How is EMDR best conducted and adapted for trauma-focused work within bipolar disorder? A Delphi study

ABSTRACT

Background: Individuals with bipolar disorder are at increased risk of exposure to traumatic events and associated psychological distress, including affective, cognitive, physiological, and mood-related difficulties. Eye movement desensitisation and reprocessing (EMDR) targets unprocessed trauma memory networks underlying these difficulties; however, guidance on adapting EMDR for this population is limited. This study aimed to establish practitioner consensus on adaptations to enhance the accessibility and effectiveness of EMDR for trauma in individuals with bipolar disorder.

Methods: A three-round Delphi study was conducted with an international sample of EMDR practitioners (34 in Round 1, 22 in Round 2, 20 in Round 3). Round 1 responses were thematically analysed to generate 131 statements for Round 2. Statements with ≥80% agreement as ‘essential’ or ‘important’ achieved consensus, while those with 60–79% agreement were re-rated in Round 3 to clarify areas of uncertainty.

Results: Practitioners identified key barriers to EMDR, including client-related factors (e.g. mood variability, inconsistent attendance), safety concerns (e.g. risk of triggering mood shifts), systemic issues (e.g. service inflexibility), and limited guidance. Consensus was reached on 71 EMDR adaptations, spanning general strategies (e.g. mood monitoring, safety planning, pacing, coping skills) and phase-specific modifications (particularly in preparation, assessment, and desensitisation). Fifteen supervision elements were also endorsed, emphasising the importance of support for clinical decision-making.

Conclusions: Trauma-focused EMDR for individuals with bipolar disorder can be delivered using the standard eight-phase protocol with phase-specific, formulation-driven adaptations. Key recommendations include structured preparation, ongoing mood monitoring, flexible pacing, and attention to safety. These findings provide practical guidance for clinicians and highlight the need for further training and research.

HIGHLIGHTS

• An international study gathered expert agreement on how trauma therapy can be safely adapted for people living with bipolar disorder.

• Key recommendations emphasise careful preparation, mood stability, and emotional safety throughout trauma treatment.

• Improved training and specialist supervision may increase access to effective trauma therapy for individuals with bipolar disorder.

1. Introduction

Bipolar disorder is characterised by a pattern of mood experiences, including periods of stability alongside episodes of depression and mania, affecting approximately 2% of the adult population (Aniserowicz et al., 2022; Quidé et al., 2020). Manic episodes involve elevated mood states, including euphoria, racing thoughts, reduced sleep, irritability, and impulsivity, while depressive episodes include loss of interest, fatigue, hopelessness, cognitive difficulties, and suicidality (World Health Organization [WHO], 2022).

These experiences can significantly impact functioning and quality of life and are a leading contributor to global disability (Grande et al., 2016). Mood patterns are often recurrent (Maj, 2000) and associated with increased risk of premature mortality, including suicide (Chan et al., 2022). Bipolar presentations are heterogeneous, and many individuals report trauma histories and post-traumatic stress symptoms, alongside other anxiety-related difficulties, further affecting psychosocial functioning (McIntyre et al., 2004; Spoorthy et al., 2019; Maina et al., 2011). Difficulties in social engagement and challenges in interpersonal relationships are also common (Bauwens et al., 1998).

1.1. Bipolar disorder and PTSD

PTSD arises following exposure to events involving actual or perceived threat to life or safety, and is characterised by intrusion, avoidance, negative alterations in mood and cognition, and hyperarousal (American Psychiatric Association [APA], 2013).

Between 60% of individuals with bipolar disorder report trauma exposure (Bedeschi, 2018), with post-traumatic stress symptom prevalence estimates ranging from 16 to 39% (Otto et al., 2004), although these are challenging to determine due to overlapping symptoms and limitations in assessment tools. Some studies suggest this may exceed 50%, influenced by factors such as childhood abuse and neglect, higher engagement in risk-taking behaviours, and involuntary hospitalisations (Katz et al., 2020). Traumatic experiences may also influence the development and course of both post-traumatic stress and bipolar disorder over time (Russell et al., 2024).

Research supports a link between childhood trauma and bipolar presentations (Misiak et al., 2018). An umbrella review by Hogg et al. (2023) found that childhood trauma is associated with an increased risk of developing mental disorders, including bipolar disorder. Bortolato et al. (2017) argue that childhood adversity is one of the most substantial risk factors for the development of the disorder. Palmier-Claus et al. (2016) showed that individuals with bipolar disorder are 2.6 times more likely to encounter childhood adversity compared to a non-clinical control group with no premorbid diagnoses.

Trauma exposure in individuals with bipolar disorder (even without a formal PTSD diagnosis) is associated with more severe symptoms and a higher frequency of manic episodes (Agnew-Blais & Danese, 2016). Goodman et al. (2001) highlight the importance of addressing adverse experiences, as co-occurring trauma is linked with more pronounced low mood and mania compared to those without trauma-related distress. Conversely, periods of mania can increase engagement in risk-taking behaviours, raising the likelihood of further traumatic experiences and potential post-traumatic stress symptoms (Otto et al., 2004). For some individuals, low mood, mania, and psychotic-like experiences may also be understood as part of trauma-related distress, and these experiences can co-occur or interact in complex ways.

1.2. Evidence-based psychological interventions for PTSD

The National Institute for Health and Care Excellence (NICE) recommends both cognitive behavioural therapy (CBT) and eye movement desensitisation and reprocessing (EMDR) as first-line, evidence-based psychotherapies for the treatment of PTSD in children and adults (NICE, 2018). A meta-analysis of 90 trials (6560 participants) demonstrated that both trauma-focused CBT (TF-CBT) and EMDR were most effective in reducing symptoms and promoting recovery in adults experiencing trauma-related difficulties (Mavranezouli et al., 2020).

The effectiveness of CBT for trauma in individuals with bipolar disorder remains unclear, with concerns about maintaining mood stability, highlighting the need for additional safeguards that support emotional regulation within treatment (Otto et al., 2004).

EMDR is an integrative therapeutic approach that is guided by the adaptive information processing model (AIP), which proposes that unprocessed trauma memories are stored dysfunctionally and contribute to current distress (Perlini et al., 2020). Bilateral stimulation (BLS) is then used to desensitise distress and process these memories, resulting in memory reconsolidation (Laliotis et al., 2021). The standard EMDR protocol consists of eight phases, with treatment length tailored to the individual’s needs.

1.3. EMDR and bipolar disorder

Given high rates of trauma exposure and post-traumatic symptoms, individuals with bipolar presentations may benefit from EMDR (Perlini et al., 2020; Quarantini et al., 2010). EMDR targets unprocessed trauma memory networks that may underlie emotional and behavioural difficulties, while supporting regulation of distressing experiences (Shapiro, 2001).

Emerging evidence suggests that EMDR may be a safe intervention in bipolar disorder, with potential benefits for both trauma-related and subsyndromal affective experiences (Novo et al., 2014). Amann et al. (2015) propose that EMDR may also act as a psychotherapeutic mood stabiliser through strengthening adaptive resources via bilateral stimulation, which may be particularly relevant for individuals with marked mood fluctuations requiring additional stabilisation to engage in trauma-focused work. However, further evidence is needed to clarify its role in bipolar presentations (Valiente-Gómez et al., 2019).

To date, only three studies have investigated the effectiveness of EMDR for trauma in people with bipolar disorder. Oh and Kim (2014) reported two cases of individuals with bipolar disorder and PTSD who received 9 and 10 weekly sessions of Shapiro’s (1989) standard EMDR protocol. Both showed remission of PTSD symptoms, maintained at 1-year follow-up, alongside ongoing support for mood regulation.

A randomised controlled pilot trial by Novo et al. (2014) compared EMDR with treatment as usual (TAU) in ten participants with subsyndromal mood experiences and a history of traumatic events. Participants in the EMDR group received 14–18 sessions of the standard protocol and showed improvements in both low and elevated mood experiences, alongside reductions in trauma-related distress. These findings suggest that EMDR’s effects on mood may occur through the processing of trauma memories, rather than directly targeting bipolar disorder.

Hogg et al. (2024) conducted a randomised controlled trial comparing EMDR with supportive therapy as an adjunctive approach for people with bipolar disorder and post-traumatic symptoms. EMDR was delivered over 20 weeks and included the EMDR bipolar protocol (Amann et al., 2015), which builds on Shapiro’s standard eight-phase EMDR protocol and incorporates five additional bipolar-specific subprotocols targeting mood stabilisation, treatment adherence, illness awareness, detection of prodromal symptoms, and de-idealisation of manic experiences (Hogg et al., 2024). These subprotocols aimed to support engagement and stability, while the standard protocol remained central for processing trauma memory networks.

Results showed reductions in trauma-related symptoms in both groups and supported the tolerability and safety of EMDR in this population (Hogg et al., 2024). However, no clear advantage over supportive therapy was found, raising questions about the added value of the bipolar-specific subprotocols and whether they may dilute core EMDR processes. This also suggests value in prioritising the established EMDR protocol alongside targeted adaptations for bipolar presentations.

Richardson and Amann (2024) propose that trauma-focused approaches for individuals with bipolar disorder should include adaptations to treatment plans that consider current mood states. These adaptations may involve mood monitoring, stabilisation strategies, reviewing hypo(manic) experiences, focusing on relapse prevention, and supporting self-efficacy. However, uncertainty remains regarding the optimal set of adaptations for EMDR in this population. Given the high rates of trauma exposure and post-traumatic stress symptoms, alongside limited guidance on the use of EMDR for trauma in bipolar presentations, the development of clear, consensus-based recommendations is a priority.

1.4. Study objectives

Establishing consensus among expert EMDR practitioners could enhance treatment accessibility, improve adherence, reduce the risk of relapse, and provide recommendations that can be used in future trials (Taylor et al., 2020).

This study aimed to establish a consensus among EMDR practitioners on two key areas: (1) adaptations to EMDR that are important when treating trauma in individuals with bipolar disorder, to improve treatment accessibility and effectiveness, and (2) recommendations for EMDR supervision for practitioners working with this population.

2. Methods

2.1. Design

This study adopted Delphi survey methodology, an iterative process used to generate expert consensus on under-researched topics through repeated surveys (Diamond et al., 2014). Participants rate the extent to which they endorse a list of statements, which are re-rated if consensus is not reached (Langlands et al., 2008).

A three-round online Delphi design was used, which reduces participant burden, maintains anonymity, and enables wider sampling (Spain et al., 2023; Spain & Happé, 2020). This approach follows prior Delphi research (Langlands et al., 2008), including studies that examined adaptations of psychological therapies, such as EMDR, for specific clinical populations (Fisher et al., 2023; Spain et al., 2023).

2.2. Ethical approval

Ethical approval was granted by the University of Southampton Research Ethics Committee (UREC; ERGO 91294). Informed consent was obtained in line with institutional guidelines.

2.3. Participants

2.3.1. Inclusion criteria

Participants were included if they were: (1) EMDR practitioners currently working in any setting, with any population, and based in any country; (2) had any level of experience (from limited to advanced) using EMDR with trauma-related difficulties (including PTSD presentations); (3) had experience working with individuals with bipolar disorder using any therapeutic intervention; and (4) proficient in English. Basic demographic information was collected about professional roles, clinical settings, and experiences of trauma and bipolar disorder cases.

2.3.2. Recruitment

Participants were recruited via convenience sampling using snowballing, online forums, social media, and professional networks, including the EMDR UK Association, EMDR Europe, regional associations, and the International Society for Bipolar Disorders Psychological Therapies Task Force. Recruitment was supported by an advertising poster.

Recruitment followed an open, self-selection approach; therefore, the total number of individuals exposed to the survey is unknown. Only participants who completed Round 1 were invited to subsequent rounds. This approach is consistent with Delphi methodology, as it ensures the same panel re-rates statements and reviews aggregated feedback to establish consensus. To minimise the risk of participant identification within this specialised expert population, and in line with the study aims, potentially identifying contextual information (e.g. personal characteristics and country of practice) was not collected.

2.4. Survey development

The Delphi methodology and survey development process were conducted under the guidance of multiple researchers and supervisors with complementary expertise in research methods and relevant experience.

2.4.1. Usability testing/community involvement

Initial piloting of the survey was conducted through patient and public involvement (PPI). Four EMDR practitioners and one expert with lived experience (with a diagnosis of bipolar disorder and experience of EMDR for trauma-related difficulties) provided informal feedback on the study’s scope and aims, survey content and potential barriers to trauma-focused therapies. PPI involvement helped shape the first survey through reflections on ethical considerations, supporting feasibility and engagement, and refining the survey focus. Feedback from piloting also informed the structure and clarity of Round 1 questions, ensuring they were clear, relevant, and captured meaningful clinical experiences.

2.4.2. Round 1 survey

The first survey comprised three sections: (1) demographic information (e.g. profession, work setting, experience providing EMDR therapy, supporting clients with bipolar disorder and trauma, and relevant training); (2) proposed barriers affecting EMDR accessibility and efficacy, including general and phase-specific adaptations used to overcome those barriers; and (3) considerations for EMDR supervision when using trauma-focused approaches with individuals with bipolar disorder.

Round 1 used open-ended free-text boxes for each EMDR phase and supervision consideration, enabling participants to provide detailed, nuanced descriptions of clinical adaptations and experiences. Despite being online, this format allowed for flexibility in responses, consistent with classical Delphi methodology. Responses were sufficiently detailed to support subsequent thematic analysis.

2.4.3. Round 2 survey

The Round 2 survey comprised 131 statements derived from the thematic analysis of the Round 1 survey (Braun & Clarke, 2021). These statements were categorised as follows: (1) 30 general adaptations; (2) 82 phase-specific adaptations; and (3) 19 considerations for clinical supervision.

2.4.4. Round 3 survey

In the Round 3 survey, a total of 27 statements were re-rated: (1) six pertaining to general adaptations, (2) 18 relating to phase-specific adaptations, and (3) three concerning clinical supervision.

2.4.5. Measurement scale

In accordance with established guidelines for Delphi surveys, general and phase-specific statements in the Round 2 and 3 surveys were rated using a 5-point Likert scale (Langlands et al., 2008). Supervision statements were rated using a newly developed Likert scale from Fisher et al. (2023), as this better aligned with the content of the statements. Both rating scales are outlined in Table 1.

Table 1.

Survey statement measurement scales.

Adaptations to EMDR	Considerations for supervision
1. Essential	1. This is essential for all
2. Important	2. This is important
3. Don’t know/depends	3. This could be useful, depending on the client group
4. Unimportant	4. This is not important
5. Should not be included	5. This is actively unhelpful

Note. EMDR = eye movement desensitisation and reprocessing.

2.5. Procedure

The study was conducted between July 2024 and December 2024, consistent with recommended timelines for Delphi studies (Iqbal & Pipon-Young, 2009). Surveys were administered via Qualtrics in a consistent format, beginning with demographic questions, followed by qualitative questions in Round 1 and quantitative statements in Rounds 2 and 3, each including optional free-text responses (Fisher et al., 2023).

Participants could complete the survey flexibly, including across multiple sittings, so completion time varied and was not standardised.

2.6. Data analysis

Data were analysed using a mixed-methods approach, combining qualitative thematic analysis with quantitative consensus analysis. Participant demographics were summarised descriptively.

2.6.1. Thematic analysis (statement generation)

Thematic analysis (Braun & Clarke, 2021) was conducted on the participants’ free-text comments from Round 1, to generate statements for the Round 2 survey. A reflexive thematic analysis approach was used, combining inductive and deductive processes, and conducted using NVivo software. Data was coded and organised into three main themes: (1) general adaptations to EMDR, (2) phase-specific adaptations, and (3) clinical supervision considerations, with sub-themes identified within each. Independent review by two team members supported consensus on codes and theme names, producing statements reflecting participants' lived clinical experiences.

The analysis was guided by a pragmatic epistemological approach (Feilzer, 2010) and interpretivist principles underpinning reflexive thematic analysis (Braun & Clarke, 2021), aiming to develop a practical, consensus-based understanding by integrating participant accounts with structured response scales. Reflexivity was maintained through written reflections documenting key decisions and interpretations during coding and theme development, ensuring transparency and awareness of researcher positionality and subjectivity in line with reflexive thematic analysis principles (Braun & Clarke, 2021).

Coding captured both semantic (explicit) and latent (underlying) meanings within an experiential framework, reflecting participants’ clinical experiences. Analysis considered how responses related across EMDR phases, including client safety and supervision needs. The lead researcher’s EMDR and trauma-informed clinical experience contributed to code interpretation and theme development, supporting the generation of clinically grounded Round 2 statements.

2.6.2. Quantitative analysis

For Rounds 2 and 3, the percentage of participants who rated each statement in the same way was calculated to assess group consensus. The criteria for determining consensus were based on guidelines from Langlands et al. (2008): (1) statements rated as essential or important by approximately 80% or more of participants were classified as fundamental components of EMDR; (2) statements rated as essential or important by 60–79% of participants were re-assessed in Round 3; and (3) statements with less than 60% agreement were excluded.

3. Results

3.1. Participant demographics

Thirty-four participants completed the Round 1 survey (Table 2). They represented a range of professional backgrounds and worked across diverse healthcare settings. Participants had between less than one and over 15 years of EMDR experience, with a mean of 7.11 years (SD = 4.82). The majority were accredited EMDR practitioners (47%), and 38% had between 5 and 9 years of EMDR experience.

Table 2.

Participant demographics.

	Round 1 (n = 34)	Round 2 (n = 22)	Round 3 (n = 20)
Profession
Psychologist (clinical, health, counsellor)	15 (44.1%)	11 (50%)	11 (55%)
Psychiatrist	4 (11.8%)	2 (9.1%)	1 (5%)
Psychotherapist (systemic, trauma, cognitive behavioural, psychosexual)	9 (26.5%)	7 (31.8%)	8 (40%)
Counsellor	3 (8.8%)	1 (4.5%)	-
Nurse	2 (5.9%)	1 (4.5%)	-
Social Worker	1 (2.9%)	-	-
Population worked with
Children and Adolescents	6 (17.6%)	1 (4.6%)	1 (5%)
Adolescents and Adults	1 (2.9%)	5 (22.7%)	4 (20%)
Adults	27 (79.4%)	16 (72.7%)	15 (75%)
EMDR training
Not yet finished basic training	2 (5.9%)	-	-
Basic adult training	16 (47.1%)	7 (31.8%)	6 (30%)
Both basic adult and child training	1 (2.9%)	1 (4.6%)	1 (5%)
Accredited EMDR Practitioner	9 (26.5%)	8 (36.4%)	6 (30%)
Accredited EMDR Consultant	6 (17.6%)	6 (27.3%)	7 (35%)
Years of experience providing EMDR
Less than a year	2 (5.9%)	-	-
1–4	10 (29.4%)	6 (27.3%)	5 (25%)
5–9	13 (38.2%)	8 (36.4%)	8 (40%)
10–14	6 (17.6%)	5 (22.7%)	5 (25%)
15–20	3 (8.8%)	3 (13.6%)	2 (10%)
Number of cases treating someone with bipolar disorder using any therapy
1–2	6 (17.6%)
3–5	8 (23.5%)
6–10	5 (14.7%)
10+	15 (44.1%)
Number of cases treating trauma with someone with bipolar disorder
0	7 (20.6%)
1–2	6 (17.6%)
3–5	5 (14.7%)
6–10	6 (17.6%)
10+	10 (29.4%)
Number of cases using EMDR with someone with bipolar disorder
0	11 (32.4%)	4 (18.2%)	4 (20%)
1–2	6 (17.6%)	4 (18.2%)	4 (20%)
3–5	10 (29.4%)	7 (31.8%)	7 (35%)
6–10	2 (5.9%)	4 (18.2%)	2 (10%)
10+	5 (14.7%)	3 (13.6%)	3 (15%)
Number of cases using EMDR to treat trauma with someone with bipolar disorder
0	12 (35.3%)	6 (27.3%)	5 (25%)
1–2	4 (11.8%)	4 (18.2%)	4 (20%)
3–5	12 (35.3%)	7 (31.8%)	7 (35%)
6–10	3 (8.8%)	2 (9.1%)	2 (10%)
10+	3 (8.8%)	3 (13.6%)	2 (10%)

Note. n = number; EMDR = eye movement desensitisation and reprocessing.

Nearly half of the participants (44%) had worked therapeutically with 10 or more individuals with bipolar disorder. Additionally, 67.6% (n = 23) had used EMDR with individuals with bipolar disorder, and 64.7% (n = 22) had used EMDR to treat trauma in this population. Regarding training, 41.2% of participants (n = 14) had received training in trauma work for bipolar disorder, whereas only 5.9% (n = 2) had received training explicitly in EMDR trauma-focused treatment with bipolar disorder.

3.2. Delphi survey outcomes

3.2.1. Round 1

3.2.1.1. Rationales for EMDR use in treating trauma in individuals with bipolar disorder

Participants identified two main rationales:

1. Trauma connections: They emphasised links between trauma and bipolar-related difficulties, suggesting that addressing trauma may support emotional regulation and stability.

2. Advantages of EMDR: They acknowledged several benefits of EMDR, including its status as an evidence-based intervention for trauma. Research supports its effectiveness, and the EMDR protocol can be adapted to suit an individual’s needs and clinical presentation. Practitioners also reported positive experiences using EMDR in this population.

3.2.1.2. Barriers to efficacy and accessibility of trauma-focused EMDR in bipolar disorder

The analysis revealed that the reported barriers could be grouped into four main categories:

1. Client factors: Features that may influence therapy engagement and processing, such as variability in attendance, mood fluctuations, and dissociation, can reduce the capacity for reprocessing. Additionally, manic and depressive experiences may affect engagement with and response to treatment.

2. Safety concerns: Participants expressed apprehensions about potential negative consequences of EMDR. For example, trauma-focused work might trigger manic, depressive, or psychotic episodes, potentially increasing distress or destabilisation during treatment.

3. Systemic issues: Inflexible care pathways, including structural barriers within services and restrictions on therapy access for individuals with a bipolar disorder diagnosis.

4. Limited provision of bipolar disorder specialist guidance: A strong emphasis on the biomedical model and pharmacological treatment can restrict access to psychological therapies. In addition, there is a lack of evidence and guidance to determine the efficacy of EMDR for this population.

Findings from Round 1 (rationales and barriers) informed the development of statements for subsequent Delphi rounds, which were organised into three domains: general adaptations, phase-specific adaptations, and supervision considerations.

3.2.2. Round 2

Twenty-two participants completed Round 2, representing a 64.7% retention rate. The Round 2 survey included 131 statements, of which 71 were rated by at least 80% of practitioners as essential or important aspects of EMDR (Table 3). Additionally, 15 elements of EMDR supervision were rated by at least 80% of respondents as essential or important (Table 4).

Table 3.

Aspects of EMDR that ≥80% of Practitioners Place as ‘Essential or Important’.

	% of participants agreeing items 1–2
General adaptations to EMDR
Review if this is an appropriate time for treatment (e.g. not at a time of extreme mood/mania)	100
Offer thorough psychoeducation about EMDR protocol	100
Create awareness about dissociation to help with stability	100
Empower the client to report any difficulties with symptoms that they may experience during therapy	100
Follow usual trauma work considerations, (window of tolerance)	96
Build strength through resourcing	96
Check that the person's mood is stable enough to continue with therapy	91
Increase responsiveness to any signs of mood relapse	91
Reinforce client stability before processing	91
Develop strategies to manage variations in client's mood	91
Develop a clear safety plan with the client in case of relapse	91
Make sure the client has a network of support around them	91
Develop thorough assessment and preparation phases before processing	91
** Use a combination of psychological therapies to help mood variations	90
Give reminders of ability to say or signal stop	87
Develop a robust, collaborative formulation of the trauma memory networks underlying the client's experiences of going high and feeling low	87
Develop a relapse plan for bipolar symptoms	82
Phase 1: Client history/treatment planning
Attention to periods of stability and instability in client history	100
Assess if the client is stable enough to proceed with therapy	100
Assess client’s view on relationship between trauma and bipolar symptoms	96
Identify signs of an acute episode (relapse signature)	96
Review what support networks they have available	96
Clarify that client has coping strategies in place to help with mood variations	96
Understand the links between bipolar history with trauma history	95
Thorough history taking of client and their presentation, including links to historic and present trauma	91
Develop clarity on what symptoms are being treated	91
Complete an in-depth risk assessment	91
Review impact of medication (past and current)	91
Review client’s safety and establish relapse plan	91
Clearer focus on expectations	91
**Ensure client can access an image that provokes calmness	90
** Offer psychoeducation on bipolar disorder	90
Identify the trauma memory networks underlying going high and feeling low	87
Review links between trauma and psychotic symptoms	87
Capture a detailed biographical timeline that encompasses all client history	86
Phase 2: Preparation Stage
Establish trust between therapist and client	100
Discuss with client the consequences of potential reprocessing	100
Encourage client to monitor their moods	96
Have more awareness of relapse indicators	95
Include safe place protocol	95
Use standard protocol stability tools (container, calm place).	91
Complete thorough resourcing by using attachment informed figures and client’s own abilities to resource.	86
Develop clear safety plan that highlights relapse indicators	86
Incorporate emotional regulation and distress tolerance skills	82
Develop more somatic resources for increasing the window of tolerance.	82
** Install protective, compassionate, wise figures/ (attachment focused EMDR)	80
Phase 3: Assessment phase
Look for similar themes in traumatic events	100
Develop a clear formulation that considers the client’s elated and low mood patterns	95
Have strategies in place to deal with abreactions	95
Identifying realistic cognitions (negative and positive) and deeper beliefs	95
Offer a clear explanation of case conceptualisation beforehand	91
Identify aspects that may trigger abreactions.	82
Ensure the client can feel the disturbance in their body.	82
** Have additional resources in place	80
Phase 4: Desensitisation
Complete regular check-in's	100
Clarify client preference around BLS	100
Making sure that the client can engage within this phase before proceeding	95
Phase 5: Installation
Make sure positive cognition is realistic and not overly grandiose	96
Encourage a PC that welcomes neutral acceptance	91
Be flexible in the approach to reduction of SUDs ratings, as you may not get them down to zero	82
Phase 6: Body scan
Evaluate the impact of any hallucinatory experiences if they are present.	96
When clients need to pay attention to their bodies, find a way that makes them more comfortable.	91
Develop clearer ways to recognise change within a client’s body	91
Make sure body scan is linked to target not general sensitivity	91
Phase 7: Closure
Normalise any disturbance if it should occur	91
Closely monitor client’s wellbeing in between sessions	86
Encourage client to make contact if they experience heightened symptoms that make them feel unsafe	86
Phase 8: Re-evaluation
Complete regular check-ins for client’s wellbeing	100
Increase the client's awareness of potential relapse warning signs	95
Have additional time for grounding at end of the session	82
Have agreed measurement tools to monitor the client’s mood levels	81

Note. **Item was re-rated in Round 3, EMDR = eye movement desensitisation and reprocessing, BLS = bilateral stimulation, SUDS = subjective units of distress scale, PC = positive cognition.

Table 4.

Aspects of EMDR supervision considered ‘essential or important’ by ≥80% of practitioners.

	% of participants agreeing items 1–2
Supervisee has good engagement skills	100
Move away from the medical model and medicalised language in supervision	100
Supervisor should have knowledge and understanding of bipolar disorder	95
Supervisee has an understanding of bipolar	95
Spend time reflecting on client formulation, which accounts for the complexity of both bipolar disorder and trauma	95
Focus on client safety throughout	95
Focus on supervising client stability throughout treatment	95
Supervisee understands bipolar psychiatric treatment	91
There should be time to review the skills that are needed by the supervisee when working with trauma and bipolar	91
Spent time discussing skilful ways to stop or pause treatment if therapy is causing adverse reactions	91
**There should be time to review supervisee's competence to work with bipolar disorder	90
Supervisor should have knowledge of what specific adaptations of EMDR to use when working with dissociation	86
There needs to be regular updates on client progress (mood and trauma symptoms)	82
Spend time discussing the aspects of the ‘reprocessing’ of traumatic events	82
There should be time to reflect on recent research and developments in clinical practice for bipolar and dissociation	82

Note. ** statements were re-rated in Round 3.

Of these statements, 24 were excluded for not reaching the 60% consensus threshold. A further 27 were considered unclear, as 60–79% of participants rated them as ‘essential’, ‘important’, or ‘don’t know/depends’, and were therefore re-rated in Round 3 to achieve consensus (Langlands et al., 2008). Items reaching consensus following re-rating are marked with an asterisk (**) in Tables 3 and 4. The percentages reported in the tables reflect agreement for items rated as ‘essential’ or ‘important’.

3.2.3. Round 3

A total of 20 participants completed the Round 3 survey, representing a 58.8% retention rate from Round 1. Of the 27 statements that were re-rated, six achieved consensus with over 80% of participants rating them as essential or important, while the remaining 21 did not reach consensus and were excluded. Full details of all statements that did not reach consensus or were excluded throughout the analysis are provided in Supplementary Table S1.

In Round 3, an additional question asked participants to indicate why they selected ‘don't know/depends’. This was included to provide a better understanding of their decision-making process and the rationale behind these ratings. Reasons for selecting ‘don’t know/depends’ are summarised in Table 5.

Table 5.

Reasons for Selecting ‘Don’t Know/Depends.’

Reasons	Number of responses
Depends on the individual and their presentation	18
Depends on the needs of the client	18
Confidence in modifying EMDR protocols	9
Lack of training in specific techniques	7
Less awareness of the research evidence base for bipolar disorder treatment	6
Amount of experience working with bipolar disorder presentations	4
Concerns about risk	4
Service restrictions	4
Preference for particular protocols	3

Note. EMDR = eye movement desensitisation and reprocessing.

3.3. Summary of results

Seventy-one statements on EMDR adaptations for trauma treatment in bipolar disorder were rated as essential or important (17 general, 54 phase-specific), along with 15 supervision recommendations, giving a total of 86 consensus statements. Forty-five statements did not reach the 80% consensus threshold, and 14 were rated as ‘should not be included’ (Supplementary Table S2). No supervision-related statements were rated as ‘actively unhelpful.’

4. Discussion

This study explored practitioners’ perspectives on barriers to the accessibility and effectiveness of trauma-focused EMDR for individuals with bipolar disorder and established consensus on key treatment adaptations and supervision considerations.

Four primary barriers were identified: client-related factors, systemic influences, safety concerns, and limited clinical guidance. Notably, 94.1% of practitioners reported receiving no EMDR training specific to treating trauma in individuals with bipolar disorder.

Client-related factors included mood fluctuations, dissociative experiences, and variability in therapy attendance, which may affect treatment engagement and reprocessing. Practitioners expressed concerns that trauma-focused work could exacerbate mood episodes, potentially limiting EMDR use (Perlini et al., 2020). Similar concerns have been reported in broader SMI populations, including fears of increased distress, suicidality, or hospitalisation, alongside limited clinician confidence (Frueh et al., 2006). However, more recent qualitative research suggests that trauma-focused EMDR can be experienced as meaningful and beneficial in individuals with psychosis, with no evidence of relapse, challenging assumptions of harm in some SMI populations (Hardwick et al., 2026).

Additional barriers included misdiagnosis or underdiagnosis of bipolar disorder in the presence of co-occurring PTSD, where overlapping clinical features may obscure distinctions between trauma-related and mood-related experiences, limiting access to trauma-informed care. This highlights the importance of comprehensive, trauma-informed assessment (Cogan et al., 2021).

Participants emphasised variability in bipolar presentations, reinforcing the need for individualised approaches. Adaptations focused on stabilisation, including safety and relapse planning, management of mood fluctuations, and therapist responsiveness. Conceptualisation changes involved integrating mood variability, relapse risk, and stabilisation needs into treatment planning, consistent with psychological approaches to mood instability (Wright et al., 2021).

These adaptations aimed to enhance EMDR’s effectiveness and accessibility while maintaining core phase functions, with most targeting early phases (1–3) and focusing on stability, risk identification, and extended preparation. History-taking included developing detailed timelines across stable and unstable periods, thorough symptom assessment, and prioritising safety.

Preparation (phase 2) incorporated mood monitoring, emotional regulation, distress tolerance, and psychoeducation. Stability was supported through standard tools (e.g. container, safe place) and somatic and attachment-informed resourcing. During assessment (phase 3), emphasis was placed on monitoring stability, managing abreactions, and strengthening formulation.

In desensitisation (phase 4), adaptations addressed client readiness and preferences for bilateral stimulation. Installation (phase 5) involved developing realistic positive cognitions and flexibility in reducing Subjective Units of Distress (SUDs) ratings. The body scan (phase 6) focused on linking sensations to target memories and, where relevant, evaluating unusual perceptual experiences. Later phases (closure and re-evaluation) prioritised stability and ongoing mood monitoring, supporting autonomy and control (Miklowitz, 2006).

Overall, trauma-focused EMDR for individuals with bipolar disorder should be flexible, responsive to safety and mood fluctuations, and guided by a comprehensive formulation.

Strong consensus emerged regarding supervision, including the importance of supervisors’ knowledge of bipolar disorder and supervisees’ understanding of its presentation and psychiatric treatment. Increased awareness of co-occurring trauma was seen as enhancing clinical skills and confidence. Supervision should include regular progress updates, formulation considerations, and a focus on stability to support complex trauma work.

Although bipolar-specific EMDR protocols have been proposed (Amann et al., 2015), consensus was not achieved, with only 32% of participants endorsing their use. This may reflect limited evidence (Hogg et al., 2024), low awareness, and confidence in adapting the standard EMDR protocol (Shapiro & Maxfield, 2002).

4.1. Limitations

This study has several limitations. The sample size was relatively small (n = 34), and attrition across rounds limited generalisability; subgroup analyses were not conducted. Participants were included based on experience with EMDR and/or bipolar disorder to avoid over-restriction of the sample. Whilst all had relevant clinical experience, some had limited experience using EMDR with this population, which may have influenced the quality and depth of responses.

Demographic data (e.g. age, gender, ethnicity) were not collected, restricting assessment of sample diversity and the perspectives informing the consensus. This constrains conclusions regarding transferability and representativeness.

Additionally, this study adopted rating scales used in previous Delphi research (Langlands et al., 2008), including a response option combining ‘don’t know’ and ‘it depends’. Collapsing these distinct forms of uncertainty into a single category may limit interpretability and introduce ambiguity (Spain & Happé, 2020).

4.2. Research implications

Further consensus-based research is warranted, including larger and more diverse samples, to inform training and guidance. Future randomised controlled trials comparing bipolar-specific and standard EMDR protocols could clarify the impact of adaptations. Research should also explore how individuals with bipolar disorder experience and engage with these adaptations.

4.3. Conclusion

Trauma-focused EMDR can be applied in bipolar disorder using the standard eight-phase protocol without disorder-specific subprotocols. Clinicians should prioritise phase-specific responsiveness, symptom-informed pacing, and mood stabilisation to support safe and effective trauma processing.

This study is the first to examine barriers, treatment adaptations, and supervision needs for EMDR in this population. It highlights how mood instability and fluctuations in emotional and physiological regulation capacity, commonly reported in bipolar presentations, can influence accessibility and clinical decision-making. Systemic challenges, including inflexible care pathways and limited specialised training, further underscore the need for professional development.

Participants recognised the value of EMDR for trauma in bipolar disorder and indicated that, given clinical variability, selected adaptations may be helpful (e.g. extended preparation, safety planning, and phase-specific adjustments), without requiring disorder-specific subprotocols. Supervision and knowledge of bipolar presentations were also considered essential.

Overall, this study provides practical guidance for clinicians working with bipolar presentations and trauma-related symptoms, supporting safe trauma processing. However, evidence for EMDR in this population remains limited, and further research is needed to strengthen the empirical foundation and inform clinical practice.

Funding Statement

This research was supported by the Doctorate in Clinical Psychology programme through NHS England.

Disclosure statement

Dr Thomas Richardson has received payment for delivering training on psychological therapies for bipolar disorder and for developing therapy-related content in this area. He also receives royalties from a book related to bipolar disorder. Professor Kim Wright declares occasional payments from Careloop Health Ltd. for consultancy work.

Declaration of generative AI and AI-assisted technologies in the manuscript preparation process

During manuscript preparation, ChatGPT (OpenAI) was used solely for minor grammar and language clarification. It was not used to generate scientific content or interpret data. The author(s) reviewed all content and take full responsibility for the manuscript.

Data availability statement

The raw data supporting the findings of this study are held by the corresponding author’s supervisor. A copy of the dataset is deposited in the University of Southampton repository (DOI: ) and is under embargo until August 31, 2026. Requests for access prior to this date can be directed to the corresponding author.

Supplemental Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/20008066.2026.2679415.