[Prev] [Next]




Obsessive-compulsive disorder (OCD) was, until recently, considered to be a rare disorder, but new data have revealed that it is in fact a common illness. "I know it's stupid. I feel like a crazy person, but I know that I'm not crazy!" Those are examples of statements from a typical patient with OCD who may spend hours a day washing his hands, although he recognizes that there is no real reason to do so. Some OCD patients check and recheck that a stove is turned off, that a door is locked, or that some disaster has not befallen their children. An internist with OCD may repeatedly call the laboratory to be absolutely certain that he heard the results correctly. Some patients do not have rituals, but they endure endless hours of intrusive obsessive thoughts. The disorder may be so severe that patients are unable to work. If untreated, patients may be disabled for life. Although a complete cure is extremely rare there are now treatments (behavior therapy and psychotropic medication) that result in considerable improvement for the majority of patients. Comorbidity of OCD with depressive disorders and other anxiety disorders is common and often complicates the diagnosis. Before beginning treatment, it is important for the clinician to understand the entire clinical picture.


The fourth edition of Diagnostic and Statistical Manual of Mental Disorders (DSM-IV) requires that a patient have either obsessions or compulsions that are a significant source of distress; are time-consuming; or interfere significantly with the person's normal routine, occupational functioning, or usual social activities or relationships. At some point during the course of the illness, the adult patient must recognize that the obsessions or compulsions are excessive or unreasonable. According to DSM-IV, obsessions are defined by the following features:

  1. Recurrent and persistent thoughts, impulses, or images that are experienced at some time during the disturbance as intrusive and inappropriate and that cause marked anxiety or distress.
  2. Thoughts, impulses, or images that are not simply excessive worries about real-life problems.
  3. Attempts to ignore or suppress such thoughts or impulses or to neutralize them with some other thought or action.
  4. Recognition that the obsessional thoughts, impulses, or images are a product of one's own mind, not imposed from without as in thought insertion.

Clinically, the most common obsessions are repetitive thoughts of violence (for example, killing one's child), contamination (for example, becoming infected by shaking hands), and doubt (for example, repeatedly wondering whether one has performed some act, such as having hurt someone in a traffic accident).

Compulsions are defined as follows:

  1. Repetitive behaviors that the person feels driven to perform in response to an obsession or according to rules that must be rigidly applied.
  2. Behaviors or mental acts aimed at preventing or reducing distress or preventing some dreaded event or situation. Those behaviors or mental acts are either unconnected realistically with what they are designed to neutralize or prevent, or clearly excessive.

Typical compulsions include handwashing, ordering, and checking. A significant change from DSM-III-R to DSM-IV is reflected in the addition of mental compulsions, such as praying, counting, and repeating words silently. In DSM-III-R, those were called obsessions, but because such repetitive mental actions generally serve to decrease anxiety, it was felt that they would be better characterized as mental compulsions. Obsessions are usually anxiety-provoking, whereas compulsions are usually anxiety-relieving (at least over the short term).

DSM-IV also notes that if another Axis I disorder is present, a diagnosis of OCD is appropriate only if the content of the obsessions or compulsions is not restricted to it (for example, preoccupation with food in the presence of an eating disorder, hair pulling in the presence of trichotillomania, concern with appearance in the presence of body dysmorphic disorder, preoccupation with drugs in the presence of a substance use disorder, preoccupation with having a serious illness in the presence of hypochondriasis, or guilty ruminations in the presence of major depressive disorder). Furthermore, to meet the diagnostic criteria for OCD, the symptoms must not be due to the direct effects of a substance, such as a drug of abuse or a medication, or to a general medical condition.

DSM-IV also allows the specification of poor insight type if, for most of the time during the current episode, the person does not recognize that the obsessions and compulsions are excessive or unreasonable. That uncommon subtype has been referred to in the past psychiatric literature as "OCD psychotic" or "schizo-obsessive" and has generally been considered to have a poor prognosis.


Centuries ago, persons with obsessive blasphemous or sexual thoughts were considered to be possessed. That religious view of obsessions

was consistent with the contemporary world view, and the logical treatment was designed to expel evil from the unfortunate soul who was possessed. Exorcism was the treatment of choice, with the person being subjected to torture in an effort to drive out the intruding entity. Surprisingly, those treatments were occasionally successful. Obsessions and hand-washing rituals resulting from guilt were immortalized in the 17th century in Shakespeare's character, Lady Macbeth.

With time, the explanation of obsessions and compulsions moved from a religious view to a medical one. OCD was first described in the psychiatric literature by Jean Etienne Dominique Esquirol in 1838, and by the end of the 19th century, it was generally regarded as a manifestation of melancholy or depression.

By the beginning of the 20th century, theories of obsessive-compulsive neurosis shifted towards psychological explanations. Pierre Janet reported successful treatment of rituals with behavioral techniques; but with Sigmund Freud's writings on psychoanalysis of the Rat Man, OCD came to be conceptualized as resulting from unconscious conflicts and from the isolation of thoughts and behaviors from their emotional antecedents. As a result of those theories, treatment of OCD turned from attempts to modify the obsessional symptoms themselves toward the resolution of the unconscious conflicts presumed to underlie the symptoms. With the rise of behavior therapy in the 1950s, learning theories which had proved useful in dealing with phobias were applied to OCD, and, although they clearly did not account for all OCD phenomenology, they led to the development of the powerful techniques of exposure and response prevention for reducing compulsive rituals.

Over the last few years research on the biology of OCD has accelerated, with ongoing studies of pharmacological agents, neurosurgical treatments, brain imaging, genetics, neuropsychological dysfunction, and the association of OCD symptoms with Tourette's disorder and other possibly related illnesses, such as trichotillomania and body dysmorphic disorder. Theories of basal ganglia and frontal lobe dysfunction have been developed that lead to testable hypotheses about the underlying pathophysiology of OCD.



Even within the past decade, OCD was considered to be extremely rare (approximately 0.05 percent of the population). But more recent studies, including the Epidemiologic Catchment Area study, have demonstrated a six-month point prevalence of about 1.5 percent and a lifetime prevalence of 2 to 3 percent. That means that in the United States alone, between five and seven million people suffer from OCD. Recent pharmaceutical data indicate that far fewer than half of those patients are being treated.


The mean age of onset for OCD in one study was in the early 20s, with over half of the patients becoming symptomatic by age 25, and three quarters by age 30. Fewer than 5 percent of the patients had onset past age 40. Another study of 83 patients arrived at similar findings: 65 percent of the sample had onset prior to 25 years of age. Another group found that 27 OCD patients had a mean onset of 25.6 years. A more recent report found a mean age of onset in 44 OCD patients of 19.8 years. From those data, it is quite apparent that OCD usually begins in early adulthood.

In an effort to see if patients with different symptoms of OCD (for example, checking, obsessions only, mixed symptoms) had different ages of onset, the author recently studied 138 consecutively evaluated OCD patients. Those with obsessions only or cleaning rituals only had a mean age of onset of about 27, while patients with checking rituals only or mixed rituals (for example, washing and checking) had an earlier onset of about age 18 or 19. In agreement with previous studies, the author found a significantly earlier mean age of onset for men (20 years) than for women (25 years).



In rare cases, one can identify a brain insult, such as encephalitis or head injury, as an antecedent to OCD, but typically there is no identifiable neurological precipitant. With the advent of precise neuroimaging techniques, such as morphometric magnetic resonance imaging (mMRI) and positron emission tomography (PET), new ways to look at the brain are now available. PET scans have indicated abnormalities in the frontal lobes, cingulum, and basal ganglia of OCD patients when compared with depressed persons and normal controls. In addition, volumetric computed tomographic measurements have demonstrated decreased caudate volumes bilaterally in OCD patients compared with normal controls (although that was not confirmed by a later magnetic resonance imaging [MRI] study), and other investigators have reported occasional patients with demonstrable lesions in the striatum. MRI has also found longer mean T1 values for frontal white matter in OCD patients than in controls. Another study compared mMRI scans of the brains of 10 OCD patients to the scans of 10 normal controls matched for age, sex, and handedness and found that OCD patients have significantly more gray matter and less white matter, suggesting a developmental abnormality. The combination of findings from several high-technology imaging studies supports a neurological hypothesis for OCD.

Clinically, there is much overlap among patients with OCD, chronic motor tics, and Tourette's disorder, and a genetic relation among those disorders seems likely. Researchers find that about 20 percent of OCD patients exhibit tics.


There is evidence that serotonin-specific reuptake inhibitors (SSRIs) are partially effective treatments for OCD. In one study comparing clomipramine (Anafranil) to nortriptyline (Aventyl) and placebo, only clomipramine was significantly superior to placebo in reducing OCD symptoms. In addition, response to clomipramine was strongly correlated with lowering of cerebrospinal fluid (CSF) concentrations of 5-hydroxyindoleacetic acid (5-HIAA), a metabolite of serotonin, suggesting that changes in the serotonergic system had something to do with the good clinical outcome. In another OCD study, clomipramine was compared to a monoamine oxidase inhibitor (MAOI) (clorgyline) in a double-blind placebo-controlled crossover manner. Again, clomipramine significantly reduced OCD symptoms in comparison with both clorgyline and placebo, with improvement correlating with plasma concentrations of clomipramine. Other researchers found clomipramine to be clearly superior to desipramine (Norpramin) in the treatment of children and adolescents with OCD. In the majority of drug studies to date, although there has been significant comorbidity with major depressive disorder, the depressed subjects showed no difference in treatment response of OCD symptoms.

Open and double-blind trials of other SSRIs, such as fluvoxamine (Luvox), paroxetine (Paxil), sertraline (Zoloft), and fluoxetine (Prozac), have yielded comparable beneficial results. Because differences in efficacy among these agents are probably quite small, only large-scale trials would be likely to demonstrate that any one drug is superior to another. It can be concluded that several agents that selectively block serotonergic uptake diminish OCD symptoms, while pharmacologically similar agents without serotonergic selectivity are not nearly as effective.

Although research suggests that alteration of brain serotonergic systems may be one mechanism through which SSRIs have their therapeutic effects, there is no evidence of baseline serotonergic dysfunction in OCD patients. No significant difference in imipramine (Tofranil) binding or serotonergic uptake in platelets has been found, but CSF 5-HIAA has inconsistently been found to be elevated in OCD patients compared with normal controls, implying a higher rate of serotonin turnover.

One study found that the Bmax (estimate of receptor density in a tissue sample) determined by 3H-imipramine binding was significantly reduced in an OCD group compared with controls. If confirmed, that would suggest a lower density of 5-hydroxytryptamine (5-HT) receptors in OCD, a conclusion inconsistent with the hypothesis that treatment with SSRIs has its effect via down regulation of 5-HT receptors. Another study of childhood OCD subjects and age-matched normal controls found no differences in platelet 5-HT binding, monoamine oxidase activity, or plasma epinephrine, or plasma norepinephrine concentrations between the two groups.

Some data suggest that serotonergic perturbations can modify symptoms of OCD. M-chlorophenylpiperazine (mCPP), a putative 5-HT

agonist, has been shown to decrease brain 5-HIAA in rats when administered intraperitoneally, but the effects of mCPP on human subjects are problematic. A comparison of the effects of intravenously administered mCPP was performed on OCD subjects and normal controls. It did not exacerbate OCD symptoms. The effects of orally administered mCPP, studied in a double-blind placebo-controlled comparison between OCD subjects and normal controls, exacerbated OCD symptoms in OCD patients, who also became significantly more anxious, depressed, and dysphoric in comparison with controls. It is difficult to explain the disparity between the various findings. Possibilities include the differential route of administration (that is, oral versus intravenous) or some difference in the population of OCD subjects studied. A subsequent experiment studied the effects of oral mCPP versus placebo in nine OCD subjects, challenged in a double-blind crossover paradigm, before and during (after four months) clomipramine treatment. The baseline effects of mCPP replicated the previous findings (that is, worsening of OCD symptoms). However, after four months of clomipramine, mCPP no longer exacerbated OCD symptoms. The investigators regarded their findings as consistent with the hypothesis that clomipramine acts via down-regulation of 5-HT receptors.

Other Neurotransmitters

A number of studies support the hypothesis that the serotonergic system is not the only system involved in the pathophysiology of OCD. The clinical ineffectiveness of the potent serotonergic agent, zimelidine, and of the anxiolytic 5-HT1A partial agonist buspirone (BuSpar), for example, are difficult to explain within a strictly serotonergic model. It is possible that a balance of adrenergic and serotonergic action is necessary. A meta-analysis of four studies of potent serotonergic agents (fluvoxamine, sertraline, fluoxetine, and clomipramine) in obsessive-compulsive patients revealed that greater effect size (that is, improvement in OCD symptoms) was actually associated with less serotonergic selectivity. The results suggest that a comprehensive model of OCD must be based on a multiple neurotransmitter system.


Although psychoanalysis and psychodynamically oriented psychotherapy are not effective in the treatment of obsessions and compulsions, a number of interesting hypotheses are raised by theorists in that area. Many of the psychoanalytic theorists do not clearly distinguish obsessive-compulsive personality disorder from OCD and may see these disorders on a continuum. The psychoanalytic theories, described by John Nemiah and Thomas Uhde in the previous edition of this textbook are summarized below.

Nemiah and Uhde noted that, from a psychoanalytic perspective, three major psychological defensive mechanisms determine the form and quality of obsessive-compulsive symptoms and character traits: isolation, undoing, and reaction formation.


Isolation is a defense mechanism that protects an individual from anxiety-provoking affects and impulses. Under ordinary circumstances, an individual experiences in consciousness both the affect and the imagery of an emotion-laden idea, whether it be a fantasy or the memory of an event. When isolation occurs, the affect and the impulse from which it derives are separated from the ideational component and pushed out of consciousness. If isolation is completely successful, the impulse and its associated affect are totally repressed, and the patient is consciously aware of only the affectless idea that is related to it. Sometimes, however, the isolation is less effective, and the total quantity of energy accruing to the impulse and its associated affect cannot be completely restrained by the repressing forces from entering the patient's consciousness. Patients experience a partial awareness of the impulse without fully recognizing its meaning or significance. For example, they may have frightening and compelling murderous impulses toward strangers or casual acquaintances; here, the impulse makes itself felt as an urge to violent action, but the direction of the urge is displaced from the true object of the patients' aggression. At the same time, isolation makes patients unaware that they are angry, so that they are puzzled and disturbed by their compulsions. Alternatively, patients may be obsessed with images and thoughts of violence and destruction; here again, the energy from the partially repressed impulse gives the thoughts their compelling quality, and the continuing partial functioning of the mechanism of isolation prevents patients from becoming aware that beneath the surface they harbor intense aggression.


Nemiah and Uhde noted that, in the face of the impulse's constant threat to escape the primary defense of isolation, further defensive operations are required to combat the impulse and to quiet the anxiety aroused by its imminent eruption into consciousness. The anxiety-allaying function of compulsive acts can readily be noted in the clinical manifestations of OCD. The compulsive act is the manifestation of a defensive operation aimed at reducing anxiety and at controlling the underlying impulse that has not been sufficiently contained by isolation. A particularly important secondary defensive operation of that sort is the mechanism of undoing. As the word suggests, undoing refers to a compulsive act that is performed in an attempt to prevent or undo the consequences that the patient irrationally anticipates from a frightening obsessional thought or impulse.

Reaction formation

Both isolation and undoing are defensive maneuvers that are intimately involved in the production of clinical symptoms. Reaction formation, a third mechanism closely associated with OCD, according to Nemiah and Uhde, results in the formation of character traits rather than symptoms. As the term implies, reaction formation involves manifest patterns of behavior and consciously experienced attitudes that are exactly the opposite of the underlying impulses. Often these patterns appear to an observer to be highly exaggerated and sometimes quite inappropriate. Reaction formation is thought to be responsible for many of the personality traits characterized by control that make up some elements of obsessive-compulsive personality disorder.

Psychogenetic factors

Nemiah and Uhde noted that one of the striking features of patients with OCD is the degree to which they are preoccupied with aggression or dirt, either overtly in the content of their symptoms or in the associations that lie behind them. That and other observations have led to the psychodynamic proposition that the psychogenesis of OCD lies in disturbances in normal growth and development related to the anal-sadistic phase. According to that conceptualization, the impulses associated with the anal-sadistic phase are normally modified in the oedipal and succeeding stages of development. If that developmental process is disturbed, unmodified anal-sadistic impulses will persist as components of the individual's psychological makeup. Ordinarily, impulses of that type are controlled and disguised by character traits and do not significantly affect the person's day-to-day functioning. However, they remain as fixation points that may give rise to difficulties under certain circumstances.


According to Nemiah and Uhde, the psychoanalytic concepts of disturbances in development and fixation points permit an understanding of the process of regression. In the classic analytic formulation, regression is the central mechanism in the formation of obsessive-compulsive symptoms and determines that a person will develop that disorder rather than

a conversion disorder. According to psychoanalytic theory, the person with conversion disorder has repressed oedipal genital libido, and the energy from that undischarged impulse is converted into somatic symptoms. A different process occurs in the obsessive-compulsive reaction. OCD patients may begin with a conflict over the oedipal genital impulse, when, for example, it is aroused by an environmental stimulus. Instead of repressing and converting that impulse, they avoid the associated anxiety by abandoning the genital impulses and regressing to the earlier anal-sadistic phase. Regression is facilitated by the fixation points that remain from the distortions that occurred during childhood development. By giving up genital urges, patients are no longer confronted with the conflicts and problems resulting from these urges.


Nemiah and Uhde noted that ambivalence is the direct result of a change in the characteristics of the impulse life. Ambivalence is an important feature during the normal anal-sadistic developmental phase. Children in that phase feel both love and murderous hate toward the same person. One emotion follows the other in such rapid succession that they appear temporarily to exist side by side. In normal development much of the aggression is neutralized, and what remains is the desire to win out over, rather than to destroy, the other person. As a result, in a mature person, love for the object is dominant, and aggression plays a minor role. When regression occurs, there is a return to the earlier level of functioning, in which ambivalence is a characteristic mode of feeling. OCD patients often consciously experience both love and hate towards others. The conflict of opposing emotions may be seen in the doing-undoing patterns of behavior and in the paralyzing doubt in the face of choices that are so frequently found in persons with OCD.

Magical thinking

In the phenomenon of magical thinking, Nemiah and Uhde reported that the regression uncovers earlier modes of thought rather than impulses; that is, ego functions, as well as id functions, are affected by regression. The phenomenon of the omnipotence of thought is inherent in magical thinking. Individuals believe that merely by thinking about an event in the external world they can cause it to occur, without intermediate physical actions. It is that feeling that makes aggressive thoughts so frightening to OCD patients. The phenomenon is related to the incantations and rituals that are central to organized magic in all ages and cultures. The same mode of thinking is present in primitive peoples who fear the evil thoughts of others and ward off the bad consequences of such thoughts by special formulas, or who try to influence natural forces, such as rain and fertility, by magic. The same kind of magical thinking can be seen in children's rituals, games, and fears, which at times reach a degree that is suggestively pathological.

Changes in the superego

Nemiah and Uhde described the psychoanalytic view of OCD as a regression to developmentally earlier stages of the infantile superego (sometimes called the archaic superego), the harsh, exacting, punitive characteristics of which now reappear in the mental functioning of neurotic adults. The appearance of symptoms in OCD is attributed to a defensive regression of the psychic apparatus to the preoedipal anal-sadistic phase, with the consequent emergence of earlier modes of functioning of the ego, superego, and id. Those factors, along with the use of specific ego defenses--isolation, undoing, displacement--combine to produce the clinical symptoms of obsessions and compulsions.



OCD patients usually present with specific complaints, such as pronounced obsessions or compulsive rituals, that allow the clinician to make the diagnosis easily. With nonpsychiatric physicians and even with psychiatrists who do not specialize in anxiety disorders, patients may be reluctant to discuss symptoms that they find embarrassing or disgusting. Some patients in intensive psychodynamic psychotherapy or psychoanalysis do not even mention their OCD symptoms. For that reason, clinicians should question new patients specifically about intrusive repetitive thoughts or rituals. Sometimes paper and pencil questionnaires, such as the Maudsley Obsessive-Compulsive Inventory (Table 17.3-1) (Table Not Available) and the Yale-Brown Obsessive-Compulsive Scale (Table 17.3-2) allow patients to respond positively to questions that the clinician can later discuss more fully. Sometimes patients cannot resist performing rituals in front of the physician, or they refuse to shake hands for fear of contamination. Most patients, however, can resist their urges when they are in public or in the physician's office. Patients usually appear completely normal to the casual observer.

Patients who divulge the nature of their obsessions may appear bizarre or irrational, but they almost always retain full insight and recognize that their thoughts and impulses are unreasonable and alien to the rest of their personality structure. No generalizations can be made about the personality types of OCD patients, and their demeanor may range from histrionic crying to obsessive fussiness and controlling. The majority meet criteria at least for mild personality disorders when first presenting to the physician, but those features usually subside as OCD symptoms improve. Conclusions cannot be drawn, therefore, about personality disorders or personality type when the patient is actively ill with OCD or with any other severe psychiatric or medical illness.

The DSM-IV diagnostic criteria for obsessive-compulsive disorder are given in Table 17.3-3 (Table Not Available) .


Symptoms can usually be placed into one of several categories: checking rituals, cleaning rituals, obsessive thoughts, obsessional slowness, or mixed rituals. Checking and cleaning rituals are the most common and multiple symptoms are the rule.

Cleaning compulsions

A 20-year-old woman feared contamination from touching various things she considered dirty. She had to wear gloves or use paper towels to touch various "dirty objects." If, however, she did happen to touch her laundry, her bed, the door handles in public restrooms, shoes, the gas cap on her car, or other "dirty" objects, she experienced vague dirty and uncomfortable feelings, and she would engage in prolonged washing of her hands and would wash any clothing that had come into contact with the object. As a result of those symptoms, she was unable to work full-time and her social life was almost nonexistent.

Checking compulsions

A 46-year-old woman checked when unsure whether she had performed an action correctly. She plugged and unplugged electric appliances many times to make sure that she actually took the plug out of the socket, and she turned light switches on and off repeatedly until she was convinced that she in fact had turned them off. She would stare at a closed door for up to 20 minutes to ensure that she had actually closed and locked it. She completely avoided financial paperwork because of a compulsion to check numbers over and over again, and she could no longer work in her previous job as a bookkeeper. She was no longer able to read because she continually reread sentences to be sure she had not missed any crucial ideas.

Primary obsessional disorder

Perhaps as many as 15 percent of OCD patients have only obsessive thoughts, with few or no rituals.


TABLE 17.3-1 -- Maudsley Obsessive-compulsive Inventory
Table from S J Rachman, R J Hodgson: Obsessions and Compulsions, p 222. Prentice-Hall, Englewood Cliffs, NJ, 1980.
(Not Available)

The thoughts are typically of an aggressive, sexual, or religious nature and are upsetting and repulsive to the patient. For example, an 18-year-old man could no longer go to public places because of obsessive thoughts and impulses to shout obscenities. Similarly, a 32-year-old woman no longer went to church because she would experience intolerable sexual thoughts about people she saw there and felt that she would blurt out obscenities at the priest.

Other less common subtypes

Some patients spend inordinate periods of time placing objects in a specific order. Others suffer with primary obsessional slowness and become stuck for hours while performing everyday tasks, such as dressing and eating. Relatively rare subtypes are being identified, such as patients with obsessions and compulsions primarily aimed at controlling an overwhelming fear of having a bowel movement or urinating in public or young women who have face-picking bouts which can last for hours. Other disorders that may be closely related to OCD are monosymptomatic hypochondriasis, body dysmorphic disorder, and obsessive fear of acquired immune deficiency syndrome (AIDS), cancer, or some other illness.



Obsessive-compulsive disorder is frequently confused with obsessive-compulsive personality disorder. OCD is an Axis I disorder in DSM-IV, while obsessive-compulsive personality disorder is an Axis II disorder. Occasionally, patients with OCD also have compulsive personality traits, and some (roughly 6 percent when assessed by a standardized structured interview generating DSM-III criteria) also meet criteria for obsessive-compulsive personality disorder.


The essential feature of specific phobia is persistent fear of a circumscribed object or situation. The essential feature of social phobia is persistent fear of humiliation or embarrassment in certain social situations. Common specific phobias include fear of small animals (for example, dogs, snakes, insects, mice), blood, closed spaces, heights, and air travel. In patients with OCD, phobic avoidance of certain situations that are associated with anxiety about dirt or contamination is frequent, but the concomitant presence of typical obsessions or rituals clarifies the diagnosis of OCD.


Classic compulsive rituals are not generally part of the picture of depression, but depressed patients occasionally ruminate about a particular topic and may appear to have obsessions. Careful history will usually reveal that the depression preceded the obsessions or ruminations. In addition, the ruminations of the depressed person are more likely to have a realistic basis. For example, a depressed patient may constantly think about losing his or her job during hospitalization and may be unable to focus on anything else. In fact, the job may be jeopardized, and the concerns may be exaggerated but well-founded. About a third of the patients with OCD develop clinically significant secondary depression; fortunately, most antiobsessional drugs are also potent antidepressant agents.


Occasionally, the obsessions in OCD become so severe that the patient seems truly uncertain whether his or her concerns are realistic. Such obsessions are called overvalued ideas; for example, patients may hold the almost unshakable belief that they are contaminating other people unless they wash their hands for three hours after urinating. However, OCD patients with overvalued ideas can, after considerable discussion, usually acknowledge the possibility that their beliefs are unfounded. In contrast, the person with a true delusion usually has a fixed conviction that cannot be shaken and is also likely to have other psychotic symptoms, such as ideas of reference, paranoia, and hallucinations.


TABLE 17.3-2 -- Yale-Brown Obsessive-compulsive Scale
  1. Time occupied by obsessive thoughts 0 = None

1 = Mild (less than 1 hr/day), or occasional intrusion (occur no more than 8 times a day).

2 = Moderate (1 to 3 hrs/day), or frequent intrusion (occur more than 8 times a day, but most hours of the day are free of obsessions).

3 = Severe (greater than 3 and up to 8 hrs/day), or very frequent intrusion (occur more than 8 times a day and occur during most hours of the day).

4 = Extreme (greater than 8 hrs/day), or near constant intrusion (too numerous to count and an hour rarely passes without several obsessions occurring)
  2. Interference due to obsessive thoughts 0 = None

1 = Mild, slight interference with social or occupational activities, but overall performance not impaired.

2 = Moderate, definite interference with social or occupational performance, but still manageable.

3 = Severe, causes substantial impairment in social or occupational performance.

4 = Extreme, incapacitating.
  3. Distress associated with obsessive thoughts 0 = None

1 = Mild, infrequent, and not too disturbing.

2 = Moderate, frequent, and disturbing, but still manageable.

3 = Severe, very frequent, and very disturbing.

4 = Extreme, near constant, and disabling distress.
  4. Resistance against obsessive thoughts 0 = Makes an effort to always resist, or symptoms so minimal doesn't need to actively resist.

1 = Tries to resist most of the time.

2 = Makes some effort to resist.

3 = Yields to all obsessions without attempting to control them, but does so with some reluctance.

4 = Completely and willingly yields to all obsessions.
  5. Control over obsessive thoughts 0 = Complete control.

1 = Much control, usually able to stop or divert obsessions with some effort and concentration.

2 = Moderate control, sometimes able to stop or divert obsessions.

3 = Little control, rarely successful in stopping obsessions, can only divert attention with difficulty.

4 = No control, experienced as completely involuntary, rarely able to even momentarily divert thinking.
  6. Time spent performing compulsions 0 = None

1 = Mild (less than 1 hr/day performing compulsions), or occasional performance of compulsive behaviors (no more than 8 times a day).

2 = Moderate (1 to 3 hrs/day performing compulsions), or frequent performance of compulsive behaviors (more than 8 times a day, but most hours are free of compulsive behaviors).

3 = Severe (spends more than 3 and up to 8 hrs/day performing compulsions), or very frequent performance of compulsive behaviors (occur more than 8 times a day and compulsions performed during most hours of the day).

4 = Extreme (more than 8 hrs/day performing compulsions), or near constant compulsive behaviors (too numerous to count and an hour rarely passes without several compulsions being performed).
  7. Interference due to compulsive behaviors 0 = None

1 = Mild, slight interference with social or occupational activities, but overall performance not impaired.

2 = Moderate, definite interference with social or occupational performance, but still manageable.

3 = Severe, causes substantial impairment in social or occupational performance.

4 = Extreme, incapacitating.
  8. Distress associated with compulsive behaviors 0 = None

1 = Mild, only slightly anxious if compulsions prevented, or only slight anxiety during performance of compulsions.

2 = Moderate, reports that anxiety would mount but remain manageable if compulsions prevented, or that anxiety increases but remains manageable during performance of compulsions.

3 = Severe, prominent, and very disturbing anxiety if compulsions interrupted, or prominent and very disturbing anxiety when performing compulsions.

4 = Extreme, incapacitating anxiety from any intervention aimed at modifying activity, or incapacitating anxiety develops during performance of compulsions.
  9. Resistance against compulsions 0 = Makes an effort to always resist, or symptoms so minimal doesn't need to actively resist.

1 = Tries to resist most of the time.

2 = Makes some effort to resist.

3 = Yields to all compulsions without attempting to control them, but does so with some reluctance.

4 = Completely and willingly yields to all compulsions.
10. Degree of control over compulsive behaviors 0 = Complete control.

1 = Much control, experiences pressure to perform the behavior, but usually able to voluntarily control it.

2 = Moderate control, strong pressure to perform behavior, must be carried to completion, can only delay with difficulty.

3 = Little control, very strong drive to perform behavior, can only delay with difficulty.

4 = No control, drive to perform behavior experienced as completely involuntary and overpowering, rarely able to even momentarily delay activity.


TABLE 17.3-3 -- Diagnostic Criteria for Obsessive-compulsive Disorder
Table from DSM-IV, Diagnostic and Statistical Manual of Mental Disorders, ed 4. Copyright American Psychiatric Association, Washington, 1994. Used with permission.
(Not Available)


Obsessive-compulsive disorder is often found in patients with Tourette's disorder, and in that case both diagnoses are given. There is much confusion in the terminology used by neurologists, who often see Tourette's disorder patients, and that used by psychiatrists, who are more likely to see OCD patients. For example, rituals are sometimes referred to as complex tics in the neurological literature.


The mean age of onset of OCD is between ages 20 and 24; over 80 percent of patients develop symptoms before age 35. Some patients describe the onset of symptoms after a stressful event, such as a pregnancy, a sexual problem, or the death of a relative, and in many cases the onset is sudden. Because many patients manage to keep their symptoms secret, there is often a delay of 5 to 10 years before patients receive psychiatric attention.

The precise course and prognosis of OCD cannot be predicted since details of its natural history are unknown. No carefully conducted studies have evaluated its longitudinal course. In general, OCD is a chronic illness that exhibits a waxing and waning course, even with treatment. Complete cures are unusual. However, approximately 90 percent of patients can expect moderate to marked improvement with optimum treatment. Some evidence indicates that good premorbid functioning is an optimistic prognostic sign, but hard evidence of this is lacking. The actual obsessional content does not seem to be related to prognosis.


Poor compliance with treatment instructions is the most common reason for failure with behavior therapy. Behavior therapists make specific demands on patients, and compliance with behavioral instructions both during treatment sessions and also during homework assignments is imperative if patients are to improve as much as possible. Family members or friends often act as surrogate therapists in helping OCD patients to carry out homework assignments.

Patients who hold overvalued ideas that their compulsive rituals are necessary to prevent a catastrophe seem to have a poor outcome with behavioral treatments. For example, the patient who believes that her daughter will die if she does not wash all of her daughter's clothes every day is unlikely to give up washing rituals with behavior therapy alone. Antiobsessional medication may produce changes in such fixed beliefs, and behavior therapy may then be helpful.

In severely depressed patients, physiological habituation to a feared stimulus does not usually occur, regardless of the length of exposure, but such patients often respond well to behavior therapy once depression is controlled pharmacologically.

Patients meeting criteria for both OCD and schizotypal personality disorder seem not to respond well to either behavior therapy or pharmacotherapy. Those patients may really believe that their rituals are necessary to prevent some terrible event. In addition, they have difficulty complying with assigned behavioral and record keeping instructions. Such patients may benefit from a structured environment, such as a day treatment center or halfway house. Behavioral treatment may produce modest decreases in their obsessive and compulsive symptoms, along with moderate improvements in overall functioning. Even though the patients themselves may benefit only slightly, treatment often allows the rest of the family to lead a more normal life.

Even when responsive to behavioral techniques, patients with checking rituals appear to respond more slowly than those with cleaning rituals. Patients with checking rituals, especially those who check excessively at home, are often unable to engage in prescribed response prevention. Patients with primary obsessional slowness respond more slowly to behavior therapy than do patients with either cleaning or checking rituals.


The treatment of patients suffering from OCD is an example of the need to integrate various approaches to maximize patient outcome. They must generally receive medication in combination with other approaches, particularly behavior therapy. That combined approach can be expected to improve the condition of most patients substantially, and occasionally completely, within a few months.


In the absence of any adequate studies of psychotherapy for OCD, it is difficult to make valid generalizations about its effectiveness. Nemiah and Uhde noted that, early in the development of psychoanalysis, psychotherapy was the treatment of choice, because, like conversion hysteria, OCD was regarded as a transference neurosis and should theoretically

therefore respond to psychoanalytical techniques. Nemiah and Uhde noted that some analysts have seen striking and lasting improvements in patients with obsessive-compulsive personality disorder traits, especially when the patients were able to come to terms with aggressive impulses behind those traits.

Traditional psychodynamic psychotherapy is not considered an effective treatment for obsessions and rituals occurring in patients who meet the criteria for OCD in DSM-IV; there are no reports of patients who stopped ritualizing when treated with that method alone. Such treatment may be helpful for patients with obsessive-compulsive personality disorder. Conversely, there is no evidence that behavior therapy and medications are helpful for patients with the personality disorder. Many traditional psychotherapists find themselves becoming more directive with OCD patients and adopt techniques similar to those used by behavior therapists.

Supportive psychotherapy is often helpful. Regular contact with a kind, warm, and understanding therapist can help the patient to comply with behavior therapy and to cope with medication side effects. Whether or not OCD patients involve family members in their rituals, families may be very troubled by patients' behaviors. Any psychotherapeutic endeavors must include attention to family members through the provision of emotional support, reassurance, explanation, and advice on how to manage and respond to the patient. In addition, family members can be very helpful to the patient by serving as surrogate home behavior therapists.


The typical randomized prospective placebo-controlled trial, which proved so useful in depression research, was until recently almost impossible because of the small numbers of OCD patients available to any one researcher. The number of controlled trials is increasing, however, as OCD clinics are seeing growing numbers of patients. Currently, pharmacotherapy (combined with behavior therapy for patients with rituals) is considered a treatment of choice for OCD.

Cyclic and atypical antidepressants

Case reports of successful treatment of OCD have involved almost every antidepressant on the market, including imipramine (Tofranil), clomipramine, amitriptyline (Elavil, Endep), doxepin (Adapin, Sinequan), desipramine, sertraline, zimelidine, fluoxetine, trazodone (Desyrel) and fluvoxamine. Recent double-blind placebo-controlled trials have revealed that clomipramine (up to 250 mg a day), fluvoxamine (up to 300 mg a day), sertraline (up to 200 mg a day), paroxetine (up to 60 mg a day), and fluoxetine (up to 80 mg a day) are effective.

The best studied antiobsessional agent is clomipramine, a tricyclic antidepressant, that has been available in Europe and Canada for many years. It has specific antiobsessional properties in addition to its antidepressant qualities. The optimum dose is unknown, but the majority of researchers believe that dosage should be increased to 250 mg a day if patients can tolerate the side effects. Many carefully controlled studies have confirmed preliminary results that clomipramine is indeed superior to placebo in the treatment of OCD. In one study, almost 60 percent of patients on clomipramine had at least a moderate response, and another 25 percent reported at least some improvement. The main drawback to clomipramine is its substantial anticholinergic side effects. Sexual difficulties are common, and a small incidence of seizures occur at higher doses. Most patients, however, tolerate it well.

Monoamine oxidase inhibitors

Anecdotal evidence suggests that MAOIs are particularly helpful for patients who suffer concomitantly from OCD and panic attacks or severe anxiety. Affective illness in patients or their families does not appear to be a good predictor of responsiveness to MAOIs.


One double-blind crossover trial of six OCD patients carried out in Denmark reported that lithium (Eskalith) was not effective in OCD. On the other hand, there are a few case reports of patients with classic OCD who improved with lithium carbonate.

Obsessive-compulsive behaviors are sometimes found in patients suffering from bipolar disorder. A recent report of two patients who met criteria for both disorders, who were treated with a combination of therapist-aided and self-administered exposure and response prevention, demonstrated that behavior therapy was effective only after their major affective disorder was effectively controlled with lithium and antipsychotics.

A 22-year-old woman did not respond to clomipramine alone, but improved greatly a few days after lithium carbonate was added with a stabilized blood level of 0.9 mEq/ L. Whether or not lithium augmentation of other tricyclic antidepressants or MAOIs for obsessive-compulsive symptoms is helpful, has yet to be shown.

Antipsychotic agents

Only a few case reports outline success with antipsychotic agents. Most of the patients were atypical, and some fit the clinical picture of schizophrenia rather than classic OCD. The schizophrenic features may have been partly, or even substantially, responsible for the good results. One group of researchers reported that antipsychotics enhanced the effects of fluvoxamine alone or in combination with lithium carbonate in OCD patients with concomitant tics, while it did not help OCD patients without tics. However, in view of the scarcity of data on the efficacy of those agents and the frequency of their toxic side effects, their use can be recommended only for acutely disturbed obsessional patients for the shortest possible period.

Anxiolytic agents

Anxiolytic agents are of little use in the treatment of obsessions or compulsions, but they do help with anxiety that many OCD patients report. If antiobsessional agents improve OCD, anxiety usually decreases without the use of anxiolytics. The literature contains a few case reports of success and a couple of controlled trials where outcome criteria were unclear. Buspirone was ineffective in one open trial, while another study reported that both buspirone and clomipramine led to similar and statistically significant improvement in OCD symptoms. The resolution of those conflicting results awaits further data, but most researchers are skeptical about the usefulness of buspirone in treating OCD symptoms.


Historically, many OCD patients received trials of electroconvulsive therapy (ECT). Most did not suffer from a major mood disorder, and the primary reason for administering ECT was for treatment of OCD.

A few studies report that ECT in combination with other treatment modalities was helpful. One atypical patient (obsessions only which developed after his wife's death) had a good response to ECT after not responding to a number of treatments, including a 12-week trial of clomipramine. One group assessed the combined effects of ECT, modified narcosis, and antidepressants on obsessional neurotics (unclear diagnostic criteria) and found that 40 percent of the patients improved; however, the relative effect of each form of treatment separately was obscure. Another author studied 100 patients with obsessional symptoms, which were also poorly defined, and concluded that

ECT had little effect on obsessional states. The general consensus is that ECT is not useful in the OCD patient who is not endogenously depressed, although scant literature exists concerning the effects of ECT alone on OCD.


With the advent of restricted and relatively safe psychosurgical operations, such as cingulotomy and capsulotomy, and the recognition that some patients are severely disabled and remain refractory to modern treatments, interest in psychosurgery has reawakened. Since most OCD patients who undergo psychosurgery have had very severe illness that has not responded to multiple therapeutic approaches (including pharmacotherapy and behavior therapy), the results of surgical intervention are impressive.

A recent study confirmed the relative safety and partial efficacy (at least 25 to 30 percent of patients improved) of stereotactic cingulotomy as a treatment for refractory, severely disabled OCD patients. The main complication, seizures, which occurred in 3 (9 percent) patients, was easily controlled by phenytoin (Dilantin). Four patients committed suicide, but each was very ill with complicating disorders, especially severe depressive disorder, in which suicide is a common complication. All four patients were known to be severely depressed with strong suicidal ideation at the time of operation. It is possible that disappointment in the failure of the last-resort treatment contributed to suicide in those patients. There is some evidence that other treatments, including pharmacotherapy and behavior therapy, are more likely to be successful after psychosurgery than before, but more research is needed in that area. Psychosurgical patients often benefit only a few weeks to months after the operation.

A number of operations are used for the treatment of disabling OCD. A review of four neurosurgical procedures (anterior cingulotomy, limbic leucotomy, tractotomy, and anterior capsulotomy) reveals that anterior cingulotomy has a very low complication rate and a moderate success rate. Limbic leucotomy combines bilateral cingulate lesions with lesions in the orbitomedial frontal areas containing fibers of a fronto-caudate-thalamic tract that may be critical in the formation of obsessive-compulsive symptoms. Anterior capsulotomy and tractotomy also produce significant improvement rates.

The identification of patient subgroups with a good prognosis after neurosurgical procedures merits further study. Currently, it is impossible to predict which patients might improve and which procedure is the best for a particular patient. It is still unclear which OCD patients should be referred for surgical procedures, and definite recommendations must await the results of ongoing prospective studies. However, currently, psychosurgery does appear to have a role for the severely disabled and treatment-refractory OCD patient.


The behavioral techniques most consistently effective in reducing compulsive rituals and obsessive thoughts are exposure to the feared situation or object, and response prevention, in which the patient resists the urge to perform the compulsion after exposure. Simple relaxation therapy is an ineffective treatment for OCD symptoms. Behavior therapy produces the most significant changes in rituals, such as compulsive cleaning or checking, whereas changes in obsessive thoughts are less predictable. That difference reflects the specific effects of behavioral treatment, in which the behaviors themselves are the targets of treatment. Behavior therapy (in combination with pharmacotherapy) is now regarded as the treatment of choice when behavioral rituals predominate.

Behavioral techniques have been understood for over a century; in fact, Pierre Janet gave a remarkably accurate description of what is now called exposure therapy, including the name itself:

The guide, the therapist, will specify to the patient the action as precisely as possible. He will analyze it into its elements if it should be necessary to give the patient's mind an immediate and proximate aim. By continually repeating the order to perform the action, that is, exposure, he will help the patient greatly by words of encouragement at every sign of success, however insignificant, for encouragement will make the patient realize these little successes and will stimulate him with the hopes aroused by glimpses of greater successes in the future. Other patients need strictures and even threats and one patient told [Janet], `Unless I am continually being forced to do things that need a great deal of effort I shall never get better. You must keep a strict hand over me.'

Inexperienced clinicians are sometimes fearful of the effects or unaware of the potential of behavior therapy. A number of common misconceptions have developed. The clinician needs to know that: behavior therapy will not lead to the formation of substitute symptoms; interrupting compulsive rituals is not dangerous in any way to the patient; the patient's thoughts and feelings are not ignored in behavior therapy; modern behavior therapists do not assume that all maladaptive behavior is learned through simple conditioning processes; the use of medication is not incompatible with behavior therapy; and behavior therapists recognize that their therapeutic techniques are not equally effective for all patients. Controlled outcome studies of exposure and response prevention for OCD over the past 15 years with more than 200 patients in various countries have found that 60 to 70 percent of OCD patients were much improved after behavioral treatment. At follow-up of two or more years, reduction in rituals were maintained in almost all patients. Preliminary results of a recent study indicate that behavior therapy may be more effective than pharmacotherapy. Patients with only obsessive thoughts and no rituals have been studied separately, with unpredictable results. Although the technique of thought-stopping is widely used to treat obsessive thoughts, there is no clear empirical support for its usefulness.

A few studies have attempted to tease apart the differential effects of the exposure and the response prevention components of behavior therapy. For example, with washers, exposure therapy was found to help mainly in reducing the anxiety component, while response prevention had its greatest effect in reducing the ritualistic washing. The combined treatment was more effective than either component in isolation.

In patients with checking rituals, combined imaginal exposure (that is, having the patient vividly imagine the most feared consequences of not ritualizing) and response prevention are superior to response prevention alone. That approach is necessary for some patients because the catastrophic consequences that many checkers fear never actually occur, so habituation must be carried out in their imagination.

Because OCD patients engage in obvious cognitive errors in inference and in assessing the probability of danger, the use of cognitive therapy to modify those cognitive processes would seem to be useful. Unfortunately, the treatment outcome with cognitive therapy has been less predictable than the combination of exposure and response prevention. The only controlled study found that cognitive therapy did not add significantly to the effects of in vivo exposure.


Neurotransmitters are discussed in Sections 1.3 and 1.4 and learning theory in Section 3.3. Personality disorders, including

obsessive-compulsive personality disorder, are discussed in Chapter 25 . Behavior therapy is discussed in Section 31.2. Biological therapies are discussed in Chapter 32 ; electroconvulsive therapy is discussed in Section 32.28, and psychosurgery in Section 32.29.


Baer L, Jenike M A, Ricciardi J, Holland A, Seymour R, Minichiello W E, Buttolph L: Personality disorders in patients with OCD. Arch Gen Psychiatry 47: 826, 1990.

*Baer L: Getting Control. Little Brown, Boston, 1991.

Baer L, Minichiello W E: Behavior therapy for OCD. In Obsess