Abstract
Introduction:
Eye movement desensitization and reprocessing (EMDR) is an evidence-based psychotherapy method designed to treat distress associated with traumatic memories. The COVID-19 pandemic has challenged providers to shift EMDR to telehealth platforms.
Objectives:
This research had to aims. (1) To compare EMDR in conjunction with cognitive behavioral therapy (CBT) with CBT alone for treatment of a patient population with prevalent anxiety and (2) to compare the efficacy of virtual EMDR with that of in-person EMDR in the primary care setting.
Methods:
Retrospective chart review of all adult patients seen by a single therapist at a primary care center was conducted from January 2018 to December 2020. Charts were reviewed for demographics, psychiatric diagnoses, number of visits, number and type of EMDR treatments, and patient health questionnaire (PHQ)-9 and general anxiety disorder (GAD-7) scores pre- and post-treatment.
Results:
Patients who underwent EMDR with CBT demonstrated greater decreases in PHQ-9 and GAD-7 scores than patients who had only CBT (2.4, 2.5 vs. 0.9, 1.1). However, after adjusting for total number of sessions, post-traumatic stress disorder, grief, and pain, the adjusted mean differences of change in PHQ-9 and GAD-7 scores between those who underwent EMDR with CBT and CBT-exclusive patients were not statistically significant. Similarly, the adjusted mean differences of change in PHQ-9 and GAD-7 scores between those who underwent virtual EMDR and those who had in-person EMDR were not statistically significant.
Conclusions:
To our knowledge, this is the first study describing the use of virtual EMDR in a primary care setting. Although our data did not provide evidence for the superiority of EMDR with CBT over CBT alone, we demonstrate that both in-person and virtual EMDR led to significantly improved GAD-7 scores.
Introduction
The COVID-19 pandemic has led to the increased prevalence of several mental health issues including anxiety, depression, and traumatic stress. 1 Eye movement desensitization and reprocessing (EMDR) is an effective evidence-based psychotherapy designed to treat distress associated with traumatic memories. 2 The pandemic has challenged providers to shift EMDR to telehealth platforms despite the dearth of literature on the validity of this treatment modality. 3
Throughout the pandemic, the department of family medicine at Mayo Clinic Arizona has offered both in-person and virtual EMDR through videoconferencing to eligible patients who were undergoing cognitive behavioral therapy (CBT) with one of our mental health providers. There are no studies to our knowledge that have evaluated the feasibility of administering any form of EMDR in primary care. Thus, we sought to retrospectively evaluate the utility of EMDR with CBT versus CBT alone to treat a variety of psychological disorders stemming from past trauma as well as compare the efficacy of virtual EMDR with that of in-person EMDR in the primary care setting.
Although EMDR is used most widely for post-traumatic stress disorder (PTSD), there is a growing body of evidence for the efficacy of the therapy for anxiety disorders. 4 Most of the patients included in this study were treated for anxiety-related disorders, offering potential insight into this emerging treatment paradigm. Thus, a secondary aim of the study was to evaluate the utility of EMDR to reduce anxiety symptoms.
Methods
EMDR METHODOLOGY
All EMDR sessions were conducted one-on-one between a patient and the provider. Virtual EMDR patients were required to use a monitor at least 30 cm in width for teleconferencing. The provider utilized an EMDR light bar measuring ∼80 cm in length with a row of 24 small light bulbs atop a tripod (Legacy EyeScan Model; NeuroTek). For virtual EMDR, the light bar was placed ∼75 cm from the provider's camera. The provider then made any necessary adjustments to ensure the patients could comfortably visualize the entire light bar. The remainder of the procedure was the same for virtual and in-person EMDR. 5
Patients were guided through five preliminary questions to help focus their attention on the negative thoughts and beliefs related to their condition. Patients also identified feelings they would like to experience as opposed to the feelings they had been experiencing. The patients then identified a comfortable speed to track the bilateral light movement on the light bar. The provider had the patients focus on specific thoughts and feelings while the patients tracked the lights. After 20 to 40 passes, called a “set,” the provider turned off the light bar and solicited feedback on what the patients were thinking and feeling. This process was repeated until the patients no longer experienced uncomfortable thoughts or feelings or body discomfort.
Next, the reprocessing of thoughts and feelings was initiated as the provider redirected the patients to what they would like to believe about themselves when thinking about the trauma, anxiety, or grief. The reprocessing of thoughts and feelings was operationalized by having the patients watch fewer passes in each set and obtaining feedback from the patients on what they were thinking and feeling. After about three or four shorter reprocessing sets, the provider then asked the patients to close their eyes and complete a body scan to identify any feelings of tension or tightness. The provider then asked the patients to focus on those areas for a shorter set. The patients and provider lastly determined whether it was appropriate to stop treatment according to how the patients felt emotionally, psychologically, and physically.
CHART REVIEW
A retrospective chart review of all adult patients who were treated by a single therapist (J.H.-D.) from January 2018 to December 2020 was performed. Charts were reviewed for patient age, gender, number of visits, number of EMDR treatments, first and last visit patient health questionnaire (PHQ-9) and general anxiety disorder (GAD-7) screening scores, and psychiatric diagnoses. Diagnoses included generalized anxiety disorder, PTSD/trauma, major depression, grief, insomnia, chronic pain, substance abuse, and other.
Secondary to the COVID-19 pandemic, nearly all visits at our center from March 2020 to October 2020 were conducted virtually, including EMDR treatments. To differentiate between in-person and virtual intervention, the number of in-person versus virtual visits as well as the number of in-person versus virtual EMDR treatments was collected. We also recorded COVID-19–related diagnoses. This study was conducted in compliance with the Health Insurance Portability and Accountability Act (HIPAA), and was approved by the Mayo Clinic Institutional Review Board.
STATISTICAL ANALYSIS
Statistical analysis was performed using SAS version 9.4. Patients who had one or more EMDR sessions were categorized as “EMDR.” Patients who had no EMDR sessions were categorized as “No EMDR.” Of the patients who had EMDR, those who had at least one virtual EMDR session were further categorized as “Virtual EMDR.” Those who had in-person EMDR but no virtual EMDR were further categorized as “In-Person EMDR only.”
Descriptive statistics were calculated for patient demographics and outcomes by the EMDR group (EMDR versus no EMDR), and by virtual versus in-person EMDR. Categorical variables are presented as counts and percentages. Between-group differences in patient characteristics were tested using Fisher's exact tests. Continuous variables are presented as mean (standard deviation), median, and range. Between-group differences in patient characteristics were tested using Wilcoxon rank-sum tests.
Differences in change in PHQ-9 and GAD-7 scores between EMDR and non-EMDR groups as well as between in-person EMDR and virtual EMDR groups were estimated using multivariable linear regression, adjusting for factors found to significantly differ between groups (total number of sessions, PTSD/trauma, pain, and grief diagnosis). In addition, differences from first to last visit among EMDR patients were estimated using paired t-tests to quantify the overall size of the change from pre-to-post on a patient level.
Results
A total of 472 unique patients were reviewed, of whom 288 (61%) had at least 1 EMDR session. Of all patients, three (1%) were missing in data on whether or not they underwent EMDR. Of the 288 EMDR patients, 44 (15%) had virtual EMDR only and 1 (0.3%) had both virtual and in-person EMDR. All patients were treated with CBT regardless of whether or not they received EMDR therapy.
Demographics, including initial GAD-7 and PHQ-9 scores as well as number of therapy sessions, are presented in Table 1. Table 2 presents patient diagnoses. On average, the patients who underwent EMDR with CBT had significantly higher initial GAD-7 scores and were significantly more likely to have PTSD/trauma, grief, and pain diagnoses than those who only had CBT. Patients who underwent virtual EMDR were significantly more likely to have a COVID-19–related diagnosis than those who underwent in-person EMDR.
Demographics, Baseline General Anxiety Disorder and Patient Health Questionnaire, and Number of Sessions by Eye Movement Desensitization and Reprocessing Status
Significant values are in bold.
Wilcoxon rank sum p-value.
Fisher exact p-value.
EMDR, eye movement desensitization and reprocessing; GAD, general anxiety disorder; PHQ, patient health questionnaire.
Diagnoses by Eye Movement Desensitization and Reprocessing Status
Significant values are in bold.
Fisher exact p-value.
PTSD, post-traumatic stress disorder.
Descriptive statistics and paired t-tests of the initial and final PHQ-9 and GAD-7 scores are listed in Table 3. Table 4 compares the mean differences in PHQ-9 and GAD-7 scores from first to last assessment. Patients who underwent EMDR with CBT demonstrated greater decreases in PHQ-9 and GAD-7 scores than patients who only had CBT (2.4, 2.5 vs. 0.9, 1.1).
Outcomes Before and After Treatment
Paired t-tests.
CI, confidence interval.
Differences in Patient Health Questionnaire-9 and General Anxiety Disorder-7 by Eye Movement Desensitization and Reprocessing Status
Multivariable linear regression.
However, after adjusting for total number of sessions, PTSD, grief, and pain, the adjusted mean differences of change in PHQ-9 and GAD-7 scores between those who underwent EMDR with CBT and CBT-exclusive patients were not statistically significant at 1.03 (CI: −0.16 to 2.21) and 0.82 (CI: −0.33 to 1.97), respectively. Similarly, the adjusted mean differences of change in PHQ-9 and GAD-7 scores between those who underwent virtual EMDR and those who had in-person EMDR were not statistically significant at 0.26 (CI: −2.66 to 3.18) and 0.16 (CI: −2.34 to 2.67), respectively.
Discussion
This study aimed to evaluate the utility of EMDR with CBT versus CBT alone to treat a variety of psychological disorders stemming from past trauma and to compare the efficacy of virtual EMDR with that of in-person EMDR in the primary care setting. The vast majority of patients who underwent EMDR in this study reported feelings of anxiety, as evidenced by the significantly higher average initial GAD-7 scores in the EMDR group. This finding reflects our interest in investigating EMDR for the treatment of anxiety.
A recent review of six randomized controlled trials on anxiety disorders including three on panic disorder, two on specific phobias (flying and dental), and one on self-esteem (considered to be a mediator factor for anxiety disorders) treated by EMDR found generally promising results. 6 The trial on dental phobia found EMDR to be more effective than waitlist control, whereas the flying phobia trial found EMDR combined with CBT to be as effective as CBT with systematic desensitization and CBT with virtual reality exposure therapy. Two of the panic disorder trials found EMDR to be more effective than waitlist control, and one found EMDR to be as effective as CBT. The last study found EMDR to be less effective than competitive memory training at improving self-esteem.
There have been no randomized control trials on EMDR for GAD to our knowledge. However, a preliminary study on four GAD patients found EMDR to decrease anxiety to levels below the diagnostic threshold in all and to result in full remission in two patients as measured by structured clinical interviews. 7 Another preliminary study on three GAD patients found EMDR to reduce pathological worry as measured by the Intolerance of Uncertainty Scale, Cognitive Avoidance Questionnaire, Penn State Worry Questionnaire, and the Worry Domains Questionnaire, which was sustained at 1-month follow-up. 8
Corroborating these studies, we found that EMDR patients had on average a 2.5-point reduction in GAD-7 scores. In contrast, CBT-exclusive patients had a mean 1.1-point reduction in GAD-7. Although the adjusted mean difference in GAD-7 score reduction between EMDR patients and CBT-exclusive patients was not significant, this may result from lack of proper sample pairing and the substantial adjustments necessary for this calculation. Thus, although EMDR in conjunction with CBT is comparable with CBT alone at reducing anxiety, it is unclear whether EMDR provides any additional benefit in this regard.
Numerous randomized controlled trials and case reports indicate that EMDR can successfully treat depression both on its own and in conjunction with standardized treatments. 9,10 Although a minority of the patients in our study had major depressive disorder, we evaluated patient PHQ-9 scores to gain a better understanding of the variety of effects of EMDR on mental health. The PHQ-9 findings followed a similar pattern as those for GAD-7 in that those who underwent EMDR had a greater average score reduction (2.4 vs. 0.9), but there was not a significant adjusted mean difference in score reduction. Again, considering the limitations of the study design, it is inconclusive as to whether EMDR combined with CBT is more effective than CBT alone at reducing symptoms of depression.
The COVID-19 pandemic has challenged therapists to adapt virtually all aspects of care to telehealth platforms, including EMDR. Although a number of telehealth EMDR techniques have been described, 11 there is a dearth of evidence for the efficacy of EMDR administered in this manner. 12 An early case report from 2007 described giving a modified form of EMDR known as alternated bilateral cerebral stimulation over videoconferencing by asking a patient with symptoms of PTSD to follow the provider's finger on the screen. 13
The patient reported immediate subjective improvement in her symptoms and sought no further treatment. A 2013 uncontrolled open trial investigated the efficacy of a 6-week online program including CBT and a Web-based EMDR tool on 15 PTSD patients. 14 The participants who completed the treatment had significant effect sizes for reducing a number of measures including the PTSD symptom scale and GAD-7, but not on PHQ-9.
A COVID-19–era study administered EMDR involving the “butterfly hug” tapping technique through a single videoconferencing session to 17 participants with a variety of indications and found significantly decreased symptoms of anxiety, depression, and disturbance compared with pretreatment as measured by the hospital anxiety depression and subjective units of disturbance scales, which was sustained at 1-week follow-up. 15 The participants were generally satisfied with the quality of treatment, although some expressed concerns over feeling free to express their feelings, which was likely due to the presence of family members at home.
Another study administered EMDR remotely using the online program, CloudEMDR, in conjunction with psychotherapy and psychoeducation to six PTSD patients for 4 consecutive days and found significantly decreased PTSD symptoms as measured by the clinician-administered PTSD scale for DSM-5 and PTSD checklist for DSM-5. 16 There have been no studies on completely self-administered EMDR without the supervision of a therapist, although such services exist. 17 Thus, our study is to our knowledge the first to directly compare the efficacy of virtual EMDR with that of in-person EMDR using demographically similar groups. We found the remote administration of EMDR to be not significantly different from traditional in-person EMDR in terms of efficacy, supporting continued treatment through the pandemic.
Furthermore, our study is to our knowledge the first to investigate virtual EMDR in a primary care setting. There is a general dearth of literature on EMDR in primary care. A U.K. report described three cases of mental health professionals successfully treating patients with obsessive-compulsive disorder with in-person EMDR in a primary care setting. 18 We found no study explicitly validating the technique in primary care. Although our study was not optimized to demonstrate the efficacy of EMDR with CBT compared with CBT alone, the methodology demonstrates the logistical feasibility of providing EMDR in this context. If the efficacy can be proven in future studies, the combination of offering EMDR to patients in the same setting as their primary care provider with the option of virtual therapy would likely make this treatment much more accessible.
Limitations to our study include small sample size and retrospective design. As mentioned earlier, patients who underwent EMDR had demographics that were significantly different from those who had CBT alone. Thus, the comparison of PHQ-9 and GAD-7 score decreases between these two groups required a great deal of adjustment and may be lacking in power and validity. Furthermore, the majority of EMDR patients in this study had one session of EMDR and also CBT, which makes it difficult to describe the specific efficacy of EMDR.
Although most patients in the study had symptoms of anxiety, the diagnoses were diverse, and therefore, PHQ-9 and GAD-7 may not be the ideal scales to quantify the effects of EMDR on all patients. Thus, future directions include prospective studies on more uniform anxiety patients or even paired samples. There should also be further studies comparing EMDR alone with CBT alone.
Future studies should also follow up on patients to evaluate maintenance of treatment effect. It is also important to survey future participants on their thoughts regarding virtual EMDR, especially regarding the limitations of this approach. A qualitative study on the effects of virtual EMDR may add to our statistical analysis of verified outcome measures to further elucidate its efficacy, as our patients expressed feeling more positive and less burdened by anxiety after treatment. Eventually, there should be a validated protocol established for a permanent telehealth option for EMDR to increase accessibility to this important treatment.
Conclusions
In our study, we demonstrated the utility of EMDR in conjunction with CBT to treat psychological disorders originating from past trauma. Our data did not offer evidence to indicate significant differences between efficacy of virtual EMDR compared with in-person EMDR in the primary care setting. In conclusion, we hope these findings promote effective and accessible virtual EMDR treatment during the COVID-19 pandemic and beyond.
Footnotes
Author Contributions
All authors contributed significantly to this article. All authors have reviewed the results and approved the final version of this article.
Disclosure Statement
No competing financial interests exist.
Funding Information
No funding was received for this article.