental health among those who became fully integrated. Concerning general functioning, the two studies that assessed for this found improvements in ability to manage affect, relationships, and social/work functioning with large effect sizes. Three studies found that participant age was a predictor of treatment outcomes: younger adults were more impaired pre-treatment but responded to treatment at a quicker pace.

Other treatments. From the three studies that delivered other treatments, two found positive outcomes on dissociation symptoms. One study found that neurofeedback training led to all participants being unified post-treatment. Another study found moderate improvements in dissociation symptoms after a 24-month web-based psychoeducation programme. However, another study found that a 20-session psychoeducation-based group had little effect on dissociation post-treatment and at 6-month follow-up. Concerning mental health symptoms, the same three studies all reported improvements in PTSD symptoms. However, one study found minimal effect of the psychoeducation-based group on overall psychological distress. Concerning general functioning, the web-based psychoeducation programme improved emotion regulation and adaptive capacities with large effect sizes. Synergetic process management therapy produced positive outcomes on identity development and autonomy, and the psychoeducation group found small-approaching-medium effects on psychosocial functioning post-treatment but large effects at 6-month follow-up.

A subjective assessment of the data indicate that all treatment types produced clinically significant improvements in at least one of the four assessed domains from pre- to post-treatment. When comparing treatment types, a visual inspection of the available data presented in Figure 3 suggests that PoT and individual psychotherapy produced the strongest effects on domains of dissociation and psychopathology (depression) from pre- to post-treatment. Multiple interventions and other treatments produced less pronounced effects on these two domains but stronger effects on the domain of general functioning.

4. Discussion

The review summarises the findings of 19 treatment studies, wherein three evaluated the efficacy of phase-oriented treatment (PoT) in dissociative disorders (DDs). Comparatively, four studies delivered individual psychotherapies, eight delivered multiple interventions concomitantly, and four used other treatments. Our most important finding is that all studies, irrespective of treatment type, found clinically significant improvements in at least one of the four assessed domains (dissociation, mental health, psychopathology, and general functioning) when comparing pre- to post-treatment.

A visual assessment of the data indicate that each treatment type has its strengths depending on the outcome of interest. The most significant improvements appear for PoT and individual psychotherapy, which produced the most favourable outcomes on dissociation and psychopathology respectively; this is in line with these treatments focusing on symptom stabilisation and trauma processing (ISSTD, 2011). Multiple interventions were most effective for general mental health, while other treatment approaches such as group interventions better targeted general functioning; this makes logical sense given that this domain taps into cognitive, social, and occupational functioning. The only RCT in the review found minimal effects on outcomes post-treatment but reported positive outcomes on mental health and general functioning at 6-month follow-up. Although this was a psychoeducation-based group as opposed to a full course of psychotherapy, these findings provide evidence against the iatrogenic theory that DID treatment is not only ineffective but harmful and dissociative symptoms would increase instead of decrease (Lilienfeld, 2007). However, most studies were uncontrolled case studies/series without a comparison/control group, so empirical conclusions regarding treatment efficacy should be interpreted with caution.

Concerning maintenance of treatment gains and longer-term outcomes, all studies except one found sustained or augmented improvements at follow-up. However, the only study to deliver PoT and collect follow-up outcomes found that improvements in schema domains of disconnection/rejection returned back to pre-treatment levels at 6-month follow-up. Nevertheless, it is important to note the nonlinearity of treatment especially of DID patients. Due to the phasic nature of treatment, many find themselves re-traumatised from new emerging knowledge about the traumatic past. The therapist explores the patient's unconscious trauma-related knowledge which may be a cause of maladaptive functioning until that information is integrated. Integrating traumatic memories brings together the different aspects of traumatic experiences, memories and sequence of events, associated affects and physiological and somatic representations (Subramanyam et al., 2020). In doing so, progress assessed through outcome measures at a short 6-month follow up does not provide an adequate representation, requiring longer timeframes of assessment. Nonetheless, this review demonstrates that DD treatment broadly is not only efficacious but can be sustained and even augmented over time. This supports the hypothesis that individuals with DDs may require more time to consolidate therapeutic material and show improvements (Jepsen et al., 2014).

Few studies reported on treatment acceptability. Four studies reported dropout rates ranging from 3–23%. Two studies gathered qualitative feedback from participants; this was extremely positive in both studies. Baekkelund et al. (2022) found that dropout was higher for participants with higher dissociation scores pre-treatment. Concerning qualitative feedback, Manchester et al. (1998) completed post-treatment interviews and 100% of participants reported that the group intervention made them feel less alone/isolated and instilled hope for recovery. Brand et al. (2019) received feedback from participants and therapists that the web-based psychoeducation programme content was relevant, clear, and useful for participants struggling with safety and symptom stabilisation. Though limited, this information suggests acceptability of DD treatment. This is of interest as many patients with a dissociative pathology share a history of years of misdiagnoses, inefficient treatment and repeated hospitalisations (Reinders et al., 2019). Evidence indicates that pathological dissociation is a negative predictor of psychotherapy response in various pathological samples (Lanius, 2015; Spitzer et al., 2007), which can be attributed to years of incorrect treatment, protracted personal suffering and high direct and indirect societal costs (Myrick et al., 2017; Reinders et al., 2019) as well as inadequate professional training and management of DDs (Chien & Fung, 2022). This maintains the need for reputable research evidence and continued research support in the field of DDs to educate and highlight the complexity trauma-related pathological dissociation.

Despite positive outcomes, most participants across studies continued to experience clinically significant symptoms post-treatment that required further intervention. Complex presentations and extensive comorbidities are common in DDs, which results in diagnostic and therapeutic challenges (Chien & Fung, 2022). This is unsurprising given that individuals with DDs typically experience years of childhood abuse, high degrees of impairment, and require many years of treatment (Raison & Andrea, 2023). Moreover, many individuals never achieve complete identity reintegration but strive towards optimal integration so they can have adequate daily functioning (Pais, 2009). Four studies in this review achieved positive outcomes in short timeframes and are of specific interest for the patient’s quality of life and cost-effectiveness.

4.1. Strengths and limitations

This is the first systematic review to investigate treatment approaches for DID/OSDD and the efficacy of PoT more broadly. A limitation of the current review was that most included studies were case studies/series and this lack of control/comparison groups means it is not possible to causally attribute symptom reduction to treatment. Such designs are inherently limited in terms of internal and external validity, which compromises the strength and generalisability of the conclusions drawn. However, given that DDs typically persist for decades (Loewenstein et al., 2024) it seems implausible that patients would improve so quickly due to spontaneous remission/other factors. Another limitation is that only three studies evaluated the effects of PoT. While all studies used pre- and post-treatment outcome measures, each study used different measures for different treatments in different settings. Therefore, results of this study are based on careful inspection of visual interpretations of the data. Methodological limitations are corroborated by EPHPP ratings; 11 studies were rated as weak and only two as strong.

4.2. Future research

The next step is to conduct high quality RCTs in order to (1) further assess treatment efficacy, (2) build an evidence base for how best to treat DDs, and (3) support the use of treatment in clinical practice. Future studies should include the voices of those with DDs by collecting post-treatment feedback via qualitative interviews or measures of treatment acceptability. The economic costs associated with DDs and their treatment is understudied (Myrick et al., 2017; Reinders et al., 2019). Future research should synthesise the evidence on the clinical effectiveness and economic outcomes of DD treatment to guide practice and policy decisions. Furthermore, it has been recommended that brain imaging be used to study the neurobiological correlates of treatment intervention. In turn, results can guide innovative psychological and pharmacological treatment approaches (Purcell et al., 2024; Reinders et al., 2022, 2023).

5. Conclusions

This systematic review explored the efficacy of PoT and other treatments for individuals with DID/OSDD. The study found that all treatment types produced clinically significant improvements in at least one domain of dissociation, psychopathology, mental health, and general functioning, and these were largely sustained/augmented at follow-up. As such, this study provides emerging evidence to support the use of psychological interventions, including PoT, for DDs.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The data supporting the findings of this systematic review are available in the published articles found in the reference list and supplementary materials.

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/20008066.2025.2545734.