Generalised Anxiety Disorder

This is often called the common cold of anxiety because of its frequency.

You do not have generalised anxiety disorder (GAD) if your anxieties and worries are realistic. If unemployment strikes and you have no income and bills to meet it is normal to feel anxious.

If your symptoms are persistent , long lasting and unrealistic then you probably have GAD. If you you are concerned that you have GAD then you should seek advice from a health professional. The first person to consult is your family doctor or GP. He or she can exclude physical causes and make the correct diagnosis.

There are four core features of Generalised Anxiety Disorder :

  1. It should have lasted for at least 6 months
  2. It is uncontrollable
  3. It is unrealistic
  4. Worry is central

The usual symptoms of GAD are as follows:

  • Restlessness, irritability, feeling on edge, fidgety
  • Presence of muscle tension especially in neck, shoulders, face, and back
  • Poor concentration
  • Difficulty with sleep. This includes going to asleep and staying asleep
  • General tiredness

The symptoms of anxiety vary from person to person, and not everyone has the typical symptoms described above.

Other symptoms experienced include:

  • Shortness of breath
  • Palpitations
  • The sensation of butterflies
  • Tremor, shaking
  • Dry mouth,difficulty swallowing
  • Excessive sweating
  • Nausea, stomach upset
  • Being easily startled

Case example of generalised anxiety disorder 

Jane had been a quiet child and had been painfully shy. During her teens she frequently felt anxious. Her first job on leaving school was working as a secretary. She wanted to have friends, but was too embarrassed to call anyone. She was married to Kevin who one day left unexpectedly. Jane developed symptoms of anxiety with nausea, palpitations and butterflies in her stomach. She was easily startled by loud noises. She was unable to control her feelings. Her worries were unrealistic and out of proportion. She lay awake at night worrying. During the day she was tired and unable to work properly. Her concentration was poor. After 7 months of struggling she was still feeling unwell and she went to see her family doctor who diagnosed her with generalised anxiety disorder. She was referred to a clinical psychologist for treatment.

Treatment of generalised anxiety disorder (GAD)

Treatment is either with medication or with psychological therapies:

Pharmacological Treatments

Three types of medication have been shown to be effective in the treatment of GAD:

  • Benzodiazepines
  • Antidepressants  
  • Azapirones

Other drugs such as anti-histamines and anti-psychotics have also been tried but there is insufficient data to recommend their use.


Examples include diazepam, lorazepam and alprazolam. They first became available in the early 1960’s and they became the most prescribed medication for anxiety. Hundreds of studies have shown them to be effective compared with placebo in 65-70% of patients.

Recurrence of anxiety is common after stopping this type of medication. These drugs work quickly with most studies showing an effect within a week. The experience of patients and doctors is that they can work very quickly - within 15-30 minutes with some benzodiazepines.

They are Gaba agonists. In other words they increase gamma-aminobutyric acid inhibitory impulses in the central nervous system, which are mediated via benzodiazepine receptors. They are more effective in treating the somatic rather than the psychic symptoms (worry etc) of anxiety.

Safety is a concern with benzodiazepines. Typical side effects include sedation, impairment of psycho-motor coordination and anterograde amnesia. There is also concern about the addictive potential of benzodiazepines.


Many antidepressants have been tried as a treatment for anxiety. Of the older tricyclic antidepressants imipramine and clomipramine have been shown to be effective in the treatment of anxiety.

Selective serotonin reuptake inhibitors have been extensively studied as treatments for generalised anxiety disorder. Paroxetine and escitalopram are approved by the US FDA as treatments for GAD in patients who have no associated symptoms of depression.

Other SSRIs such as citalopram, fluoxetine, sertraline and fluvoxamine are also prescribed as treatments for GAD but do not have FDA approval.

The SNRI Venlafaxine has FDA approval as a treatment for GAD.


In 1996 Buspirone was approved by the US FDA for the treatment of of GAD. Other azapirones have been studied in the treatment of GAD, but none have shown be as effective as buspirone. This drug is thought to work by binding to serotonin 5-HT1A receptors in the brain. 

Buspirone generally takes longer to work than benzodiazepines. It is also thought to have more of an effect on psychic symptoms than somatic symptoms, in contrast to benzodiazepines. Studies have shown that buspirone has little dependence or abuse potential.

Psychosocial Treatments

Various psychological treatments have been tried for GAD. One of the most widely used is cognitive behavioural therapy (CBT). This has shown to help in about 40-60% of patients. Other treatments that have been tried are “worry exposure”, and mindfulness.

Treatments such as CBT and mindfulness are now widely available and help large numbers of people. But unfortunately still, a lot of anxiety sufferers are not helped by these therapies.

Generalised Anxiety Disorder in more detail - technical


In the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV) (American Psychiatric Association, 1994) generalised anxiety disorder (GAD) is currently defined as excessive anxiety and worry (apprehensive expectation) occurring for a majority of days during at least a 6-month period, about a number of events or activities (such as work or school performance; see DSM-IV). In individuals with GAD, the anxiety and worry are accompanied by at least three of six somatic symptoms (only one accompanying symptom is required in children), which include restlessness or feeling keyed up or on edge, being easily fatigued, difficulty concentrating or mind going blank, irritability, muscle tension and sleep disturbance. In addition, the affected individual has difficulty controlling his/her worry, and the anxiety, worry, or somatic symptoms cause clinically significant distress or impairment in social, occupational, and/or other important areas of functioning. Further, the GAD symptoms should not be due to the direct physiological effects of a substance such as drugs or alcohol or a general medical condition, and should not occur exclusively during a mood disorder, psychotic disorder, or pervasive developmental disorder.

Worry and anxiety are part of normal human behavior and it may be difficult to define a cutoff point distinguishing normal or trait anxiety (i.e., a relatively stable tendency to perceive various situations as threatening) from GAD. However, as described in the DSM-IV definition of GAD, individuals suffering from a “disorder” exhibit significant distress and impairment in functioning as a result of their anxiety symptoms.

Etiology and Pathophysiology

Family Studies

Family studies suggest a familial (and probably a genetic) basis for certain anxiety disorders such as panic disorder. Genetic transmission of a disorder suggests that certain gene-encoded changes in proteins and the resulting biological abnormalities may play a role in the pathophysiology of specific disorders. Skre and collaborators (1993) examined 20 monozygotic and 29 dizygotic twins with DSM-III-R-defined GAD. They found GAD to be diagnosed in 22% of first-degree relatives of 33 probands with anxiety disorders. In the largest twin study to date which included 1033 female twin pairs, Kendler and associates (1992) found that genetic factors play a significant, but not overwhelming role in the etiology of GAD, with the heritability of GAD estimated ataround 30% in comparison to 70% heritability in major depression. In addition, the authors found that the vulnerability to GAD and major depression is influenced by the same genetic factors. In short, the available data suggest at most a modest genetic contribution to the etiology of GAD.

Biological Studies

Relatively few studies have addressed issues regarding the biological aspects of GAD. Existing studies have focused on the evaluation of catecholamine and autonomic responses, neuroendocrine measures, sleep, neuroanatomical/neuroimaging studies, infusion studies and evaluation of other neurotransmitter systems. There is not strong evidence for abnormalities in catecholamine or thyroid function in GAD patients. Some studies have shown a higher prevalence of an “escape” (nonsuppression) response in following dexamethasone administration (that was not attributable to the presence of depression) in GAD patients when compared with normal comparison subject. These data indicate that there may be dysregulation of the HPA axis in these patients, as observed following dexamethasone.

Although restless and decreased sleep are common complaints in GAD patients, there have been only a few polysomnographic studies in this patient population. There is some evidence suggesting that patients with GAD have a longer rapid eye movement (REM) latency, shorter REM duration, increased sleep onset latency and less total sleep time compared with control subjects (Papadimitriou et al., 1988). These findings may differentiate patients with GAD from patients with depression, who show shorter REM latencies.

Alterations in different neurotransmitter systems have been implicated in the pathophysiology of various anxiety disorders. It is generally accepted that anxiety disorders are not associated with abnormalities in only one neurotransmitter system; rather dynamic interactions among several different neurotransmitter systems are believed likely to underlie different anxiety states. Presently, there are data suggesting that the catecholamine serotonin and GABA-benzodiazepine systems may be involved in the pathophysiology of anxiety disorders.


Benzodiazepines have been the treatment of choice for many patients with GAD. They act at specific recognition sites in the brain, the benzodiazepine receptors, which are located in a subunit of a receptor for gamma-aminobutyric acid (GABA), the major inhibitory neurotransmitter in the brain. Several lines of evidence suggest that the GABA-benzodiazepine receptor complex may be involved in the mediation of anxiety responses. Studies with animals suggest a relationship between benzodiazepine receptors, and fear and anxiety. Models using gamma-2 knockout mice have shown a reduction in GABAA receptor clustering in the hippocampus and cerebral cortex along with behavioral inhibition to aversive stimuli and increased responsiveness in trace fear conditioning (Lesch, 2001).


Alterations in serotonergic (5-HT) neurotransmission have been implicated in the mediation of fear and anxiety responses in animal models and in humans. Specifically, researchers hypothesize that anxiety may represent dysregulated serotonergic activity in critical brain areas. Given the available data, whether overactivity or underactivity of the 5-HT system is the mechanism for GAD development remains unclear.


Cholecystokinin (CCK), a highly abundant neurotransmitter in the brain, has also been implicated in anxiety in humans. CCK may be possibly involved in the pathophysiology of panic disorder and may also play a role in the biology of GAD. Corticotropin-releasing factor (CRF), a major physiological regulator of adreno corticotropic hormone (ACTH), appears to be involved in stress and anxiety responses. Administration of CRF to various parts of animal brains has elicited anxiety and fear responses, e.g., suppression of exploratory behavior, shock-induced freezing. Interestingly, both these peptides are functionally antagonized by benzodiazepines. Neuropeptide Y, glutamate and tachykinins may also play a role in anxiety.

Neuroanatomic Sites of Anxiety

Several potential neuroanatomic anxiogenic sites in the central nervous system (CNS) have been proposed based on brain imaging and neuroanatomic studies. The areas potentially involved in anxiety are the parts of the limbic system involving the hippocampus, prefrontal cortex, occipital lobes, basal ganglia and brain stem structures, specifically the locus coeruleus, nucleus paragigantocellularis and periaqueductal gray (Gray, 1988). These structures are rich in noradrenergic, GABAergic and serotonergic receptors which are believed to be involved in the pathophysiology of different anxiety states such as GAD.

Cerebral Blood Flow and Metabolism Studies

Only a few imaging studies in GAD have appeared in the literature. In one study, patients with GAD displayed decreases in cortical blood flow compared with control subjects. Significant negative correlations between state anxiety and cerebral blood flow in most brain regions were observed. Wu and colleagues (1991) evaluated 18 patients who met DSM-III criteria for GAD using positron emission tomography (PET) measurements of cerebral glucose at “baseline” (during a passive viewing task), following a cognitive vigilance task designed to stimulate anxiety, and following treatment with benzodiazepines. They found a higher relative metabolic rate for GAD patients in parts of the occipital, temporal, frontal lobes and cerebellum relative to normal control subjects during a passive viewing task. The authors also found a decrease in absolute basal ganglia metabolic activity in GAD patients. During the vigilance task, GAD patients showed a significant increase in relative basal ganglia metabolism. The authors did not find a global decrease in cortical metabolism, as had been predicted by blood flow studies. Finally, benzodiazepine treatment resulted in a significant decrease in glucose metabolism in cortical surface (especially in the occipital cortex), the limbic system and basal ganglia compared with patients receiving placebo.

MRI and SPET studies have found that those with GAD have decreased benzodiazepine receptor binding in the left temporal pole as compared with matched healthy controls. In a study using functional MRI in GAD patients, Lorberbaum and colleagues (2001) found greater activity in the right cingulate, right medial prefrontal and orbitofrontal cortex, right temporal poles and right dorsomedial thalamus, during periods of anticipatory anxiety, compared with rest periods, than matched control subjects. Further, only matched control subjects displayed increased activity in the medial prefrontal cortex.

Psychological Mechanisms

Cognitive–Behavioral Model

Psychological models, which emphasize the cognitive–behavioral processes involved in the onset and maintenance of GAD, have emerged in recent years. Individuals’ thoughts, cognitive style and behaviors are thought to instigate and maintain episodes of anxiety. In support of cognitive theories of GAD, anxious individuals with GAD are more likely to perceive ambiguous information as threatening and/or negative, and to perceive that they are more likely than others to experience threatening situations.Patients with GAD also pay more attention to the detection of potentially threatening information and incorporate this information into highly elaborate cognitive schemas, thus lowering their threshold for activation of an anxiety response. The threatening information then elicits anxious affect and the individual begins to worry in an attempt further to define the problem.

Patients with GAD are likely to be characterized by a perception of lack of control over threatening. In addition, these patients are likely to believe that they have little control over their emotions, especially their worrying, leading to further distress. It has been suggested that the interaction between perceived uncontrollability and a cognitive focus on negative/threatening stimuli may amplify the general worry to pathological.

Psychodynamic Theories

Freud’s concept of signal anxiety followed the development of Freud’s structural model of the mind, which proposes three interacting psychological functions: ego (which mediates between the demands of primitive drives, the social and parental prohibitions, and reality), superego (representing the internalized parental and social prohibitions) and id (representing the primitive drives and urges). Freud believed that anxiety serves as a signal to the ego of a threat (in the form of an unconscious drive or wish arising from the id), which, if enacted, may be dangerous to the ego, signaling the potential punishment by the superego or the external world. According to this model, the ego can activate defense mecha- nisms, such as repression, and prevent the actualization of the forbidden urge either by preventing the expression of the wish or by avoiding the life situations in which the wish might be potentially expressed. Ideally, repression into the unconscious (i.e., out of subject’s awareness) should successfully contain the drives. However, if the defenses fail, one may experience symptomatic anxiety and other distressing psychological symptoms. Implicit in this model is the concept that the individuals themselves are not consciously aware of these processes. Therefore, the promotion of subjects’ insight into unconscious conflicts and the uncovering of the unconscious origins of anxiety through interpretation and other techniques is the primary goal (and method) of the psychoanalytic treatment approach.

Sullivan developed a theory of anxiety based on the importance of interpersonal relationships. He viewed affects (such as anxiety) as forms of interpersonal communication. According to this model, anxiety communicates the sense of insecurity in interpersonal relationships. For example, a mother who is insecure in her role may communicate her insecurity to the infant when she is anxious in her child’s presence. The child in turn identifies her anxiety and expresses anxious affect himself. Another approach to the understanding of the origins of anxiety was offered by object relations theorists such as Klein and Bowlby. They believed that anxiety reflects a fear of the loss of the nurturing object or fear of being hurt by the antagonistic object.

Finally, self-psychology theorists, such as Kohut, believed that the individual strives to achieve and maintain an integrated, cohesive sense of self. Beginning in early age, the individual devel- ops this sense of self through idealization of important others, such as important caregivers (called self-objects), and through a proc- ess of positive interaction with caregivers (called mirroring). He believed that inadequate provision of these experiences can lead to anxiety (fear of disintegration) and the loss of the cohesive self.

Assessment and Differential Diagnosis

GAD patients frequently report that they have been anxious all their lives. Typically, they were moderately anxious during childhood, later developing full-blown GAD when their stress levels increased through activities such as attending college or starting to work. Patients with early onset of symptoms report experiencing significant anxiety and fears, social isolation, obsessionality, more academic difficulties and disturbed home environment during their childhood. The social maladjustment and emotional overreactivity persist into adulthood. Epidemiological studies and clinical studies suggest that the onset of GAD typically begins between the late teens and late twenties. However, not all GAD patients have a lifelong history of excessive anxiety. Some patients develop their disorder at a later age, that is, in one’s thirties or later. These patients frequently report identifiable, precipitating stressful events, specifically unexpected, negative, important events in the year preceding development of GAD.

Patients with GAD experience chronic anxiety and tension. They find the worry as being uncontrollable. However, some patients intentionally initiate and maintain worry with an almost superstitious assumption that, by doing so, they can avert a negative event. Patients tend to worry predominantly about family, personal finances, work and illness. They are also likely to report worrying over minor matters, such as making a slight social faux pas. The majority report being anxious for at least 50% of the time during an average day. In children and adolescents, the worries often revolve around quality of performance in school/competitive areas; catastrophic events; and physical/mental inadequacies. They typically require excessive reassurance and often appear shy, overcompliant and perfectionistic. Frequent multiple physical complaints are common. They may have an unusually mature and serious manner and appear older than their actual age. These children are often the eldest in small, competitive, achievement-oriented families.

Individuals with GAD commonly complain of feeling tense, jumpy and irritable. They have difficulty falling or staying asleep, and tire easily during the day. Particularly distressing to patients is the difficulty in concentrating and collecting their thoughts. Cognitions appear to play a central role in GAD, as well as other anxiety disorders. Patterns of cognitions, however, appear to be disorder-specific. Cognitions about interpersonal conflict or acceptance by others are quite common.

Patients may present complaining of muscular tension, especially in their neck and shoulders and headaches which frequently are described as frontal and occipital pressure or tension. Patients commonly experience sweaty palms, feel shaky and tremulous, complain of dryness of the mouth, and experience palpitations and difficulty breathing. They may also experience gastrointestinal symptoms such as heartburn and epigastric fullness and approximately 30% of patients experience severe gastrointestinal symptoms of irritable bowel syndrome.

Physical complaints frequently lead patients to seek medical attention, and most will initially consult a primary care physician. Although chest pain is more frequently reported by patients with panic disorder, Carter and Maddock (1992) observed that 34% of patients with GAD without panic attacks experienced chest pain. They also found that these patients were predominantly males and many had undergone extensive cardiac evaluations that revealed no demonstrable cardiac pathology.

Differential Diagnosis

Psychiatric Conditions

Anxiety can be a prominent feature of many psychiatric disorders. In addition, the substantial overlap of symptoms between GAD and other psychiatric disorders such as major depressive disorder, often creates diagnostic and treatment dilemmas for the clinician and may complicate the difficult task of differential diagnosis and treatment planning. This section will highlight the major disorders that should be considered in the differential diagnosis of GAD. 

Major Depressive Disorder and Dysthymic Disorder

Several symptom profiles discriminate between major depressive disorder or dysthymic disorder and GAD. Patients with major depressive disorder exhibit higher rates of dysphoric mood, psychomotor retardation, suicidal ideation, guilt, hopelessness and helplessness, as well as more work impairment than patients with GAD. In contrast, patients with GAD show higher rates of somatic symptoms, specifi cally, muscle tension and autonomic symptoms (e.g., respiratory or cardiac complaints) than depressed patients.

Panic Disorder With/Without Agoraphobia

Some researchers have suggested that GAD is attributable to panic disorder. However, clear differences exist between GAD and panic disorder. For example, panic disorder is characterized by the presence of panic attacks; that is, recurrent, discrete episodes of intense anxiety or fear associated with a cluster of somatic symptoms reflecting autonomic hyperactivity such as rapid heartbeat, dizziness, numbness or tingling, trouble breathing or choking, and nausea or vomiting. In contrast, patients with GAD experience predominantly symptoms of muscle tension and vigilance such as fatigue, muscle soreness, insomnia, difficulty concentrating, restlessness and irritability. Patients with panic disorder tend to seek treatment earlier in life than patients with GAD. Additionally, reports of types of worry differ between those diagnosed with GAD and those with panic disorder. For example, panic patients worry about having additional panic attacks, whereas GAD patients worry unrealistically about a number of everyday issues.

Obsessive–Compulsive Disorder

Anxiety is part of the clinical picture of obsessive–compulsive disorder (OCD) and may be a central factor in initiating and maintaining obsessions and compulsions. Interestingly there are also some data suggesting that OCD and GAD may be related. For example, Black and associates (1992) found an increased prevalence of GAD among relatives of patients with OCD. However, several features distinguish the excessive worry that accompanies GAD from the obsessional thoughts of OCD. Obsessive thoughts are described as ego-dystonic intrusions that often take the form of urges, impulses, or images. They are often senseless and are frequently accompanied by time-consuming compulsions designed to reduce mounting anxiety. In contrast, the worries in GAD are about realistic concerns, such as health and finances.

Other Anxiety Disorders

In phobic disorders, the anxiety is characteristically associated with a specific phobic object or situation that is frequently avoided by the patient. Such is the case with social anxiety disorder as well, in which the individual is afraid of or avoids situations in which he or she may be the focus of potential scrutiny by others. Anxiety is also a characteristic part of the presentation of post traumatic stress disorder (PTSD) and acute stress disorder. However, unlike in GAD, the principal symptoms experienced in PTSD and acute stress disorder follow exposure to a traumatic event and are characterized by avoidance of reminders of the event and persistent reexperiencing of the traumatic event. In addition, in contrast to GAD which must last at least 6 months, acute stress disorder does not persist for more than 4 weeks. Finally, in adjustment disorders anxiety when present occurs in response to a specific life stressor or stressors and generally does not persist for more than 6 months (American Psychiatric Association, 1994).

Normal Anxiety

Worry and anxiety are part of normal human behavior, and it may be difficult to define a cutoff point distinguishing normal or trait anxiety (i.e., a relatively stable tendency to perceive various situations as threatening) from GAD. However, individuals suffering from a “disorder” exhibit significant distress and impairment in functioning as a result of their anxiety symptoms.

Anxiety Disorder due to a General Medical Condition

Many general medical conditions may present with prominent anxiety symptoms. If not identified and properly addressed, these conditions may adversely affect the treatment outcome of the anxious patient. In this section we will highlight important medical conditions in the differential diagnosis of generalized anxiety.

Cardiovascular Disorders

Patients with GAD may complain of palpitations, skipped heartbeats and chest pain. In addition, many GAD patients, especially males, fear having an acute myocardial infarction and often present to the emergency room for evaluation. However, most patients with GAD without a concomitant cardiovascular disease do not experience severe chest pain. Following the controversial evidence suggesting an association between mitral valve prolapse (MVP) and panic disorder, researchers evaluated the prevalence of MVP in patients with GAD and found no evidence of increased prevalence in patients with GAD. Nevertheless, patients with anxiety symptoms associated with unexplained chest pain should be evaluated for possible cardiovascular disease.


Anxiety is a prominent feature of hyperthyroidism with some overlap in the symptomatology of thyrotoxicosis and GAD. Symptoms such as tachycardia, tremulousness, irritability, weakness and fatigue are common to both disorders. In GAD, however, the peripheral manifestations of excessive concentrations of circulating thyroid hormones are absent, including symptoms such as weight loss, increased appetite, warm and moist skin, heat intolerance and dyspnea on effort. Presence of goiter makes the diagnosis of hyperthyroidism likely; however, the absence of thyroid enlargement does not exclude it. Thus, confirmatory laboratory tests (free T4, T3 and TSH) assume significant diagnostic importance. In mild cases, laboratory tests may be within the upper limit of the normal range, in which case a thyroid releasing hormone stimulation test is indicated.


Pheochromocytomas, also known as chromaffin tumors, produce, store and secrete catecholamines. They are derived most often from the adrenal medulla, as well as the sympathetic ganglia and occasionally from other sites. The clinical features of these tumors, most commonly hypertension and hypertensive paroxysms, are predominantly due to the release of catecholamines. Patients may also experience diaphoresis (excessive sweating), tachycardia, chest pain, flushing, nausea and vomiting, headache and significant apprehension. Although the clinical presentation frequently mimics spontaneous panic attacks, pheochromocytomas should also be considered in the differential diagnosis of GAD. The diagnosis of pheochromocytoma can be confirmed by increased levels of catecholamines (epinephrine i.e. adrenaline and norepinephrine) or catecholamine metabolites (metanephrines and vanillylmandelic acid) in a 24-hour urine collection.

Other Medical Conditions

Menopause is commonly referred to as the period that encompasses the transition between the reproductive years and beyond the last episode of menstrual bleeding. Frequently associated with significant anxiety, menopause should be considered in the differential diagnosis of GAD. However, other associated symptoms such as vasomotor instability, atrophy of urogenital epithelium and skin, and osteoporosis make the diagnosis of menopause probable. Another endocrinologic disorder, hyperparathyroidism, can present with anxiety symptoms, and the initial evaluation of serum calcium levels may be indicated. Finally, certain neurologic conditions such as complex partial seizures, intracranial tumors and strokes, and cerebral ischemic attacks may be associated with symptoms typically observed in anxiety disorders and may require appropriate evaluation.

Substance-induced Anxiety Disorder

Anxiety disorders can occur frequently in association with intoxication and withdrawal from several classes of substances. Excessive use of caffeine, especially in children and adolescents, may cause significant anxiety. Cocaine intoxication may be associated with anxiety, agitation and hypervigilance. During cocaine withdrawal, patients may also present with prominent anxiety, irritability, insomnia, fatigue, depression and cocaine craving. Adverse reaction to marijuana includes extreme anxiety that usually lasts less than 24 hours. Mild opioid withdrawal presents with symptoms of anxiety and dysphoria. However, accompanying symptoms such as elevated blood pressure, tachycardia, pupilary dilatation, rhinorrhea, piloerection and lacrimation are rare in patients with GAD.

The clinical phenomenology observed both in alcohol and sedative–hypnotic drug withdrawal and in GAD, although variable, may be highly similar. In both conditions, nervousness, tachycardia, tremulousness, sweating, nausea and hyperventilation occur prominently. Additionally, the same drugs (i.e., benzodiazepines) can be used to treat anxiety symptoms, and some patients may use alcohol in an attempt to alleviate anxiety. Thus, the symptoms of an underlying anxiety disorder may be difficult to differentiate from the withdrawal symptoms associated with the use of benzodiazepines or alcohol.

The use of many commonly prescribed medications may produce side effects manifesting as anxiety. Such medications include sympathomimetics or other bronchodilators such as theophilline, anticholinergics, antiparkinsonian preparations, corticosteroids, thyroid supplements, oral contraceptives, antihypertensive, and cardiovascular medications such as digitalis, insulin (secondary to hypoglycemia), and antipsychotic and antidepressant medications. Finally, heavy metals and toxins such as organophosphates, paint and insecticides may also cause anxiety symptoms.

Epidemiology and Comorbidity

Current data indicate that GAD is probably one of the more common psychiatric disorders. The National Comorbidity Survey (NCS) of psychiatric disorders (Kessler et al., 1994) found prevalence rates of 1.6% for current GAD (defined as the most recent 6-month period of anxiety), 3.1% for 12-month GAD, and 5.1% for lifetime GAD, with lifetime prevalence higher in females (6.6%) than males (3.6%). The more recent National Comorbidity Survey replication (NCS-R) found a lifetime prevalence of 5.7% (Kessler et al., 2005).

GAD appears at even higher rates in clinical settings, particularly in primary care settings. For example, Shear and colleagues (1994) found prevalence rates of GAD, using DSM-III-R criteria, reported by patients at four primary care centers, to be twice as high as those reported in community samples (i.e., 10 versus 5.1%). Similarly, a collaborative study by the World Health Organization (WHO) across 15 international sites reported prevalence rates of GAD at approximately 8% in primary care settings (Maier et al., 2000).

Those with anxiety symptoms meeting criteria for GAD in the Epideiological Catchment Area (ECA) study reported receiving more outpatient mental services during the previous year than those diagnosed with other psychiatric disorders (Blazer et al., 1991). Many of those with GAD in the National Comorbidity Survey sought professional help for GAD (66% of participants) and used medications to reduce their symptoms of GAD (44% of participants). Over 80% of the GAD group in the Harvard/ Brown Anxiety Disorders Research Program (HARP) data indicated that they received psychotherapy and/or pharmacotherapy (Yonkers et al., 1996).

Rates of GAD appear similar in special populations such as children and the elderly. GAD appears to be less prevalent in children than in adults. Data from the NCS indicate that for both lifetime and 12-month prevalence rates, GAD occurs at the lowest rate in younger ages and at the highest rate in older adults Epidemiological data on prevalence of GAD in childhood using DSM-IV criteria are currently lacking. In the elderly, GAD appears to account for the majority of anxiety disorders, with prevalence rates ranging from 0.7 to 7.3%.

Although it is unclear whether childhood GAD predisposes to the development of adult GAD or represents an early manifestation of adult GAD, these studies further suggest that generalized anxiety is highly prevalent both in the community and in the clinical population.

GAD: Comorbidity with Other Disorders

Despite different methodological approaches in early studies, the available studies report a high prevalence of psychiatric comorbidity in patients with GAD. For example, in some studies more than 90% of GAD patients had additional symptoms that fulfi lled criteria for at least one or more concurrent disorders (range of 45–91%). An examination of the relative frequencies of various comorbid diagnoses in patients with GAD obtained from the available studies reveals that other anxiety and mood disorders frequently complicate the course of GAD.

The National Comorbidity Survey showed 90% of respondents with lifetime GAD had at least one other lifetime disorder and of those with current GAD, 66% had at least one other current disorder. The most common comorbidities (specifically that criteria for both disorders were met) were found for mood disorders (major depression and dysthymia), panic disorder and (for current comorbidity only) agoraphobia. High 12 month rates for comorbidity for GAD and major depression were reaffi rmed in the NCS-R (Kessler et al., 2005). Other studies have also found that the highest comorbidities were with depressive disorders and panic disorders. GAD usually has an earlier onset than other anxiety and depressive disorders when comorbid disorders are present. Brawman-Mintzer and associates (1993) found that GAD had an onset before dysthymia and panic disorder, and after simple and social phobia. Further, onset of major depression seemed to follow the onset of anxiety. Similar findings have been reported by other investigators.

As in adult GAD, childhood GAD (or overanxious anxiety disorder as it was earlier labeled) is also characterized by an unusual degree of comorbidity. Kashani and coworkers (1990) observed that over 50% of children with overanxious disorder had symptoms that met criteria for at least one additional psychiatric diagnosis. Among the most prevalent current comorbid diagnoses are social phobia (16–59%), simple phobia (21–55%), panic disorder (3–27%) and depression (8–39%). Furthermore, Masi and coworkers (1999) found, in those children and adolescents they sampled, that 87% had a comorbid disorder. In particular, high rates of separation anxiety, social anxiety and depressive disorders were found.

Alcoholism also complicates the clinical course of GAD for some patients; however, the available literature suggests that the diagnosis of alcohol abuse is not as prevalent in GAD as in other anxiety disorders, and the pattern of abuse is often a brief and nonpersistent one. GAD onset is usually later than that of the alcohol use disorder. Personality disorders have been observed to co-occur in approximately 50% of patients with GAD. For example, rates of GAD and personality disorders in clinical populations have ranged from 31 to 46%. Cluster C personality disorders, specifically avoidant personality disorder, dependent personality disorder and obsessive–compulsive personality disorder are common. Interestingly, Cluster A personality traits, in particular suspiciousness and mistrust, may be prominent in GAD as well.

Comorbid GAD is associated with increased severity of comorbid disorders (Kessler, 2000; Kessler et al., 2005). Additionally, the presence of comorbid disorders in GAD patients is related to increased rates of negative outcomes such as disability, impairment and cost of care. Rates of relapse for GAD patients with comorbid depression appear higher than in noncomorbid GAD patients. Further, comorbidity is also associated with greater treatment seeking. Unsurprisingly, data indicate that patients with comorbid GAD and depression may have poorer response to treatment than patients with either disorder only.


Retrospective and prospective reports indicate that the typical course of GAD is chronic, nonremitting, and that it often persists for a decade or longer. Rickels and colleagues (1986) reported that two-thirds of patients treated initially with diazepam relapsed within 1 year of discontinuation of treatment. Other studies utilizing criteria prior to the DSM-III-R for GAD found comparable levels of chronicity, with almost half the patients reporting moderate symptoms at follow-up (Yonkers et al., 2000).

HARP, a prospective, naturalistic study of 711 adults with DSM-III-R anxiety disorders, recruited initially from psychiatric clinics and hospitals in the Boston Metropolitan area, indicated that only 15% of those with GAD at baseline experienced a full remission for 2 months or longer at any time during the fi rst year after baseline, and only 25% had a full remission in the 2 years after baseline (Yonkers et al., 1996). Further, among patients who experienced full or partial remission, 27 and 39% respectively, experienced a full relapse during a 3-year follow-up period (Yonkers et al., 2000). Chronicity of GAD was also associated with cluster B and C personality disorders or a concurrent Axis I comorbidity. Wittchen and coworkers (1994) found that approximately 80% of subjects with GAD reported substantial interference with their life, a high degree of professional help seeking, and a high prevalence of taking medications because of their GAD symptoms. The disability associated with GAD was found to be similar to that found in individuals with panic disorder or major depression.

Treatment Approaches

GAD is a chronic, relapsing illness, which means that most treatments do not cure the patient. Furthermore, it also suggests that when treatments are discontinued, symptoms may return. It follows that a thorough understanding of the long-term benefits and risks associated with the different treatments available is important. Thus, each case must be considered individually according to the severity and chronicity of the disorder, the severity of somatic symptoms, the presence of stressors, and the presence of specific personality traits. The clinician may also need to work with the patient to determine how much improvement is sufficient. For example, a reduction in disability may occur without a marked change in symptoms. Symptoms may persist but occur less frequently, or their intensity may be reduced. All these variations have important treatment implications, including decisions regarding the need for long-term treatment. Patients with milder forms of GAD may respond well to simple psychological interventions, and require no medication treatment. In more severe forms of GAD, it may become necessary to see the patient regularly and to provide both more specific psychological and pharmacological interventions.

During the early (acute) phase of treatment, an attempt should be made to control the patient’s symptomatology. It may take 3 to 6 months for an optimal response to be achieved. However, there may be a considerable variation in the length of the initial treatment phase. For example, clinical response to benzodiazepines occurs early in treatment. Response to other anxiolytic medications or to cognitive–behavioral treatment generally requires longer periods of time. During the maintenance phase, treatment gains are consolidated. Unfortunately, studies suggesting how long treatment should be continued are limited. Routinely, pharmacological treatment is continued for a total of 6 to 12 months before attempting to discontinue medications. Recent data indicate that “maintenance” psychotherapeutic treatments such as cognitive–behavioral therapy may be helpful in maintaining treatment gains in patients with anxiety disorders following the discontinuation of pharmacotherapy. It is clear that many patients may experience chronic and continuous symptoms that require years of long-term treatment.

Doctor–Patient Relationship

The vast majority of patients with GAD who present for treatment have been ill for many years and frequently have received a variety of treatments. Some patients have been sent to psychiatrists for treatment as a “last resort” in order to learn how to cope with their various ill-defined somatic and emotional complaints. Patients may feel shame and guilt over their inability to control symptoms. They are often demoralized and angry, and feel that their symptoms are not taken seriously. Thus, it is important to help the patient understand their illness and to conceptualize it as a health problem rather than a “personal weakness”. Once the burden of perceived responsibility is lifted from the patient, and they believe that effective treatment of their illness is possible, a working alliance with the treating physician can begin. The treatment plan should be outlined clearly, and the patient cautioned that recovery may have a gradual, variable course. Finally, during the critical early stages of treatment, the clinician should make a special effort to be available in person or by phone to answer questions and provide support.


Below, we will discuss the use of various anxiolytic agents in the treatment of GAD.


Benzodiazepines are commonly used for the treatment of GAD and are still considered by some clinicians to be the first-line treatment for GAD. Several controlled studies have demonstrated the efficacy of different benzodiazepines such as diazepam, chlordiazepoxide and alprazolam in the treatment of GAD. The available placebo-controlled studies found that diazepam, alprazolam and lorazepam were effective in the treatment of GAD. The benzodiazepines have a broad spectrum of effects including sedation, muscle relaxation, anxiety reduction and decreased physiologic arousal (e.g. palpitations, tremulousness, etc.). Interestingly, available studies indicate that benzodiazepines have the most pronounced effect on hypervigilance and somatic symptoms of GAD, but exhibited fewer effects on psychic symptoms such as dysphoria, interpersonal sensitivity and obsessionality (Hoehn-Saric et al., 1988).

The main difference between individual benzodiazepines is potency and elimination half-life. These differences may have important treatment implications. For example, benzodiazepines with relatively short elimination half-lives such as alprazolam (range of 10–14 hours) may require dosing at least three to four times a day in order to avoid interdose symptom rebound. Conversely, the use of longer-acting compounds such as clonazepam (range of 20–50 hours) may minimize the risk of interdose symptom recurrence. In comparative studies of different benzodiazepines, alprazolam appeared to perform somewhat better than lorazepam. Data from the HARP study indicate that the most frequently reported medication used by GAD patients was alprazolam (31%), followed by clonazepam (23%) (Yonkers et al., 1996).

Benzodiazepines exert their therapeutic effects quickly, often after a single dose. However, concern has emerged over the use of benzodiazepines, particularly long-term benzodiazepine use. Side effects of benzodiazepines, such as sedation, psychomotor impairment and memory disruption, were noted by treating clinicians, and confirmed in research studies. Further, although it was suggested that the use pattern of benzodiazepines by patients with anxiety disorders may not represent abuse, addiction, or drug dependence as typically understood, the chronic use of benzodiazepines in the treatment of GAD has been increasingly discouraged in recent years.

When initiating treatment with benzodiazepines, it is helpful for patients to take an initial dose at home in the evening to see how it affects them. Gradual titration to an effective dose allows for limiting unwanted adverse effects. A final daily dosage of alprazolam between 2 and 4 mg/day, 1 and 2 mg/day for clonazepam, or 15 and 20 mg/day of diazepam is usually suffi cient for the majority of patients. Upon treatment discontinuation, it is important to consider appropriate taper in order to avoid withdrawal symptoms. Possible factors that may contribute to the severity of withdrawal and the ultimate outcome of benzodiazepine taper include the dosage, duration of treatment, the benzodiazepine elimination half-life and potency, and the rate of benzodiazepine taper (gradual versus abrupt). Additionally, patient factors such as premorbid personality features have been implicated. It appears that a taper rate of 25% per week is probably too rapid for many patients. We, therefore, recommend a slow benzodiazepine taper of at least 4 to 8 weeks, with the final 50% of the taper conducted even more gradually, with the patient decreasing by the lowest possible daily dose of the benzodiazepines during this period. We also recommend continuing to use divided doses of short to intermediate half-life benzodiazepines (alprazolam, lorazepam) during taper to minimize fluctuations in benzodiazepine levels over a 24-hour period, or using longer half-life benzodiazepines (such as clonazepam) which have the advantage of maintaining a once- or twice-daily dosing schedule.

Tricyclic Antidepressants

Clinical trials conducted in the early 1990s have confirmed that tricyclic antidepressants (TCAs) may also be effective in the treatment of GAD. These studies, as well as other trials, suggest that TCAs may be especially effective in the treatment of psychic symptoms of GAD (Brawman-Mintzer and Lydiard, 1994).The relationship between plasma levels of TCAs and their anxiolytic efficacy in patients with GAD has not been studied. Until this relationship is clarified, decisions regarding the total daily doses and the monitoring of plasma levels should be based on the patient’s treatment response and side-effects profile. Further, due to potential jitteriness, restlessness and agitation during the initial stages of treatment, we suggest that the initial dose of the TCAs in patients with GAD may need to be low (for example 10 mg/day of imipramine), and increased gradually.

Adverse effects commonly associated with the use of TCAs include anticholinergic effects (dry mouth, blurred vision, constipation), cardiovascular effects (orthostatic hypotension, slightly increased heart rate), sexual side effects and weight gain. As mentioned, patients may also experience significant jitteriness, restlessness and agitation during the initial stages of treatment. These side-effects often limit the acceptability of TCAs by many patients. Potential toxicity in overdose has been of concern to clinicians as well. Due, in part, to the side-effect profile, need for dose titration and importantly the emergence ofnew and effective agents (as described below), the use of TCAs in the treatment of GAD has been reserved for those resistant to these newer agents.

Serotonin Reuptake Inhibitors

Selective serotonin reuptake inhibitors (SSRIs) are rapidly becoming a key tool in the treatment of GAD. SSRIs are generally well tolerated. The most problematic side effect associated with SSRI use is interference with sexual function (e.g., delayed orgasm or abnormal ejaculation) in women and men. A variety of treatment strategies have been suggested for the management of SSRI-induced sexual dysfunction. Such strategies include waiting for tolerance to develop, dosage reduction, drug holidays and various augmentation strategies with 5-hydroxytryptamine-2 (5- HT2), 5-HT3 and alpha-2-adrenergic receptor antagonists, 5-HT1A and dopamine receptor agonists, and phosphodiesterase (PDE5) enzyme inhibitors.

Serotonergic and Noradrenergic Reuptake Inhibitors

The antidepressant venlafaxine extended release (XR) is an inhibitor of both 5-HT and NE reuptake (SNRI). Several large, placebo-controlled trials have evaluated it in the treatment of patients with DSM-IV-diagnosed GAD. As a result, venlafaxine XR was the first antidepressant approved by the FDA for the treatment of GAD. Results from two short-term studies indicate that venlafaxine XR (75, 150 and 225 mg/day) was significantly more effective than placebo and superior to buspirone on certain anxiety measures and in the prevention of relapse (Sheehan, 2001). The adverse events for GAD patients treated with venlafaxine XR resembled those in depression trials. The most common adverse events included nausea, somnolence, dry mouth, dizziness, sweating, constipation and anorexia.

Other Antidepressants

Both trazodone and imipramine have comparable efficacy to diazepam. In addition, trazodone and imipramine exhibit higher effi cacy in the treatment of psychic symptoms such as tension, apprehension and worry. The alpha-2-adrenoreceptor antagonist mitrazapine, which is also a 5-HT2, 5-HT3 and H(1) receptors antagonist, has been evaluated as a potential anxiolytic in the treatment of patients with major depressive disorder and comorbid GAD in an 8-week, open-label study (Goodnick et al., 1999). Results suggest that this antidepressant may be useful in the treatment of anxiety symptoms.


Buspirone hydrochloride, the only currently marketed azapirone, was the first non-benzodiazepine anxiolytic agent approved for the treatment of persistent anxiety by FDA. Results have been mixed about the efficacy of buspirone over placebo. For example, in four placebo-controlled studies that compared buspirone with a standard benzodiazepine, two showed no benefit for diazepam and buspirone over placebo, and two showed no benefit for buspirone over placebo. Benzodiazepines may also be slightly more effective than buspirone in the treatment of somatic symptoms of anxiety but no significant differences appear to exist between buspirone and benzodiazepines in measures of psychic anxiety (Rickels et al., 1997). Side effects most frequently associated with buspirone use included gastrointestinal system-related side effects, such as appetite disturbances and abdominal complaints, and dizziness. Prior use of benzodiazepines may adversely affect the therapeutic response to buspirone. DeMartinis and colleagues (2000) found that buspirone treatment was less effective for patients who had been taking benzodiazepines within 30 days of initiating buspirone treatment. Delle Chiaie and colleagues (1995) reported that a gradual 2-week taper of lorazepam with a simultaneous addition of buspirone for 6 weeks prevents the development of clinically significant rebound anxiety or benzodiazepine withdrawal. This approach was shown to provide clinically significant relief of anxiety symptoms in GAD patients previously treated with benzodiazepines for 8 to 14 weeks.

Perhaps the most significant problem with the use of buspirone has been that experts have advocated too low a dose to produce symptom reduction. In order to achieve optimal response, buspirone dosing in the range of at least 30 to 60 mg/day is currently recommended.

Other Agents

Hydroxyzine is a histamine-1 receptor blocker and a muscarinic receptor blocker. Recent controlled trials with the antihistamine have suggested that this compound may be effective in the acute treatment of GAD symptoms. Finally, the potential use of the anticonvulsant gabapentin in the treatment of anxiety symptoms has also been suggested.

Nonpharmacological Treatments

Numerous studies have shown that psychological interventions are beneficial in the comprehensive management of anxiety disorders. However, data suggesting that specific psychotherapeutic techniques yield better results in the treatment of patients with GAD are inconclusive and more evidence is needed on the comparative efficacy and long-term effects of different psychological treatments.

Cognitive–Behavioral Therapy

In recent years specific cognitive–behavioral therapy (CBT) interventions for the treatment of patients with anxiety disorders have been developed. Components of CBT include teaching patients to identify and label irrational thoughts and to replace them with positive self-statements or modify them by challenging their veracity. The cognitive modification approaches are combined with behavioral treatments such as exposure or relaxation training. There is currently evidence suggesting that CBT may be more effective in the treatment of GAD than other psychotherapeutic interventions, such as behavioral therapy alone or nonspecific supportive therapy (Chambless and Gillis, 1993). Six additional studies confi rmed the efficacy of CBT compared with waiting list or pill placebo and patients tend to maintain improvement following CBT over 6 to 12 months of follow-up.

CBT targeting intolerance of uncertainty, erroneous beliefs about worry, poor problem orientation and cognitive avoidance demonstrated effectiveness at post treatment (no change in the delayed treatment control group) 6- and 12-month follow-up, with 77% of the treatment group no longer having symptoms meeting criteria for a GAD diagnosis (Ladouceur et al., 2000).

Cognitive therapy was also compared with analytic psychotherapy, and was found to be significantly more effective (Borkovec and Costello, 1993). Overall, two-thirds in the cognitive therapy group achieved clinically significant improvements and cognitive therapy was associated with significant reductions in medication usage.

A meta-analytic review of controlled trials examining CBT and pharmacotherapy for GAD, which included 35 studies, demonstrated the robustness of CBT in the treatment of GAD (Gould et al., 1997). Overall, both modalities offered clear effi cacy to patients, with the effect size for CBT not being statistically different from psychopharmacological approaches. CBT demonstrated greater effects in reducing depression and was associated with clear maintenance of treatment gains, whereas long-term effi cacy of pharmacologic treatment was attenuated following medication discontinuation.

Barlow and colleagues (1986) developed a CBT approach to GAD which concentrates on the behavioral element of direct exposure to the contents of patients’ worry and apprehension (i.e., a deconditioning strategy) in addition to relaxation techniques (progressive muscle relaxation) and cognitive restructuring. The authors found that this technique is effective in reducing anxiety symptoms in patients with GAD.

Supportive Psychotherapy

Many patients with milder forms of GAD will benefit from simple psychological interventions such as supportive psychotherapy. They may experience lessening of anxiety when given the opportunity to discuss their diffi culties with a supportive clinician and to become better informed about their illness. Thus, basic supportive techniques such as reassurance, clarification of patient concerns, direct suggestions and advice are often effective in reducing anxiety symptoms.

Relaxation and Biofeedback

Relaxation techniques such as progressive muscle relaxation and biofeedback have also been utilized in the treatment of patients with anxiety symptoms. Few controlled studies have examined their effectiveness. In a recent controlled study, Borkovec and Costello (1993) compared a comprehensive relaxation treatment and cognitive–behavioral therapy in the treatment of patients with DSM-III-R-defined GAD. The authors found that both treatments were equally effective and superior to a nonspecific supportive treatment intervention.

Psychodynamic Psychotherapy

Although there are no controlled studies evaluating the efficacy of psychodynamic psychotherapies in the treatment of patients with GAD, some of its important principles in understanding patients may be helpful. First, it is important to note that the psychoanalytic theories view anxiety as an indicator of certain unconscious conflicts, rather than as a primary target symptom to be alleviated. It is, therefore, the clinician’s task to use various techniques to help the patient uncover these unconscious conflicts. It is believed that the newly gained understanding of the underlying reasons for symptoms will have a therapeutic effect, thereby reducing anxiety. Through interpretation of previously unconscious conflicts and unconscious origins of anxiety, the patient will be able to utilize new insights and find more adaptive outlets or solutions to problems.

In Harry Stack Sullivan’s theory, anxiety reflects the failure to develop secure interpersonal interactions, such as an emphatic and secure mother–infant relationship. He believed that the child learns to identify anxiety states in himself and signifi cant others and develops protective defensive strategies which enable him to avoid experiencing anxiety. However, the defenses employed (termed security operations) are generally restrictive and may result in limiting the subject’s interpersonal interactions. Therefore, the task of the therapist, according to Sullivan’s model, is to trace the patterns of interpersonal interactions throughout the patient’s developmental stages (rather than to uncover the unconscious drives), thereby promoting a more accurate perception of self and others and subsequently better social adaptation.

Another therapeutic approach to the treatment of anxiety symptoms was offered by object relation and self-psychology theorists. They view anxiety as a result of the loss of or inadequate emotional relationships with significant others. Therefore, the primary focus of therapy shifts to emphasize the importance of the relationship to the therapist, who functions as an empathic object providing emotionally corrective experiences. For example, the patient may learn that an important person may be imperfect but still be trusted and nurturing.

Finally, most psychodynamically oriented therapists agree that the outcome of psychodynamic psychotherapy is determined by factors reflecting a patient’s maturity and strength. Specifically factors such as the patient’s capacity for introspection, intelligence, ability to relate to the therapist, and ability to bear painful feelings should be carefully evaluated.

Case study 

RJ is a 27-year-old graduate student who presented to his primary care physician with complaints of recurrent frontal headaches. He asked his physician to order a CT scan or an MRI because he feared that this may be a sign of a serious and a life-threatening illness. However, the headaches appeared to worsen during times of stress, and abated when the patient was on vacation with a friend. During the interview the patient admitted that he is very worried about the quality of his research at the university, even though he had received consistently positive evaluations. He attributed the perceived academic difficulties to his forgetfulness and diminished ability to concentrate during the last 6 months. He also described feeling fatigued, as well as significant indigestion, abdominal pain associated with bouts of changes in bowel movements (alternating diarrhea and constipation). For the past 6 months he had few social contacts because of his symptoms and his girlfriend demanded that he seek professional help. Following complete physical evaluation, including consultation with a gastroenterologist, a diagnosis of generalized anxiety disorder and irritable bowel syndrome was made.

Long-term Management of Generalized Anxiety Disorder

As mentioned, GAD is a chronic, continuous condition in the majority of patients. Frequently beginning in adolescence or early adulthood, the course of GAD can persist for decades, with relatively low remission rates. The HARP data indicate that among the 164 GAD patients followed, only 15% had a full remission for 2 months or longer at any time during the first follow-up year, 25% had a full remission during the 2-year follow-up, and only 35% had a full remission after 5 years (Yonkers et al., 1996). In a 5-year follow-up study of 64 GAD patients, only 18% of GAD subjects achieved a full remission compared with 45% of panic disorder subjects (Woodman et al., 1999).

Despite these findings, less research has been conducted to assess the efficacy of chronic long-term anxiolytic therapy. To our knowledge, only three double-blind, controlled studies evaluating the long-term pharmacological treatment of GAD have been conducted. The first, a 1-year follow-up of patients who participated in a 6-month diazepam maintenance study, found that two-thirds of all patients relapsed within a 1-year period after diazepam discontinuation. In the second study, GAD patients were treated for 6 months with buspirone or clorazepate, and reevaluated them after discontinuation at 6 and 40 months. The authors found that the improvement achieved in both treatment groups was sustained during the 6-month maintenance phase, with no need for an increase in medication intake, and no evidence of tolerance or abuse. At follow-up, approximately 60% of patients treated initially with clorazepate compared with 30% of patients treated initially with buspirone were experiencing at least moderate anxiety symptoms. Finally, in a recent study, Stocchi and colleagues (2001) found that many patients with GAD are not able to remain symptom-free for long periods without treatment. Therefore, a long-term therapy may be needed in many patients.

Generally, the current recommendation for GAD treatment suggests a treatment period of approximately 1 year after response has been established prior to considering treatment discontinuation. Stress management and problem-solving techniques, along with specific psychotherapeutic approaches such as cognitive–behavioral therapy, should be attempted in addition to medication treatment.

Comorbidity and the Treatment of GAD

GAD is often accompanied by other concurrent psychiatric disorders, specifically anxiety and mood disorders. The presence of these comorbid conditions may reflect more severe loading for psychopathology, and may have important implications on the course and treatment response of the primary disorder. In the National Comorbidity Survey, more patients with comorbidity experienced interference with daily activities than did patients with pure GAD (Wittchen et al., 1994). The presence of a comorbid anxiety disorder and major depressive disorder is frequently associated with a poorer overall outcome than for patients with a single psychiatric disorder (Kessler et al., 1999). Currently, there are treatment options that can target both GAD and major depressive disorder simultaneously. The use of SSRIs and the SNRI venlafaxine is recommended as the first-line treatment for comorbid GAD and depression.

Social anxiety disorder frequently complicates the course GAD takes. With the recent data indicating that the SSRIs such as sertraline and paroxetine are effective in the treatment of social anxiety, these agents may be useful in the treatment of comorbid social anxiety disorder and GAD. The benzodiazepines that have been established as effective in the treatment of patients with GAD also appear to be effective in the treatment of patients with social anxiety disorder. Thus, these agents (considering the caveats associated with their use described earlier) may have a therapeutic role in patients with GAD and coexisting social phobia. Finally, patients with GAD and concurrent panic disorder or panic attacks may be effectively treated with SSRIs, TCAs and benzodiazepines. However, buspirone is probably ineffective in the treatment of panic disorder.

Concurrent alcohol and substance abuse tend to confuse the clinical picture of GAD and can interfere with the therapeutic efforts. Additionally, symptoms associated with alcohol withdrawal or other sedative–hypnotics may mimic the underlying anxiety disorder. If a substance abuse problem exists, the clinician and the patient should take the necessary steps to discontinue the use of the abused substance. This may well include specific substance abuse treatment. Specifically, the need for detoxification should be assessed and discussed with the patient. Following cessation of substance abuse the patient’s symptomatology should be reevaluated. The use of benzodiazepines in these patients may be contraindicated, and alternative treatments with SSRIs, SNRIs, TCAs, buspirone, or gabapentin may be needed.

Treatment of GAD in the Elderly

Epidemiological data suggests that GAD is highly prevalent in the geriatric population (prevalence rates ranging from 0.7 to 7.1%), accounting for the majority of anxiety disorder cases in this group. GAD is the most common of the pervasive late-life anxiety disorders. In the elderly, anxiety symptoms are often associated with depression, medical conditions and cognitive dysfunction. Thus, a careful differential diagnosis to eliminate exogenous causes of anxiety and identification of other coexisting conditions is necessary. For example, treatment of medical illness, depression, or underlying dementia may reduce anxiety symptoms. Dose reductions or elimination of anxiety-inducing medications as well as reducing stressful life circumstances may also reduce anxiety symptoms. However, if these interventions are not effective in reducing anxiety, pharmacotherapy may be necessary. Several factors influencing pharmacologic treatment in the elderly should be considered. These factors include alterations in pharmacokinetics and pharmacodynamics of psychotropic drugs, primarily because of reduced hepatic clearing efficiency, alterations in the response of the central nervous system to drugs, such as changes in receptor sensitivity, and concurrent medical conditions that may alter drug effect, side-effect profile, and toxicity.

Benzodiazepines can be effective in the treatment of anxiety symptoms. However, older patients are often sensitive to their effects. Adverse effects may include increased sedation, tendency to fall, psychomotor discoordination and cognitive impairment. Older patients may become disinhibited by benzodiazepines and experience agitation and aggression. The administration of long acting benzodiazepines such as diazepam and chlorazepate may result in increased accumulation of the drug predisposing the patient to these side effects. Conversely, the use of short half life high potency benzodiazepines such as alprazolam may be associated with more severe withdrawal symptoms following rapid discontinuation. Because of these factors, benzodiazepines should be prescribed for the briefest period of time, at the lowest therapeutic dose, giving preference for the short half-life, low potency benzodiazepines such as oxazepam. We recommend initiating treatment with oxazepam at low doses (10 mg t.i.d.), to be increased gradually, while carefully monitoring for the emergence of side effects.

Buspirone has been extensively used in the treatment of GAD symptoms. The lack of associated sedation, discoordination and dependence with the use of buspirone makes its use in the elderly less problematic. However, additional research is needed to determine its long-term efficacy in the GAD elderly population. The average therapeutic doses of buspirone for elderly patients range from 5 to 20 mg/day.

The use of TCAs in the anxious elderly patient should be viewed in light of the side-effect profile of TCAs. Side effects commonly associated with the use of TCAs, such as the anticholinergic effects and orthostatic hypotension, may be especially troublesome in these patients. We therefore recommend the use of TCAs with low anticholinergic and hypotensive effects such as desipramine and nortryptiline, starting at low doses (10 mg/day) that are raised slowly and gradually.

Finally, despite the widespread use of the newer antidepressant agents, specifically the SSRIs and the SNRIs, in the treatment of adult GAD patients, very limited data exist regarding their use in the anxious elderly population. However, preliminary evidence suggests that they can decrease symptoms, improve quality of life and potentially promote healthier outcomes in geriatric patients who have comorbid anxiety and depression and/or comorbid mental and physical illness. A potential drawback of venlafaxine in this population is the need to monitor for drug-induced blood pressure elevation in those taking the medication.

Most controlled studies examining CBT in older adults have focused on the treatment of GAD. This literature suggests that CBT is effective in the treatment of GAD in this population. For example, group-administered CBT was found to be effective in reducing GAD and coexistent symptoms in older adults. In conclusion, several agents may play an important role in the treatment of anxiety in the elderly. However, until more studies in the elderly GAD population are available, treatment choices should be guided by clinical judgment and specific factors relevant to this patient population, such as medical comorbidity and age-associated changes in the drug metabolism.

Treatment of GAD in Adolescents and Children

The treatment of GAD in adolescents and children should be multifaceted, including psychotherapy for the patient, psychoeducation for the parents, and pharmacotherapy when other psychotherapeutic interventions have not produced satisfactory outcome. CBT is successful in treating GAD in children. The addition of family anxiety management skills taught to parents appears to increase treatment success.

Pharmacotherapy in children and adolescents differs from that of the adult population primarily because of the difference in the hepatic biotransformation and elimination of many psychotropic drugs that may require some adjustments in treatment regimen. Hepatic metabolic rate is faster in children and adolescents than in adults, reaching adult values around 15 years of age. Thus a particular milligram per kilogram (mg/kg) dose will yield a lower blood level in a child than in an adult, and higher mg/kg doses than based on those for adults may be necessary. This applies for all liver-metabolized drugs, such as antidepressants, anxiolytics, anticonvulsants and neuroleptics. In addition, the higher clearance of these drugs requires more frequent administration of medications (i.e., small divided doses rather than one large dose).

Over the years, a number of medications have been used in the treatment of childhood GAD (previously classified as overanxious anxiety disorder). Unfortunately, only a few studies have been conducted in children with overanxious anxiety disorder. Given these limited findings and the reported occurrence of significant behavioral activation, other side effects, and their addictive potential, the use of benzodazepines to treat children with GAD is suspect.

The use of SSRIs in the treatment of GAD in children and adolescents appears promising; however, no controlled trials with large samples have been conducted. Buspirone may also be effective in the treatment of GAD in children and adolescents. No controlled studies have been completed for the treatment of GAD with buspirone in children to date. TCAs have not been studied in overanxious anxiety disorder. It should be noted that cardiovascular side effects may occur more frequently in children and adolescents than in adults, and preexisting conduction abnormalities may be associated with significant TCA effect on cardiac conduction. However, the clinical significance of TCA-induced changes in cardiac conduction is not clear. Given this lack of efficacy data for GAD and the concern of significant cardiovascular side effects in children, TCAs should not be a first-line treatment in children and adolescents with GAD.

Treatment Resistance

The clinician is frequently faced with a patient whose anxiety symptomatology is not responding satisfactorily to the standard treatment. Different factors, such as inadequate length of treatment, low dose and noncompliance, may contribute to treatment failure in the management of patients with GAD. Pharmacologic treatment is often complicated by the occurrence of side effects, which may impair quality of life, deter clinicians from prescribing adequate doses and contribute to noncompliance. For example, some antidepressants, including SSRIs, TCAs and venlafaxine, are associated with activation, overstimulation, or “jitters” primarily during the initial stage of treatment. When evaluating noncompliance, clinicians should also assess for akathisia and worsening of anxiety and hypomania or mania. Further, the presence of comorbid general medical and psychiatric conditions in GAD patients may be associated with nonresponse or lower response rates and should be carefully assessed during patient evaluation. Finally, the use of concurrent medications that can precipitate anxiety symptoms may affect the response to treatment.

The clinician should always evaluate whether an adequate treatment trial was complete. We believe that an attempt should be made to maintain the patient on medication for at least 6 weeks. Although there are no data suggesting that certain doses may be particularly effective in the treatment of GAD, it is advisable to titrate the medication up to maximally tolerated doses prior to discontinuing the medication for nonresponse. It is important to inquire about the presence of side effects such as sedation, anticholinergic effects, or sexual side effects, which may limit the attainment of a therapeutic dosage and reduce compliance. Additionally, many patients with GAD fear that they may become “drug-dependent” and thus avoid dose increases. Some estimate of the patient’s compliance may be helpful in determining whether a treatment was adequate, as indicated by blood plasma levels or pill counts. Drug plasma levels may also be useful to identify patients who are rapid metabolizers. A careful evaluation for the presence of psychiatric comorbid conditions that may contribute to treatment refractoriness should follow. As mentioned, comorbidity which may reflect more severe loading for psychopathology is often associated with increased severity of illness and poorer response to treatment in comparison to patients with an uncomplicated (i.e., single) disorder. Thus, treatment strategies in GAD patients with a concurrent disorder may differ from those in an uncomplicated disorder, often requiring multiple drug therapy. The clinician should also be alert to the presence of underlying general medical conditions such as hyperthyroidism which may present with refractory anxiety, or conditions/medications which may alter the effects of treatment such as hepatic disease or medications (e.g., steroids) that affect hepatic clearance.

The use of psychotherapy, such as cognitive–behavioral therapy, in conjunction with pharmacotherapy may also enhance response in the treatment-resistant patient. Finally, education and psychological support for patients and their families may help them better to understand and deal with their illness, especially during periods of increased stress, and consequently improve treatment outcome.


In conclusion, GAD appears to be a chronic, frequently comorbid condition, often requiring long-term management. Benzodiazepines were, at one time, the first-line treatment choice for GAD. However, newer medication may offer more hope for those with GAD. SSRIs and SNRIs are promising new pharmocotherapies for GAD. Additionally, these treatments may be especially effective for patients presenting with depressive symptomatology. Further, buspirone and TCAs have been shown to be an important treatment alternative to benzodiazepines, and the TCAs may be especially effective in patients presenting with depressive symptomatology. Finally psychotherapy, specifically cognitive–behavioral therapy, is an additional treatment strategy, potentially effective in maintaining treatment gains. Further information is needed on long-term treatment, medication discontinuation and comparative efficacy of pharmacotherapies, psychotherapies and combination in the treatment of uncomplicated and comorbid GAD.

Comparison of DSM-IV/ICD-10 Diagnostic Criteria

The ICD-10 Diagnostic Criteria for Research specify that four symptoms from a list of 22 be present. In contrast, DSM-IV-TR requires three out of a list of six (of which five are included among the ICD-10 list of 22).


American Psychiatric Association (1994) Diagnostic and Statistical Manual of Mental Disorders, 4th edn. APA, Washington DC.

Barlow DH, Blanchard RB, Vermilyea JB et al. (1986) Generalized anxiety and generalized anxiety disorder: Description and reconceptualization. Am J Psychiatr 143, 40–44.

Black DW, Noyes R, Goldstein RB et al. (1992) A family study of obsessive– compulsive disorder. Arch Gen Psychiatr 49, 362–368.

Blazer DG, Hughes D, George LK et al. (1991) Generalized anxiety disorder, in Psychiatric Disorders in America: The Epidemiologic Catchment Area Study (eds Robins LN and Regier DA). Free Press, New York, pp. 180–203.

Borkovec TD and Costello E (1993) Efficacy of applied relaxation and cognitive–behavioral therapy in the treatment of generalized anxiety disorder. J Consult Clin Psychol 61(4), 611–619.

Brawman-Mintzer O, Lydiard RB, Emmanuel N et al. (1993) Psychiatric comorbidity in patients with generalized anxiety disorder. Am J Psychiatr 150, 1216–1218.

Brawman-Mintzer O, Lydiard RB (1994) Psychopharmacology of anxiety disorders. Psychiatr Clin N AM 1, 51–79.

Carter CS and Maddock RJ (1992) Chest pain in generalized anxiety disorder. Int J Psychiatr Med 22(3), 291–298.

Chambless DL and Gillis MM (1993) Cognitive therapy of anxiety disorders. J Clin Consult Psychol 61(2), 248–260.

Delle Chiaie R, Pancheri P, Casacchia M et al. (1995) Assessment of the efficacy of buspirone in patients affected by generalized anxiety disorder, shifting buspirone from prior treatment with lorazepam: A placebo-controlled, double-blind study. J Clin Psychopharmacol 15(1), 12–19.

DeMartinis N, Rynn M, Rickels K et al. (2000) Prior benzodiazepine use and buspirone response in the treatment of generalized anxiety disorder. J Clin Psychiatr 61(2), 91–94.

Goodnick PJ, Puig A, DeVane CL et al. (1999) Mirtazapine in major depression with comorbid generalized anxiety disorder. J Clin Psychiatr 60, 446–448.

Gould RA, Otto MA, Pollack MH, et al. (1997) Cognitive–behavioral and pharmacological treatment of generalized anxiety disorder: A preliminary meta-analysis. Behav Ther 28(2), 285–305.

Gray JA (1988) The neuropsychological basis of anxiety. In Handbook of Anxiety Disorders, Last CG and Hersen M (eds). Pergamon Press, New York, pp. 10–37.

Hoehn-Saric R, McLeod DR and Zimmerli WD (1988) Differential effects of alprazolam and imipramine in generalized anxiety disorder: Somatic versus psychic symptoms. J Clin Psychiatr 49, 293–301.

Kashani JH, Vaidya AF and Soltys SM (1990) Correlates of anxiety in psychiatrically hospitalized children and their parents. Am J Psychiatr 147, 319–323.

Kendler KS, Neale MC, Kessler RC et al. (1992) Generalized anxiety disorder in women. Arch Gen Psychiatr 49, 267–272.

Kessler RC (2000) The epidemiology of pure and comorbid generalized anxiety disorder: A review and evaluation of recent research. Acta Psychiatr Scand 406(Suppl), 7–13.

Kessler RC, McGonagle KA, Zhao S et al. (1994) Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: Results from the National Comorbidity Survey. Arch Gen Psychiatr 51, 8–19.

Kessler RC, DuPont RL, Berglund P et al. (1999) Impairment in pure and comorbid generalized anxiety disorder and major depression at 12 months in two national surveys. Am J Psychiatr 156, 1915–1923.

Kessler RC, Berglund MBA, Demler O et al. (2005) Lifetime prevalence and age-of-onset distributions of DSM-IV disorders in the National Comorbidity Survey Replication. Arch Gen Psychiatr 62, 593–602.

Ladouceur R, Dugas MJ, Freeston MH et al. (2000) Efficacy of a cognitive–behavioral treatment for generalized anxiety disorder: Evaluation in a controlled clinical trial. J Clin Consult Psychol 68(6), 957–964.

Lesch KP (2001) Genetic dissection of anxiety and related disorders, in Anxiety Disorders (eds Nutt D and Ballenger J). Blackwell Science, Oxford.

Lorberbaum JP, Varon D, Brawman-Minzter O et al. (2001) The functional neuroanatomy of anticipatory anxiety in healthy adults and patients with generalized anxiety disorder. Presented at the 21st National Conference of Anxiety Disorders Association of America, Atlanta.

Maier W, Gaeniscke M, Freyberger HJ et al. (2000) Generalized anxiety disorder (ICD-10) in primary care from a cross-cultural perspective: A valid diagnostic entity? Acta Psychiatr Scand 101, 29–36.

Masi G, Mucci M, Favilla L et al. (1999) Symptomatology and comorbidity of generalized anxiety disorder in children and adolescents. Compr Psychiatr 40, 210–215.

Papadimitriou GN, Kerkhofs M, Kempenaers C et al. (1988) EEG sleep studies in patients with generalized anxiety disorder. Psychiatr Res 26, 183–190.

Rickels K, Case WG, Downing RW et al. (1986) One-year follow-up of anxious patients treated with diazepam. J Clin Psychopharmacol 6, 32–36.

Rickels K, Schweizer E, DeMartinis N et al. (1997) Gepirone and diazepam in generalized anxiety disorder: A placebo-controlled trial. J Clin Psychopharmacol 17(4), 272–277.

Shear MK, Schulberg HC and Madonia M (1994) Panic and generalized anxiety disorder in primary care. Paper presented at a meeting of the Association for Primary Care, Washington DC.

Sheehan DV (2001) Attaining remission in generalized anxiety disorder: Venlafaxine extended release comparative data. J Clin Psychiatr 62(9S), 26–31.

Skre I, Torgersen S, Lygren S et al. (1993) A twin study of DSM-III-R anxiety disorders. Acta Psychiatr Scand 88, 85–92.

Stocchi F, Jokinen R and Lepola U (2001) Efficacy and tolerability of paroxetine for long-term treatment of generalized anxiety disorder. Presented at 2001 Annual Meeting of the American Psychiatric Association (May 5–10), New Orleans.

Wittchen HU, Zhao S, Kessler RC et al. (1994) DSM-III-R generalized anxiety disorder in the National Comorbidity Survey. Arch Gen Psychiatr 51, 355–364.

Woodman CL, Noyes R, Black DW et al. (1999) A 5-year follow-up study of generalized anxiety disorder and panic disorder. J Nerv Ment Disord 187, 3–9.

Wu JC, Buchsbaum MS, Hershey TG et al. (1991) PET in generalized anxiety disorder. Biol Psychiatr 29, 1181–1199.

Yonkers KA, Dyck IR, Warshaw M et al. (2000) Factors predicting the clinical course of generalized anxiety disorder. Br J Psychiatr 176, 544–549.

Yonkers KA, Massion A, Warshaw M et al. (1996) Phenomenology and course of generalized anxiety disorder. Br J Psychiatr 168, 308–313.