ABSTRACT

BACKGROUND: Age-related macular degeneration (ARMD) is the leading cause of blindness in the United States among people aged 65 years or older. Observational and experimental data suggest that antioxidant or zinc supplements may delay progression of ARMD and visual loss.

POPULATION STUDIED: Eleven retinal specialty clinics enrolled participants aged 55 to 80 years in 4 ARMD categories determined by the size and extent of drusen and retinal pigment epithelial abnormalities in each eye, the presence of advanced ARMD (each determined by evaluation of color photographs at a reading center), and visual acuity. Persons in category 1 had no ARMD; those in category 2 had mild or borderline ARMD; those in category 3 had moderate ARMD; and those in category 4 had advanced ARMD. At least 1 eye had a best corrected visual acuity of 20/32 or better (the study eye). Among participants, 56% were women, 96% were white, and the median age was 69 years. Potential participants were excluded for illness or disorders (history of cancer with a poor 7-year prognosis, major cardiovascular or cerebrovascular event within the previous year, or hemochromatosis) that would have made long-term follow-up or compliance with the study protocol unlikely or difficult.

STUDY DESIGN AND VALIDITY: This was a randomized, double-masked, placebo-controlled trial (concealed allocation assignment). Participants were assigned to 1 of 4 treatment groups: (1) antioxidants (500 mg vitamin C, 400 IU vitamin E, 15 mg beta carotene); (2) 80 mg zinc as zinc oxide and copper, 2 mg as cupric oxide; (3) antioxidants plus zinc; or (4) placebo. The groups did not differ in their baseline characteristics. Average follow-up was 6.3 years, with 2.4% lost to follow-up. Analysis was by intention to treat. The judicial assessors of outcomes were masked to treatment group assignment.

OUTCOMES MEASURED: Two primary outcomes were defined for study eyes in the ARMD trial: (1) progression to advanced ARMD and (2) at least a 15-letter decrease in visual acuity score.

RESULTS: Patients with no ARMD (category 1) and mild or borderline ARMD (category 2) did not benefit from antioxidant and/or zinc supplementation. However, participants in the moderate and advanced ARMD groups (categories 3 and 4) had a lower risk of progression to advanced ARMD and visual acuity loss in the good eye if they took both zinc and antioxidants compared with placebo for 7 years (35.7% vs 26.7%, respectively; P < .001; number needed to treat = 11).

ABSTRACT

BACKGROUND: Age-related macular degeneration (ARMD) is the leading cause of blindness in the United States among people aged 65 years or older. Observational and experimental data suggest that antioxidant or zinc supplements may delay progression of ARMD and visual loss.

POPULATION STUDIED: Eleven retinal specialty clinics enrolled participants aged 55 to 80 years in 4 ARMD categories determined by the size and extent of drusen and retinal pigment epithelial abnormalities in each eye, the presence of advanced ARMD (each determined by evaluation of color photographs at a reading center), and visual acuity. Persons in category 1 had no ARMD; those in category 2 had mild or borderline ARMD; those in category 3 had moderate ARMD; and those in category 4 had advanced ARMD. At least 1 eye had a best corrected visual acuity of 20/32 or better (the study eye). Among participants, 56% were women, 96% were white, and the median age was 69 years. Potential participants were excluded for illness or disorders (history of cancer with a poor 7-year prognosis, major cardiovascular or cerebrovascular event within the previous year, or hemochromatosis) that would have made long-term follow-up or compliance with the study protocol unlikely or difficult.

STUDY DESIGN AND VALIDITY: This was a randomized, double-masked, placebo-controlled trial (concealed allocation assignment). Participants were assigned to 1 of 4 treatment groups: (1) antioxidants (500 mg vitamin C, 400 IU vitamin E, 15 mg beta carotene); (2) 80 mg zinc as zinc oxide and copper, 2 mg as cupric oxide; (3) antioxidants plus zinc; or (4) placebo. The groups did not differ in their baseline characteristics. Average follow-up was 6.3 years, with 2.4% lost to follow-up. Analysis was by intention to treat. The judicial assessors of outcomes were masked to treatment group assignment.

OUTCOMES MEASURED: Two primary outcomes were defined for study eyes in the ARMD trial: (1) progression to advanced ARMD and (2) at least a 15-letter decrease in visual acuity score.

RESULTS: Patients with no ARMD (category 1) and mild or borderline ARMD (category 2) did not benefit from antioxidant and/or zinc supplementation. However, participants in the moderate and advanced ARMD groups (categories 3 and 4) had a lower risk of progression to advanced ARMD and visual acuity loss in the good eye if they took both zinc and antioxidants compared with placebo for 7 years (35.7% vs 26.7%, respectively; P < .001; number needed to treat = 11).

ABSTRACT

BACKGROUND: Age-related macular degeneration (ARMD) is the leading cause of blindness in the United States among people aged 65 years or older. Observational and experimental data suggest that antioxidant or zinc supplements may delay progression of ARMD and visual loss.

POPULATION STUDIED: Eleven retinal specialty clinics enrolled participants aged 55 to 80 years in 4 ARMD categories determined by the size and extent of drusen and retinal pigment epithelial abnormalities in each eye, the presence of advanced ARMD (each determined by evaluation of color photographs at a reading center), and visual acuity. Persons in category 1 had no ARMD; those in category 2 had mild or borderline ARMD; those in category 3 had moderate ARMD; and those in category 4 had advanced ARMD. At least 1 eye had a best corrected visual acuity of 20/32 or better (the study eye). Among participants, 56% were women, 96% were white, and the median age was 69 years. Potential participants were excluded for illness or disorders (history of cancer with a poor 7-year prognosis, major cardiovascular or cerebrovascular event within the previous year, or hemochromatosis) that would have made long-term follow-up or compliance with the study protocol unlikely or difficult.

STUDY DESIGN AND VALIDITY: This was a randomized, double-masked, placebo-controlled trial (concealed allocation assignment). Participants were assigned to 1 of 4 treatment groups: (1) antioxidants (500 mg vitamin C, 400 IU vitamin E, 15 mg beta carotene); (2) 80 mg zinc as zinc oxide and copper, 2 mg as cupric oxide; (3) antioxidants plus zinc; or (4) placebo. The groups did not differ in their baseline characteristics. Average follow-up was 6.3 years, with 2.4% lost to follow-up. Analysis was by intention to treat. The judicial assessors of outcomes were masked to treatment group assignment.

OUTCOMES MEASURED: Two primary outcomes were defined for study eyes in the ARMD trial: (1) progression to advanced ARMD and (2) at least a 15-letter decrease in visual acuity score.

RESULTS: Patients with no ARMD (category 1) and mild or borderline ARMD (category 2) did not benefit from antioxidant and/or zinc supplementation. However, participants in the moderate and advanced ARMD groups (categories 3 and 4) had a lower risk of progression to advanced ARMD and visual acuity loss in the good eye if they took both zinc and antioxidants compared with placebo for 7 years (35.7% vs 26.7%, respectively; P < .001; number needed to treat = 11).

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.

When you suspect a patient has obsessive-compulsive disorder (OCD) (Box 1), how can you differentiate OCD from psychosis? Once you have made the diagnosis, what critical factors suggest treatment will be successful—or unsuccessful? Is behavioral therapy more effective than medication? Which medications are most likely to be effective? The answers to these questions can help you improve the well-being of your patients with OCD.

Differential diagnosis

Unfortunately, many patients with severe OCD are misdiagnosed with psychosis or schizophrenia spectrum disorder and spend many years suffering without proper treatment.³ Despite many similarities between patients with severe OCD and psychosis—including rigid belief systems, unrealistic concerns, magical thinking, and odd behaviors—patients with OCD can recognize the irrational nature of their beliefs when they are not in the throes of anxiety.

Patients with OCD also will usually respond to behavioral interventions within a few weeks while patients who are psychotic usually get progressively worse. Treatment must be given time as both cohorts will get anxious or increase their negative symptoms initially, but patients with OCD should soon habituate and find symptom relief.

Some patients have OCD with psychotic features and tend to have more difficulty responding to behavior therapy without medication. Patients with both OCD and schizotypal personalities respond poorly to both behavior therapy and psychotropic medications.⁴

Box 1

Predictors for successful treatment

Insight Researchers³ found that about 52% of patients with insight into the reasonableness of their obsessions responded to medications, while none who lacked insight responded. Therefore, it pays to assess patients’ insight and ability to recognize the long-term consequences of OCD to themselves and those around them.

Some patients who have suffered with treatment-refractory OCD for most of their lives lack a premorbid high level of functioning to serve as a reference for normalized behavior. Educating these patients to see the advantages of living without certain negative behaviors improves their receptivity to treatment.

Patients who lack insight often refuse to acknowledge that many of their behaviors are manifestations of OCD. Such patients, however, are usually more amenable to giving up or modifying their dysfunctional behaviors—and the clinician more likely to avoid confrontations—if they are shown how certain behaviors undermine their goals.

Cost-benefit analysis Because of the aversive nature of exposure and response prevention therapy (ERP) and the negative side effects of many medications, some patients may find it easier to live with their symptoms, as painful as they are, rather than undergo the discomfort of behavior therapy. Because the prognosis is poor in such cases, patients need to be convinced that the discomfort of treatment is merely short-term, while the discomfort of the illness could last forever if left untreated.

Motivation In our experience, motivation has played a crucial role in determining treatment outcome for severe refractory OCD. And regardless of the severity of their symptoms, patients who are fed up with their symptoms, or are tired of living a life controlled by their obsessions, usually are excellent candidates for treatment.

Conversely, those who enter treatment as a result of external pressure from spouses or family face a less positive prognosis. High emotional expressiveness, overinvolvement, and hostility by relatives is related to higher attrition rates in treatment.³ Because ERP is so aversive, these patients will find ways to dilute the treatment’s effectiveness. In many cases, they do the minimal amount of work required to stay in treatment to avoid whatever consequences their families would impose for not adhering to treatment.

One marker to assess compliance is whether the clinician feels he or she is investing more time and effort into the patient’s treatment than the patient is. If so, this should be addressed in a timely manner. Also, sporadic attendance at sessions and noncompliance with medications, homework, and behavior therapy assignments may also portend a poor outcome. Remember, though, that noncompliance and lack of motivation are fluid states; many previously noncompliant patients later return to treatment better motivated and more compliant.

Predictors for a lower success rate

Secondary gain Researchers⁴ found that patients who were enabled by their families had more severe symptoms than those who were not. These relational and environmental factors should be discussed openly. If the patient finds that many of his or her life needs are being met secondary to the illness, that patient might not agree to an aversive treatment.

To overcome this, urge family members or other individuals who provide dysfunctional reinforcers to remove them from the environment. Meet with the patient and family/friends and frankly point out dysfunctional gains and the ways in which family members unknowingly allow the gains to continue (e.g., giving the patient more money after he or she overspent his or her allowance). A family behavioral contract should be devised to address how these gains will be reasonably eliminated.

Recognize, too, that a patient may find it difficult to give up the secondary gains, detrimental as they may be, without adequate skills or coping mechanisms to fill the void. So in some cases, it is best not to remove all the secondary gains at once; this can cause many patients to terminate treatment prematurely.

Trauma or abuse history Many patients with treatment refractory OCD have trauma histories and cannot habituate to the behavioral tasks because of dissociation, emotional numbing, or some form of distraction that mediates their anxiety and prevents proper habituation. If the patient is adequately complying with the exposures, yet still is unable to confront every feared stimulus, inquire about a trauma or abuse history (Box 2).

Substance abuse The stress that is inherent to ERP can cause many patients to relapse or abuse illicit substances to manage their anxiety. Therefore, patients with severe substance abuse problems often have great difficulty handling ERP, as they are asked to experience the very discomfort that initially caused them to abuse drugs and alcohol.

Box 2

In such cases, a patient cannot realistically be asked to give up a coping mechanism, faulty as it may be, until a more functional reinforcer takes its place. Hence, skills training is a crucial part of treatment for this group.

Residential treatment for OCD patients with comorbid substance abuse in remission may be necessary to ensure a positive outcome. Patients should continue recovery work concurrent to behavior therapy to prevent relapse.

High-risk OCD symptoms Patients who have more traditional OCD symptoms usually have a good prognosis. Unfortunately some symptoms do not respond to ERP treatment. These include:

• Repeating, hoarding, and symmetry. Though evidence suggests that hoarding is predictive of poor outcomes,⁵ treatment carried out in the home can be effective over a 24-week trial.⁶
• Incompletion, or the need for things to feel right.
• Rigid and overvalued belief systems.
• Sexual and religious obsessions. These appear to be more resistant to behavior therapy and selective serotonin reuptake inhibitors (SSRIs).⁷

More research needs to be conducted to offer patients with these symptoms better respite.

Researchers also found that patients with childhood and adolescent onset of symptoms, tics, history of hospitalization, and terminated treatment against medical advice are more likely than other OCD patients to develop more severe symptoms in adulthood.⁸ Patients with OCD who also suffer from generalized anxiety disorders are more likely than those without GAD to drop out of treatment.⁹

Behavior therapy: first choice

ERP is considered the premier treatment for OCD and is suitable for both adults and children.¹⁰ Exposure forces patients to confront their feared stimuli. Response prevention blocks patients from engaging in compulsions or avoidance behaviors intended to reduce their discomfort. Patients are asked to identify situations that trigger their obsession and compulsions and rank them along a fear hierarchy. Patients confront a moderately rated situation and, once they become habituated to it, move up the fear hierarchy to the next situation.

ERP has been proven effective for OCD not only as an individual behavior treatment, but also when done in a group setting¹¹ or when delivered online or by telephone.¹²

Table 1

Dosage levels for SRIs in OCD

Some clinicians prefer cognitive behavioral therapy (CBT) to ERP because it is less aversive. Researchers found that patients who were treated with either CBT or ERP improved. Patients treated with ERP, however, were more likely to maintain their gains in recovery 3 months after treatment concluded.¹³ Evidence suggests that ERP or CBT when implemented alone, or when applied in conjunction with fluvoxamine,¹⁴ are equally effective.

ERP should be managed only by clinicians specially trained in this modality. Several treatment centers across the country provide specialized care for OCD patients. For the nearest treatment center in your community that accepts referrals for ERP, contact the OC Foundation in North Branford, Conn. (See Related Resources.)

Medication for OCD: SRIs as first-line therapy

Experts agree that first-line somatic treatments for OCD include not only behavior therapy but also serotonin reuptake inhibitors (SRIs),¹⁵ that is, clomipramine or selective serotonin reuptake inhibitors, (SSRIs) (Table 1).

Caution: Many patients who “respond” to treatment in clinical studies remain symptomatic and meaningfully affected by their residual illness. Therefore, it is critical that you inform patients at the outset that 100% reduction in symptoms is rare.

SRIs Overwhelming evidence from multiple randomized, double-blind, placebo-controlled studies support the efficacy of SRIs. In adults, well-designed and controlled trials have demonstrated the relative efficacy of clomipramine, fluoxetine, sertraline, paroxetine, and fluvoxamine vs. placebo.

SRIs also have been shown to be significantly more effective than tricyclic antidepressants (TCAs) in both placebo-controlled and non-placebo-controlled studies.

Despite initial reports that clomipramine may be more effective than SSRIs, a growing number of studies and a recent comprehensive literature review suggest that the SRIs all have comparable efficacy.¹⁶ Because clomipramine has significantly more anticholinergic- and antiadrenergic-mediated side effects than the SSRIs, however, many clinicians choose SSRIs as the initial agent.

When using SRIs, remember that response is typically delayed; an adequate trial requires at least 10 weeks. Indeed, a meaningful proportion of responders continue to emerge past the 8-week mark. Experts suggest that optimal dosages of SRIs for OCD may exceed those typically used for major depression. Guidelines for SRI dosage ranges for OCD appear in (Table 1).

Data regarding treatment duration also suggest that discontinuation of SRIs results in a high relapse rate, though the use of lower maintenance dosages of SRIs is still debated.

Table 2

Ratings of SRI-augmenting agents for OCD treatment

Augmentation of SRIs When first-line interventions fail, second-line pharmacological approaches include augmentation of SRIs with additional medications (Table 2). Numerous agents have been tried for patients who were unresponsive or only partially responsive to SRIs alone.¹⁷ Few controlled trials of such strategies have been conducted, however. The most impressive data document the benefits of adding low doses of dopamine antagonists (both conventional and atypical neuroleptics).

Table 3

Using alternative monotherapies

Recent uncontrolled studies of augmentation with atypical neuroleptics have yielded encouraging preliminary results, as has one controlled trial of augmentation of an SRI with risperidone. Other data suggest that lithium, buspirone, and clonazepam may also be effective.

Numerous other agents have been tried in combination with SRIs, including clonidine, tryptophan, fenfluramine, pindolol, trazodone, nortriptyline, and other antidepressants. The small number of subjects, lack of sufficient controls, and mixed results preclude drawing even preliminary conclusions as to the potential efficacy of such strategies.

Alternative Monotherapies For patients who do not respond satisfactorily to trials of SRIs alone or to augmentation strategies, consider alternative monotherapies in place of SRIs (Table 3). In addition to uncontrolled data, positive controlled studies lend some support for trials of clonazepam, monoamine oxidase (MAO) inhibitors, and buspirone.

Pertinent negative findings are worthy of mention. In contrast to promising results with risperidone as an augmenter, an open trial of the atypical antipsychotic clozapine suggests inefficacy as a monotherapy. Several case reports suggest that clozapine can actually precipitate obsessive-compulsive symptoms in patients with psychotic disorders.¹⁸ Controlled trials have not demonstrated the efficacy of trazodone, clonidine, and diphenhydramine as monotherapies.

Pharmacotherapy + or vs. behavioral therapy

Only a few studies directly comparing behavior therapy vs. medication have been reported. In practice, the two are routinely used in concert. Experts have long recommended this treatment approach. Two recent studies^19,20 have demonstrated that the combination is more effective than either treatment alone.

In another study, behavior therapy significantly outperformed clomipramine; no significant incremental benefit was seen from the two treatments in combination.²¹ However, the dosages of clomipramine were relatively low (mean=164 mg/d and maximum=225 mg/d) and of inadequate duration (6 weeks). Still another older head-to-head comparison of behavior therapy and clomipramine showed that medication was better for reducing obsessional doubt, whereas behavior therapy more effectively reduced compulsive rituals.

Third-line treatments may include the unproven augmentation therapies described above, or intravenous clomipramine if available.²²

Treatments of last resort

Finally, other nonpharmacologic treatments, including neurosurgery and electroconvulsive therapy (ECT), have remained controversial and are reserved for particular clinical situations or as treatments of last resort.

Despite a large body of uncontrolled data reporting antiobsessional benefits from a variety of neurosurgical procedures, ethical considerations and technical limitations have precluded the performance of sham-controlled studies to definitively establish the efficacy of these strategies.

Neurosurgical treatment of OCD is reserved for patients with severe and debilitating illness who have failed an exhaustive array of other available treatment options and who provide informed consent or assent. Currently, the most commonly employed neurosurgical treatments for OCD include anterior cingulotomy, anterior capsulotomy, subcaudate tractotomy, and limbic leukotomy. In recent prospective trials of cingulotomy and capsulotomy, approximately 45% of patients experienced a 35% or more symptom reduction.

With the advent of innovative surgical devices that allow functional neurosurgery without craniotomy (e.g., by gamma knife), the performance of ethical, double-blind, sham-controlled trials of neurosurgery for OCD is now feasible. A team of investigators from Brown University and Massachusetts General Hospital is conducting one such study that tests the efficacy of anterior capsulotomy.

There are no controlled data regarding the efficacy of ECT for OCD. Given the high comorbidity of major affective illness in OCD and the well-established efficacy of ECT for major depression, it is not surprising that some patients with OCD have reportedly shown clinical improvement after ECT. Several limited case series and anecdotal reports suggest that ECT may help in some circumstances, and such intervention would seem prudent in some cases where severe, comorbid affective illness is present.²³

Related resources

• Jenike MA, Baer, L, Minichiello WE, eds. Obsessive Compulsive Disorders: Practical Management. 3rd ed. Boston: Mosby, 1998.
• Jenike MA. An update on obsessive-compulsive disorder. Bulletin of the Menninger Clinic. 2001;65:4-25.
• Obsessive-Compulsive Foundation, (203) 315-2190, www.ocfoundation.org

Drug brand names

• Buspirone • BuSpar, BuSpar DIVIDOSE
• Citalopram • Celexa
• Clomipramine • Anafranil
• Clonazepam • Ativar, Diastat, Halcion
• Clonidine • Catapres, Catapres TTS-1
• Clozapine • Clozaril
• Fenfluramine • Pondimin
• Fluoxetine • Prozac, Prozac Weekly
• Fluvoxamine • Luvox
• Paroxetine • Paxil
• Phenelzine • Nardil, Parnate
• Pindolol • Inderol, Corgard, Betaloc
• Risperidone • Risperidal
• Sertaline • Zoloft
• Trazodone • Desyrel
• Tryptophan* • L-Tryptophan, Alti-trytophan

Disclosure

Dr. Boxill and Ms. Shapiro report no affiliation or financial arrangement with any of the companies whose products are mentioned in this article.

Dr. Dougherty reports conflicts of interest with Pfizer Inc., Forest Pharmaceuticals, and Solvay Pharmaceuticals.