Abstract
Introduction
Individuals with a personality disorder (PD) often report high exposure to adverse events. Given the strong link between negative life events and PD pathology, it is essential to evaluate trauma-focused treatments in this population. This study evaluated the effectiveness of eye movement desensitization and reprocessing (EMDR) therapy versus waitlist in reducing post-traumatic stress disorder (PTSD) symptoms and diagnostic status, distinguishing between patients with and without PTSD. Second, it examined the impact of EMDR therapy on adverse event memories, both meeting and not meeting Criterion A for PTSD.
Methods
Participants (n = 159) were randomly assigned to EMDR therapy or a waitlist. PDs were assessed using the Structured Clinical Interview for DSM-5. PTSD symptoms and diagnoses were evaluated using the Clinician-Administered PTSD Scale for DSM-5. Secondary analyses examined the effectiveness of EMDR on various types of memory.
Results
Significant group-by-time interactions showed a greater reduction in PTSD symptoms in the EMDR therapy group, regardless of the baseline PTSD diagnosis. In patients with PTSD, large effects were found post-treatment (d = 1.26; control: d = 0.28) and at follow-up (d = 1.5; control: d = 0.69). In patients without PTSD, moderate to large effects emerged post-treatment (d = 0.77; control: d = 0.18) and at follow-up (d = 1.09; control: d = 0.46). EMDR therapy reduced symptoms associated with Criterion A trauma and other adverse events. Post-treatment, 65.5% of EMDR patients lost their PTSD diagnosis, increasing to 73.1% at the follow-up.
Conclusion
EMDR effectively reduced PTSD symptoms in individuals with a PD, regardless of PTSD diagnosis, and proved effective for memories that did not meet Criterion A.
Keywords: Personality disorders, Eye movement desensitization and reprocessing, Adverse events, Trauma, Post-traumatic stress disorder
Introduction
Individuals with a personality disorder (PD) experience high levels of distress that significantly affect multiple areas of their lives [1]. Although several treatments have been shown to be effective in individuals with a PD, they are often prolonged and resource-intensive [2], highlighting the need to enhance therapeutic options for this patient group.
The levels of exposure to adverse (childhood) events are high, with 73% reporting abuse and 82% reporting neglect [3]. Additionally, comorbid post-traumatic stress disorder (PTSD) is common, with 20–56% of patients meeting the criteria for PTSD [4]. A significant body of research links distressing and traumatic life events, such as physical abuse, emotional abuse, and non-Criterion A events such as neglect, to the development of all types of PDs [5, 6]. However, trauma-focused treatments are not routinely provided to this population [7].
Possible barriers to trauma-focused care for this patient group include the frequent underdiagnosis of PTSD [7, 8], partly attributable to severe difficulties with affect regulation, such as suicidal behavior and intense emotional responses that can mask the underlying trauma symptoms [9, 10]. However, even in the absence of a formal PTSD diagnosis, untreated memories of adverse events seem to contribute to the exacerbation and maintenance of PD pathology [5, 11], such as low self-esteem, difficulties in relationships, and emotional dysregulation [12, 13]. Therefore, trauma-focused therapy may be beneficial for this patient group.
Recent studies have found that eye movement desensitization and reprocessing (EMDR) therapy, a first-line evidence-based trauma-focused therapy [14], is effective in individuals with a PD and comorbid PTSD [4, 15]. Knowledge of trauma-focused psychotherapy in patients with a PD without PTSD is limited. However, a few case studies [16–18] and a randomized controlled trial (RCT) yielded promising results [19]. In the latter RCT, EMDR therapy targeted unprocessed memories of adverse events, most of which did not meet Criterion A, contributing to current symptoms and dysfunctional behavior patterns, with positive effects on psychological symptoms and personality functioning. The theoretical framework for this approach is based on the Adaptive Information Processing (AIP) model, which posits that unprocessed memories of adverse events can be triggered, driving current psychopathological symptoms [13, 20]. Furthermore, research has shown that non-Criterion A events, such as emotional neglect and emotional abuse, are likely to trigger PTSD symptoms of similar severity [21, 22] and are linked to the development of PDs [23]. Despite the potential benefits of trauma-focused treatment, no previous study has specifically examined its effectiveness in patients with a PD, differentiating between those with and without PTSD diagnosis.
The primary aim of this RCT was to assess the effectiveness of EMDR therapy, compared to a waitlist control, in reducing PTSD symptoms and diagnostic status in patients with a PD, distinguishing between those with and without PTSD. This study also examined how EMDR therapy affects various types of memories, including those related to events that do not meet Criterion A. It was hypothesized that EMDR therapy would yield positive outcomes even when focusing on memories of events that do not meet the Criterion A for PTSD and in the absence of a comorbid PTSD diagnosis. This study offers valuable insights into the potential of trauma-focused treatment for patients with PD and may enhance the treatment outcomes in this population.
Methods
This study was part of the Trauma-Focused EMDR for Personality Disorders among Outpatients (TEMPO) study [24]. In this single-blind, multicenter RCT, EMDR therapy was compared with a waiting list. Participants in the EMDR group received ten bi-weekly EMDR sessions over 5 weeks. Assessments were performed at baseline (pre-randomization, T0), post-treatment (after 5 weeks, T1), and at the 3-month follow-up (T3). Demographic data were collected at baseline. All researchers were blinded to the randomization process. The study was approved by the Medical Research Ethics Committee (MREC) of Erasmus Medical Center in Rotterdam (number MEC-2020-0583). This study was conducted and reported in accordance with the CONSORT guidelines. The completed CONSORT checklist is available in the Supplementary Material. The original outcomes were maintained throughout the study.
Participants
Participants were recruited from the outpatient clinics of two mental health institutions, GGZ Delfland and Parnassia Psychiatric Institute, in the Netherlands. The recruitment and follow-up period spanned from December 2020 to the third follow-up in February 2025. Patients were eligible if they were 18 years or older and diagnosed with a PD, as confirmed by the Structured Clinical Interview for DSM-5 Personality Disorders (SCID-5-PD). Participants with an “other specified” PD were required to present with at least ten PD symptoms. The exclusion criteria were inadequate Dutch proficiency and an estimated IQ of <70. Power analysis determined a sample size of 159 participants [24].
Interventions
After baseline measurements, participants were randomly assigned in a 1:1 ratio to either the EMDR therapy or waiting-list group using a randomization table in Data Manager. This study was a randomized controlled superiority trial with a parallel design. A case report form is generated in Data Manager, without blocks. The non-blind researcher reveals the participant’s allocation via the randomization button, informing participants through telephone or video calls. The principal investigator (K.S.) has access to all allocations, ensuring that unblinding is never necessary. Further details of the randomization process have been published previously [24].
EMDR Therapy
Half of the participants received a total of ten bi-weekly EMDR sessions of 90 min. EMDR therapy is a trauma-focused treatment in which attention is sequentially directed toward the distressing memory while simultaneously engaging the patient in another concurrent (dual-attention) task [14, 20]. First, the patients were informed about EMDR therapy. Next, a case conceptualization was conducted based on the patient’s symptoms, such as intrusions and the current core symptoms of PD. All memories considered to play a significant role in the patients’ current core symptoms were identified and processed. These included those related to Criterion A events and other adverse experiences, such as emotional abuse, neglect, the death of a loved one, or illness. A detailed case example and theoretical framework for this approach have been published previously [13].
Waiting List
The remaining half was assigned to the waiting-list control group. Patients in this group did not receive EMDR therapy or any other PD treatment. They were informed of their assignments and invited for post-treatment assessment (T1). At the 3-month follow-up (T3), participants from both groups met with their initial diagnostic therapist to determine the most appropriate treatment as usual for PD.
Treatment Training and Integrity
Protocol adherence and treatment fidelity were ensured in this study. EMDR therapy was provided by certified therapists who completed the Dutch Basic EMDR training and attended a 1-day workshop by an EMDR Europe-accredited trainer to ensure uniformity. The therapists received monthly supervision from EMDR Europe-approved consultants. All case conceptualizations and session reports were submitted to supervisors. Session recordings were used for feedback on technique and protocol adherence, and a subset was assessed by blinded reviewers for therapist competence and compliance with the protocol.
Outcome Measures
Structured Clinical Interview for DSM-5 Personality Disorders
The SCID-5-PD is a diagnostic tool for assessing DSM-5 PDs [25]. It rates each PD symptom on a 3-point scale to determine its presence or absence [26]. The interview includes 135 questions, some overlapping to capture different aspects of a symptom. Only trained clinicians should administer the SCID-5-PD because of the need for clinical expertise. The assessors were trained accordingly. Although Dutch reliability and validity data are unavailable, they are expected to match the nearly identical previous version, the SCID-II [27], which showed substantial to almost perfect inter-rater reliability (κ = 0.78–0.98) [28, 29].
Clinician-Administered PTSD Scale for DSM-5
The Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) is a structured diagnostic instrument designed to evaluate the severity of PTSD symptoms according to DSM-5 criteria [30]. It rates 20 PTSD symptoms based on the DSM-5's four main criteria: Criterion A (exposure to traumatic events), Criterion B (intrusive symptoms such as flashbacks and nightmares), Criterion C (avoidance of reminders), Criterion D (negative alterations in mood and cognition), and Criterion E (arousal symptoms such as irritability, sleep disturbances, and hypervigilance). These symptoms were rated on a scale of 0–5, with total scores reflecting overall PTSD severity. Additionally, ten questions addressed symptom duration, functional impairment, and dissociative symptoms. Psychometric evaluations have indicated strong internal consistency (α = 0.90) and excellent inter-rater reliability (ICC = 0.98) [31]. Assessors were trained in the administration of the CAPS-5.
Childhood Trauma Questionnaire – Short Form
The Childhood Trauma Questionnaire – Short Form (CTQ-SF) is used to screen for childhood trauma in both the clinical and general population [32]. Respondents evaluate 25 statements about their childhood experiences using a five-point Likert scale (e.g., “I had to wear dirty clothes”). The questionnaire includes five subscales: physical neglect, emotional neglect, and physical, emotional, and sexual abuse. The Dutch version, which demonstrates strong to excellent internal consistency (α = 0.89–0.95), will be used, although the physical neglect subscale shows lower reliability (α = 0.63) [33].
Statistical Analysis
Baseline differences were assessed using parametric tests. Outcome comparisons followed the intention-to-treat principle and included all participants. A linear mixed model (LMM) [34] examined fixed effects of time (baseline, post-treatment, and follow-up), group (EMDR vs. control), PTSD diagnosis, and their interactions. Since baseline PTSD symptom severity was expected to explain a significant part of the error variance, it was categorized1 and included as a covariate in the model to increase statistical power. The LMM included two levels. Level 1 examined within-person changes in CAPS-5 scores over time, incorporating fixed effects for PTSD severity, group, and diagnosis, along with random slopes for individual variations. Level 2 modeled between-person differences with random intercepts and slopes for baseline scores and rates of change. Cohen’s d was calculated to quantify the effect size, providing a standardized measure for easier interpretation and comparison with other studies. This was calculated by dividing the mean difference (within and between groups) by the pooled standard deviation. Effect sizes were categorized as small (0.2), medium (0.5), or large (0.8). A second LMM analyzed the impact of EMDR on memory types (emotional abuse, physical abuse, and others), using time and memory type as fixed effects and their interactions. A Type I error rate of 0.05 was maintained using two-tailed tests. Data were analyzed using SPSS 27 and R 4.4.2 with the nlme package.
Results
Patient Flow and Sample Characteristics
Table 1 presents the baseline demographic and clinical characteristics, including PTSD symptom severity and diagnostic status, by treatment group. No significant differences were found between the groups in any of the baseline characteristics.
Table 1.
Demographic characteristics at baseline according to the allocated group
| Variable | Total (n = 159) | EMDR group (n = 79) | Control group (n = 80) | Statistic |
|---|---|---|---|---|
| Mean age, years | 35.4 (12) | 34.3 (11.7) | 36.5 (12.2) | t(157) = 1,115, p = 0.266 |
| Gender | | | | χ2(1, N = 159) = 0.59, p = 0.808 |
| Male | 29 | 15 (19%) | 14 (17.5%) | |
| Female | 130 | 64 (81%) | 66 (82.5%) | |
| Personality cluster | | | | χ2(1, N = 159) = 0.45, p = 0.929 |
| A | 3 | 1 (1.3%) | 2 (2.5%) | |
| B | 54 | 26 (32.9%) | 28 (35%) | |
| C | 77 | 39 (49.9%) | 38 (47.5%) | |
| OS | 25 | 13 (16.4%) | 12 (15%) | |
| PTSD diagnosis | 62 | 32 (40.5%) | 30 (38%) | χ2(1, N = 158) = 0.11, p = 0.745 |
| PTSD criteria | | | | χ2(1, N = 155) = 2.37, p = 0.668 |
| Criterion A | 101 | 47 (59.5%) | 54 (68.4%) | |
| Criterion B | 126 | 63 (79.8%) | 63 (79.7%) | |
| Criterion C | 119 | 61 (77.2%) | 58 (73.4%) | |
| Criterion D | 142 | 71 (89.9%) | 71 (89.9%) | |
| Criterion E | 98 | 50 (63.3%) | 48 (60.8%) | |
| Criteria B + C + D + E | 22 | 12 (54.5%) | 10 (45.5%) | χ2(1, N = 158) = 0.21, p = 0.646 |
EMDR, eye movement desensitization and reprocessing therapy; OS, other specified.
Note: Personality cluster was determined using the Structured Clinical Interview for DSM-5 Personality Disorders (SCID-5-PD) and post-traumatic stress disorder diagnosis and criteria were determined using the Clinician-Administered PTSD Scale for DSM-5 (CAPS).
Figure 1 shows the participant flow throughout the trial. No adverse events were reported. Missing post-treatment CAPS-5 data were higher in the control group (18 vs. 7 in the EMDR group; p = 0.01), with a greater proportion of PTSD patients having higher baseline scores among missing cases. As data were missing not at random (MNAR), case-wise deletion was applied.
Fig. 1.
Participant flow.
Exposure to Adverse Events
Table 2 shows the frequency of specific adverse events reported at baseline in patients with different PD clusters categorized by the presence or absence of PTSD. No significant differences in trauma exposure were observed between the EMDR therapy and control group. Emotional neglect (94.7%) and emotional abuse (89.5%) were the most commonly reported adverse events, with similar distributions across the PTSD and non-PTSD group. Physical abuse (37.9%), physical neglect (60.8%), and sexual abuse (36.5%) were also commonly reported.
Table 2.
Descriptive statistics of reported exposure to adverse events indexed by CTQ-SF
| CTQ-SF | Total, N (%) | PTSD group, N (%) | Non-PTSD group, N (%) | Statistic |
|---|---|---|---|---|
| Total | 153 | 61 (39.9%) | 92 (60.1%) | t(151) = −1.60, p = 0.112 |
| Emotional abuse | 137 (89.5%) | 55 (40.1%) | 82 (59.9%) | t(135) = −0.28, p = 0.783 |
| Personality cluster | ||||
| A | 3 (100%) | 3 (100%) | – | |
| B | 51 (94.4%) | 24 (47.1%) | 27 (52.9%) | |
| C | 64 (86.5%) | 22 (34.4%) | 42 (65.6%) | |
| OS | 19 (82.6%) | 6 (31.6%) | 13 (68.4%) | |
| Emotional neglect | 145 (94.7%) | 58 (40%) | 87 (60%) | t(143) = 1.44, p = 0.151 |
| Personality cluster | ||||
| A | 3 (100%) | 3 (100%) | – | |
| B | 50 (92.6%) | 23 (46%) | 27 (54%) | |
| C | 72 (97.3%) | 27 (37.5%) | 45 (62.5%) | |
| OS | 20 (86.9%) | 5 (25%) | 15 (75%) | |
| Physical abuse | 58 (37.9%) | 26 (44.8%) | 32 (55.2%) | t(56) = 0.41, p = 0.684 |
| Personality cluster | ||||
| A | 1 (33.3%) | 1 (100%) | – | |
| B | 20 (37%) | 12 (60%) | 8 (40%) | |
| C | 30 (40.5%) | 12 (40%) | 18 (60%) | |
| OS | 7 (30.4%) | 1 (14.3%) | 6 (85.7%) | |
| Physical neglect | 93 (60.8%) | 39 (41.9%) | 54 (58.1%) | t(91) = −1.19, p = 0.237 |
| Personality cluster | ||||
| A | 1 (33.3%) | 1 (100%) | – | |
| B | 34 (63%) | 17 (50%) | 17 (50%) | |
| C | 47 (63.5%) | 19 (40.4%) | 28 (59.6%) | |
| OS | 11 (47.8%) | 2 (18.2%) | 9 (81.8%) | |
| Sexual abuse | 56 (36.5%) | 29 (51.8%) | 27 (48.2%) | t(54) = −0.53, p = 0.599 |
| Personality cluster | ||||
| A | 1 (33.3%) | 1 (100%) | – | |
| B | 20 (37%) | 11 (55%) | 9 (45%) | |
| C | 31 (41.9%) | 16 (51.6%) | 15 (48.4%) | |
| OS | 4 (17.3%) | 1 (25%) | 3 (75%) | |
OS, other specified; CTQ-SF, Childhood Trauma Questionnaire Short Form.
Note: Personality cluster was determined using the Structured Clinical Interview for DSM-5 Personality Disorders (SCID-5-P).
Six patients did not complete the CTQ-SF at baseline.
Effectiveness of EMDR in Reducing PTSD Symptoms
Table 3 shows the mean scores at baseline, post-treatment, and follow-up by the allocated group, with and without PTSD diagnosis, measured by the CAPS-5. In Figure 2, the changes in PTSD symptoms for both the EMDR and control group are illustrated, differentiated by baseline PTSD status.
Table 3.
Raw mean scores at baseline, post-treatment, and follow-up by allocated group (with and without PTSD diagnosis measured by the CAPS-5)
| | Total group | EMDR group | Control group | ||||||
|---|---|---|---|---|---|---|---|---|---|
| CAPS-5 | baseline (T0) | post-treatment (T1) | follow-up (T3) | baseline (T0) | post-treatment (T1) | follow-up (T3) | baseline (T0) | post-treatment (T1) | follow-up (T3) |
| Total scores | |||||||||
| N | 158 | 133 | 125 | 79 | 72 | 68 | 79 | 61 | 57 |
| Mean (SD) | 26.0 (12.5) | 18.5 (12.4) | 15.1 (13.3) | 26.7 (13.1) | 15.6 (12.1) | 11.6 (12.8) | 25.2 (11.8) | 22.0 (11.8) | 19.3 (12.8) |
| PTSD group | |||||||||
| N | 62 | 47 | 46 | 32 | 29 | 26 | 30 | 18 | 20 |
| Mean (SD) | 34.9 (9.5) | 24.8 (13) | 20.9 (15.2) | 35.1 (11.0) | 20.2 (12.6) | 15.3 (15.6) | 34.7 (7.8) | 32.3 (10.0) | 28.2 (11.4) |
| Non-PTSD group | |||||||||
| N | 96 | 86 | 79 | 47 | 43 | 42 | 49 | 43 | 37 |
| Mean (SD) | 20.2 (10.6) | 15.1(10.5) | 11.7 (10.8) | 21.1 (11.4) | 12.5 (10.8) | 9.2 (10.3) | 19.4 (10) | 17.6 (9.8) | 14.6 (10.8) |
EMDR, eye movement desensitization and reprocessing therapy; PTSD, post-traumatic stress disorder; CAPS-5, Clinician-Administered PTSD Scale for DSM-5; SD, standard deviation.
Fig. 2.
Differences in PTSD symptoms over time between groups (LMM estimates), stratified by baseline PTSD diagnosis (CAPS-5) with standard errors. PTSD, post-traumatic stress disorder; CAPS-5, Clinician-Administered PTSD Scale for DSM-5; LMM, linear mixed model.
Table 4 presents the LMM estimates examining the effects of group allocation, time, and PTSD status on CAPS-5 scores. There was no significant main effect of group, indicating that baseline differences between the EMDR therapy and control group were not statistically significant. PTSD diagnosis and symptom severity were significant predictors of the CAPS-5 scores at baseline. Significant group-by-time interactions at post-treatment and follow-up showed that the EMDR group exhibited greater symptom reduction over time than the control group. Interactions involving PTSD status were not significant (p > 0.05), indicating that the effectiveness of treatment did not differ significantly based on the presence or absence of an initial PTSD diagnosis.
Table 4.
Effects of the linear mixed model analyses of the CAPS-5
| | β | SE | df | t | p value |
|---|---|---|---|---|---|
| CAPS-5 | |||||
| Intercept | 2.79 | 0.72 | 226 | 3.85 | <0.001 |
| Group | 0.44 | 0.73 | 151 | 0.59 | 0.551 |
| Post-treatment | −0.28 | 1.64 | 226 | −0.17 | 0.864 |
| Follow-up | −3.60 | 1.94 | 226 | −1.85 | 0.065 |
| PTSD | 2.67 | 0.91 | 151 | 2.93 | 0.003 |
| Symptom severity | 11.00 | 0.33 | 151 | 33.04 | <0.001 |
| Group * post-treatment | −7.65 | 2.34 | 226 | −3.25 | 0.001 |
| Group * follow-up | −8.26 | 2.72 | 226 | −3.02 | 0.002 |
| Group * PTSD | −0.05 | 1.17 | 151 | −0.04 | 0.966 |
| Follow-up * PTSD | −0.96 | 2.96 | 226 | −0.32 | 0.746 |
| Post-treatment * PTSD | −2.74 | 3.33 | 226 | −0.82 | 0.411 |
| Group * post-treatment * PTSD | −6.64 | 3.91 | 226 | −1.69 | 0.091 |
| Group * follow-up * PTSD | −5.42 | 4.51 | 226 | −1.20 | 0.231 |
CAPS-5, Clinician-Administered PTSD Scale for DSM-5; PTSD, post-traumatic stress disorder.
Change in PTSD Diagnostic Status
Post-treatment, 65.5% of patients in the EMDR group lost their PTSD diagnostic status, increasing to 73.1% by follow-up. In the control group, 27.8% no longer met the diagnostic criteria for PTSD post-treatment, increasing to 40.0% at follow-up, with a significant difference between groups over time (p = 0.03).
Within-Group and between-Group Effect Sizes
Within-Group Effect Sizes of EMDR Therapy on PTSD Severity
For the entire sample in the EMDR group, a within-group effect size of Cohen’s d = 0.60 was found at post-treatment (95% CI [0.36, 0.84]), which further increased at the 3-month follow-up (d = 0.85, 95% CI [0.60, 1.09]). In patients with PTSD, EMDR therapy significantly reduced PTSD severity at post-treatment (EMDR: Cohen’s d = 1.26, 95% CI [0.70, 1.80]; control: d = 0.28, 95% CI [−0.31, 0.86]), with effects sustained at 3-month follow-up (EMDR: d = 1.5, 95% CI [0.90, 2.06]; control: d = 0.69, 95% CI [0.10, 1.26]). No significant differences were found between patients with and without PTSD. EMDR also reduced PTSD severity in patients without PTSD (EMDR: d = 0.78, 95% CI [0.34, 1.20]; control: d = 0.18, 95% CI [−0.23, 0.59]), with the effects maintained at follow-up (EMDR: d = 1.09, 95% CI [0.64, 1.53]); control: d = 0.46, 95% CI [0.03, 0.89]).
Between-Group Effect Sizes of EMDR Therapy on PTSD Severity
For the entire sample, a Cohen’s d between-group effect size of −0.53 was found post-treatment (95% CI [−0.88, −0.18]), with effects sustained at 3-month follow-up (d = −0.60, 95% CI [−0.96, −0.24]). In the PTSD group, Cohen’s d for between-group effect sizes comparing EMDR therapy to the control group was −1.04 (95% CI [−1.64, −0.39]) at post-treatment and −0.93 (95% CI [−1.52, −0.20]) at follow-up. For the non-PTSD group, effect sizes were −0.49 (95% CI [−0.92, −0.06]) at post-treatment and −0.51 (95% CI [−0.96, −0.06]) at follow-up.
Differential Effectiveness of EMDR Therapy on Memory Types
Of the 77 patients who received EMDR therapy, the most frequently treated memories were related to physical abuse (29.9%), emotional abuse (28.6%), and other types of adverse events (37.7%). Sexual abuse was not analyzed as it was the most frequently treated trauma in only three patients (n = 3; 3.9%). Figure 3 shows the mean differences in CAPS-5 scores for each memory type measured over time. Table 5 presents the statistics of the LMM analysis of the memory types of adverse events in patients. The intercept represents the average baseline score for emotional abuse. Significant PTSD symptom reductions were found for memories related to emotional abuse (p < 0.001), with no significant differences between the memory types.
Fig. 3.
Mean differences between memory types of adverse events in patients (LMM estimates) on the CAPS-5, measured over time. CAPS-5, Clinician-Administered PTSD Scale for DSM-5; LMM, linear mixed model.
Table 5.
Effects of linear mixed model analyses on memory types of adverse events
| | β | SE | df | t | p value |
|---|---|---|---|---|---|
| CAPS-5 | |||||
| Intercept | 22.33 | 3.12 | 122 | 7.15 | <0.001 |
| Physical abuse | 7.72 | 4.42 | 61 | 1.75 | 0.086 |
| Other memories | 2.77 | 4.00 | 61 | 0.69 | 0.491 |
| Post-treatment | −10.89 | 2.66 | 122 | −4.09 | <0.001 |
| Follow-up | −12.67 | 2.92 | 122 | −4.33 | <0.001 |
| Physical abuse * post-treatment | 3.5 | 3.76 | 122 | 0.93 | 0.354 |
| Other memories * post-treatment | −0.25 | 3.41 | 122 | −0.07 | 0.941 |
| Physical abuse * follow-up | −1.67 | 4.13 | 122 | −0.40 | 0.687 |
| Other memories * follow-up | −2.51 | 3.75 | 122 | −0.67 | 0.504 |
CAPS-5, Clinician-Administered PTSD Scale for DSM-5.
Discussion
This study demonstrated that EMDR therapy led to significant reductions in PTSD symptoms compared to a waitlist control condition, with sustained effects at the follow-up. A substantial proportion of participants in the EMDR group lost their PTSD diagnostic status post-treatment (65.5%) and at follow-up (73.1%), with a low treatment dropout rate (5.1%). Furthermore, the results support our hypothesis that EMDR therapy is effective in reducing distress linked to various types of memories, including memories related to events that do not meet Criterion A for PTSD. The results support the utility of trauma-focused therapy for PDs, even in the absence of a PTSD diagnosis.
This study highlights the high prevalence of adverse childhood experiences, with emotional neglect (94.7%) and emotional abuse (89.5%) being particularly common across all PD types, in line with previous research [3, 35, 36]. Importantly, although almost 40% of participants had a PTSD diagnosis at baseline, 38.6% of patients whose index trauma did not meet Criterion A showed elevated PTSD symptoms across all symptom clusters B–E on the CAPS-5. These findings highlight the high prevalence of PTSD symptoms in patients with PD, both with and without a formal PTSD diagnosis.
In line with previous research [15, 37], this study confirmed the effectiveness of EMDR therapy in reducing PTSD symptoms in patients with PD and PTSD. Notably, this effect was also observed in individuals with PD but without PTSD. Although the effect size from baseline to follow-up was higher in the PTSD group (d = 1.5), large effects were observed in the non-PTSD group (d = 1.09), whereas no significant differences were found between these groups. These findings are consistent with those of an earlier RCT [38], which showed significant improvements in psychological symptoms in patients with a PD without PTSD, focusing on memories of events that did not meet the Criterion A. The findings of our study further support the notion that EMDR therapy can be applied broadly and transdiagnostically, proving effective for individuals with a PD, both with and without a PTSD diagnosis at baseline.
Supported by the AIP model [13, 20], which posits that unprocessed memories drive psychopathological symptoms, our study results confirm that non-Criterion A events, such as emotional abuse, significantly contribute to PTSD symptoms in patients with PD. For 28.6% of the patients, EMDR sessions primarily focused on memories of emotional abuse, whereas memories of primarily sexual abuse were processed in only three individuals. This highlights the need to also address non-Criterion A memories in trauma-focused therapy. Previous studies have shown a strong link between non-Criterion A memories, such as emotional abuse, PTSD severity [39], and the development of PD [23]. Emotional abuse, which frequently leads to rumination and reduced emotional acceptance [40], is a common risk factor for both PTSD [41] and PD [42]. It may also exacerbate PTSD symptoms by limiting support [43, 44], as observed in the current RCT, in which a lack of support was often reported as more impactful than the trauma itself. Additionally, other non-Criterion A events such as neglect (reported by 94.7% of patients in this study), have been found to be associated with both PTSD [45] and PD, which may thereby contribute to greater clinical severity, particularly in individuals with BPD [23].
It is essential to address the limitations of this study. First, the focus on PTSD-related outcome measures may have limited the generalizability of the findings to other aspects of PDs. Future studies should examine the effects of EMDR therapy on other PD-specific symptoms. This topic will be further explored on data from the TEMPO study [24]. Future research should assess changes in core symptom severity reported at baseline to more precisely evaluate whether specific symptoms indeed decrease after EMDR therapy. The use of a waitlist control group limits insights into comparative treatment effectiveness, highlighting the need for future studies to include an active treatment group for a better comparison of the effects of EMDR therapy. Finally, at post-treatment, missing data were more prevalent in the control group, particularly among participants with higher baseline CAPS-5 scores, suggesting data were likely not MNAR. Consequently, case-wise deletion might have introduced bias and limited generalizability, as individuals with greater symptom severity, possibly underrepresented due to dropout in the control group, may have skewed follow-up estimates. Analyses based solely on complete cases assume representativeness, an assumption that may not hold under MNAR conditions. This common limitation in clinical trials necessitates cautious interpretation, particularly concerning treatment effect size. While this limitation poses methodological challenges, it also indicates that untreated patients experiencing more severe symptoms may have been more likely to discontinue participation. In contrast, the dropout rate in the EMDR group was significantly lower, suggesting higher patient motivation to remain in treatment. This may have been influenced by factors such as positive treatment experiences, perceived effectiveness, and the therapeutic alliance.
Despite these limitations, the relatively low total study dropout rate (16.4%) compared to other trauma-focused [4, 46] and PD studies [47, 48] enhances the internal validity of the findings. A notable strength of this multicenter study with participants from specialized departments is that the patients with varying PDs provided a strong representation of clinical practice. Additionally, this is the first study to explore the effectiveness of EMDR therapy within a single RCT, distinguishing between individuals with a PD diagnosis with and without PTSD. Future research should focus on exploring the long-term effects.
Conclusion
This study demonstrates the effectiveness of EMDR therapy in reducing PTSD symptoms in individuals with a PD, regardless of an initial PTSD diagnosis, and in treating memories that do not meet Criterion A. This highlights the need to address such memories in trauma-focused therapy, as emphasized in integrative EMDR frameworks that support the processing of a broad range of distressing experiences beyond Criterion A trauma [49]. Given the high prevalence of PTSD symptoms in patients with a PD and the role that unprocessed memories seem to play in current symptoms and dysfunctional patterns, EMDR therapy may be a valuable addition to routine care. These findings highlight the importance of bridging diagnostic boundaries and urge clinicians to assess and actively address memories of adverse events in patients with a PD, even without a formal PTSD diagnosis or the presence of Criterion A events.
Acknowledgments
We would like to express our sincere thanks to all the patients who participated in this study, as well as to the clinicians at Parnassia Group and GGZ Delfland for their valuable contributions. We would like to thank Dr. Bouwmeester for her essential assistance with the statistical analysis, which was critical for the completion of this study.
Statement of Ethics
The design of this study was approved by the Medical Research Ethics Committee (MREC) of the Erasmus Medical Center in Rotterdam (Address: Dr. Molewaterplein 40, 3015 GD Rotterdam, The Netherlands; Email: metc@erasmusmc.nl; phone: +31 (0) 10-703 4428, Approval No. MEC-2020-0583, Netherlands Trial Register: NL9078, date of approval: November 31, 2020). The study followed the Declaration of Helsinki (64th WMA General Assembly, October 2013) and the International Conference on Harmonization – Good Clinical Practice (ICH-GCP) guidelines. Before any procedures began, the investigator explained the study’s objectives, methods, benefits, and risks to each participant. Written informed consent was obtained, ensuring participants understood they could withdraw or refuse at any time. Only individuals legally capable of providing consent were included.
Conflict of Interest Statement
Prof Dr. De Jongh and Hafkemeijer, MSc, report personal fees from teaching activities about EMDR therapy. Prof. Dr. De Jongh reports personal fees from books about trauma and its treatment (e.g., EMDR therapy) outside the submitted work and has been a board member of the Dutch EMDR Association and the EMDR Europe Association. The other authors have nothing to disclose.
Funding Sources
The study was internally funded by GGZ Delfland and the Parnassia Group and externally funded by the Vereniging EMDR Nederland, EMDR Europe and EMDR Research Foundation. None of the funders had any role in the study design, data collection, data analysis, or reporting.
Author Contributions
L.H. led the drafting of the manuscript, conducted data curation and formal analyses, and contributed to all stages of the study. A.J., A.S., S.H., K.S., D.R. and M.V. were involved in conceptualization and manuscript review. M.V. and S.H. participated in investigation and project administration. D.R., A.S., S.H., A.J. and K.S. contributed to methodology development. S.H. and M.V. assisted with the data curation.
Funding Statement
The study was internally funded by GGZ Delfland and the Parnassia Group and externally funded by the Vereniging EMDR Nederland, EMDR Europe and EMDR Research Foundation. None of the funders had any role in the study design, data collection, data analysis, or reporting.
Footnotes
The dimensional CAPS-5 scores at baseline were standardized and then categorized as follows: Z <−1.5 = 0, Z <−0.5 and Z >−1.5 = 1, Z <0.5 and Z >−0.5 = 2, Z <1.5 and Z >0.5 = 3, Z >1.5 = 4.
Data Availability Statement
The data that support the findings of this study are not publicly available due to restrictions related to patient confidentiality agreements but are available from the corresponding author (L.H.) upon reasonable request.
Supplementary Material.
References
- 1. Oltmanns TF, Melley AH, Turkheimer E. Impaired social function and symptoms of personality disorders assessed by peer and self-report in a nonclinical population. J Pers Disord. 2002;16(5):437–52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Setkowski K, Palantza C, van Ballegooijen W, Gilissen R, Oud M, Cristea IA, et al. Which psychotherapy is Most effective and acceptable in the treatment of adults with a (sub) clinical borderline personality disorder? A systematic review and network meta-analysis. Psychol Med. 2023;53(8):3261–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Battle CL, Shea MT, Johnson DM, Yen S, Zlotnick C, Zanarini MC, et al. Childhood maltreatment associated with adult personality disorders: findings from the collaborative longitudinal personality disorders study. J Pers Disord. 2004;18(2):193–211. [DOI] [PubMed] [Google Scholar]
- 4. Slotema CW, van den Berg DP, Driessen A, Wilhelmus B, Franken IH. Feasibility of EMDR for posttraumatic stress disorder in patients with personality disorders: a pilot study. Eur J Psychotraumatology. 2019;10(1):1614822. [Google Scholar]
- 5. Hengartner MP, Ajdacic-Gross V, Rodgers S, Müller M, Rössler W. Childhood adversity in association with personality disorder dimensions: new findings in an old debate. Eur Psychiatr. 2013;28(8):476–82. [Google Scholar]
- 6. Lobbestael J, Arntz A, Bernstein DP. Disentangling the relationship between different types of childhood maltreatment and personality disorders. J Pers Disord. 2010;24(3):285–95. [DOI] [PubMed] [Google Scholar]
- 7. Hofman S, Slotema CW. Underdiagnosis of posttraumatic stress disorder among outpatients with personality disorders in clinical practice despite the use of a diagnostic instrument. J Pers Disord. 2024;38(5):477–92. [DOI] [PubMed] [Google Scholar]
- 8. Cloitre M, Garvert DW, Weiss B, Carlson EB, Bryant RA. Distinguishing PTSD, complex PTSD, and borderline personality disorder: a latent class analysis. Eur J psychotraumatology. 2014;5(1):25097. [Google Scholar]
- 9. Ford JD, Courtois CA. Complex PTSD, affect dysregulation, and borderline personality disorder. Bord Personal Disord Emot Dysregul. 2014;1:9–17. [Google Scholar]
- 10. Luyten P, Fonagy P. The neurobiology of mentalizing. Personal Disord Theor Res Treat. 2015;6(4):366–79. [Google Scholar]
- 11. Erazo MB, Krygsman AL, Vaillancourt T. The cumulative effects of bullying victimization in childhood and adolescence on borderline personality disorder symptoms in emerging adulthood. Int J Bullying Prev. 2023;5(2):121–34. [Google Scholar]
- 12. Bohus M, Schmahl C, Fydrich T, Steil R, Müller-Engelmann M, Herzog J, et al. A research programme to evaluate DBT-PTSD, a modular treatment approach for Complex PTSD after childhood abuse. Bord Personal Disord Emot Dysregul. 2019;6:7–16. [Google Scholar]
- 13. De Jongh A, Hafkemeijer L, Hofman S, Slotema K, Hornsveld H. The AIP model as a theoretical framework for the treatment of personality disorders with EMDR therapy. Front Psychiatry. 2024a;15:1331876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. De Jongh A, De Roos C, El-Leithy S. State of the science: Eye Movement Desensitization and Reprocessing (EMDR) therapy. J Trauma Stress. 2024b;37(2):205–16. [DOI] [PubMed] [Google Scholar]
- 15. Snoek AC, van den End A, Beekman ATF, Dekker J, Aarts I, Blankers M, et al. Eye movement desensitization and reprocessing with and without dialectical behavior therapy for posttraumatic stress disorder and comorbid borderline personality disorder symptoms: a randomized controlled trial. Psychother Psychosom. 2025;94(3):175–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Annesley P, Alabi A, Longdon L. The EMDR DeTUR protocol for the treatment of self-injury in a patient with severe personality disorder: a case report. J Criminological Res Pol Pract. 2019;5(1):27–38. [Google Scholar]
- 17. Brown S, Shapiro F. EMDR in the treatment of borderline personality disorder. Clin Case Stud. 2006;5(5):403–20. [Google Scholar]
- 18. Safarabad NM, Farid AAA, Gharraee B, Habibi M. Treatment of a patient with borderline personality disorder based on phase-oriented model of eye movement desensitization and reprocessing (EMDR): a case report. Iran J Psychiatry. 2018;13(1):80. [PMC free article] [PubMed] [Google Scholar]
- 19. Hafkemeijer L, de Jongh A, van der Palen J, Starrenburg A. Eye Movement Desensitization and Reprocessing (EMDR) in patients with a personality disorder. Eur J Psychotraumatol. 2020;11(1):1838777. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Shapiro F Eye movement desensitization and reprocessing: basic principles, protocols and procedures. 3rd ed. New York, NY: Guilford Press; 2018. [Google Scholar]
- 21. Anders SL, Frazier PA, Frankfurt SB. Variations in Criterion A and PTSD rates in a community sample of women. J Anxiety Disord. 2011;25(2):176–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Bjornsson AS, Hardarson JP, Valdimarsdottir AG, Gudmundsdottir K, Tryggvadottir A, Thorarinsdottir K, et al. Social trauma and its association with posttraumatic stress disorder and social anxiety disorder. J Anxiety Disord. 2020;72:102228. [DOI] [PubMed] [Google Scholar]
- 23. Bierer LM, Yehuda R, Schmeidler J, Mitropoulou V, New AS, Silverman JM, et al. Abuse and neglect in childhood: relationship to personality disorder diagnoses. CNS Spectr. 2003;8(10):737–54. [DOI] [PubMed] [Google Scholar]
- 24. Hofman S, Hafkemeijer L, de Jongh A, Starrenburg A, Slotema K. Trauma-focused EMDR for personality disorders among outpatients (TEMPO): study protocol for a multi-centre, single-blind, randomized controlled trial. Trials. 2022;23(1):196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. First MB. Structured clinical interview for the DSM (SCID). The encyclopedia Clin Psychol. 2014:1–6. [Google Scholar]
- 26. American Psychiatric Association . SCID-5-P: structured clinical interview for DSM-5® personality disorders (Dutch translation) and user’s guide for the structured clinical interview for DSM-5® personality disorders (SCID-5-P). Boom. 2017. [Google Scholar]
- 27. Weertman A, Arntz A, Kerkhofs MLM. Gestructureerd diagnostisch interview voor DSM-IV persoonlijkheidsstoornissen (SCID II) [structural and clinical interview for DSM-IV personality disorders (SCID II)]. Swets Test. 2000. [Google Scholar]
- 28. Lobbestael J, Leurgans M, Arntz A. Inter-rater reliability of the structured clinical interview for DSM-IV axis I disorders (SCID I) and axis II disorders (SCID II). Clin Psychol Psychoth. 2011;18(1):75–9. [Google Scholar]
- 29. Maffei C, Fossati A, Agostoni I, Barraco A, Bagnato M, Deborah D, et al. Interrater reliability and internal consistency of the Structured Clinical Interview for DSM-IV axis II personality disorders (SCID-II), version 2.0. J Pers Disord. 1997;11(3):279–84. [DOI] [PubMed] [Google Scholar]
- 30. Boeschoten MA, Bakker A, Jongedijk RA, Olff MPTSD. PTSD checklist for DSM-5-Nederlandstalige versie. Diemen Arq Psychotrauma Expert Groep. 2014. [Google Scholar]
- 31. Boeschoten MA, Van der Aa N, Bakker A, Ter Heide FJJ, Hoofwijk MC, Jongedijk RA, et al. Development and evaluation of the Dutch clinician-administered PTSD scale for DSM-5 (CAPS-5). Eur J Psychotraumatol. 2018;9(1):1546085. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Bernstein DP, Stein JA, Newcomb MD, Walker E, Pogge D, Ahluvalia T, et al. Development and validation of a brief screening version of the childhood trauma questionnaire. Child Abuse Neglect. 2003;27(2):169–90. [DOI] [PubMed] [Google Scholar]
- 33. Thombs BD, Bernstein DP, Lobbestael J, Arntz A. A validation study of the Dutch childhood trauma questionnaire-short form: factor structure, reliability, and known-groups validity (2022) 23:196 Page 14 of 15 reliability, and known-groups validity. Child Abuse Neglect. 2009;33(8):518–23. [DOI] [PubMed] [Google Scholar]
- 34. Pinheiro J, Bates D, DebRoy S, Sarkar D, R Core Team . Nlme: Linear and nonlinear mixed effects models. R Package Version. 2023;3:1–162. https://CRAN.R-project.org/package=nlme [Google Scholar]
- 35. Massaal-van der Ree LY, Eikelenboom M, Hoogendoorn AW, Thomaes K, van Marle HJ. Cluster B versus cluster C personality disorders: a comparison of comorbidity, suicidality, traumatization and global functioning. Behav Sci. 2022;12(4):105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Yen S, Shea MT, Battle CL, Johnson DM, Zlotnick C, Dolan-Sewell R, et al. Traumatic exposure and posttraumatic stress disorder in borderline, schizotypal, avoidant, and obsessive-compulsive personality disorders: findings from the collaborative longitudinal personality disorders study. The J nervous Ment Dis. 2002;190(8):510–8. [Google Scholar]
- 37. Kolthof KA, Voorendonk EM, Van Minnen A, De Jongh A. Effects of intensive trauma-focused treatment of individuals with both post-traumatic stress disorder and borderline personality disorder. Eur J Psychotraumatol. 2022;13(2):2143076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Hafkemeijer L, Starrenburg A, van der Palen J, Slotema K, de Jongh A. Does EMDR therapy have an effect on memories of emotional abuse, neglect and other types of adverse events in patients with a personality disorder? Preliminary data. J Clin Med. 2021;10(19):4333. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Hoeboer C, De Roos C, van Son GE, Spinhoven P, Elzinga B. The effect of parental emotional abuse on the severity and treatment of PTSD symptoms in children and adolescents. Child Abuse Neglect. 2021;111:104775. [DOI] [PubMed] [Google Scholar]
- 40. Burns EE, Jackson JL, Harding HG. Child maltreatment, emotion regulation, and posttraumatic stress: the impact of emotional abuse. J Aggress Maltreat Trauma. 2010;19(8):801–19. [Google Scholar]
- 41. Tull MT, Barrett HM, McMillan ES, Roemer L. A preliminary investigation of the relationship between emotion regulation difficulties and posttraumatic stress symptoms. Behav Ther. 2007;38(3):303–13. [DOI] [PubMed] [Google Scholar]
- 42. Solmi M, Dragioti E, Croatto G, Radua J, Borgwardt S, Carvalho AF, et al. Risk and protective factors for personality disorders: an umbrella review of published meta-analyses of case–control and cohort studies. Front Psychiatry. 2021;12:679379. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Cohen LR, Hien DA. Treatment outcomes for women with substance abuse and PTSD who have experienced complex trauma. Psychiatr Serv. 2006;57(1):100–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Sanchez MM, McCormack KM, Howell BR. Social buffering of stress responses in nonhuman primates: maternal regulation of the development of emotional regulatory brain circuits. Soc Neurosci. 2015;10(5):512–26. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Nikulina V, Widom CS, Czaja S. The role of childhood neglect and childhood poverty in predicting mental health, academic achievement and crime in adulthood. Am J Comm Psychol. 2011;48(3–4):309–21. [Google Scholar]
- 46. Kredlow MA, Szuhany KL, Lo S, Xie H, Gottlieb JD, Rosenberg SD, et al. Cognitive behavioral therapy for posttraumatic stress disorder in individuals with severe mental illness and borderline personality disorder. Psychiatry Res. 2017;249:86–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47. Iliakis EA, Ilagan GS, Choi-Kain LW. Dropout rates from psychotherapy trials for borderline personality disorder: a meta-analysis. Personal Disord Theor Res Treat. 2021;12(3):193–206. [Google Scholar]
- 48. Gülüm İV, Soygüt G, Safran JD. A comparison of pre-dropout and temporary rupture sessions in psychotherapy. Psychother Res. 2018;28(5):685–707. [DOI] [PubMed] [Google Scholar]
- 49. Balbo M, Cavallo F, Fernandez I. Integrating EMDR in psychotherapy. J Psychotherapy Integration. 2019;29(1):23–31. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are not publicly available due to restrictions related to patient confidentiality agreements but are available from the corresponding author (L.H.) upon reasonable request.
