Relapse Prevention Program for Treatment of Obsessive–Compulsive Disorder

Relapse Prevention Program for Treatment of Obsessive–Compulsive Disorder
[Regular Articles]
Hiss, Hella1; Foa, Edna B.2,3; Kozak, Michael J.2
1Department of Psychiatry, Hamburg University, Hamburg, Germany
2Department of Psychiatry, Medical College of Pennsylvania.
3Correspondence concerning this article should be addressed to Edna B. Foa, Department of Psychiatry, Medical College of Pennsylvania, 3200 Henry Avenue, Philadelphia, Pennsylvania 19129.
This research was supported in part by National Institute of Mental Health Grant R01 MH45404 awarded to Edna B. Foa. Therapy of patients participating in this study was conducted by Blanche Freund, Hella Hiss, Michael J. Kozak, and David Riggs. David Penkower did data management and SPSS runs.
Received Date: August 6, 1993; Revised Date: September 30, 1993; Accepted Date: October 19, 1993
Abstract
Eighteen participants with obsessive–compulsive disorder received 3 weeks of intensive treatment by exposure and response prevention, which were followed by either a relapse prevention (RP) program or associative therapy (AT; an attention-control program). Independent evaluators conducted assessments of obsessive–compulsive symptoms, anxiety, and depression, before and after intensive behavior therapy, after the week of intensive RP or AT and at a 6-month follow-up. Results indicated that the RP program was effective in preventing relapse: Both treatment groups improved immediately after the intensive treatment, but the RP group remained improved at follow-up, whereas the AT group showed some return of symptoms.
________________________________________


Prolonged exposure and response prevention are effective treatment procedures for obsessive–compulsive disorder (OCD), with the majority of patients who receive this therapy showing significant improvement that persists over time. A review of outcome studies that included over 300 persons who practiced obsessive–compulsive rituals and were treated by exposure and response prevention revealed that from 60% to 100% (M = 76%) of these patients remained improved from 3 months to 6 years posttreatment. (Foa & Kozak, in press). For example, Foa, Steketee, Grayson, and Doppelt (1983) investigated 50 patients with OCD who received this treatment. Ninety-six percent benefited (defined as 30% improvement over pretreatment) immediately after treatment, and 76% remained improved at follow-up (M = 16 months). Moreover, Ost (1989), in reviewing seven studies on the long-term effects of exposure treatment for OCD, noted that during 1–3 years posttreatment, 85% of the patients maintained their gains, and over 50% received no further therapy. The estimates ranging from 75% to 85% maintenance after treatment by exposure and response prevention are quite impressive when compared with the 10% rate found after withdrawal from clomipramine (Pato, Zoka-Kadouch, Zohar, & Murphy, 1988), which is the principal pharmacotherapy for OCD.
Despite the relatively low rate of relapse after behavioral treatment of OCD and other anxiety disorders, the potential contribution of relapse prevention techniques has been addressed by clinical researchers who study behavior therapy. Ost (1989) described a program of maintenance techniques that was introduced in the final treatment session, and was monitored by the therapist by means of telephone contacts during the follow-up period. This procedure required daily practice of skills learned in therapy for coping with anxiety and was thought to be applicable with various anxiety disorders. A program directed specifically at persons who practice obsessive compulsive rituals and who had relapsed was described by Espie (1986). The treatment components of this 10-week program were training in identifying early signs of irrational thinking and compulsive behavior, setting goals, and involving the spouse.
Although relapse prevention techniques for various psychiatric disorders have been described and their effectiveness has been documented (e.g., Marlatt & Gordon, 1985), their efficacy with anxiety disorders, particularly with OCD, has not been subjected to controlled investigations. Ost (1989) reported low relapse rates for patients who had received his maintenance program, but there were no comparable data from patients who had not received the program, so its efficacy was not established. Similarly, Espie's (1986) results indicating the long-term success of his maintenance program are inconclusive because of the absence of a comparison group.
At least some behavioral programs for OCD routinely include techniques aimed at relapse prevention. For example, the behavior therapy program for OCD at the Medical College of Pennsylvania consists not only of intensive exposure and response prevention but also of discussion of the difference between lapse and relapse. It also includes instructions for posttreatment exposure, correction of cognitive distortions, and the enlisting of family support and telephone contacts after the intensive treatment. Thus it is unclear whether the high maintenance rate derives from exposure and response prevention itself, or from the addition of techniques aimed at relapse prevention. The present study was designed to separate the long-term effects of exposure treatment from those of procedures specifically aimed at relapse prevention. To this end, gains at follow-up for a group that received intensive behavior therapy followed by a control therapy were compared with those of a group that received behavior therapy followed by a relapse prevention program.
The relapse prevention program consisted of four daily sessions and combined training in self-exposure with anxiety management techniques (Meichenbaum, 1974), goal setting, recruitment of social support, and additional interventions that have been advocated by other researchers (Brownell, Marlatt, Lichtenstein, & Wilson, 1986; Espie, 1986; Ost, 1989). The attention–control condition, called associative therapy, was adapted from a procedure found to be credible for patients with anxiety disorders but ineffective in reducing anxiety (Butler, Cullington, Munby, Amies, & Gelder, 1984; Gelder et al., 1973).
We hypothesized that the relapse prevention program would enhance the maintenance of behavioral treatment gains of persons who practiced obsessive–compulsive rituals. Specifically, we predicted that although both groups would profit from the intensive behvior therapy, patients who received the associative therapy would show the return of their obsessive–compulsive symptoms between the posttreatment and six-month follow-up assessments, whereas those who received the relapse prevention program would maintain their gains.
METHOD
Design
A 3 × 2 factorial design was used in this study with one within-subject factor, Assessment Occasion (pretreatment, posttreatment, and follow-up) × Type of Treatment (relapse prevention and associative therapy). Persons who practiced obsessive–compulsive rituals received 3 weeks of intensive exposure and response prevention (Phase 1); this was followed by either a relapse prevention treatment or an associative therapy (Phase 2). Patients were assigned randomly to the Phase 2 conditions. A 3-hour screening evaluation and two 1-hour posttreatment interviews were conducted at four points by doctoral-level psychologists who were unaware of the treatment assignment: before treatment, after Phase 1 intensive behavior therapy, after the week of intensive relapse prevention or associative therapy, and at 6-month follow-up.
Participants
Twenty patients, 8 women and 12 men, who met Diagnostic and Statistical Manual of Mental Disorders. (Third ed.; rev.; DSM–III–R; American Psychiatric Association, 1987) criteria for OCD participated in the study. Diagnosis was determined with the Structured Clinical Interview for Diagnosis (SCID DSM–III–R; Spitzer, Williams, & Gibbon, 1987), which was conducted by doctoral-level clinical psychologists who were trained to use this instrument. Each patient was interviewed separately by two psychologists. Only patients who were diagnosed with OCD as their primary Axis I diagnosis by both interviewers were entered into the study. Mean age was 31 years; mean duration of symptoms was 11 years, and mean age of onset of symptoms was 19 years.
One patient did not complete the relapse prevention homework assignments, and 1 missed half of the weekly telephone contacts. The data from these 2 patients were dropped from the analyses. Therefore, the analyses included data from 8 patients in the relapse prevention group and from 10 in the associative therapy group. Six patients had primarily washing compulsions (3 in each group), 8 had primarily checking compulsions (4 in each group), 3 had both washing and checking (1 in the relapse prevention group and 2 in the associative therapy group), and 1 had mainly cognitive rituals (associative therapy group).
All subjects improved by over 50% from pretreatment to posttreatment and therefore none were excluded from the analyses of relapse at follow-up.
Measures
Obsessive–Compulsive Symptoms
Assessor Ratings of Obsessive–Compulsive Symptom Severity
During a semistructured 1-hr interview, an independent assessor, who was a doctoral-level psychologist trained in the use of this measure, rated each patient using three 9-point Likert scales ranging from none (0) to extreme (8) for severity of obsessive fear, avoidance, and ritualistic behavior. The sum of the severity scores for the main obsession, avoidance, and ritual for each patient was computed for each assessment point. The combined obsessive–compulsive symptom severity scores ranged from 0 to 24. These ratings had satisfactory interrater reliability (r = .92–.97) and were sensitive to treatment effects (Foa et al., 1983).
Yale–Brown Obsessive–Compulsive Scale (Y–BOCS)
Independent assessors administrered the Y–BOCS scale, a 10-item scale ranging from none (0) to extreme (4) that rates the severity of the following symptoms: time occupied by obsessions and compulsions, interference, distress, resistance, and control. The scale yields a total severity score (ranging from 0 to 40) that has satisfactory psychometric characteristics. Interrater reliability for the severity score has been estimated at .95 (Goodman, Price, & Rasmussen, 1989) and has been found to be sensitive to treatment effects (DeVeaugh-Geiss, Landau, & Katz, 1989).
Mood State
Because of the high comorbidity of anxiety and depression with OCD (Rachman & Hodgson, 1980), we measured these mood states.
Hamilton Rating Scale for Depression (HAM–D)
This is a 17-item scale, ranging from 0 to 50, that is used to evaluate vegetative symptoms, depressive mood, and thoughts (Hamilton, 1960).
Beck Depression Inventory (BDI)
This is a widely used 21-item questionnaire, ranging from 0 to 66, that emphasizes cognitive symptoms of depression. It has been found (Beck, Ward, Mendeleson, Mock, & Erbaugh, 1961) to have satisfactory psychometric characteristics (split-half reliability of .93 and interitem correlations ranging from .62 to .66).
Spielberger State–Trait Anxiety Inventory (STAI)
The STAI contains 20 items for state anxiety and 20 items for trait anxiety. Only the state anxiety scale was administered in the present study. Spielberger, Gorsuch, and Lushene (1970) reported internal consistency estimates ranging from .83 to .92; test–retest reliability was estimated to be .40.
Patient Expectancies About Treatment Efficacy
Self-ratings of expectancy for the success of the relapse prevention and the associative therapy programs on obsessions, compulsions, and general discomfort were recorded before participants entered the relapse prevention program or associative therapy (at Week 3). Each self rating ranged from “I don't think it will help me” (0) to “I'm quite sure it will help me” (8).
Treatment
Treatment was conducted by four doctoral-level clinical psychologists with expertise in exposure and response prevention with OCD. Their postdoctoral experience ranged from 3 to 8 years. Therapist assignment was balanced across treatment groups, such that each therapist treated 2 or 3 patients in each group. Biweekly group supervision was conducted by Edna B. Foa to promote treatment integrity. Treatment consisted of two phases: In Phase 1, all participants received two 3-hr sessions of information gathering and treatment planning, followed by an intensive 3-week behavioral treatment for OCD, consisting of 15 daily sessions of exposure and response prevention. In Phase 2, participants were randomly assigned to either a 1-week relapse prevention program or to associative therapy.
Phase I: Intensive Behavior Therapy
Treatment consisted of intensive imaginal exposure, in vivo exposure, and response prevention (cf. Foa & Wilson, 1991). Treatment included 15 daily sessions of exposure with 45 min devoted to imaginal exposure and 45 min to in vivo exposure. In the remaining time, homework assignments were discussed. During the third week of therapy, home visits on two consecutive days were conducted to promote generalization of treatment gains from the clinic to the patient's home and work environment.
Imaginal Exposure
Six scenes, consisting of descriptions of feared situations and related feared disasters, were created by the therapist with the patient. The therapist described the scenes for the patient to imagine, one each day, for 45 consecutive minutes, in increasing order of difficulty. The most disturbing scene was imagined on Day 6 of treatment and repeated on Days 7–15. If additional feared material emerged during sessions, this was incorporated into the scenes. During imagery, patients were instructed not to fantasize themselves carrying out rituals. Patients were instructed to imagine, as vividly as possible, the scenes described by the therapist. Scenes were audiotaped during each session for later imagery practice at home.
Exposure in Vivo
This consisted of daily confrontation with situations that provoke obsessional distress. Prolonged exposure was graded so that moderately disturbing situations were encountered before severely upsetting ones. Exposure to each situation was repeated until distress decreased considerably. For homework, patients exposed themselves to situations that were similar to those they had confronted with the therapist.
Response Prevention
This consisted of abstinence from ritualistic behavior. Previously designated relatives or friends monitored compliance at home and were available to encourage the patient to remain in the fear-evoking situation until the urge to perform a ritual decreased noticeably. In addition, patients recorded any violations of the response prevention rule and delivered the self-monitoring records to the therapist for discussion at the beginning of each session. Therapists were prohibited from discussing relapse prevention procedures during the exposure and response prevention stage of the therapy.
Phase 2: Relapse Prevention or Associative Therapy
Relapse Prevention
The relapse prevention program consisted of four 90-min sessions conducted over a 1-week period during which the participants received training in self-exposure and cognitive restructuring and planned for the changes in their lifestyle.
Session 1.
First, the rationale for the relapse prevention program was presented as follows:
Treatment by exposure and response prevention modifies your obsessive–compulsive habits and reduces your obsessive–compulsive symptoms. However, your symptoms do not go away completely. In stressful situations you may experience increased anxiety and urges to ritualize. Therefore we designed a program that will help you cope with anxiety and will prevent you from returning to your obsessive–compulsive habits.
Second, the patient was asked to identify possible stressors or risk situations that might cause him or her to experience a setback. The therapist helped the patient develop a list of potentially problematic situations (e.g., specific stressors, such as extra responsibilities or interpersonal conflicts).
Third, setbacks were presented as a normal reaction to anxiety and the patient was instructed to view setbacks as an opportunity to practice the skills that he or she had learned during treatment. Outlines for “what to do when a setback occurs” (adapted from Ost, 1989), “how to cope with stress” (Meichenbaum, 1974), and “guidelines for maintaining change” (developed in our Center) were presented to the patients. Patients were instructed to use these self-management skills in everyday life.
Session 2.
A significant other was present in this session and issues pertinent to both the patient and their friend or family member were discussed. Maladaptive patterns of interpersonal interaction (e.g., anger, criticism) were identified and remedial actions were considered. Unrealistic expectations about treatment results (i.e., obsessive–compulsive symptoms will completely disappear) were challenged by the therapist. The therapist explored ways in which the friend or family member could assist the patient when he or she experiences setbacks: Hostile criticism was discouraged. The need for activities to occupy time that was formerly spent on obsessions and rituals was discussed, and systematic planning for job-seeking or leisure activities (or both) was begun.
Session 3.
The therapist presented a rationale for cognitive restructuring to counter anxiety by reevaluating misconceptions. The therapist first trained the patient to identify cognitive distortions (e.g., exaggerating the probability of harm, all-or-nothing thinking: “If my performance is not perfect, it is a total failure”). Next, an A–B–C (A = antecedent; B = belief; C = consequences) mnemonic for understanding automatic, irrational thoughts was taught. To teach the A–B–C technique, the therapist had the patient describe a situation in which he or she was likely to become upset. The therapist first identified the antecedent (A) event and the consequences (C), that is, how the patient felt. The patient was then asked to identify the relevant belief (B; e.g., “I am a failure”). This exercise was repeated until the patient could reproduce the A–B–C identifications independent of the therapist. A list of cognitive distortions (Burns, 1980) was provided to the patient to assist him or her in identifying maladaptive thoughts.
Session 4.
Cognitive restructuring was continued with a focus on beliefs that were directly related to the obsessive–compulsive symptoms. Examples are “If I make a mistake, I will be blamed or punished” and “One should be perfect, otherwise others will disapprove of him.”
The steps for cognitive restructuring were as follows: (a) complete A and C and identify B, as learned in the previous session; (b) identify cognitive distortions underlying the belief; and (c) correct distortions and substitute a realistic belief.
Each session included homework assignments that were discussed in the beginning of the following session. These homework assignments consisted of practice applications of the techniques introduced in the sessions.
Nine telephone follow-up contacts, each lasting approximately 15 min, were spaced over a 12-week period. During each telephone session, the therapist reviewed the patient's use of the maintenance procedures and addressed difficulties the patient experienced when applying the techniques to problem situations that had arisen since the last contact. No office booster sessions were provided.
Associative Therapy
The control treatment consisted of deep muscle relaxation (tension relaxation contrast training by Jacobson, 1938), which has not been found effective for treatment of OCD (Marks, Stern, Mawson, Cobb, & McDonald, 1980), and associative therapy, which has been found to be credible to patients but has not been found to reduce anxiety (Butler et al., 1984; Gelder et al., 1973).
The treatment consisted of four 90-min sessions conducted over a 1-week period. Each session was conducted as follows: (a) discussion of homework (10 min), (b) deep muscle relaxation training (30 min), (c) verbalized associations (45 min), and (d) homework assignments. All association sessions were taped for review as homework.
Session 1.
In this session, the rationale for associative therapy was introduced. The patient was told that
This will teach you to think more clearly about your obsessive–compulsive symptoms. You will be asked to think about anything that comes into your mind when you think about your symptoms. This therapy will enable you to gain a broader view of your obsessive–compulsive behavior and, in turn, will allow you to face problematic situations more effectively.
The patient then received training in progressive muscle relaxation and described to the therapist for 45 min any thoughts that freely occurred about the history of his or her obsessive–compulsive symptoms.
Session 2.
In the second session, a significant other participated to help both the patient and the friend or family member “view the obsessive–compulsive problem in a wider context.” The significant other was asked to free associate about obsessive–compulsive symptoms. The patient was told that listening to a friend's thoughts about the obsessive–compulsive symptoms would help him or her to pay more attention to aspects of the disorder that might have been ignored in the past or perhaps had not been recognized as related to the obsessive–compulsive symptoms. The patient listened to the significant other's associations and then described the thoughts that had been evoked by the friend or family member's associative narration.
Sessions 3 and 4.
The next two sessions consisted of relaxation practice and the therapist's attending to patient's associations about the past and present and of his or her obsessive–compulsive behavior as described in Session 1.
During each of the association periods, the therapist encouraged the patient to follow his or her train of associations and prompted the patient if silence persisted more than 3 min. No other therapy techniques were used.
Homework consisted of (a) relaxation practice with the aid of audiotaped instructions, (b) playing back an audiotape of the previous associations, and (c) audiotaping at home a period tape of free association about obsessive–compulsive symptoms.
The associative therapy procedure did not include telephone contacts during the follow-up period.
RESULTS
Means and standard deviations for all measures were calculated for each group at each assessment point. These data are presented in Table 1. Examination of the within-group variability of the outcome measures revealed sizable differences between groups in the standard deviations of the five outcome measures at different assessment points. This apparent heterogeneity of variance was statistically significant (Bartlett–Box F) for some measures, so nonparametric statistics were used for hypothesis testing.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Table 1 Means and Standard Deviations for the Relapse Prevention (n = 8) and the Associative Therapy Groups (n = 10)
________________________________________
First, we examined possible pretreatment differences between the groups by means of Mann–Whitney U statistics that were calculated separately for each of the five symptom measures. No significant pretreatment group differences emerged. Second, we compared patients' expectancies about the efficacy of the relapse prevention program with their expectations for the associative therapy program by means of a Mann–Whitney U on combined expectancy scores for obsessions, compulsions, and general anxiety. No difference between the groups emerged.
Before group differences in relapse can be detected, improvement in obsessive–compulsive symptoms must be achieved. To examine whether patients indeed improved after Phase 1 treatment, we performed Wilcoxon matched pairs signed ranks tests by comparing each symptom measure before and after Phase 1 treatment (i.e., 3 weeks of intensive behavior therapy). As expected, significant improvement was detected on each of the five symptom measures. These tests are summarized in Table 2. In addition, assessor severity ratings of main obsessions, avoidance, and rituals indicated that all patients in both groups were responders (i.e., showed 50% improvement) immediately after treatment. Therefore, no patients were excluded from the analysis of relapse at follow-up.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Table 2 Wilcoxon Matched Pairs Signed Ranks Tests on Change Scores From Pretreatment to Immediately After Intensive Behavior Therapy
________________________________________
We explored possible differences between the groups immediately after Phase 1 (but before Phase 2) by means of Mann–Whitney U comparisons on each of the five symptom measures. No significant group differences in symptom severity emerged between the groups after Phase 1. On one measure, however, assessor combined ratings of obsessive–compulsive symptoms, there emerged a nonsignificant tendency for lower posttreatment severity in the relapse prevention group (U = 21.5, p = .095).
To examine whether the intensive relapse prevention or associative therapy sessions differentially affected symptoms, we performed Mann–Whitney U tests to compare the two groups on change in symptom severity from immediately after intensive behavior therapy to immediately after the intensive relapse prevention or associative therapy sessions. No differences emerged on any of the five outcome measures.
We examined the differences in relapse between the groups by means of Mann–Whitney U tests in which the groups were compared on change in symptom severity from immediately after intensive behavior therapy to 6-month follow-up. As is evident from the means in Table 1, patients in the associative therapy group showed some return of symptoms at follow-up, whereas patients in the relapse prevention group maintained therapeutic gains. The relapse prevention group showed significantly less relapse than did the associative therapy group on four of the five outcome measures, and they showed a marginal tendency for less relapse on the fifth measure. These tests are summarized in Table 3.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Table 3 Mann-Whitney U Tests on Change Scores From Posttreatment to 6-Month Follow-Up
________________________________________
We used a categorical analysis to examine the hypothesis that the associative therapy group would show clinically significant relapse in obsessive–compulsive symptoms at follow-up and that the relapse prevention group would maintain its gains. For this analysis, participants were divided into responders and nonresponders on each of the 2 measures of obsessive–compulsive symptoms. Responding was defined as improvement of 50%1 or more over pretreatment obsessive–compulsive symptom severity, and group differences were evaluated with Fisher's Exact Tests. A tendency emerged for a few nonresponders in the relapse prevention group (assessor's combined ratings, p = .12; Y–BOCS, p = .11). The results are presented in Figures 1 and 2.

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Figure 1. Percentage of nonresponders on Yale–Brown Obsessive–Compulsive Scale symptom severity at posttreatment and at follow-up in the response prevention (RP) and the associative therapy (AT) groups. Nonresponding was defined as < 50% improvement over pretreatment scores.
________________________________________

________________________________________

[Help with image viewing]
[Email Jumpstart To Image]
Figure 2. Percentage of nonresponders on assessor ratings symptom severity at posttreatment and at follow-up in the response prevention (RP) and the associative therapy (AT) groups. Nonresponding was defined as <50% improvement over pretreatment scores.
________________________________________
DISCUSSION
The present results are consistent with the hypothesis that relapse prevention procedures enhance the long-term efficacy of therapy by exposure and response prevention for OCD. Immediately after treatment, patients in both groups responded well to behavior therapy: 100% improved according to assessor's combined ratings, and 72% improved according to Y–BOCS severity scores. At 6-month follow-up, more patients maintained their gains in the relapse prevention group than in the associative therapy group. Thus, it is not surprising that the latter group showed greater variance on four of the five outcome measures. As indicated by the assessor's ratings, 87.5% of the relapse prevention group remained improved at follow-up compared with 50% of the associative therapy group. According to the Y–BOCS severity scores, 75% of the relapse prevention group remained improved at follow-up compared with 33% of the associative therapy group. In addition to showing better maintenance of improvement in obsessive–compulsive symptoms, the relapse prevention group was significantly less depressed and anxious than the associative therapy group at follow-up.
The percentage of patients who benefited from therapy both immediately after treatment and at follow-up in this study is somewhat lower than is customary at our clinic and at other clinics. As mentioned earlier, Foa and Kozak's (in press) review of long-term outcome of exposure and response prevention treatment revealed that overall, 76% of patients were responders at follow-up. Long-term outcome results from the Foa, Kozak, Steketee, and McCarthy (1992) study, conducted at our clinic indicated that 89% of the patients were responders 3 months after intensive treatment, and 81% were still responders at a 16-month follow-up.
This inconsistency is readily explainable by the stringent criterion for treatment response used in the present study. Many studies used 25% or 30% improvement over pretreatment as the criterion for clinically significant outcome, whereas in the present study a criterion of 50% improvement was used. When a 30% improvement criterion was applied to the present data, then posttreatment response rate was similar to that of previous studies. With respect to the 6-month response rate, a 30% improvement criterion rendered the outcome of the relapse prevention group similar to that of previous studies; however, outcome for the associative therapy group was less favorable.
One factor that might account for the poorer long-term response found in our associative therapy group, compared with that found for exposure and response prevention in previous studies is the exclusion from the associative therapy program of interventions directed at relapse prevention. We noted earlier that such interventions are routinely included in behavior therapy programs for OCD. The present findings suggest that such interventions contribute to long-term maintenance of gains.
Of particular interest are the brief therapy contacts following the relapse prevention sessions. Whereas both groups received four 90-min sessions (total = 6 hr) delivered within the week after intensive therapy, only the relapse prevention group also received nine 15-min telephone contacts over the next 12 weeks. Thus, this group received 2.25 hr more contact than the associative therapy group. As noted earlier, such contacts are routine in our clinic, but were prohibited in the present study for the associative therapy group. It is unlikely that the additional 2.25 hr itself can account for the superiority of the relapse prevention program, because this additional time amounted to only about 5% of the total contact time that a patient received in this study.
Although it seems unlikely that group differences in contact time influenced outcome, the content of the telephone contacts with the therapist may have been important in promoting maintenance of treatment gains. Notably, in the Foa et al. (1992) study, all patients received brief weekly contacts with the therapist during the 3 months following intensive exposure. In the present study, the relapse prevention group had 9 brief telephone contacts with the therapist over the 3 months following intensive exposure. Because we hypothesized that continued contact but not contact time itself was important, we did not incorporate into the associative therapy telephone contacts that were devoid of helpful content. The weekly telephone conversations in the relapse prevention group were intended to reinforce learning that had occurred during the intensive phase of therapy and to guide patients past unanticipated obstacles that arose shortly after they returned to daily living. Furture research, designed to dissect the effects of individual relapse prevention interventions, can shed light on the relative contributions of continued contact with therapists and contact time itself.
One could speculate that the relatively poor long-term response rate of the associative therapy group occurred because these procedures were noxious, and thus detrimental to outcome, but this seems unlikely. The present findings that intensive relapse prevention and associative therapy did not differentially affect symptom severity mitigate against the hypothesis that the associative therapy component was noxious. These results converge with those of two previous studies. Butler et al. (1984) found that immediately after treatment, associative therapy groups did not differ from groups that did not include this treatment. If associative therapy were deterimental, inferior postreatment outcome would have been expected for those who received this treatment. Moreover, in the study by Gelder et al. (1973), a control group that received associative therapy alone showed no deterioration between posttreatment and 6-month follow-up. Such deterioration would have been expected if associative therapy were detrimental.
The likelihood that differences in patient expectations account for the differential efficacy of the two programs is small, because no difference emerged in the measure of treatment expectancy. However, therapist expectancies for relapse prevention and associative therapy were not assessed. It is likely that therapists expected the relapse prevention program to be more effective than the associative therapy procedure, and such differences in therapist expectancy may have contributed to the superiority of the relapse prevention program.
Whereas the two treatment groups in the present study showed essentially equivalent improvements immediately after treatment, the relapse prevention group showed marginally better immediate outcome on 1 of 5 symptom measures. This difference was small, however, and did not represent an overall pattern of differential immediate posttreatment gains in the two groups. Therefore, it seems unlikely that the superior maintenance of gains at follow-up observed in the relapse prevention group resulted from this marginal difference in response to the intensive behavior therapy. In addition, because comparisons were conducted on change scores from posttreatment to follow-up, potential effects of individual differences in posttreatment severity of symptoms were controlled statistically.
Although it can be concluded that the relapse prevention procedures used in the present study were effective in promoting maintenance of gains, the separate contributions of the individual components of the relapse prevention program cannot be distinguished here. The program consisted of several interventions: cognitive restructuring for general thinking distortions as well as for thinking errors specific to obsessive–compulsive symptoms; discussion of obsessive–compulsive fear and its reduction with a friend or family member; instructions for continuing self-exposure and response prevention; and nine brief telephone contacts. In light of the present encouraging results about the efficacy of the overall relapse prevention program, the contributions of the individual components merit further exploration.
References
American Psychiatric Association. (1987). Diagnostic and statistical manual of mental disorders (3rd ed. rev.). Washington, DC: Author. [Context Link]
Beck, A. T., Ward, C. H., Mendeleson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561–571. Bibliographic Links [Context Link]
Brownell, K. D., Marlatt, G. A., Lichtenstein, E., & Wilson, G. T. (1986). Understanding and preventing relapse. American Psychologist, 41, 765–782. Ovid Full Text [Context Link]
Burns, D. D. (1980). Feeling good: The new mood therapy. New York: William Morrow. [Context Link]
Butler, G., Cullington, A., Munby, M., Amies, P., & Gelder, M. G. (1984). Exposure and anxiety management in the treatment of social phobia. Journal of Consulting and Clinical Psychology, 52, 642–650. [Context Link]
DeVeaugh-Geiss, J., Landau, P., & Katz, R. (1989). Treatment of OCD with clomipramine. Psychiatric Annals, 19, 97–101. Bibliographic Links [Context Link]
Espie, C. A. (1986). The group treatment of obsessive–compulsive ritualizers: Behavioral management of identified patterns of relapse. Behavioral Psychotherapy, 14, 21–33. [Context Link]
Foa, E. B., & Kozak, M. J. (in press). Obsessive–compulsive disorder: long-term outcome of psychological treatment. In Mavissakalian & Prien (Eds.), Long-term treatments of anxiety disorders. Washington, DC: American Psychiatric Press. [Context Link]
Foa, E. B., Kozak, M. J., Steketee, G. S., & McCarthy, P. M. (1992). Treatment of depressive and obsessive–compulsive symptoms in OCD by imipramine and behavior therapy. British Journal of Clinical Psychology, 31, 279–292. Bibliographic Links [Context Link]
Foa, E. B., Steketee, G. S., Grayson, J. B., & Doppelt, H. G. (1983). Treatment of obsessive–compulsives: When do we fail? In E. B. Foa & P. M. G. Emmelkamp (Eds.), Failures in behavior therapy (pp. 10–34). New York: Wiley. [Context Link]
Foa, E. B., & Wilson, R. (1991). Stop obsessing! How to overcome your obsessions and compulsions. New York: Bantam Books. [Context Link]
Gelder, M. G., Bancroft, J. H. J., Gath, D. H., Johnston, D. W., Mathews, A. M., & Shaw, P. M. (1973). Specific and non-specific factors in behavior therapy. International Journal of Psychiatry, 123, 445–462. [Context Link]
Goodman, W. K., Price, L., & Rasmussen, S. (1989). The Yale–Brown obsessive–compulsive scale (Y–BOCS): Past development, use, and reliability. Archives of General Psychiatry, 46, 1006–1016. [Context Link]
Hamilton, M. (1960). A rating scale for depression. Journal of Neurological and Neurosurgical Psychiatry, 23, 56–62. [Context Link]
Jacobson, E. (1938). Progressive Relaxation. Chicago: University of Chicago Press. [Context Link]
Marks, I. M., Stern, R. S., Mawson, D., Cobb, J., & McDonald, R. (1980). Clomipramine and exposure for obsessive–compulsive rituals—1. British Journal of Psychiatry, 136, 1–25. Bibliographic Links [Context Link]
Marlatt, G. A., & Gordon (Eds.). (1985). Relapse prevention: Maintenance strategies in addictive behavior change. New York: Guilford Press. [Context Link]
Meichenbaum, D. (1974). Self-instructional methods. In F. H. Kanfer & A. P. Goldstein (Eds.), Helping people change. Elmsford, NY: Pergamon Press. [Context Link]
Ost, L. G. (1989). A maintenance program for behavioral treatment of anxiety disorders. Behaviour Research and Therapy, 27, 123–130. [Context Link]
Pato, M. T., Zoka-Kadouch, R., Zohar, J., & Murphy, D. L. (1988). Return of symptoms after desensitization of clomipramine and patients with obsessive–compulsive disorder. American Journal of Psychiatry, 145, 1521–1525. [Context Link]
Rachman, S., & Hodgson, R. (1980). Obsessions and compulsions. Englewood Cliffs, NJ: Prentice Hall. [Context Link]
Spielberger, C. D., Gorsuch, R. L., & Lushene, R. E. (1970). Manual for the State–Trait Anxiety Inventory (self-evaluation questionnaire). Palo Alto, CA: Consulting Psychologists Press. [Context Link]
Spitzer, R. L., Williams, J. B. W., & Gibbon, M. (1987). Structured Clinical Interview for DSM-III–R(SCID). New York: Biometrics Research Department, New York State Psychiatric Institute. [Context Link]
1We adopted a 50% improvement criterion for successful outcome to provide a conservative estimate of clinically significant improvement. In previous treatment studies of OCD we have used a 30% criterion (e.g., Foa, Steketee, Grayson, & Doppelt, 1983). A 30% improvement criterion yields the following responder rates for the present data: relapse prevention group: assessor ratings, 87% responders; Yale–Brown Obsessive–Compulsive Scale, 75% responders; associative, therapy group: assessor ratings, 60% responders; Yale–Brown Obsessive–Compulsive Scale, 67% responders. [Context Link]
________________________________________

Accession Number: 00004730-199408000-00017
Copyright (c) 2000-2007 Ovid Technologies, Inc.
Version: rel10.5.8, SourceID 1.13281