Fifteen Month Follow-Up of Eye Movement Desensitization and Reprocessing (EMDR) Treatment for Posttraumatic Stress Disorder and Psychological Trauma
Sandra A. Wilson The Spencer-Curtis Foundation
Lee A. Becker University of Colorado at Colorado Springs
Robert H. Tinker The Spencer-Curtis Foundation
(abstract) The present study is a 15-month follow-up of the effects of eye movement desensitization and reprocessing (EMDR) therapy on the functioning of 66 participants, 32 of whom were diagnosed with posttraumatic stress disorder (PTSD) prior to treatment. PTSD participants imporved as much as those without the diagnosis, with both groups maintaining their gains at 15 months. At 15-month follow-up, the three 90-minute sessions of EMDR previously administered (S.A. Wilson, L.A. Becker, & R.H. Tinker, 1995) produced an 84% reduction in PTSD diagnosis and a 68% reduction in PTSD symptoms. The average treatment effect size was 1.59; the average reliable change index was 3.37. Implications of the maintenance of EMDR treatment effects are discussed.
Posttraumatic stress disorder (PTSD) research and eye movement desensitization and reprocessing (EMDR; Shapiro, 1989a, 1989b, 1995) research to date have provided limited information about the long-term effectiveness of EMDR or other psychological interventions for PTSD.
EMDR research using follow-up periods of 1-3 months has shown maintenance of treatment effects (Renfrey & Spates, 1994; Rothbaum, in press; Scheck, Schaeffer, &Gillette, in press; Vaughan et all., 1994). Our original study (S.A. Wilson, Becker, & Tinker, 1995) used a delayed-treatment design with 80 participants. We found that participants immediately receiving EMDR (n = 40) showed decreases in presenting complaints (including anxiety, depression, and PTSD symptoms), whereas those assigned to the delayed-treatment condition (n = 40) showed no improvement on any of those measures across the 30 days before treatment, but after treatment showed similar improvement on all measures. All gains for both conditions were maintained at 90 days posttreatment.
Two controlled studies on EMDR have published follow-up data of more than 3 months. Carlson, Chemtob, Rusnak, Hedlund, and Muraoka (in press) studied 35 combat veterans diagnosed with combat-related PTSD, comparing 12 sessions of EMDR (n = 10) to 12 sessions of biofeedback-assisted relaxation (n = 13). A control condition (n = 12) received routine clinical care. Carlson et al. (in press) obtained clinically and statistically significant treatment effects in the EMDR condition on a variety of self-report, psychometric, and standardized interview measures, which were greater than those for the other two conditions.
(discussion) The present results suggest that the initial treatment effects (S.A. Wilson et al., 1995) were maintained 15-months post-treatment for both the PTSD and the partial PTSD participants; that there was a substantial (84%) reduction in PTSD diagnoses; and that there was a 68% reduction in PTSD symptoms for all participants. These findings compare reasonably well with the 75-80% reduction in PTSD symptoms found by Richards et al. (1994) with their longer behavioral exposure method (60 hr of exposure vs. 4 ½ hr in the S.A. Wilson et al., 1995, study). This suggests that the cost-effectiveness of EMDR can be a factor to be considered in selecting a treatment modality for traumatized individuals in the future.
The maintenance of treatment effects over 15 months in the present study corresponds well with other research on EMDR, which has demonstrated maintenance of treatment effects from 1 to 12 months (Carlson et al., in press; Rengrey & Spates, 1994; Rothbaum, in press; Vaughan et al., 1994; D. Wilson et al., 1996). In the present study, when the treatment ESs were computed by using SUDS, SCL-90-R GSI, and STAI Trait Anxiety, they were found to be substantial initially (Mes = 1.43) and to remain substantial at 3 months (Mes = 1.60) and 15 months (Mes = 1.59) following treatment.
Limitations of the present study include not investigating the effects of comorbidity, having no control condition for the 15-month follow-up period because of the delayed-treatment design of the original study (S.A. Wilson et al., 1995), and attrition of participants at the 15-month follow-up.
Clinical significance was demonstrated by using the objective criteria of movement into a functional range and reliability of change. The RCIs indicate that at 15-month follow-up all measures denoted reliable change (i.e., are greater than 1.96). Further, for the PTSD participants, initial scores on all dependent variables were well above the cutoff scores, whereas at the 15-month follow-up all means (save one) were below the cutoff scores, indicating movement from the dysfunctional range to the functional range. When we looked at clinical significance based on data from individual participants, we found that no individuals were reliably worsened on all measures by the three sessions of EMDR. Exposure methods with PTSD have sometimes been found to have detrimental or limited effects (Pitman et al., 1991; Solomon, Gerrity, & Muff, 1992) and to be relatively difficult for participants compared to EMDR (Boudewyns, Stwerka, Hyer, Albrecht, & Sperr, 1993). These findings suggest that EMDR may be less aversive than those exposure therapies. On the other hand, about one third of the participants showed a degree of improvement classified as uncertain. Most of the remainder manifested reliable improvement, principally of a recovered nature. These results indicate that for some individuals, three sessions of EMDR do not constitute complete treatment.
It appears that posttreatment psychotherapy had a negligible effect on follow-up outcomes as only two differences were found; The 7 participants who sought additional therapy related to the original traumatic memory had higher SUDS scores and more intrusion symptoms at 3 months, suggesting that these symptoms motivated their seeking additional therapy. This therapy seems to have helped, in that the differences in functioning of these individuals had disappeared by 15 months.
Overall, the present results provide encouraging evidence of the durability of the effects of EMDR on a heterogeneous, predominantly civilian sample of chronically traumatized adults.
Sandra A. Wilson and Robert H. Tinker, The Spencer-Curtis Foundation, Colorado Springs, Colorado; Lee A. Becker, Department of Psychology, University of Colorado at Colorado Springs.
Correspondence concerning this article should be addressed to Lee A. Becker, 1420 Austin Bluffs Parkway,
P.O. Box 7150, Department of Psychology, University of Colorado, Colorado Springs, Colorado 80933-7150. Electronic mail may be sent via Internet to email@example.com.