Eye Movement Desensitization and Reprocessing for Posttraumatic Stress Disorder Following Acquired Brain Injury: A Multiple Baseline Single-Case Experimental Design Study Across Four Cases

Eye Movement Desensitization and Reprocessing for Posttraumatic Stress Disorder Following Acquired Brain Injury: A Multiple Baseline Single-Case Experimental Design Study Across Four Cases - Eye movement desensitization and reprocessing for posttraumatic stress disorder following acquired brain injury: A multiple baseline single case experimental design study across four cases

 

Authors: Ellen P. J. Janssen, Peggy J. J. Spauwen, Sophie J. M. Rijnen & Rudolf W. H. M.

Ponds

Translated by: Specialist Psychologist Gizem Pozam

 

ABSTRACT
 Posttraumatic stress disorder (PTSD) is common among individuals with acquired brain injury (ABI). This study investigated the effectiveness and feasibility of Eye Movement Desensitization and Reprocessing (EMDR) therapy in individuals with ABI and posttraumatic stress disorder. Data were collected using a nonconcurrent multiple baseline single-case experimental design (SCED) involving four cases. The study comprised baseline, treatment, maintenance, and 3-month follow-up phases. EMDR treatment was delivered using a standardized, manualized EMDR protocol.

The primary outcome variable was PTSD symptoms. Secondary outcome variables were general mental health and cognitive functioning. Visual analyses, TAU-U analyses, and analyses using the Reliable Change Index were performed. All four clients participating in the study (two with traumatic brain injury and two with a history of stroke) showed a significant reduction in PTSD symptoms; this improvement continued during the maintenance phase and was sustained at follow-up. Participants no longer met the diagnostic criteria for PTSD at post-treatment and follow-up and showed reliable improvement in PTSD severity scores. No adverse events occurred, and no adaptations to the EMDR protocol were required. However, no consistent improvement in general mental health or cognitive functioning was observed. This study provides empirical support for the effectiveness and feasibility of EMDR in four clients with stroke or traumatic brain injury.

 

Introduction

Individuals with acquired brain injury (ABI), following conditions such as traumatic brain injury (TBI) or stroke, are at increased risk of developing posttraumatic stress disorder (PTSD). PTSD prevalence is estimated at 16% following TBI (van Praag et al., 2019) and 17% following stroke (Tang et al., 2022), compared with a reported lifetime prevalence of 4% in the general population (Koenen et al., 2017).

PTSD can develop following exposure to actual or threatened death, serious injury, or sexual violence and is characterized by four core symptom clusters: (1) re-experiencing (intrusions), (2) avoidance, (3) negative alterations in cognition and mood, and (4) alterations in arousal and reactivity that emerge following the traumatic event, persist for at least one month, and cause significant distress in social or occupational functioning (American Psychiatric Association, 2013). In addition, PTSD in the general population has been associated with cognitive impairment, particularly in attention, executive functioning, and verbal memory (Flaks et al., 2014; Gilbertson et al., 2001; Johnsen & Asbjornsen, 2008; Polak et al., 2012; Yehuda et al., 2004). In individuals with ABI, PTSD has been associated with reduced health-related quality of life and poorer functional outcomes (Haagsma et al., 2015; Stein et al., 2018). Recognizing and treating PTSD in individuals with ABI is therefore of great importance.

Eye Movement Desensitization and Reprocessing (EMDR) therapy is an approach with strong evidence supporting its effectiveness in treating PTSD in adults (Cusack et al., 2016; Mavranezouli et al., 2020) and is recommended in guidelines for PTSD management (Bisson et al., 2020; World Health Organization, 2013). EMDR is a standardized, protocol-based treatment grounded in the assumption that distress associated with traumatic events arises from unprocessed information (Shapiro, 2001). During EMDR, the client focuses on distressing material while simultaneously attending to a stimulus, typically therapist-guided eye movements and/or auditory or tactile stimuli. During this desensitization process, distress decreases and dysfunctional or negative beliefs change (de Jongh & ten Broeke, 2011). Currently, the strongest empirical support for EMDR’s mechanism of action is provided by the “working memory hypothesis” (Landin-Romero et al., 2018; Wadji et al., 2022). According to this hypothesis, because simultaneous tasks compete for limited working memory capacity, negative memories become less vivid and less emotional and are stored in this form (van den Hout & Engelhard, 2011). Successful PTSD treatment has been shown to be associated with improvements in cognitive functions such as memory, inhibition, and cognitive flexibility (set-shifting) (Haaland et al., 2016; Nijdam et al., 2018; Walter et al., 2010).

Given the strong support for EMDR’s effectiveness in adults with PTSD, this treatment may also be a suitable option for individuals with ABI. In particular, because EMDR does not rely heavily on cognitive functioning (de Jongh & ten Broeke, 2011), it may offer an advantage for individuals with ABI who experience cognitive impairment. Several observational studies support the feasibility and effectiveness of EMDR in individuals with ABI and PTSD (Janssen et al., 2023; Moore, 2023; Pagani et al., 2018; Smart, 2022; Yaşar et al., 2022). However, to our knowledge, no controlled study has examined the effectiveness and feasibility of EMDR in individuals with ABI.

Given the limited research on the effectiveness and feasibility of EMDR for PTSD in individuals with ABI, we conducted a study in a clinical setting using single-case experimental design (SCED) methodology. SCED methodology is considered an appropriate method for individuals with comorbid conditions, balancing methodological rigor with an individualized approach (Vlaeyen et al., 2020).

The primary research questions addressed in this study were:

1. Does EMDR lead to a significant reduction in PTSD symptoms?


2. Are adaptations to the standard EMDR protocol necessary?

Our secondary research questions were:
3. Does EMDR lead to improvements in general mental health?
4. Does EMDR improve cognitive functioning (particularly memory, inhibition, and cognitive flexibility/set-shifting)?

Materials and Methods

Design

This study used a nonconcurrent multiple baseline design comprising baseline (A), intervention (B), maintenance (C; following phase B, Krasny-Pacini & Evans, 2018), and follow-up (D; 3 months after the end of phase B) phases (Coon & Rapp, 2018; Watson & Workman, 1981). A single-case experimental design (SCED) was implemented individually for each participant, with replication across participants. The duration of the baseline phase (A) (2, 3, or 4 weeks) was randomly assigned to each participant using randomized intervention start points (Onghena & Edgington, 2005; Smith, 2012; Michiels & Onghena, 2019).

The design was nonconcurrent for ethical and practical reasons. A concurrent design would have required all participants to begin baseline measurements at the same time, unnecessarily delaying treatment for some participants. As participants entered the study, one of the baseline durations of 2, 2, 3, 3, 4, 4 weeks (six participants were initially planned) was randomly selected. The treatment phase (B), consisting of weekly EMDR sessions for a maximum of 10 weeks, then began. If treatment was completed earlier than scheduled, maintenance measurements (C) replaced treatment measurements.

The study design was developed based on the Risk of Bias in N-of-1 Trials (RoBiNT) Scale (Tate et al., 2008; Tate et al., 2013). Critical appraisal using the RoBiNT scale yielded a total score of 22 out of 30 (see Appendix A, online Supplementary Material, for details). The study achieved a strong external validity score but scored lower on some internal validity criteria, primarily because the design was based on practical requirements. The study report was prepared in accordance with the Single-Case Reporting Guideline in BEhavioural Interventions (SCRIBE) Statement Checklist (Tate et al., 2016).

This study was approved by the scientific review board of GGZ Oost Brabant and was conducted in accordance with the 1964 Declaration of Helsinki and its subsequent revisions. The Medical Ethics Review Committee (METC) AZM/UM issued a statement that the study did not fall within the scope of the Medical Research Involving Human Subjects Act (WMO) (METC AZM/UM 2018-0542).

Participants and Setting

Potential participants were selected through consecutive case finding at an outpatient brain injury clinic between July 2018 and October 2019. The study was conducted within a specialized tertiary mental health institution in the Netherlands providing multidisciplinary rehabilitation treatment and mental health care.

A total of nine individuals were assessed for eligibility, four of whom met the inclusion criteria.

Inclusion Criteria:

1. Meeting the diagnostic criteria for PTSD on the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) (Weathers, Blake et al., 2013).


2. A history of a cerebrovascular accident (stroke) or traumatic brain injury diagnosed by a neurologist and confirmed by CT or MRI (diagnosed at least 6 months before inclusion).


3. Being 18 years of age or older.


4. Fluency in Dutch.


5. Sufficient cognitive capacity to complete the questionnaires.


6. Access to the internet and a personal smartphone.


7. If taking psychotropic medication, a stable dose for the 4 weeks before the study and throughout the study.


8. Providing informed consent.

Exclusion Criteria:

1. Current suicide risk.


2. A history of substance dependence or psychosis.


3. Inability to attend the outpatient clinic weekly.


4. Previous PTSD treatment for the same traumatic experience.


5. A diagnosis of the dissociative subtype of PTSD.


6. Current participation in Cognitive Behavioral Therapy (CBT) or other psychotherapeutic interventions (occasional use of supportive services, such as those provided by a social worker, case manager, or counselor, was exempted).

Participants’ demographic data, history of acquired brain injury, and PTSD characteristics were obtained from the electronic patient record system (EPR).

Measures

Primary Outcome Measures: PTSD Symptoms

Three different instruments were used to assess PTSD symptoms:

1. Daily measure: Used to provide sufficient data points for SCED studies.


2. Weekly measure: A validated scale measuring intrusion and avoidance symptoms was used to provide a more robust assessment across phases.


3. Gold-standard structured clinical interview: Used as the criterion measure for PTSD to assess symptoms over the past month.

Daily Measure

Participants were asked to rate the severity of the following symptoms experienced during the preceding night and day on a 5 point scale:

• Re-experiencing (intrusions)


• Avoidance


• Heightened arousal (hyperarousal)


• Negative alterations in cognition and mood

The total score calculated from these four items ranges from 0 to 16 (see Appendix B, online Supplementary Material, for details). For this purpose, the first four items of the Dutch version (Bakker et al., 2014) of the Primary Care PTSD Screen 5 (Prins et al., 2016) were adapted to a rating-scale format.

Impact of Event Scale (IES)

The Impact of Event Scale (IES) (Brom & Kleber, 1985; Horowitz et al., 1979) is a self-report scale measuring psychological distress associated with a specific event. It contains 15 items rated on a 4-point scale, with a total score ranging from 0 to 75. The scale specifically measures intrusion and avoidance symptoms.

The IES has been used to screen for PTSD symptoms in individuals with acquired brain injury and has demonstrated valid psychometric properties (van der Ploeg et al., 2004; Huenges Wajer et al., 2018; Utz et al., 2019). A total score of 26 or higher indicates a clinically significant posttraumatic stress response (Rodenburg et al., 2009).

Clinician-Administered PTSD Scale for DSM-5 (CAPS-5)

The CAPS-5 (Boeschoten et al., 2014; Weathers, Blake et al., 2013) is a structured interview used to assess clinician-rated PTSD symptom severity and diagnostic classification. With strong psychometric properties, the CAPS-5 is the most widely accepted criterion measure for PTSD (Weathers et al., 2001). PTSD is diagnosed when all DSM-5 criteria A–G are met.

 

Secondary Outcome Measures

General Mental Health

The Outcome Questionnaire-45 (OQ45) (de Jong et al., 2007; Lambert et al., 2004) is a self-report scale assessing psychological distress, interpersonal functioning, and social role dysfunction over the past week.

• The scale contains 45 items rated on a 5-point Likert scale.


• The total score ranges from 0 to 180.


• The scale has adequate psychometric properties, and the clinical cutoff is 56; scores below this indicate normal functioning (Timman et al., 2017).

The OQ45 was used as a generalization measure.

Cognitive Functioning

The following tests were used to assess cognitive functioning:

• Working memory and attention: The scaled score of the Wechsler Adult Intelligence Scale–IV Digit Span (DS) subtest was used (Wechsler, 2008).


• Information processing speed and response inhibition: The Stroop Color–Word Test (SCWT) (Hammes, 1973; Stroop, 1935) was used. This test measures information processing speed using color and word completion times and response inhibition using the interference condition completion time.


• Psychomotor speed and cognitive flexibility: The Trail Making Test (TMT) (Reitan, 1956) was used:
  
  
  
  • Part A: A response-time measure of psychomotor speed (Time A).
  
  
  • Part B: A measure of set-shifting/cognitive flexibility (Time B).
  
  
  
  


• Verbal memory: The Rey Auditory Verbal Learning Test (RAVLT) (Brand & Jolles, 1985; Rey, 1958) was used. This test was evaluated using the total score across trials 1–5 and the delayed recall score.


• Visuospatial memory: The Location Learning Test (LLT) (Bucks & Willison, 1997; Kessels et al., 2006) was used. This test was assessed using the total displacement score.

Alternate versions of the RAVLT and LLT were used to minimize practice effects (van der Elst et al., 2008; van den Berg et al., 2010).

Intervention

EMDR therapy was delivered using the standardized 8-phase EMDR protocol (de Jongh & ten Broeke, 2011; Shapiro, 2001).

Treatment included all eight phases of EMDR. However, no safe-place installation or other stabilization technique was used. See Appendix C in the Online Supplementary Material for a detailed description of the procedure.

• Eye movements were preferred as the primary stimulus. However, alternative stimuli were used when horizontal eye movements were difficult because of hemianopia or oculomotor problems.


• The “index trauma” was defined as the traumatic event causing the client the greatest distress and central to their intrusion symptoms. One or more targets related to this event were addressed first during EMDR.


• Treatment was delivered in weekly sessions, each lasting a maximum of 90 minutes.


• When cognitive impairments made longer sessions challenging, sessions were shortened to a minimum of 45 minutes.


• The number of sessions was determined by the client’s individual needs and the therapist’s clinical assessment of PTSD severity; a maximum of 10 sessions was delivered.


• If changes to the standard EMDR protocol were required (for example, because of difficulty understanding instructions), these were documented in the electronic patient record (EPR).

EMDR treatment ended when the SUD (Subjective Units of Disturbance) score reached zero for all intrusions and, if present, images associated with trauma-related anticipatory fear.

Every EMDR session was video-recorded. Any serious adverse events observed during EMDR (e.g., self-harm, suicide attempts, situations requiring crisis intervention) were documented.

EMDR treatment was delivered at a desk in a therapy room at the outpatient clinic. The therapists were the first author (EJ) and another neuropsychologist, both specialists experienced in working with clients with ABI who had completed an EMDR Europe–approved Basic Training Program.

Treatment Integrity

To ensure treatment integrity and reliability, a randomly selected 25% sample of each participant’s video-recorded sessions was evaluated (Tate et al., 2013). The assessment was performed by an EMDR-trained clinical psychologist independent of the study and the therapists. The assessor was blinded to treatment outcomes and used a checklist containing 16 key elements of the EMDR treatment protocol. Each item was scored 0 or 1, yielding a total score of 0–16 for each session.

 

Procedure

• Clients were screened for eligibility when diagnosed with PTSD by a psychiatrist or psychologist at the outpatient clinic.


• Clients who met inclusion criteria 2–7 were informed about the study.


• After written informed consent was obtained, the CAPS-5 assessment was conducted by an independent neuropsychologist trained in administering this measure.


• Participants with a confirmed PTSD diagnosis were assigned a random baseline duration, and treatment sessions were scheduled.


• The treatment approach was explained using an informational brochure on EMDR (Vereniging EMDR Nederland, 2018), provided as part of the informed consent process.


• See Table 1 for the measurement schedule and phase durations.


• To remind participants to complete the daily measure, a notification was sent to their smartphones through an e-mental health application (Minddistrict, 2020) in the evening between 18:30 and 20:30, at the time they considered most convenient.


• The Impact of Event Scale (IES) was completed weekly using the same method.


• Participants received an explanation of how to use the e-mental health platform and application and practiced using them.


• Baseline assessments began using the daily measure and IES and continued throughout phases A–C and follow-up.


• The daily measures and IES were automatically processed by the e-mental health platform and linked to the electronic patient record system (EPR).

Analyses

• Demographic data, disease-related variables, and treatment characteristics (e.g., total number of sessions, session duration, stimuli used during desensitization) were analyzed using descriptive statistics.

Primary Outcome Measures

Visual Analyses

• Daily measure and IES scores were plotted for each participant in R using the Shiny SCDA application (Shiny Single-Case Data Analysis), and visual analyses were performed (Bulté & Onghena, 2013).


• Intervention-related changes in PTSD symptoms and the duration of these changes were determined.


• Based on the procedure described by Lane and Gast (2014), analyses focused on the following factors:
  
  
  
  • Level, trend, variability, and overlap.
  
  
  • The proportion of overlapping data between phases.
  
  
  
  


• No adjustments were made for missing data.


• Because an immediate response to treatment was not expected (Krasny-Pacini & Evans, 2018), immediacy of response was not included as a primary assessment factor.


• To evaluate the robustness of the intervention, maintenance and follow-up scores were examined to analyze the duration of change (whether sustained or temporary) (Manolov et al., 2016).

Level Analysis

• The baseline and treatment phases were evaluated using the median, range, and stability envelope.


• Within-phase level change was calculated using the following formula:
  
  
  
  • Relative change = (Median of the second half of the phase - median of the first half).
  
  
  
  


• The stability envelope criterion was defined as follows:
  
  
  
  • At least 80% of baseline data were required to fall within ±25%.
  
  
  • The same stability envelope was applied to the treatment phase (Barton et al., 2018).
  
  
  
  

Between-Phase Level Change

• This was calculated using the change in median:
  
  
  
  • (Treatment phase median - Baseline phase median).
  
  
  
  

Trend Analysis

• Trend lines for the baseline and treatment phases were calculated using the Theil–Sen method (Sen, 1968; Tarlow, 2017; Theil, 1950).


• Within-phase trends were visualized using the Shiny Architecta application (Bouwmeester, 2021).

Variability Analysis

• A stability envelope (25%) was applied to the trend lines in the baseline and treatment phases.


• Variability was determined by calculating the percentage of data points within the envelope.

Statistical Analyses

Primary Outcome Measures

Tau-U Analysis

• Tau-U analysis was used to determine the degree of overlap between data points across phases A, B, C, and D for the daily measure and the Impact of Event Scale (IES).


• Tau-U combines overlap and trend to control for an improving baseline trend (Tate & Perdices, 2019).


• It is highly robust to autocorrelation effects inherent in single-case studies (Barnard-Brak et al., 2021; Parker et al., 2011).


• For the daily measure, if a significant (p < .05) monotonic baseline trend was present, an adjusted Tau-Uadjusted value was calculated using the method recommended by Tarlow (2017) to reduce the risk of type I error (Tarlow, 2016).


• For the IES, when there were 5 or fewer baseline measurement points, the trend was corrected using the Tau-Utrend A method (Fingerhut et al., 2021; Parker et al., 2011).


• Tau-U analyses were performed using the online calculator (Vannest et al., 2016) at singlecaseresearch.org. However, a different calculation method was used when baseline trend correction was required.


• p-values were reported, but it was noted that they should be interpreted cautiously because of susceptibility to autocorrelation effects inherent in single-case studies.


• Effect size interpretations (Vannest & Ninci, 2015):
  
  
  
  • < .2: Small
  
  
  • .2 - .6: Moderate
  
  
  • .6 - .8: Large
  
  
  • > .8: Large to very large
  
  
  
  

Reliable Change Index (RCI) Analysis

• The Reliable Change Index (RCI) was calculated using the method of Jacobson and Truax (1991).


• The statistical significance of individual changes in IES scores was analyzed between baseline (median) and follow-up.


• A two-tailed z-test was used to determine whether the measured change was not attributable to measurement error (considered statistically significant at p < .05).


• Statistical significance criterion: Z < -1.96 or Z > 1.96


• RCI calculations for the IES used data from the largest sample in the validation study (van der Ploeg et al., 2004).


• Clinically significant change (Jacobson & Truax, 1991) was determined by whether the final treatment-phase score, median maintenance-phase score, and follow-up scores fell below the clinical cutoff (< 26).

CAPS-5 Analysis

• CAPS-5 total scores were compared using the RCI method to evaluate the treatment effect and its maintenance (T0–T1, T0–T2).


• The RCI for the CAPS-5 total severity score was calculated using validation study data (Boeschoten et al., 2018) (see Appendix D, Online Supplementary Material, for details).


• PTSD remission rates were reported for each individual based on whether they continued to meet DSM-5 PTSD diagnostic criteria at T1 and T2 (yes/no).

Secondary Outcome Measures

General Mental Health

• OQ45 total scores were analyzed over time using a J&T RCI value of 18 (T0–T1, T0–T2) (Timman et al., 2017).

Cognitive Functioning

• J&T RCI calculations were performed for the cognitive measures (T0–T1 comparison).


• The following validation studies were used as sources:
  
  
  
  • DS (Sayı Dizisi Alt Testi): Wechsler (2012)
  
  
  • LLT (Konum Öğrenme Testi): van den Berg ve ark. (2010), Kessels ve ark. (2012)
  
  
  • SCWT (Stroop Renk-Kelime Testi) - Renk ve Kelime koşulları: Levine ve ark. (2004)
  
  
  • SCWT (Stroop) - Girişim koşulu: van der Elst ve ark. (2008)
  
  
  • RAVLT (Rey İşitsel Sözel Öğrenme Testi): van der Elst ve ark. (2008)
  
  
  • TMT (İz Sürme Testi) A ve B: Dikmen ve ark. (1999)
  
  
  
  

Results

Participant Characteristics

• Participants ranged in age from 32 to 53 years.


• Time since injury ranged from 6 months to 24 years.


• All participants had a diagnosis of a neurocognitive disorder and were living independently at the time of participation.


• Participants’ demographic data, acquired brain injury (ABI) characteristics, and treatment variables are presented in Table 2.


• No participants dropped out of the study.

 

Case Formulation

Participant AA

• AA experienced a stroke in hospital following heart surgery and subsequently received outpatient rehabilitation.


• Persistent cognitive-communication impairments made communication with their spouse and three children difficult.


• There were mild impairments in sustained and divided attention.


• AA had successfully returned to work but had moved to a different position requiring less communication.


• A relative had recently experienced a heart attack and sustained a severe brain injury.


• Four months before this event, AA had developed late-onset PTSD related to their own stroke.


• PTSD symptoms included:
  
  
  
  • Distressing memories of the stroke when trying to fall asleep
  
  
  • Sleep problems
  
  
  • Avoidance of the relative
  
  
  • Anxiety and feeling detached from others
  
  
  • Irritability, hypervigilance, and difficulty concentrating
  
  
  
  

 

Participant BB

• BB sustained a severe traumatic brain injury (TBI) in a traffic accident.


• Initially, BB exhibited severe cognitive impairments:
  
  
  
  • Aphasia
  
  
  • Slowed information processing
  
  
  • Attention and memory impairments
  
  
  • Posttraumatic epilepsy
  
  
  • Sleep problems
  
  
  
  


• BB received intensive inpatient and outpatient rehabilitation.


• BB was recently divorced and had two children.


• One and a half years earlier, BB had experienced a severe sexual assault and received hospital treatment for the injuries.


• PTSD symptoms included:
  
  
  
  • Distressing memories when interacting with men or watching television
  
  
  • Nightmares
  
  
  • Avoidance of reminders by isolating at home
  
  
  • Negative beliefs about self and others
  
  
  • Feelings of anxiety, anger, guilt, and shame
  
  
  • Hypervigilance
  
  
  • Loss of interest in previously enjoyable activities
  
  
  • Difficulty experiencing positive emotions
  
  
  • Increased concentration difficulties and sleep problems
  
  
  
  

 

Participant CC

• CC experienced a stroke at work and subsequently showed the following persistent symptoms:
  
  
  
  • Paralysis of the left leg and face
  
  
  • Neglect of the left arm and leg
  
  
  • Impaired fine motor skills in the left arm and hand
  
  
  • Left hemianopia
  
  
  • Dysarthria
  
  
  • Persistent headaches, fatigue, and balance problems
  
  
  
  


• Impairments in working memory, executive functioning, social cognition, and communication skills were identified.


• Outpatient rehabilitation focused on motor skills and energy management.


• CC was married with three children.


• PTSD symptoms included:
  
  
  
  • Distressing flashbacks when confronted with stroke-related limitations
  
  
  • Avoidance of stroke-related reminders
  
  
  • Feelings of anxiety, anger, and shame
  
  
  • Irritability and sleep problems
  
  
  • Difficulty concentrating
  
  
  
  

Participant DD

• DD sustained a traumatic brain injury (TBI) as a result of a violent assault.


• Since the event, DD had experienced the following PTSD symptoms:
  
  
  
  • Assault-related intrusions (flashbacks), particularly triggered when interacting with people or seeing related images on television
  
  
  • Nightmares and sleep problems
  
  
  • Avoidance of other people by isolating at home
  
  
  • Negative beliefs about self and others
  
  
  • Feelings of anxiety, anger, guilt, and shame
  
  
  • Inability to enjoy previously pleasurable activities
  
  
  • Difficulty experiencing positive emotions
  
  
  • Hypervigilance, concentration difficulties, and memory problems
  
  
  • Sleep difficulties
  
  
  
  


• DD was single and ran their own business.


• Both criminal and civil proceedings were ongoing at the time of participation.

Treatment Characteristics

• Median number of treatment sessions: 4.5 (range = 3–9) (see Table 2).


• For participant BB, the second session was scheduled 2 weeks after the first because of scheduling difficulties.


• For participant DD, there were 2-week intervals between sessions 3 and 4 and between sessions 4 and 5.
  
  
  
  • These intervals occurred because the therapist judged that an EMDR session would be inappropriate at those times, as DD was emotionally preoccupied with ongoing events.
  
  
  
  


• No changes were made to the standard EMDR protocol.


• No serious adverse events were reported.


• Overall treatment integrity rate: 94% (75 of 80 sessions fully adhered to the protocol).

 

Outcome Measures

• Daily measure and IES scores across the four phases (A, B, C, D) are presented graphically (see Figure 1).


• Participants’ baselines were conducted at different times (Harvey et al., 2004).


• See Appendix E for percentages of missing data and Appendices F and G for trend lines and stability envelopes.


• Results of within-phase visual analyses of daily measure and IES scores are presented in Table 3.


• Tau-U statistical analysis results are presented in Table 4.


• See Appendix H for CAPS-5 and secondary outcome measure results.

 

Results for Participant AA

Daily Measure

• Visual analysis (Figure 1) showed that the data were variable (<80% within the stability envelope) and worsened during baseline.


• During the intervention, scores were stable and decreased, indicating improvement.


• Scores remained stable at the lowest level during maintenance and follow-up.


• Theil–Sen median regression lines (see Table 3 and Appendix F) showed:
  
  
  
  • An accelerating (worsening) trend at baseline,
  
  
  • A leveling trend during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Baseline data were considered variable,
  
  
  • Intervention-phase data were found to be stable.
  
  
  
  


• A decrease in median level (improvement) was observed between phases (A–B, B–C).


• Scores remained at the lowest level during follow-up.


• Tau-U analysis results:
  
  
  
  • Large and significant effect sizes from baseline to treatment and from treatment to maintenance (large Tau-U, p < .05).
  
  
  • A very small (nonsignificant) Tau-U value from maintenance to follow-up, indicating complete score overlap and no change.
  
  
  
  

Impact of Event Scale (IES)

• Scores were stable and decreased, indicating improvement, during baseline and intervention.


• Theil–Sen median regression lines (Table 3) showed:
  
  
  
  • A decelerating (improving) trend at baseline,
  
  
  • A more pronounced improving trend during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data in both the baseline and intervention phases were found to be stable.
  
  
  
  


• A decrease in median level was observed across the first three phases, and scores remained at the lowest level during maintenance and follow-up.


• Tau-U results:
  
  
  
  • Large to very large effect sizes from baseline to treatment and from treatment to maintenance (both significant).
  
  
  
  


• A statistically significant improvement was observed at follow-up compared with baseline (Reliable Change (RC) = 4.68, improvement).


• The final treatment-phase score, median maintenance-phase score, and follow-up score represented clinically significant change (IES score < 26).

CAPS-5, General Mental Health, and Cognitive Functioning Results

• The CAPS-5 total severity score decreased significantly from baseline (T0) to post-treatment (T1) and follow-up (T2) (RC = 5.08).


• PTSD was diagnosed at T0, but diagnostic criteria were no longer met at T1 and T2.


• The general mental health measure, OQ45, showed reliable change from T0 to T1 (RC = 3.05), with a stronger change from T0 to T2 (RC = 4.14).


• A significant change in cognitive performance was observed only on the Digit Span (DS) subtest (RC = 2.38, improvement).

 

Results for Participant BB

Daily Measure

• Visual analysis:
  
  
  
  • Baseline was stable, whereas variability was observed during the intervention.
  
  
  • Symptoms temporarily increased at the start of treatment.
  
  
  • The baseline level was stable, with a decrease (improvement) during the intervention.
  
  
  
  


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • An accelerating (worsening) trend was observed at baseline,
  
  
  • A decelerating (improving) trend was identified during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Baseline was considered stable, but the intervention phase was considered variable.
  
  
  
  


• There was no change in median level from baseline to treatment, but scores decreased during the post-treatment maintenance phase.


• Scores remained at the lowest level from maintenance to follow-up.


• Tau-U analysis results:
  
  
  
  • A small (nonsignificant) effect size from baseline to treatment
  
  
  • A large to very large effect size from treatment to maintenance (significant)
  
  
  • A very small (nonsignificant) Tau-U value from maintenance to follow-up, indicating complete score overlap and no change.
  
  
  
  

Impact of Event Scale (IES)

• Baseline was stable, whereas the intervention phase showed variability.


• Scores decreased (improved) during both baseline and intervention.


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • A decelerating (improving) trend was observed at baseline,
  
  
  • A more pronounced improving trend was observed during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data in both baseline and intervention were found to be stable.
  
  
  
  


• A decrease in median level (improvement) was observed across the first three phases, but there was a small increase from maintenance to follow-up.


• Tau-U results:
  
  
  
  • A large (but nonsignificant) effect size from baseline to treatment
  
  
  • A large to very large effect size from treatment to maintenance (significant)
  
  
  
  


• A statistically significant improvement was observed at follow-up compared with baseline (Reliable Change (RC) = 11.09, improvement).


• The final treatment-phase score, median maintenance-phase score, and follow-up score represented clinically significant change (IES score < 26).

CAPS-5, General Mental Health, and Cognitive Functioning Results

• The CAPS-5 total severity score decreased significantly from baseline (T0) to post-treatment (T1) and follow-up (T2) (RC = 4.10 for T0–T1, RC = 3.12 for T0–T2).


• PTSD was diagnosed at T0, but diagnostic criteria were no longer met at T1 and T2.


• No reliable change was observed in the OQ45 score or cognitive test results.

 

Results for Participant CC

Daily Measure

• Visual analysis:
  
  
  
  • Scores were variable at baseline but stable during the intervention.
  
  
  • A decrease (improvement) was observed at baseline, which leveled off during the intervention.
  
  
  
  


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • A significant decelerating (improving) trend was observed at baseline,
  
  
  • The trend remained flat during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data were considered variable at baseline but stable during the intervention.
  
  
  
  


• A decrease in median level (improvement) was observed from baseline to treatment, and scores remained low during post-treatment maintenance.


• However, there was a small increase from maintenance to follow-up.


• Tau-U analysis results:
  
  
  
  • Adjusted Tau-U (Tau-Uadjusted) from baseline to treatment was moderate and significant in the direction of deterioration.
  
  
  • This result contradicted the visual analysis (Figure 1).
  
  
  • Tau-U from treatment to maintenance was moderate and significant in the direction of improvement.
  
  
  • Tau-U from maintenance to follow-up was moderate and significant in the direction of deterioration.
  
  
  
  

Impact of Event Scale (IES)

• Visual analysis:
  
  
  
  • Scores remained stable during baseline and intervention.
  
  
  • A decrease (improvement) was observed during baseline and intervention.
  
  
  
  


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • A decelerating (improving) trend was observed at baseline,
  
  
  • A less pronounced improving trend was observed during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data in both baseline and intervention were found to be stable.
  
  
  
  


• A decrease in median level (improvement) was observed between phases.


• Tau-U results:
  
  
  
  • Large to very large effect sizes from baseline to treatment and from treatment to maintenance (both significant).
  
  
  
  


• A statistically significant improvement was observed at follow-up compared with baseline (Reliable Change (RC) = 7.45, improvement).


• The final treatment-phase score, median maintenance-phase score, and follow-up score represented clinically significant change (IES score < 26).

 

CAPS-5, General Mental Health, and Cognitive Functioning Results

• The CAPS-5 total severity score decreased significantly from baseline (T0) to post-treatment (T1) and follow-up (T2) (RC = 3.12 for T0–T1, RC = 2.30 for T0–T2).


• PTSD was diagnosed at T0, but diagnostic criteria were no longer met at T1 and T2.


• No reliable change was observed in the OQ45 score.


• A significant change in cognitive tests was observed only on the Location Learning Test (LLT) (RC = 2.74, improvement).

Results for Participant DD

Daily Measure

• Visual analysis:
  
  
  
  • Scores remained stable during baseline and intervention.
  
  
  • Levels decreased (improved) during baseline and intervention.
  
  
  
  


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • A decelerating (improving) trend was observed at baseline.
  
  
  • A less pronounced improving trend was observed during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data in both baseline and intervention were found to be stable.
  
  
  
  


• A decrease in median level (improvement) was observed between phases:
  
  
  
  • A decrease was observed from baseline to treatment,
  
  
  • Stability was achieved from treatment to maintenance,
  
  
  • A small level change was observed from maintenance to follow-up.
  
  
  
  


• Tau-U analysis results:
  
  
  
  • A large (significant) effect size from baseline to treatment.
  
  
  • A moderate (significant) effect size from treatment to maintenance.
  
  
  • A large to very large (significant) effect size from maintenance to follow-up.
  
  
  
  

Impact of Event Scale (IES)

• Visual analysis:
  
  
  
  • Scores remained stable during baseline and intervention.
  
  
  • Levels decreased (improved) during baseline and intervention.
  
  
  
  


• Theil–Sen median regression trend lines (Table 3):
  
  
  
  • A decelerating (improving) trend was observed at baseline.
  
  
  • A less pronounced improving trend was observed during the intervention.
  
  
  
  


• After applying a stability envelope to the trend lines:
  
  
  
  • Data in both baseline and intervention were found to be stable.
  
  
  
  


• A decrease in median level (improvement) was observed between phases.


• Tau-U results:
  
  
  
  • A large to very large effect size from baseline to treatment (significant).
  
  
  • A large to very large effect size from treatment to maintenance (significant).
  
  
  
  


• A statistically significant improvement was observed at follow-up compared with baseline (Reliable Change (RC) = 4.33, improvement).


• The final treatment-phase score, median maintenance-phase score, and follow-up score represented clinically significant change (IES score < 26).

 

 

CAPS-5, General Mental Health, and Cognitive Functioning Results

• The CAPS-5 total severity score decreased significantly from baseline (T0) to post-treatment (T1) and follow-up (T2) (RC = 3.28 for T0–T1, RC = 5.09 for T0–T2).


• PTSD was diagnosed at T0, but diagnostic criteria were no longer met at T1 and T2.


• The general mental health measure, OQ45, showed a significant decrease from T0 to T1 (RC = 3.05), but no significant change was observed from T0 to T2.


• A significant change in cognitive tests was observed only on Trail Making Test A (TMT A) (RC = 1.98, improvement).

Discussion

The results of this study provide empirical evidence supporting the effectiveness of EMDR in reducing PTSD symptoms in individuals with acquired brain injury (ABI).

• For weekly measured PTSD symptoms (IES):
  
  
  
  • The effect size from baseline to treatment was large to very large and significant for three participants.
  
  
  • The effect size was large but nonsignificant for one participant.
  
  
  
  


• For daily measured PTSD symptoms:
  
  
  
  • The effect size from baseline to treatment was small and nonsignificant for one participant.
  
  
  • The effect size was large and significant for two participants.
  
  
  • One participant’s results were inconsistent with visual analysis and inconclusive.
  
  
  
  


• Notably, all four participants showed moderate to very large improvements (significant) from treatment to maintenance.


• For daily measured PTSD symptoms at follow-up:
  
  
  
  • The treatment effect was maintained for three participants.
  
  
  • A moderate increase in symptoms (significant) was observed in one participant.
  
  
  
  

All participants:

• Showed reliable (significant) improvement in weekly PTSD scores from baseline to follow-up.


• Demonstrated clinically significant change in PTSD.


• No longer met the diagnostic criteria for PTSD at post-treatment and follow-up (as assessed with the CAPS-5).


• Showed a reliable (significant) reduction in PTSD severity scores (as assessed with the CAPS-5).

 

Treatment Safety and Feasibility

• There were no serious adverse events or dropouts.


• No changes to the standardized EMDR treatment protocol were required.


• This supports the feasibility and acceptability of EMDR.


• The therapist shortened session duration for two participants because of fatigue.


• All participants completed treatment within a maximum of ten sessions.

Effects on General Mental Health and Cognitive Functioning

• Significant improvement in the general mental health measure (OQ45) was observed:
  
  
  
  • In two participants from baseline to post-treatment.
  
  
  • In one participant from baseline to follow-up.
  
  
  • These findings suggest that treatment effects may generalize to overall mental health.
  
  
  
  


• Three participants showed significant improvement on cognitive tests:
  
  
  
  • Psychomotor speed
  
  
  • Visuospatial memory
  
  
  • Working memory/attention
  
  
  
  

 

Differences Between Daily and Weekly PTSD Measures

• The results of the daily and weekly measures did not fully align.


• The main reasons for this may include:
  
  
  
  1. The time frame covered by the instruments:
     
     
     
     • The daily measure assesses symptoms experienced over the past 24 hours, whereas the weekly measure covers a broader time frame.
     
     
     
     
  
  
  2. The scope of symptoms measured:
     
     
     
     • The daily measure provides a brief assessment of the four core PTSD symptom domains (intrusions, avoidance, hypervigilance, and alterations in cognition and mood).
     
     
     • The weekly measure assesses only two symptom domains (intrusions and avoidance) in greater detail.
     
     
     
     
  
  
  
  

These differences may explain why some participants showed different symptom patterns across the daily and weekly measures.

Discussion

The small effect size observed for one participant on daily measures from baseline to treatment may be explained by a temporary worsening of symptoms at the start of EMDR. This is not uncommon in trauma-focused therapies and may result from more intensive exposure to traumatic memories (Burger et al., 2023). However, symptoms decreased as treatment progressed, and a large (significant) change was observed during the maintenance phase.

• The participant who showed a moderate increase in PTSD symptoms from baseline to treatment already had a decreasing symptom trend during baseline, and this trend was adjusted for in the analysis (TauUadjusted).


• However, visual analysis showed that symptoms did not increase during treatment compared with baseline.


• This may suggest that the information provided during study enrollment positively affected the participant’s daily measured PTSD symptoms during baseline.


• Nevertheless, a similar decreasing baseline trend was not observed in weekly measured PTSD symptoms.


• Furthermore, the bounded range of the data (0 to 16) and the inability of baseline trend correction to adjust these out-of-range predictions may have produced results that contradicted the visual analysis (Chen et al., 2019).


• However, the large to very large improvement observed on the weekly measure (significant) suggests that treatment was effective.

Overall, EMDR treatment is considered to have been effective for this individual, but this finding should be interpreted cautiously because the daily measure results were inconclusive.

 

Post-Treatment PTSD Symptoms

• For most participants, the reduced PTSD symptoms remained stable or decreased further after EMDR treatment.


• This outcome has previously been observed in trauma-focused treatments (Berg et al., 2018; van der Kolk et al., 2007).

Effects on General Mental Health

• Three participants showed no significant change in general mental health (OQ45) from baseline to follow-up.


• However, two of these three participants showed positive changes that were not statistically significant.


• Given that the OQ45 measures general mental health rather than specifically targeting PTSD symptoms, brain injury and/or additional psychiatric problems may have contributed to persistent general mental health complaints in these individuals.


• For example, work-related problems persisted at follow-up for three participants, whereas general mental health improved markedly in the participant who was able to return to work.

Effects on Cognitive Functioning

• No consistent, significant improvement in cognitive functioning was observed across all participants.


• This finding is inconsistent with previous group studies (Haaland et al., 2016; Nijdam et al., 2018; Walter et al., 2010).


• However, a recent systematic review found improvement only in memory functioning, with no significant changes in attention, executive functioning, or information processing speed (Susanty et al., 2022).


• In group studies, small mean changes can become statistically significant even when individual changes are not significant (Hays & Peipert, 2021).


• The use of individual change scores in this study may have failed to capture the effect reported in previous group studies.


• Furthermore, previous studies (Haaland et al., 2016; Nijdam et al., 2018; Walter et al., 2010) did not include individuals with moderate or severe brain injury.


• In the current study, cognitive functioning was thought to be affected by brain injury, such that a reduction in PTSD symptoms would not directly improve cognitive functioning.


• The absence of changes in cognitive functioning despite reduced PTSD symptoms in individuals with brain injury may indicate persistent cognitive problems of organic origin.


• Future larger-scale studies should address this issue in individuals with ABI and PTSD.

 

Comparisons in the Context of EMDR and Dementia

• The number of studies on EMDR in individuals with ABI is very limited.


• Studies on the effectiveness of EMDR in individuals with dementia were therefore reviewed.


• A recent systematic review provided some evidence that EMDR is feasible and effective in patients with dementia. However, these findings were based on only three observational case studies involving five cases (Ruisch et al., 2023).


• In addition, a Delphi study found expert consensus that EMDR may be an appropriate intervention for patients with mild, moderate, and possibly severe dementia who experience intrusions (Driessen et al., 2023).

Strengths and Limitations

The main strength of this study is that it is the first randomized single-case experimental design (SCED) study to investigate the effectiveness of standard EMDR for treating PTSD in individuals with acquired brain injury (ABI).

• The study is highly rigorous in terms of external validity within SCED methodology and meets current scientific standards.


• A maintenance phase and a 3-month follow-up were included to assess whether effects persisted after EMDR treatment ended.


• A brief daily measure provided sufficient measurement points, and diagnosis was established using a gold-standard instrument.

Limitations

• The methodological quality assessment revealed weaknesses in internal validity.
  
  
  
  • In particular, the nonconcurrent multiple baseline design did not allow control of the threat to internal validity posed by shared historical effects (Michiels & Onghena, 2019).
  
  
  • However, random assignment to multiple baseline durations can be considered a design advantage that reduces systematic bias and strengthens internal validity.
  
  
  
  


• There is no agreed-upon standard method for statistical analysis of SCED data (Kratochwill et al., 2013; Smith, 2012).
  
  
  
  • This study used Tau-U, a method commonly employed in SCED studies.
  
  
  • However, results may differ if alternative statistical methods are applied (Brossart et al., 2006).
  
  
  
  


• No stand-alone performance validity tests were used in the cognitive assessment.
  
  
  
  • Nevertheless, a retrospective analysis of embedded performance validity indicators, such as RAVLT recognition performance (Boone et al., 2005) and WAIS Reliable Digit Span (Schroeder et al., 2012), was conducted.
  
  
  • Because no participant failed both indicators, the data were considered reliable (Larrabee, 2022).
  
  
  
  


• The study participants may not fully represent the general ABI population.
  
  
  
  • This is because sufficient cognitive and communication abilities were required for inclusion.
  
  
  • For example, patients with global aphasia were not included.
  
  
  • However, with some adaptations to the treatment protocol, EMDR is likely to be feasible and effective for individuals with more severe cognitive or communication impairments as well (Amano & Toichi, 2014; Guina & Guina, 2018; Janssen & van Donzel, 2023).
  
  
  
  

 

Recommendations for Future Research

• Future SCED research should use a validated instrument suitable for weekly administration.
  
  
  
  • In addition, the number of measurements should be increased by using a baseline phase lasting at least 5 weeks.
  
  
  • Ideally, a brief PTSD instrument with established validity and reliability for daily use should be developed.
  
  
  • No such instrument was available when the current study was designed, and none is currently available.
  
  
  • However, shortened versions of the PTSD Checklist for DSM-5 (PCL-5), namely the PCL-4 and PCL-8, appear promising in this respect (Alting van Geusau et al., 2021; Borst et al., 2024).
  
  
  
  


• The use of gold-standard structured clinical interviews is also recommended.
  
  
  
  • This approach was used in the current study and should be adopted in future research.
  
  
  
  

Conclusions

• This SCED study provides empirical evidence for the effectiveness of EMDR.
  
  
  
  • PTSD symptoms decreased, and all participants no longer met PTSD diagnostic criteria after treatment.
  
  
  • Treatment effects persisted at the 3-month follow-up.
  
  
  
  


• A standardized EMDR treatment protocol was implemented, with no changes to the protocol required.


• No consistent improvement in cognitive functioning or general mental health was observed following EMDR treatment.


• Clinicians working with individuals with ABI and PTSD are encouraged to consider EMDR as a treatment option for PTSD.


• Future studies should focus on further strengthening internal validity, and findings should be replicated using randomized controlled group designs in addition to SCED studies.

Acknowledgments

The authors thank the participants who took part in this research.

Conflict of Interest Statement

The authors reported no potential conflicts of interest.

Funding

This study was supported by Vereniging EMDR Nederland under the project entitled “Effectiveness of Eye Movement Desensitization and Reprocessing (EMDR) Therapy in Individuals with Acquired Brain Injury.”

ORCID iD

Ellen P. J. Janssen: http://orcid.org/0000-0001-9626-3881