Individual, Group, and Multifamily Cognitive-Behavioral Treatments

Abstract

Back to table of contents Previous article Next article INFLUENTIAL PUBLICATIONFull AccessIndividual, Group, and Multifamily Cognitive-Behavioral TreatmentsBarbara L. Van Noppen, M.S.W., and Gail Steketee, Ph.D.Barbara L. Van NoppenSearch for more papers by this author, M.S.W., and Gail SteketeeSearch for more papers by this author, Ph.D.Published Online:1 Jul 2004https://doi.org/10.1176/foc.2.3.475AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack Citations ShareShare onFacebookTwitterLinked InEmail Since the early 1980s, the psychosocial treatment of choice for obsessive-compulsive disorder (OCD) has been exposure for obsessions and prevention of rituals, or “response prevention,” for compulsions conducted mainly in an individual format. This treatment method, based on clinical observations that obsessions increase anxiety and compulsions reduce it, has now been incorporated into the diagnostic criteria for OCD. That is, obsessions “cause marked anxiety or distress” and compulsions “are aimed at preventing or reducing distress” provoked by obsessions (American Psychiatric Association 1994). Not surprisingly, behavioral treatment based on this model includes procedures to reduce anxiety associated with obsessions and to prevent or curtail ritualistic behavior. In this chapter we briefly describe the theoretical model for exposure and response prevention (ERP) and then review the empirical literature supporting the efficacy of this method. In addition, cognitive conceptualizations and interventions have gained considerable recent attention and the limited literature on this method will also be reviewed. Promising alternative treatment strategies to deliver ERP, group, and multifamily formats offer added advantages that may be particularly beneficial to some patients, so we present literature relevant to group and multifamily behavioral treatment as well.Behavioral modelsFoa and Tillmanns (1980) articulated a definition of OCD based on the functional relationship between obsessions and compulsions—that is, the thoughts, images, impulses, or actions that generate obsessive anxiety may be prompted by external (environmental) or internal (thoughts, images) triggers for fear. Obsessive fears may be accompanied by fears of potential disaster (e.g., disease, death, going to hell), or they may occur without fears of catastrophic consequences. Most sufferers try to avoid the feared situation or stimuli (passive, phobic-like avoidance), but when this is difficult or impossible, they usually perform overt rituals or covert mental events to restore safety or prevent harm (Rachman 1976). Both behavioral and mental rituals are functionally equivalent in that both are intended to reduce obsessive fear (Rachman 1976; Rachman and Hodgson 1980).Why obsessions become highly anxiety provoking in the first place is the subject of some debate that remains unresolved. Possible etiologic models include parental teachings and modeling, biologic sources, cultural factors, historical experiences, religious teachings, cognitive beliefs and appraisals, and many other variables. ERP treatment, however, is based on the assumption that thoughts and behaviors are learned responses that have become conditioned and generalized to various contexts despite their seeming irrationality.Behavioral theorists (e.g., Dollard and Miller 1950; Mowrer 1960) have proposed a two-stage theory of acquisition of fear in which individuals first associate fear or other emotional discomfort with particular situations for various reasons and then find that escaping from or avoiding those contexts reduces discomfort. Because most patients cannot easily avoid many fear-provoking situations (e.g., use of toilets or stoves, perverse religious ideas), they develop ritualistic behaviors such as washing, checking, or praying to prevent or reduce discomfort, even if only minimally or briefly. Such actions are reinforced and repeated precisely because they reduce discomfort. Supporting this hypothesis is substantial evidence from early studies of OCD that obsessions increase both subjective and physiologic anxiety or discomfort and that compulsions reduce it (e.g., Boulougouris et al. 1977; Hodgson and Rachman 1972; Hornsveld et al. 1979; Rabavilas and Boulougouris 1974; Roper et al. 1973). The treatments that logically derive from this learning theory model are exposure to foster habituation of obsessive fears and blocking of rituals to prevent escape and avoidance.Assessment of symptomsBefore beginning behavioral (or cognitive) treatment, it is important for the clinician to gain a full picture of the OCD symptoms and their function for the patient. A complete assessment of symptoms consists of interview data (from the patient and, if possible, from family members or close others), clinician assessment of symptom types and severity, and standardized self-report measures.In an initial evaluation interview, preparatory to conducting a behavioral treatment, clinicians should assess obsessions and compulsions separately, along with mood state and general functioning. Rating scales include the Yale-Brown Obsessive Compulsive Scale (Y-BOCS; Goodman et al. 1989a, 1989b), used either as a rater-administered measure (Woody et al. 1995) or as a self-report (Steketee et al. 1996). Other necessary self-report instruments include a comprehensive list of feared and avoided situations as well as internal thought images and impulses.Outstanding among standardized clinician ratings of severity is the Y-BOCS, favored because of its detailed assessment of the severity of idiographic symptoms without regard to particular types of obsessions or compulsions. This measure begins with the Y-BOCS Symptom Checklist to determine which obsessions and compulsions occur most frequently for a particular patient and to identify other, less frequent or disturbing OCD symptoms that may prove problematic during behavioral treatment. After administering the Symptom Checklist, the clinician uses the Y-BOCS scale to assess five aspects each of obsessions and compulsions: time spent, distress, interference, resistance, and control. Scores range from 0 to 40. Scores of 16 and above are considered to be in the clinically significant range, with scores above 28 in the severe to extreme range. Positive evidence for the reliability and validity of the Y-BOCS is available in recent studies (e.g., Woody et al. 1995). A self-report version of this measure has demonstrated good reliability and validity in studies of clinical and nonclinical samples (Steketee et al. 1996). However, for patients with very poor insight into the irrationality of their symptoms, the clinician-rated Y-BOCS may be more valid.Observational measures of the frequency and duration of ritualistic behavior are recommended because these bring the clinician closest to an understanding of the impact and role of obsessions and compulsions in the patient’s everyday life. Self-reported minutes spent on compulsive activity have been collected in some studies to provide an independent assessment of symptom severity before and after therapy (Emmelkamp and van Kraanen 1977; Foa et al. 1980b), although the reliability and validity of such measures have not been established.Behavioral treatmentVariants of exposure and blocking procedures have been used very successfully for patients with OCD who have covert and overt rituals. Direct exposure techniques require the patient to directly confront fearful or disturbing ideas or situations and may be accompanied by exposure in imagery to feared catastrophic outcomes. Response prevention or blocking methods halt the patient’s ritualistic behaviors. To block “mental compulsions,” strategies such as thought stopping or distraction can be useful. These treatment strategies follow from the conceptualization of OCD described earlier; procedures that reduce anxiety (e.g., exposure) are applied specifically to anxiety-provoking obsessional content, whereas blocking strategies (e.g., response prevention) are used to prevent cognitive and behavioral rituals, thus allowing for habituation of obsessive fears. Both interventions are necessary for patients with obsessions accompanied by compulsions, as discussed later (Steketee 1993b; Steketee and White 1990).The combining of exposure for obsessions with response prevention for compulsions was first employed by Meyer in 1966 in patients with washing and cleaning rituals. In this program, compulsions were prevented while the patient was required to repeatedly touch objects that evoked anxiety about “contamination” and consequent urges to wash (Meyer and Levy 1973; Meyer et al. 1974). Studies using variants of direct exposure techniques were compared with relaxation training, both in conjunction with response prevention. Of the patients treated with ERP, 75% were improved or much improved after 15 sessions and maintained their gains after 2 years; by contrast, relaxation training had no effect (Marks et al. 1975). In two studies from Greece, an average of 11 sessions of in vivo and imaginal exposure plus response prevention produced good results in 85% of patients (Boulougouris and Bassiakos 1973; Rabavilas et al. 1976), but a long-term follow-up indicated that only 60% were still improved (Boulougouris 1977). Several studies by investigators in the Netherlands used 10–15 sessions of in vivo exposure and blocking of compulsions. Overall, about 70%–80% of a large sample of patients with OCD improved and most remained so at follow-up, although some patients required additional treatment sessions (Boersma et al. 1976; Emmelkamp and van Kraanen 1977; Emmelkamp et al. 1980; Hoogduin and Duivenvoorden 1988).In the initial studies by Foa and Goldstein (1978) in the United States, after 10 sessions of daily imagined and in vivo exposure treatment, 85% of patients were nearly symptom-free on rituals and only one patient failed to show improvement on obsessions, but fewer (57%) were asymptomatic. At follow-up, approximately 15% of patients relapsed. These findings indicated that treatment was somewhat more effective with compulsions than with obsessions, results that have held up in subsequent studies of exposure therapy. Later studies showed very positive gains with 15 sessions of treatment (Foa et al. 1992); most treatment regimens generally provide between 12 and 15 sessions of ERP.Although “pure obsessionals” (OCD patients without overt rituals) have traditionally been considered more difficult to treat with ERP, studies suggest that careful application of this behavioral treatment may result in substantial improvement for some patients. For instance, Hoogduin et al. (1987) treated 26 obsessive patients with a systematic program of deliberate evoking of obsessional thoughts (exposure) combined with strategies for refraining from neutralizing thoughts and cognitive rituals (response prevention). Nineteen subjects (73%) showed improvement of greater than 30%, and 61% of these maintained their gains at a 1-year follow-up. Salkovskis and Westbrook (1989) outlined some helpful approaches to invoking an ERP treatment paradigm with pure obsessionals, including the use of tape-recorded obsessional thoughts to allow for deliberate exposure.To date, prolonged ERP has been used to treat hundreds of patients with OCD, with most data derived from group studies. The remarkable convergence of results from studies conducted in many centers attests to the generalizability of the treatment effects. It is not surprising that, at present, ERP is considered the psychologic treatment of choice for OCD.Although the basic components of ERP have been well established, further work has been done to explore the relative importance of the various components, specifically, the relative need for exposure as well as response prevention, the required duration of the exposure, and the need for a therapist to model the behavior. Meyer’s original treatment consisted of two basic components: exposure to discomfort-evoking stimuli and prevention of ritualistic responses. Theoretically, exposure should be necessary to reduce anxiety associated with obsessions, and ritualistic behavior should be blocked because it terminates confrontation with the fearful stimuli, thus preventing extinction of anxiety. The research data support these assumptions. However, both in case studies and in controlled comparisons, obsessive anxiety declined more after prolonged exposure rather than after blocking of rituals, and compulsions were reduced mainly by response prevention but not by exposure (Foa et al. 1980a, 1984; Mills et al. 1973; Turner et al. 1980). Thus, not surprisingly, combined treatment led to the best results. From a clinical standpoint, therapists should gradually expose patients to situations that provoke obsessions while at the same time preventing the rituals that usually occur in these circumstances.According to clinical studies of patients with OCD, prolonged exposure to fear-provoking stimuli is superior to brief exposure: 80 minutes of continuous direct in vivo exposure proved superior to eight 10-minute segments (Rabavilas et al. 1976). Surprisingly, however, duration of the imagined exposure did not affect outcome. How quickly the therapist moves up the hierarchy of disturbing stimuli has not proved to be important in the treatment of OCD. Hodgson et al. (1972) exposed some patients gradually and others immediately to the most feared situation. The two procedures were equally effective, although patients reported feeling more comfortable with the gradual approach. We suspect from clinical experience that progressing too slowly will be unhelpful for most patients whose motivation and sense of accomplishment may wane. In general, then, clinicians are advised to extend patients’ exposure experiences in the office and at home as long as feasible and to encourage them to confront their fears as rapidly as they can tolerate.A combination of response prevention and participant modeling, in which the patient copied the therapist’s demonstration of exposure, yielded better results than passive modeling in which the patient only observed the therapist (Roper et al. 1975). However, other investigators found that adding modeling did not improve outcome (Boersma et al. 1976; Rachman et al. 1973). Nonetheless, some patients have reported that observing the therapist helped them overcome their resistance and avoidance of exposure. How the therapist models or conducts exposure may influence patients’ willingness to continue in treatment. Marks et al. (1975) proposed that ERP treatment requires a good therapeutic relationship and often a sense of humor. The very limited research on the qualities of a good therapist for OCD indicated that therapists who were respectful, understanding, interested, encouraging, challenging, and explicit were able to help patients achieve greater gains than those who gratified dependency needs or were permissive or tolerant (Rabavilas et al. 1979). In practice, a combination of support, encouragement, humor, and firm insistence that the patient follow therapeutic instructions for ERP seems to be optimal.Although the personal style of the therapist may be important, his or her presence during exposure may not be required, at least in some cases. Emmelkamp and van Kraanen (1977) found no differences in outcome for self-controlled versus therapist-controlled exposure, although subjects in the therapist-led group required more treatment sessions at follow-up than did the other group. The authors suggested that the self-controlled exposure patients may have gained greater independence in handling their fears. Consistent with this earlier study, the addition of therapist-aided exposure after 8 weeks of self-exposure instructions yielded only transient benefits that were lost at week 23 (Marks et al. 1988). Preliminary trials of computer-aided exposure suggest that such treatment may prove very useful for selected individuals with OCD (Greist 1996; Griest et al. 1996). The findings of these studies do not suggest that therapists are dispensable but do indicate that direct exposure may be implemented without their immediate presence. Whether this is especially true for patients with mild to moderate (rather than severe) symptoms remains to be tested.In conclusion, it seems that both exposure and blocking of mental and overt rituals are needed for successful outcome. Imagined treatment may be especially useful when fears of disasters are prominent features of a patient’s OCD symptoms. From a clinical standpoint, research suggests that therapists may begin treatment by conducting prolonged exposure in office and then assigning more exposure as homework between sessions. Only if the patient has serious difficulty with homework should the therapist insist on being present through the process. Most patients are likely to prefer graduated exposure, but some circumstances may require more rapid confrontation. Modeling may be used whenever patients feel it would be useful.Cognitive modelsIt is apparent from the phenomenology and characteristics of OCD that patients with this disorder exhibit some disturbances in cognitive functioning. Accordingly, several cognitive models for OCD have been proposed, many of which emphasize similar features of the disorder (see Steketee et al. 1998 and summary below). Pitman’s (1987) cybernetic model suggested that faulty beliefs and pathologic symptoms of OCD stem from signals experienced internally, such that a perceptual mismatch is registered in the perception of the input. This faulty perception leads to pervasive uncertainty and ritualistic efforts to correct it, along with difficulty withdrawing attention from intrusive thoughts. Pitman proposed neuroanatomic underpinnings for these processes.From a more traditional cognitive perspective, Warren and Zgourides (1991) emphasized the role of irrational beliefs in a rational-emotive treatment (RET) model of OCD. They hypothesized that biologic vulnerability influenced by developmental and learning experiences determined which thoughts a person considers unacceptable and what meaning he or she attaches to the thoughts. Common irrational thoughts include assumptions about the need to make correct decisions, the need to be perfectly certain to avoid causing harm, and the unacceptableness of bizarre thoughts and impulses. According to the RET model, under stress, negative emotions tend to provoke such intrusive thoughts. Thereafter, attention narrows on these thoughts, with accompanying hypervigilance and efforts to avoid or escape them (see also Wegner 1989).Salkovskis (1985) and Rachman (1993) formulated cognitive models focused on the salience of common intrusive thoughts associated with negative automatic thoughts. Discomfort arises from mistaken assumptions about responsibility for endangering oneself or others, leading to self-blame and precautions to avoid guilt, shame, and depression. Neutralization (mental and behavioral rituals) serves to reduce discomfort, responsibility, and the possible consequences of having the thought. Freeston et al. (1996) broadened this formulation to include additional types of faulty appraisals, including overestimation of the consequences of thoughts and of anxiety, the presumption that thinking can lead directly to doing an act (thought–action fusion), and perfectionism and the need for control.Additional hypothesized cognitive distortions include overestimation of threat or harm (Carr 1974; McFall and Wollersheim 1979), problems with epistemologic reasoning associated with safety (Kozak et al. 1987), a need for certainty (Beech and Liddell 1974), ideas that one must be perfectly competent and that failure to do so should be punished (Guidano and Liotti 1983; McFall and Wollersheim 1979), feelings of loss of control of thoughts (Clark and Purdon 1993) and consequent efforts at suppression (Wegner 1989), and underestimates of coping capacity (Carr 1974; Foa and Kozak 1986; Guidano and Liotti 1983). Experimental findings have supported some of the above assertions, particularly with respect to overspecification, the need for certainty (Makhlouf-Norris and Norris 1972; Makhlouf-Norris et al. 1970; Milner et al. 1971; Persons and Foa 1984; Reed 1985; Volans 1976), and excessive responsibility (e.g., Lopatka and Rachman 1995; Rheaume et al. 1995; Salkovskis 1989). Evidence is now accumulating to substantiate several aspects of these theoretical ideas, but it will undoubtedly be some time before the relationship among these concepts and their importance for effective treatment is clearly articulated.Cognitive treatmentTo date, only a handful of studies, most of them uncontrolled, have attempted to determine whether treatments derived from cognitive models are fruitful for OCD. In an earlier study, cognitive methods proved minimally helpful in reducing OCD symptoms (Emmelkamp et al. 1980). In contrast to these disappointing findings, a study of RET compared with self-controlled ERP showed that both treatments improved OCD symptoms equally (Emmelkamp and Beens 1991; Emmelkamp et al. 1988).However, the above-mentioned cognitive therapies did not appear to be designed specifically for cognitive distortions typical of patients with OCD. If certain cognitions (e.g., excessive responsibility, overestimation of harm, need for control) are particularly germane to OCD, cognitive treatment focused on these patient-specific thoughts and beliefs may be even more effective (Beck and Emery 1985). Several case studies demonstrated good effects of a traditional Beckian cognitive therapy tailored specifically for participating OCD patients (e.g., Ladouceur et al. 1993; Salkovskis and Warwick 1986; Van Noppen et al. 1995). Treatment included socratic dialogue and the triple column technique, which consisted of listing thoughts/beliefs, and rating the strength of conviction in the belief, associated emotions, and possible alternative beliefs. Experimental testing of beliefs and other cognitive strategies were intended to dispute various OCD-associated distorted beliefs. This cognitive treatment was highly successful, reducing Y-BOCS scores by 11 points after treatment and 12 points at 6-month follow-up. This outcome matched the effects of ERP (Van Noppen et al. 1995). These findings suggest that cognitive therapy is a promising adjunctive or, perhaps, alternative treatment to ERP.Many clinicians providing behavioral treatment informally incorporate cognitive techniques into the therapy. Psychoeducation and the labeling of OCD symptoms are standard in our treatment and likely to alter cognitive misinterpretations. We also encourage patients to separate affect (“I feel as though I have to wash”) from distorted perceptions, assumptions, and beliefs (“I have to wash or I’ll get AIDS”) and to challenge faulty assumptions about harm, perceived responsibility, and unacceptableness of bizarre thoughts and impulses. A group context (see below) is particularly suitable for providing a normative consensus to test beliefs and rehearse alternative ways of thinking. The case example below illustrates ERP with an individual patient.Case exampleRick, a 40-year-old computer systems support analyst, sought individual behavioral treatment after 10 months of pharmacologic treatment with limited benefits. He was married, the father of an 18-month-old son, and of Italian-Catholic descent. Although he was raised in a devout family, he described himself as not religious and “liberal” in his political views.Upon initial evaluation, Rick spoke about “disturbing” thoughts that interfered with his ability to enjoy his wife, Susan, and their son, Nate. Completion of the Y-BOCS Symptom Checklist revealed primary aggressive obsessions, a need to know, considerable avoidance, reassurance seeking, and mental rituals. Rick described “worrying that I might have the ability to impulsively hurt my son or wife....Suppose I just do it for no reason at all.” He was very bothered by the constant distraction of these “horrible” thoughts, which arose when he was in the company of his wife and son. Rick dreaded the days he had to drive Nate home from day care and be with him alone at home; when his wife was home he felt reassured that his “impulses” might stay in check. Although Rick had no previous history of difficulty with loss of impulse control or aggressive outbursts, he was worried that “what if, one day, I might just lose control and do something awful?” When questioned about insight, Rick wavered, acknowledging that his fears and behaviors were unreasonable but uncertain whether he had reason to be concerned. Rick’s initial Y-BOCS score was 25, reflecting moderate severity of symptoms, 3–8 hours a day of obsessions and compulsions, and little sense of control over the OCD.Rick described mild childhood obsessive-compulsive symptoms that included an excessive need for reassurance, a fear of “germs,” and some body dysmorphic symptoms (e.g., a preoccupation with his appearance, concern that he was “ugly,” checking in mirrors). In his late teens and early 20s, Rick’s fear of germs became more predominant, and on his own he used confrontation to help his fear “fade away.” He sought psychiatric treatment at age 19 because he was having difficulty with social relationships at college and was feeling insecure and inadequate. Aggressive obsessions began to emerge. In his psychodynamic approach, he examined his relationships in his family of origin and intrapsychic conflicts. After 10 years of weekly psychotherapy and pharmacotherapy, Rick felt more “normal” socially and was able to have some meaningful relationships, graduate from college, secure a job, and live on his own. When he first presented for treatment, he knew he had OCD but spoke about the aggressive thoughts as though they were reflective of suppressed anger.An individual behavioral treatment was outlined for Rick, who agreed to a protocol of two 90-minute information-gathering sessions, 12 weekly 2-hour treatment sessions, and six monthly check-in sessions. He decided to be maintained on a stable dose of a selective serotonin reuptake inhibitor (SSRI) throughout the therapy. During the first two information-gathering sessions, a more detailed history was taken; the intrapersonal behavior therapy was described, including a definition of ERP; and a more detailed description of Rick’s OCD was elicited. Rick seemed motivated, engaged, and eager to get started. The therapist instructed him to read When Once Is Not Enough (Steketee and White 1990) during the first few weeks. He came to the first session with his exposure homework hierarchy folded into a tiny square. Rick stated that he was so ashamed of his thoughts and fearful that “if anyone knew” just what he thought, they might think he was capable of doing these terrible things and “put him away.” He said he had never disclosed the exact content of his thoughts before and felt anxious to do so. Worried that someone would find the paper, he had put it in an envelope, placed the envelope into a jar, and hid the jar inside a bag of fertilizer, which he had then put into the trunk of his car and covered with a blanket. Thus, the very process of articulating the internal cues/triggers required Rick to expose himself to the obsessive thoughts and feared catastrophic consequences—in this case, that if he told someone about his obsessions, he would require hospitalization and possibly face divorce and loss of custody.Rick’s therapy was based on his exposure hierarchy, which was constructed around his fears of harming his wife and son. Table 1 lists obsessive thoughts and situations and the subjective discomfort they provoked on a scale of 0–100. Similarly, a hierarchy of situations that Rick avoided or endured with anxiety was constructed; this is shown in Table 2.Treatment, which consisted of ERP in vivo, imagined exposure in vivo, homework ERP assignments, and self-monitoring, proceeded based on these hierarchies. The first few sessions also contained psychoeducation on OCD, reading assignments (finish reading When Once Is Not Enough), and the viewing of a videotaped discussion by Michele T. Pato, M.D., on the neurobiology of OCD.Because of the nature of Rick’s aggressive obsessions, most of the in vivo therapy involved the use of scripted imagery that he read aloud, audiotaped, and replayed. He was given instructions to write the scripts in the first person and to be as descriptive and detailed as possible, as though the obsessive idea or image were happening. Whenever possible, the therapist encouraged Rick to bring “props” to the sessions to heighten his discomfort. For example, to confront the fear of smashing his son in the head with a hammer, Rick brought in a hammer and pictures of his son. The exposure task was for Rick to look at the pictures while swinging the hammer toward the photos and saying “I will smash Nate in the head with a hammer.” Initially, Rick said “I’m afraid I will hit my son in the head,” but the therapist reminded him of the scripting instructions, adding that to make the technique effective he needed to confront the exact fear. As the behavioral treatment continued, Rick combined in vivo and imaginal exposure using a bat, scissors, and knives (first small then large). Rick found the in vivo practice, coupled with the exposure homework, to be highly effective in reducing his anxiety. The day Rick entered the session with a baseball bat hidden under his coat, he laughed and joked with the therapist about becoming a “bat killer.”The therapist used modeling, often participating in the exposure challenge when Rick expressed difficulty getting started. For instance, the therapist took out a picture of her daughter, jabbed a knife at the photo, and said “I will stab Jill!” Observing this, Rick asked with puzzlement, “Doesn’t th