Applying eye movement desensitization and reprocessing (EMDR) to the treatment of traumatized children: Five case studies.

Citation: Greenwald, R. (1994). Applying eye movement desensitization and reprocessing (EMDR) to the treatment of traumatized children: Five case studies. Anxiety Disorders Practice Journal, 1, 83-97.

Note: You are reading a somewhat different paper than the one cited above. The editor of that journal made major cuts (e.g., the entire final paragraph), changes, and errors, without allowing me to review the changes prior to publication. This is the final draft as submitted. However, the graph of treatment effects is omitted here, because I don't know how to program it on.

Applying eye movement desensitization and reprocessing (EMDR) to the treatment of traumatized children: Five case studies.

Eye movement desensitization and reprocessing (EMDR) is a recently developed psychotherapy method which appears to increase efficiency in treating trauma-based psychological disturbance. Applications to child treatment were explored in five case studies of children suffering from post-traumatic symptoms several months after Hurricane Andrew. Subjects were treated with one or two EMDR sessions, until Subjective Units of Disturbance (SUDS) went to 0. Follow-up parent interviews at one and four weeks post-treatment found all subjects returning to pre-trauma levels of functioning, with additional improvement in some cases. Further study is recommended.

Eye Movement Desensitization and Reprocessing (EMDR) is a psychotherapy method which features having the client focus on a disturbing memory image while moving the eyes rapidly from side to side. EMDR is used to access, neutralize, and bring to adaptive resolution the upsetting memories at the root of current psychological disturbances.

Eye movement desensitization and reprocessing (EMDR) is a recently developed psychotherapy procedure that has been reported to dramatically increase efficiency in the treatment of traumatic memories (Shapiro, 1989a, 1991a). Briefly, the procedure involves having the client focus intensively on the traumatic memory while moving the eyes rapidly from side to side, by visually tracking the therapist's moving hand. This seems to render the traumatic memory accessible to the healing resources of the rest of the personality, where it may be worked through and integrated. The treatment may involve many 'sets' of eye movements (10 seconds to a minute or more), with the focus changing according to the client's status. For example, if the client reports, "Now I'm feeling more anger," the therapist may suggest concentrating on the anger in the next set. Frequently, earlier memories emerge in the course of treatment, which are thematically and functionally related to the initially targeted memory. Such related memories must also be processed for maximum treatment effect. One or more traumatic memories can usually be fully processed within a single session. The procedure, which requires supervised training for maximum effect and client safety (Shapiro, 1991c), has been described in detail elsewhere (Shapiro, 1989a).

In a controlled study, Shapiro (1989a) reported very successful brief EMDR treatment of Viet Nam veterans and rape victims suffering from PTSD, with gains maintained at three year follow-up (Shapiro, 1991a). A pilot study of EMDR treatment with Viet Nam veterans on an inpatient PTSD unit also showed promising results (Boudewyns et al, 1991). Others have presented case reports in which EMDR appeared to play a key role in treatment success (Kleinknecht & Morgan, 1992; Lipke & Botkin, 1992; Puk, 1991; Wolpe & Abrams, 1991). Marquis' (1991) report on a larger number of cases (N=78) suggests that EMDR can enhance the treatment of many types of problems. The variable magnitude of effect he reported may be, roughly, a function of the number of traumatic memories on which the disturbance is based. Solomon's (1991) consistent and very positive results with "hundreds" of clients supports this view, as his subject group was composed entirely of those recovering from critical incident trauma.

The only published reports of EMDR use with children have been case studies. Shapiro (1991a) briefly described the treatment of an 8 year old boy suffering from PTSD following an earthquake, and of two children whose parents had died. Mendoza-Weitman (1992) described the treatment of a 10 year old boy who was depressed since his father abandoned the family five years before. Greenwald (in press) reported single session treatment of several children in the second week following a hurricane, but provided few details or follow-up data. Many practitioners who use EMDR with children report enhanced treatment efficiency in a wide range of applications, while others report more mixed results (Greenwald, 1993). This discrepancy may be attributable to EMDR's variable efficacy in different applications, and/or to practitioners' differing repertoires of specialized child EMDR techniques (Greenwald, 1993). Treatment of children has been reported to be quite rapid, especially when the focus is a single traumatic event (Greenwald, 1993, in press).

This paper reports on the EMDR treatment of five children who developed post-traumatic symptoms following a natural disaster. It is the largest report of EMDR child treatment to date that includes detailed case information and follow-up data, and the first to utilize the child EMDR technical manual (Greenwald, 1993) as the standard of care. These preliminary findings are commensurate with reports of EMDR treatment of adults with critical incident trauma, warranting further research.

Design

On August 24, 1992, Hurricane Andrew devastated much of South Dade County in Florida. Almost four months later, many area children were still suffering the psychological consequences, generally related to prolonged fear during the event, and subsequent loss experiences. Five such children between the ages of 4 and 11 were treated with one or two sessions of EMDR. The mother was interviewed prior to the first session and asked to describe and rate the changes in her child since the hurricane, and to respond similarly at one week and four week follow-up telephone interviews. All treatment and assessment was conducted by the author.

Subjects

Subjects (N=5) were Caucasian and Hispanic children between the ages of 4 and 11 referred by their parents for emotional and/or behavioral difficulties which began or worsened following the hurricane. Free treatment was provided, between 12/11/92 and 12/17/92. No children were rejected for treatment. Two additional cases are not discussed here: a 2 year old who was not sufficiently cooperative for the EMDR procedure was treated with play therapy only; and a child who claimed absence of trauma was given a different kind of treatment focus.

Assessment

As treatment was the first priority, formal, standardized measures were unfortunately underutilized. Measures included the Structured Interview, the Problem Rating Scale, and the Subjective Units of Disturbance Scale, as described below.

Structured Interview.

The subject's mother completed a brief (about 10-15 minutes) structured interview covering current living situation and family structure, and subject's trauma history, including pre-hurricane. In the follow-up interviews, the mother was asked to report any new traumatic experiences or changes in life situation, and any other negative or positive changes she noticed in her child.

Problem Rating Scale (PRS).

The PRS is a modification of the SUDS (described below), and was designed for this study. During the interview the mother was asked to describe, and then rate, current symptoms/complaints, "with 10 being the worst the problem could be, and 0 being no problem at all." Ratings were to take into account the week prior to the rating interval. The PRS' face validity is strong, as it represents a quantification of the process by which many children may normally be referred for treatment: according to the judgment of the parent as to the magnitude of the problem.

Subjective Units of Disturbance Scale (SUDS).

The other measurement, used during treatment only, involved the subject rating current emotional reactivity to the memory serving as a treatment focus, on a 0 to 10 Subjective Units of Disturbance Scale (SUDS; adapted from Wolpe as described in Shapiro, 1989a; further modified for children as described in Greenwald, 1993). This measure is taken at several intervals during the treatment of each upsetting memory, as treatment is generally continued until the SUDS reaches 0. Non-reactivity to the traumatic memory is considered an indicator of recovery (Horowitz, 1986).

Data collection.

Data was collected as follows. The mother was interviewed just prior to the first treatment session. At that time the PRS ratings were taken, on each problem/concern, for the following intervals: the week prior to the trauma (hurricane), the second week following the trauma, and the week prior to treatment. Follow-up telephone interviews were conducted one week and four weeks after the final treatment session. At each of these interviews, a review of prior PRS ratings was given for each problem before asking for a rating covering the previous week, to enhance consistency in the response set. SUDS was assessed during treatment only, and served primarily as a clinical tool, an indicator of treatment progress within the session.

One subject (#2) had experienced a trauma two months prior to the hurricane, which was identified as the primary treatment focus, although the hurricane had aggravated symptoms to original post-trauma levels. In this case, the baseline problem rating was given for the time prior to the earlier trauma, and post-trauma problem ratings were given for both events, and then combined in a single rating. The other subject (#4) with major prior trauma was scored in the standard manner, with the hurricane as a focus. This decision was made according to the mother's determination as to the primary source of current disturbance.

Procedures

The EMDR procedure, though complex and variable, can be described roughly as follows. The child is typically asked to concentrate on a particularly upsetting image while moving the eyes rapidly from side to side (e.g., by visually following the therapist's moving hand) for a brief period. Following a set of eye movements, the therapist 'checks in' with the child, in order to monitor progress and assist in refocussing for the next set of eye movements. At certain points the therapist might request a SUDS rating, request a rating for level of confidence in a positive statement, or make other interventions. During treatment the client may experience decreased distress, decreased vividness of the image, imaginal reexperiencing, physiological (e.g., tension) release, progression of emotional responses (e.g., fear, anger, sadness, acceptance, humor), and the supplanting of negative beliefs and self-statements by positive ones. Whenever possible, treatment continues until SUDS is eliminated and positive beliefs are strongly held.

Treatment was conducted according to the standards presented by Shapiro (1989b, 1991b, 1992a, 1992b) and Greenwald (1993). Following a brief parent interview, the child was treated with EMDR, in sessions ranging from about 20 to 90 minutes. No other modalities (e.g. play or art therapy) were used, although one subject (#1) alternated between EMDR and using her coloring book. Sometimes prior traumatic experiences were also treated in the session, and in one case, a prior trauma was the sole focus. Two children, for whom SUDS did not reach 0 on all targeted memories in the first session, returned for an additional session. Telephone follow-up interviews were conducted at one week and four weeks after the final treatment session.

Rather than following the stringent requirements for the PTSD diagnosis, a broader, more inclusive and clinically useful conceptualization of traumatization was used, as suggested by Terr (1991). All five subjects were determined to have been traumatized, by virtue of their acquisition and maintenance of distressing symptoms following an upsetting event.

SUDs did reach 0 on targeted upsetting memories for all subjects, in either one or two sessions. This measure was not repeated at follow-up, and served primarily as the therapist's indicator of completion for a given treatment focus.

The three or four problems listed for each subject in the PRS were averaged at each rating interval, as shown in figure 1. While some individual subjects had shown gradual improvement from interval 2 (post-hurricane) to interval 3 (pre-treatment), and others had gotten worse in that period, all showed a marked improvement at interval 4 (1 week post-treatment), and varied from maintenance of gains to substantial continued improvement at interval 5 (4 week post-treatment). All subjects recovered by interval 5 to at least within one rating unit of their pre-trauma level, on each problem and on average, and two subjects (#4 and #5) showed marked improvement (3 or more units) in certain areas.

One subject (#2), who ended the first session with SUDs having gone from 10 to 6, experienced some emotional volatility until the second session, as well as some symptom relief. No other negative effects were reported. Many additional positive effects were reported, including increased sense of happiness, improved concentration and school performance, more cooperative and responsible behavior, less emotional reactivity, and better sibling relationships. Three of the subjects experienced significant life changes during the follow-up period, one positive (#5 - returning to a favored school) and two negative (#3 - father's increased absence for new job; and #4 - having to move from an apartment to a friend's backyard, and changing schools) with no apparent effect on treatment results.

Subject #1: "Jamie"

Jamie was a precocious 4 year old girl with increased obstinance, clinging, and whining since the hurricane four months before. The first session included completed (i.e., SUDS down to 0) EMDR treatment of night fears and upsetting hurricane-related memories. The death of a cat one year before was touched on, but perhaps not fully addressed. She came back for a second session, mainly because she had enjoyed the first one, and because she was along for the ride when her brother (Subject #2) came back. The second session was short, amd touched briefly on many items, including the death of the cat. At one week follow-up her mother reported that her symptoms were back down to pre-hurricane levels, and that she was more able to play contentedly by herself than ever before. At four week follow-up her mother reported that Jamie was still sleeping alone, as she had just started to do around the beginning of treatment, and that the gains were holding steady.

Subject #2: "Andy"

Andy was an intelligent 9 year old boy who, two months before the hurricane, experienced the accidental death of a disliked classmate (he was not present for the incident). His mother and he agreed that the death was the primary precipitant, and the hurricane just a minor intensifier, of his symptoms, including being very pushy and testy, racing thoughts interfering with getting to sleep, trouble concentrating, and bad moods. Six months after the death, the symptoms were not much less than they started. He believed that "It was my fault" and "I'm next [to die]."

In the first session Andy apparently got over his guilt and need for punishment, and expressed full confidence in the self-generated statements "It wasn't my fault," and "I'm going to live a long life." He was too tired to complete processing on the entire memory, and ended with a SUDS of 6, including feelings of confusion and possible helplessness. Over the next several days, his parents reported that Andy was extremely volatile, but also that he began to express and articulate difficult emotions that he had always held in before. In the second session, the death incident was processed until the SUDS reached 0. The hurricane-related events were not addressed.

At one week follow-up, his mother reported that his symptoms were dramatically down to just above the pre-trauma baseline, with a bit of volatility remaining. Ability to express himself had continued to develop, and he was more cooperative, and took more responsibility in his family, than he had before. At four week follow-up, he was back to pre-trauma baseline, and his mother said he was "doing pretty well."

Subject #3: "Paul"

Paul was an eager, intelligent 11 year old boy who changed markedly following the hurricane, becoming apathetic, easily frustrated, pessimistic, and irritable. Although Paul's mother reported that the family was in counseling, she felt that Paul's changes were hurricane-related due to the timing of onset. Towards the end of the session Paul was asked to describe his most recent bad dream. It was two months prior to the session, and consisted of a radio announcement that a hurricane was coming. This seemed to indicate that Paul's mother's assessment was correct.

Although Paul expressed no motivation for treatment, and checked his watch frequently, he was co-operative. The treatment focus was the scariest hurricane memory, of huddling in a closet with the house falling apart around him. Paul found himself reexperiencing many associated upsetting memories, in a sequential manner. In addition to fear he expressed helplessness, saying "I can't do anything." However, he eventually expressed full confidence in the statements, "We're gonna make it better than before," and "I'm gonna get through this." His SUDS level went from 8 to 0, very gradually as he reexperienced different memory segments. The reexperiencing of a segment was often preceded by a physical sensation. For example, when asked what he noticed in his body, he might say "my ears are hurting," and then in the next set of eye movements he would reexperience the noise and the air pressure from the storm. He commented that "I get better and better at handling this the more times I go through it." Finally, he was able to imagine weathering another hurricane, but with less fear and more comfort and companionship.

At one week follow-up, Paul's mother reported a great drop in symptoms, "just about back to normal." At four week follow-up, Paul's mother reported that a sadly anticipated family change had taken place: Paul's father took a job out of state, and was with his family only rarely. Despite this, the symptoms were down to nearly pre-hurricane baseline, a small additional improvement. His mother reported that he had caught up on his schoolwork and was doing well again, and that he had a "better attitude" in general. For example, he was more forgiving of himself, and didn't feel so badly after making a mistake.

Subject #4: "Sam"

Sam was a mildly retarded 8 year old boy who was scared and uncomfortable during the hurricane, and experienced loss and displacement afterwards. His parents reported that he was having trouble sleeping, was emotionally volatile, frequently angry, and overactive as an expression of his distress. He had been traumatized two years earlier by a teacher who would frequently lock him in a closet for punishment, but since the hurricane his distress had roughly doubled. He was treated in a single hour and a half session. He was cooperative but inarticulate, and required special technical adaptations in order to maintain effective participation.

Treatment began with a nightmare, and during eye movements Sam was able to visualize using a sword to protect himself against the scary dream image. The sword was invoked periodically throughout the session, and he would occasionally smile and say, "Now I'm going to have a good dream with my sword." The hurricane material was treated next. Finally, experiences with the 'bad teacher' were addressed. At one point during the eye movements, Sam spontaneously visualized the teacher's face becoming a bad person with a sword, and having a swordfight with Sam winning. Asked what thoughts went with the image of the teacher's face, he said "I can't do anything [right]." Asked if he could do things now, he said yes, and that thought was used as a focus for eye movements. Later, asked whose fault it was that the teacher was mean to him, he said (during eye movements), "She told me it was my fault she was mean, but it wasn't. It was her fault."

At one week follow-up his mother reported that symptoms were around pre-hurricane levels, and that "he seems happier." At four weeks follow-up his mother reported that the symptoms were very low or gone, representing an improvement over the pre-hurricane baseline. In this period, the family had to leave their temporary lodgings, move into someone's backyard, and Sam had to change schools, leaving one he liked. His mother said "He's taking [the displacement] better than the first time." She also reported that Sam had become more responsible, and was dealing better with his problems. For example, "When he gets angry, he either avoids [the situation] or handles it [instead of acting out]."

Subject 5: "Mike"

Mike was a four year old boy who had been extremely angry and hostile since the hurricane. He had already had these tendencies before, which his mother attributed to the birth of his younger sister and their sibling rivalry. Treament went very quickly, covering all the hurricane and post-storm incidents. Following a search for times when he was angry at various family members, a couple of apparently minor examples were treated. Mike reported being able to find no more bad feelings to treat.

At one week follow-up, symptoms were down to pre-hurricane level, and his mother reported simply, "He's back to normal." He had also returned to his old school in this period, which he preferred. At four weeks follow-up, symptoms had continued to decrease slightly, essentially back to baseline, except for unhappiness which was much lower than baseline. Mike's mother reported that he was "really happy... like his old, normal self." It's possible that, in addition to treating hurricane-related trauma, some family issues were effectively addressed.

By using every case of child trauma in a particular treatment setting, selection bias in reporting cases is eliminated, thus providing a more realistic estimate of EMDR's efficacy in this context, and of the potential value of additional study.

The quality of the data can be questioned on several grounds. A single type of measurement was given by a single rater, relying on recall for the first two intervals. The mothers did express confidence in these ratings, whereas they expressed lack of confidence in a rating at a midpoint interval between post-trauma and pre-treatment, which was later discarded. Indirect support for reliability of such recall data is provided by Norris and Kaniasty (1992), who also point out that memory tends to be organized around landmark events. While the in-session SUDS declines have maintained over time in the other published data reviewed above, and additional treatment outcomes were actively sought in the follow-up interviews, multiple raters and measures would allow for more confidence in the results.

Investigator bias is also an issue, as the author was responsible for all treatment and assessment. Awareness of the potential for bias led to concerted efforts to encourage reporting of negative phenomena. In the author's past experience, mothers have been quite open in reporting their child's negative symptoms. However, this issue must ultimately be addressed in controlled studies with blind assessment.

Additional problems include: lack of long-term follow-up (which may not be appropriate, as children are so subject to ongoing environmental influences); potential unreported environmental changes which may have affected the subjects (e.g., changes in marital relationship or parental stress level); use of a single therapist for all cases; the small number of cases; and the lack of control which is typical of case reports. Of course, the subjects provided their own no-treatment control, insofar as the PRS recall data may be usable.

The rather dramatic response to this brief treatment was consistent with expectations, as the subjects' presenting problems were based on a small number of known traumatic events. The fact that two of the subjects made substantial improvements beyond pre-trauma baseline in at least one problem area suggests that the baseline included some dysfunction rooted in earlier trauma, which was addressed in this treatment. The continued post-treatment progress of some subjects suggests that a process of integration or generalization may occur over time. This is consistent with other reports of EMDR treatment. This finding may also have been influenced by an interaction between the child's recovery and responsive environmental factors, such as supportive parents.

As those who remain traumatized from prior experiences often have special difficulty handling new upsetting experiences, previously traumatized children may have been disproportionately represented in the group of still-distressed from which the subjects came. Conversely, following EMDR treatment, the subjects in this study appeared to cope well with new challenges. If further investigations yield similar results regarding post-treatment coping capacity, this would suggest that EMDR treatment can facilitate 'complete' integration of the traumatic memory, and concomitant elimination of the trauma burden.

This report provides preliminary evidence of the efficacy of EMDR in treating children with post-traumatic reactions. Every subject showed the same trend of substantial and sustained improvement following the brief treatment. However, given the serious design limitations, these results can only be considered suggestive. In future studies, treatment should continue to be conducted by practitioners formally trained in this specialty, with additional training for using EMDR with children when applicable, to maximize effectiveness and to ensure meaningful results. As EMDR appears to have great potential in the treatment of trauma, further research in this area is urgent.

Boudewyns, P. A., Stwertka, S. A., Hyer, L. A., Albrecht, J. Q., & Sperr, E. V. (1991, August). Eye Movement Desensitization for PTSD of combat: A pilot study. Presented at the American Psychological Association annual convention, San Francisco.

Greenwald, R. (1993, Spring). Treating children's nightmares with EMDR. EMDR Network Newsletter, 3(1), 7-9.

Greenwald, R. (in press). Using EMDR with children: A manual. Pacific Grove, CA: EMDR.

Horowitz, M. J. (1986). Stress Response Syndromes (2nd ed.). Northvale, NJ: Jason Aronson.

Kleinknecht, R. A. & Morgan, M. P. (1992). Treatment of post-traumatic stress disorder with eye movement desensitization. Journal of Behavior Therapy and Experimental Psychiatry, 23, 43-49.

Lipke, H. J. & Botkin, A. L. (1992). Case studies of eye movement desensitization and reprocessing (EMD/R) with chronic post-traumatic stress disorders. Psychotherapy, 29, 591-595.

Marquis, J. N. (1991). A report on seventy-eight cases treated by eye movement desensitization. Journal of Behavior Therapy and Experimental Psychiatry, 22, 187-192.

Mendoza-Weitman, L. (1992). Case study. EMDR Network Newsletter, 2(1), 11-12.

Norris, F. H. & Kaniasty, K. (1992). Reliability of delayed self-reports in disaster research. Journal of Traumatic Stress, 5, 575-588.

Puk, G. (1991). Treating traumatic memories: A case report on the eye movement desensitization procedure. Journal of Behavior Therapy and Experimental Psychiatry, 22, 149-151.

Shapiro, F. (1989a). Efficacy of the eye movement desensitization procedure in the treatment of traumatic memories. Journal of Traumatic Stress, 2, 199-223.

Shapiro, F. (1989b). Eye Movement Desensitization and Reprocessing: Level I Basic workshop manual. Available from EMDR, P.O. Box 51010, Pacific Grove, CA 93950-6010 with formal training.

Shapiro, F. (1991a). Eye movement desensitization and reprocessing procedure: From EMD to EMD/R - A new treatment model for anxiety and related traumata. The Behavior Therapist, 14, 133-135.

Shapiro, F. (1991b). Eye Movement Desensitization and Reprocessing: Level II workshop manual. Available from EMDR, P.O. Box 51010, Pacific Grove, CA 93950-6010 with formal training.

Shapiro, F. (1991c). Eye movement desensitization and reprocessing: A cautionary note. The Behavior Therapist, 14, 188.

Shapiro, F. (1992a, May). Eye Movement Desensitization and Reprocessing: Level I Basic training[workshop]. Honolulu, HI.

Shapiro, F. (1992b, July). Eye Movement Desensitization and Reprocessing: Level II training [workshop]. San Jose, CA.

Solomon, R. (1991, October). Eye movement desensitization and reprocessing: Treatment of choice for critical incident trauma. Paper presented at 8th Users' Stress Workshop, San Antonio, TX.

Terr, L. (1991). Childhood traumas: An outline and overview. American Journal of Psychiatry, 148, 10-20.

Wolpe, J. & Abrams, J. (1991). Post-traumatic stress disorder overcome by eye-movement desensitization: A case report. Journal of Behavior Therapy and Experimental Psychiatry, 22, 39-43.