Basic Principles and Techniques of Cognitive-Behavioral Therapies for Comorbid Psychiatric Symptoms Among Headache Patients


  • Gay L. Lipchik PhD,

  • Todd A. Smitherman PhD,

  • Donald B. Penzien PhD,

  • Kenneth A. Holroyd PhD

  • From the Saint Vincent Health Psychology Services, Erie, PA (Dr. Lipchik); Head Pain Center, Department of Psychiatry and Human Behavior, University of Mississippi Medical Center, Jackson, MS (Drs. Smitherman and Penzien); and Psychology Department, Ohio University, Athens, OH (Holroyd).

Address all correspondence to Gay L. Lipchik, PhD, Saint Vincent Health Psychology Services, 145 West 23rd Street, Suite 202, Erie, PA 16502.


Recent research on headache has focused on identifying the prevalence of psychiatric disorders in headache patients and discerning the impact of psychiatric comorbidity on treatment of headache. The presence of comorbid psychiatric disorders, especially anxiety and depression, in headache patients is now a well-documented phenomenon. Existing but limited empirical data suggest that psychiatric comorbidity exacerbates headache and negatively impacts treatment of headache. Problematically, these findings have not yet eventuated in improved treatments for individuals suffering from both headache and a psychiatric disorder(s). The present article is an attempt to describe the application of cognitive-behavioral therapies (CBT) for depressive and anxiety disorders to headache patients who present with psychiatric comorbidity. We discuss the origins of the chronic care model in relation to CBT, review basic cognitive-behavioral principles in treating depression and anxiety, and offer clinical recommendations for integrating CBT into existing headache treatment protocols. Directions for future research are outlined, including the need for treatment outcome studies that examine the effects of treating comorbid psychiatric disorders on headache (and vice versa) and the feasibility of developing an integrated CBT protocol that addresses both conditions simultaneously.


cognitive-behavioral therapy


Diagnostic and Statistical Manual of Mental Disorders (4th ed.)




Mini-International Neuropsychiatric Interview


obsessive-compulsive disorder


Primary Care Evaluation of Mental Disorders


posttraumatic stress disorder

Although most individuals with headache in the general population do not have comorbid psychiatric disorders, psychiatric comorbidity is relatively commonplace among headache patients who present for treatment, especially in specialty centers.1,2 Patients with migraine or tension-type headache exhibit psychiatric illness at a disproportionately higher rate than individuals with no headache history.3 Depressive disorders occur with at least a threefold greater frequency among migraineurs than in the general population. This prevalence increases further in clinical populations, especially among those with chronic daily headache or with medication overuse headache.4,5 Similarly, anxiety disorders (eg, generalized anxiety disorder and panic disorder) occur with a fivefold greater frequency in migraineurs than in the general population.3 Migraine with comorbid depression is often complicated by the presence of an anxiety disorder, with the onset of anxiety often preceding the onset of migraine and possibly present as early as childhood.6,7

Clinical wisdom suggests that the presence of a comorbid psychiatric disorder portends a poorer prognosis; however, empirical data are limited. A recent longitudinal study suggests that the presence of psychiatric disorders (including sleep disorders) may predict a poorer outcome for headache treatment, especially if multiple psychiatric disorders are present.8 In this 8-year prospective follow-up study of 100 young patients with headache, Guidetti and colleagues found that 86% of those with headache who had 2 or more comorbid psychiatric disorders in childhood or adolescence had either no improvement or deterioration in their headache condition (migraine or tension-type headache).8 In 62% of patients with a single psychiatric disorder in childhood or adolescence, migraine remained unchanged or worsened. In contrast, the absolute absence of psychiatric disorders was associated with remission of headaches after 8 years. Psychological distress is also associated with the progression from episodic to chronic and daily headache9–11 and may play a greater role in the transformation and chronification of headache than does analgesic overuse/abuse.12 A recent study of migraine suggests that the presence of a comorbid psychiatric disorder may be associated with long-term (6 to 7 years) relapse, even if it is not associated with a poor initial response to drug treatment.13 In sum, there is limited evidence to suggest that psychiatric comorbidity predicts a poorer headache prognosis; however, the study samples have been small.

The exact nature of the relationship between psychiatric disorders and recurrent headache disorders is unclear. Psychiatric comorbidity is a risk factor for nonadherence with medical treatment,14–16 and this nonadherence could play a significant role in the poorer outcome of headache patients with psychiatric comorbidity. Additionally, depression and anxiety may contribute to head pain through corticolimbic sensitization in which limbic activation may lead to a complicated intertwining and progression of chronic pain and affective distress.17

The negative prognosis associated with psychiatric comorbidity emphasizes the importance of the identification of psychopathology among those with headache beginning at an early age and suggests that the treatment of psychiatric comorbidity is necessary to improve the outcome of headache management. Unfortunately, to date, there is no cogent evidence that treating depression or anxiety improves headache outcomes. However, differential headache outcomes have been observed with treatments such as antidepressant medications and cognitive-behavioral stress management, modalities that would be presumed to impact psychological or psychiatric processes as well as headache. For example, Holroyd and colleagues evaluated the separate and combined effects of cognitive-behavioral therapy (CBT) and tricyclic antidepressant medication for the treatment of chronic tension-type headache in a large randomized, placebo-controlled trial.18 Combined therapy was more likely to produce clinically significant (≥50%) reductions in headache (64% of participants) than antidepressant medication alone (38% of participants), stress management alone (35% of participants), or placebo (29% of participants). Planned moderator analysis examined the influence of baseline headache severity and psychiatric comorbidity (anxiety or mood disorder) on treatment outcome.19 Psychiatric comorbidity had no influence on improvements in headache activity observed with the 3 active treatments or with placebo. However, psychiatric comorbidity influenced the effectiveness of treatments in reducing headache-related disability (as assessed by the Headache Disability Inventory). When no psychiatric comorbidity was present, no placebo effect was observed, and all 3 active treatments produced similar substantial reductions in headache-related disability. However, in the presence of psychiatric comorbidity, notable improvements were observed with placebo; in fact, improvements observed with placebo were similar in magnitude to improvements observed with antidepressant medication alone, and with CBT alone. Only the combined treatment was superior to placebo.

In short, the limited existing empirical data suggest that psychiatric disorders negatively impact headache prognosis. The lack of evidence indicating that treating psychiatric comorbidity improves headache outcomes is not a function of repeated null findings; rather, there is simply a paucity of research on this topic. Accordingly, clinical experience and related empirical findings suggest that identification and treatment of comorbid psychiatric conditions are warranted, as psychiatric disorders add impairment and have negative consequences in their own right. Below we describe how CBT fits within a chronic care model of disease, review basic principles of screening for psychiatric disorders, and discuss the application of CBT for treating comorbid psychiatric disorders in headache patients.


In the past 50 years, chronic disease has overtaken acute illness as the primary cause of morbidity and mortality.20 Chronic conditions—illnesses that last longer than 3 months and are not self-limiting (including headache and psychiatric disorders)—affect almost half of the US population. Unfortunately, the conventional medical model is designed for the treatment of acute illness, which emphasizes the diagnosis of potentially serious conditions and the goal of relieving acute symptoms. This acute disease model leaves little time for patient education or adequate follow-up of patients with chronic illnesses in order to monitor and improve their adherence to recommended medical regimens, or to measure the outcome of this management. This results in inadequate treatment for patients with chronic illnesses.21

Health care for many chronic illnesses is shifting toward a chronic care model that emphasizes collaborative care and self-management, which are based on behavioral principles and typically employ cognitive-behavioral interventions.21–24 The essential elements of self-management specifically for chronic headache and other pain disorders have been formulated.25,26 Collaborative care emphasizes a patient-provider relationship that focuses on shared decision making. This partnership paradigm implies that while professionals are experts about diseases, patients are experts about their own lives. The self-management model recognizes that the day-to-day responsibilities of chronic illness fall most heavily on patients and their families, and provides them with patient education that teaches problem solving and coping skills to manage their illness (eg, optimize medication adherence, identify/respond to triggers, make/maintain lifestyle changes, improve functioning/limit disability, cope with affective distress associated with chronic illness, improve stress management). Self-management complements traditional patient education and allows patients to identify their problems and provides techniques to help patients make decisions, take appropriate actions, and alter these actions as they encounter changes in circumstances or disease.27

Evidence from controlled trials suggests that self-management skills improve clinical outcomes and reduce health-care costs in a variety of chronic illnesses, such as asthma and arthritis.28 The chronic care self-management approach also has been applied successfully to the treatment of psychiatric illnesses such as major depression, panic disorder, and generalized anxiety disorders in medical settings.29–32 These collaborative self-management interventions for psychiatric illness have demonstrated not only positive treatment results, but also cost-effectiveness and cost offset, as well as improvement in employment and work outcomes.32,33


Because of the high rates of psychiatric comorbidity in headache patients, it is highly recommended that all headache patients be screened, at a minimum, for depression and anxiety.1,34–36 Among the most useful screening instruments for the major psychiatric disorders that are well suited to primary care settings are the Mini-International Neuropsychiatric Interview (MINI)37 and the Primary Care Evaluation of Mental Disorders (PRIME-MD),38 both of which were developed for use with medical populations. Two brief screening measures that patients can quickly complete in the waiting room are the 7-item Beck Depression Inventory for Primary Care39 and the 21-item Beck Anxiety Inventory.40 Other useful broad-based and disorder-specific screening measures are discussed by Maizels, Smitherman, and Penzien in this supplement.35 In screening for psychiatric disorders, clinicians must be aware of the overlap of somatic symptoms associated with headache, depression, and anxiety disorders in order to avoid inappropriate or incorrect psychiatric diagnoses. Using somatic symptoms such as insomnia, fatigue, or difficulty concentrating as indicators of depression in headache patients may confound the assessment of depression with headache symptoms and severity.41 Similarly, muscle tension, irritability, and sleep disturbance are commonly reported symptoms in both headache and anxiety disorders. Thus, it is recommended that screening be followed by a diagnostic interview based on the diagnostic criteria outlined in the American Psychiatric Association's Diagnostic and Statistical Manual (DSM-IV),42 with an emphasis on the core symptoms of various psychiatric disorders (for depression: feeling down and loss of pleasure in previously enjoyed activities most days during the past 2 weeks; for anxiety: feeling anxious, panic attacks, excessive rumination/worry during the past 6 months).


A distinction can be made between behavioral and cognitive-behavioral treatments for various disorders. Behavioral treatments focus exclusively on modifying environmental contexts/contingencies and patterns of behavior that presumably exacerbate symptoms and functional impairment. Cognitive-behavioral treatments, by comparison, include a focus on behavior modification as well as an emphasis on modifying maladaptive patterns of thinking. This is an admitted oversimplification, as an exhaustive discussion of these 2 therapeutic approaches is beyond the scope of this article. For the purposes of simplicity and consistency, we refer to CBT in this article, because we review both behavioral and cognitive techniques for treating psychiatric comorbidity in headache patients. In clinical practice, however, behavioral and cognitive strategies may or may not be used in conjunction with one another.

There are central elements of CBT and self-management that cut across treatment for specific psychiatric disorders as well as headache disorders. The central elements include patient education, a collaborative patient-provider relationship, self-monitoring, problem solving, cognitive restructuring, behavioral assignments and action plans, and abdominal breathing and relaxation training. A primary goal of these elements is to increase the patient's sense of self-efficacy and thereby improve clinical outcomes and long-term disease management. Self-efficacy is the confidence that one is capable of carrying out a behavior necessary to reach a desired goal.43 Self-efficacy is believed to be critical because it determines the degree of effort and persistence a patient will put forth toward behavior change.43

Patient Education.— Education of patients with headache and psychiatric disorders begins with a biopsychosocial explanation of the disorder(s). The patient is provided with an explanation of the disorder(s)' genetic predisposition that contributes to the vulnerability to the disorder, as well as the role of stress (environmental, social, psychological) in triggering/exacerbating the disorder. Education also includes a discussion of problematic cognitions and poor coping skills. This is deemed to be a critical step in increasing the patient's openness to treatment.44 Motivational interviewing may be employed to assess the patient's readiness to change and participate in treatment. Together, patients and providers determine the costs/benefits of treatment options, and the patient makes a decision about participation in treatment. Patient education also teaches patients cognitive restructuring and problem-solving skills (see below) to enhance their ability to manage their illness and to improve their coping skills.

Patient-Provider Partnership and Collaborative Care.— In traditional care models, the health-care provider is seen as an expert, with patients bringing little to the table other than their illness.28 Thus, it is not surprising that many patients enter medical or behavioral treatment assuming that they will be passive recipients of treatment. However, cognitive-behavioral therapists and proponents of disease self-management programs recognize that patients with chronic illnesses are their own principle caregivers and hold unique information about their experiences, while the provider holds knowledge of general strategies and treatment interventions. In this collaborative patient-provider partnership, the patient and provider are each considered experts in their own right. Typically, the provider may need to orient or socialize patients to expectations for mutual collaboration. A collaborative relationship encourages patients to ask questions and learn as much as possible to make informed choices. In a collaborative relationship, provider and patient jointly set therapeutic goals, determine priorities among these goals, set the agenda for each treatment visit, and mutually design out-of-session homework activities that facilitate the acquisition and practice of self-management skills.

Self-Monitoring.— For most psychiatric disorders, as well as headache disorders, self-monitoring is recommended for assessing episode frequency, intensity, and symptoms. Self-monitoring also assists patients with identification of triggers for episodes or exacerbations of their disorder(s) as well as for identification and assessment of the patient's attempts at self-management. Accuracy of self-monitoring is enhanced when it is prospective, when instructions and descriptions of monitored behaviors are clear, and when record forms are provided regularly and reviewed with the patient at subsequent office visits. Ongoing self-monitoring throughout treatment is considered therapeutic in treating psychiatric disorders, just as it is in the management of headache, because it contributes to self-awareness and provides regular feedback about progress.

Problem Solving.— Problem solving, when utilized in self-management protocols, helps patients identify problems they are having in the management of their illness and provides techniques to improve decision making, take appropriate actions, and alter these actions as they encounter changes in their circumstances or in their illness.27 The problem-solving approach helps patients to create alternative solutions, especially those that may not have been entertained in the past. The steps include identification of triggers (eg, for stress, headache, anxiety, depressed mood, nonadherence) and past maladaptive “solutions,” generation of alternative solutions, evaluation of each solution (eg consideration of time, effort, and likelihood of success), choice and implementation of the solution, and evaluation of the outcome.

Behavioral Assignments/Action Plans.— Out-of-office (homework) behavioral assignments/action plans are central to the success of behavioral interventions and self-management programs. Behavioral assignments/activities are not only important in the acquisition and generalization of new skills, but help patients to learn to take an active role in the management of their headaches or psychological disorder. Activities typically involve collecting information that will be needed in subsequent treatment sessions (eg, monitoring thoughts, emotions, physical reactions to headache-related stressors; monitoring somatic symptoms of panic attacks), practicing the various skills learned in treatment sessions (eg, relaxation training, cognitive restructuring, problem solving, exposure to feared situations/stimuli), or other lifestyle modifications and behavioral changes (eg, exercise). A central feature of these activities is that the patient works collaboratively with the therapist in their development; these activities are referred to as patient-generated short-term action plans by some.28 Treatment typically begins with activities that are small and easy to complete. Action plans are specific and of short duration—1 to 2 weeks. For example, “This week, I will practice full-length relaxation training Monday through Friday before bed.” Or, “I will walk for 15 minutes before lunch on Monday, Wednesday, and Friday.” The activity should be a behavior patients are confident they can complete. According to Lorig, confidence can be measured by asking the patient, “On a scale from 0 to 10, how sure are you that you can walk 15 minutes before lunch Monday, Wednesday, and Friday?” If the answer is 7 or higher, the behavior or action plan is likely to be accomplished, but if the answer is below 7, the action plan needs to be modified in order to avoid failure.28 Audiotapes and patient workbooks are often used to repeat and extend what is learned in treatment sessions and to present homework assignments and action plans. Patients who engage in these action plans between office visits are typically the most likely to make significant progress in treatment.45–47

Cognitive Restructuring.— Cognitive restructuring teaches patients to change their reaction to situations by counteracting stress-generating thoughts or cognitive distortions/errors (eg, catastrophic thinking, overgeneralization, dismissing the positive) and identifying and challenging the accuracy of the underlying distorted/inaccurate beliefs. Self-talk strategies are used to counteract inaccurate and maladaptive statements and to replace them with realistic and adaptive coping statements. Cognitive restructuring is a mainstay of treatment in stress management, headache self-management, and cognitive therapy for a wide array of psychiatric disorders. Its application in mood and anxiety disorders is discussed in more detail below.

Relaxation and Breathing Retraining.— Relaxation and breathing retraining are typically used in conjunction to teach patients to exert control over physiological responses and lower sympathetic arousal in anxiety disorders as well as headache disorders. Abdominal breathing and relaxation training are common interventions for many anxiety disorders, especially panic disorder and generalized anxiety disorder. With anxiety patients, these techniques are aimed at reducing symptoms of hyperarousal, especially those related to respiratory symptoms (dyspnea, chest tightness, hyperventilation), cardiac symptoms (increased heart rate, dizziness), and chronic muscle tension. These techniques are also useful in patients who endorse frequent physical symptoms of anxiety but do not meet criteria for a specific disorder.

Instruction in abdominal breathing is typically provided prior to the initiation of progressive muscle relaxation. Abdominal breathing involves taking slow, deep breaths using the diaphragm during inhalation. Slow, deep breaths are taken at a rate of approximately 10 breaths per minute, with exhalation longer than inhalation. Patients are instructed to subvocalize the word “relax” each time they exhale and to practice their skills regularly until diaphragmatic breathing becomes relatively automatic. After learning abdominal breathing, progressive muscle relaxation begins with having the patient alternately tense and release 16 different muscle groups, noticing the difference in sensation between muscles that are relaxed and those that are tense. As treatment progresses, muscle groups are consolidated and patients are instructed to incorporate brief relaxation skills into everyday experiences. The ultimate goal is for the patient to be able to relax the entire body simultaneously and to be able to apply this training to stressful situations that might trigger headaches or anxiety.


Over the past 3 decades, several widely used cognitive-behavioral interventions for migraine headache have been shown to be effective.2,25,48 In most instances these interventions emphasize prevention of headache episodes as opposed to aborting acute headache. Although behavioral modalities can be highly effective as monotherapy, they are most commonly used in conjunction with pharmacological management. The most extensively researched and most frequently used behavioral therapies for headache disorders are various muscle relaxation training techniques, thermal and electromyographic (EMG) biofeedback training, and cognitive-behavioral (stress management) therapy—similar treatment modalities employed in the management of various psychiatric disorders.

For headache, the most commonly used relaxation procedure in the United States is progressive muscle relaxation, which involves alternately tensing and releasing selected muscle groups throughout the body. Relaxation training skills are thought to enable patients with recurrent headache disorders to exert control of headache-related physiologic responses such as increased muscle tension, and more generally, to lower sympathetic arousal.

Biofeedback refers to any procedure that provides information about physiological processes (usually through the use of electronic instrumentation) in the form of an observable display such as an audio tone or visual display. The patient utilizes this “feedback” to self-regulate the response being monitored. The two most widely used biofeedback modalities for headache are (1) thermal biofeedback—hand warming feedback quantifying skin temperature from a finger, and (2) EMG biofeedback—feedback of electrical activity from muscles of the scalp, neck, and sometimes the upper body (trapezius muscles). Biofeedback training for headache is commonly administered in conjunction with relaxation training, either concurrently or sequentially.

CBT for headache management focuses on the cognitive, emotional, and behavioral components of headache and is typically administered in conjunction with relaxation training. CBT educates patients about the relationships between stress, coping, and headaches, and alerts them to the role that their cognitions play in their response to stress. Patients are taught to identify the specific psychological or behavioral factors that trigger or aggravate their headaches and to employ more effective strategies for coping with headache-related stress. Behavioral therapy for headache also includes identification and modification of headache precipitants, pain management techniques, and adherence-enhancing strategies for medication management of their headaches. Readers interested in a more detailed description of behavioral treatment for headache are referred to Blanchard and Andrasik,49 Penzien and Holroyd,50 Holroyd, Lipchik, and Penzien,51 and Lipchik, Holroyd, and Nash.52

For headache patients with comorbid mood or anxiety disorder, the addition of short-term CBT specific to their mood or anxiety disorder may be particularly useful. Briefly stated, CBT for depression or anxiety, like the CBT employed for headache management, educates patients about the roles their cognitions and behavior play in their psychological problems. Cognitive-behavioral interventions for depression53–55 are in many ways similar to cognitive behavioral interventions for headache disorders; thus, integrating these interventions for headache patients with comorbid depression is logical. Similarly, there are several empirically supported, cognitive-behavioral interventions for the treatment of anxiety disorders that can easily be adapted to a behavioral headache management protocol.56,57 Combining treatment protocols for anxiety or depression with behavioral treatment protocols for headache management may yield positive treatment outcomes for headache patients with comorbid anxiety and depressive disorders. However, there are currently no published controlled trials of behavioral therapies for the treatment of comorbid psychological and headache disorders.

Brief, focused attention to the specific psychological problems that precipitate or exacerbate headache episodes, interfere with treatment adherence, or interfere with the use of self-regulatory skills may be sufficient, at least in cases in which anxiety or depression is of mild to moderate severity. The relaxation techniques and/or biofeedback employed in headache management can easily be adapted to focus on the reduction of symptoms of physiological arousal associated with anxiety, especially those related to hyperventilation (eg, dizziness, numbness and tingling, shortness of breath, increased heart rate, chest tightness). The use of relaxation techniques (and biofeedback) may provide a nonthreatening way to introduce the patient to the process of psychological treatment and thus, to encourage the patient to acknowledge psychological difficulties and accept treatment for psychiatric disorders.

Described below are specific cognitive-behavioral interventions typically incorporated into the empirically supported treatment protocols for depression and anxiety.


A large body of literature attests to the efficacy of CBT in treating mood and anxiety disorders.58,59 CBT has consistently produced outcomes that equal (and often exceed) those of pharmacotherapy for depression (regardless of depression severity),60,61 panic disorder,62 generalized anxiety disorder,63 obsessive-compulsive disorder (OCD),64–66 and posttraumatic stress disorder (PTSD).67 CBT is recommended as first-line treatment for unipolar depression of mild to moderate severity,68 and for various anxiety disorders, such as panic disorder, generalized anxiety disorder, and obsessive compulsive disorder.69–72 CBT has been found consistently to have strong relapse prevention effects in the long-term treatment of mood and anxiety disorders.73

Across diagnostic domains, there is some suggestion that CBT is more cost effective than pharmacotherapy, especially when the long-term outcomes provided by CBT are considered.73 Specifically, short-term CBT is associated with strong maintenance of treatment gains (preventing relapse), compared with pharmacotherapy that typically requires ongoing treatment to prevent relapse.74,75 Although individual CBT has been found to be more expensive than pharmacologic treatments during the acute treatment phase ($1357 vs $839, respectively), it has proved to be more cost effective over a 1-year period: CBT was 59% of the cost of pharmacologic treatments over the same interval.75 Additionally, there is strong evidence that CBT can be successfully exported from research settings to real world settings, such as community clinics and primary care offices. For example, a 15-session manualized CBT protocol for panic disorder has been effectively administered in a community mental health center with short- and long-term treatment gains equivalent to those in the clinical trials.76,77


In the treatment of mood and anxiety disorders, it is important to consider the severity and chronicity of symptoms as well as the phase of treatment in order to optimize treatment outcome and cost-effectiveness with combined treatment.

Unipolar Depression.— In the acute treatment of unipolar depression, the rates of treatment response tend to be numerically higher for combined treatment relative to either modality alone; however, these differences tend to be small and often do not meet statistical significance.73 For example, Murphy and associates compared the efficacy of nortriptyline alone to CBT alone, CBT with placebo, and CBT with nortriptyline for the treatment of depression.78 No statistically significant differences were found between groups. However, patients who received nortriptyline with CBT had a slightly higher response rate (72%) than those who received CBT alone (63%) or nortriptyline alone (50%), and this rate of response was in the same range as that of the combined CBT and placebo group (76%). Conversely, several studies have shown that there are significant additive effects when CBT is combined with pharmacotherapy for the treatment of patients with chronic depression.73 Moreover, there is strong evidence to suggest that patients failing one modality of treatment (CBT or pharmacotherapy) can achieve an adequate treatment response when switched to another modality because alternate brain mechanisms are involved.73

There is consistent evidence that adding CBT to the acute phase of pharmacotherapy offers comparable efficacy to maintenance pharmacotherapy in preventing relapse.73 For example, Simons and colleagues examined 1-year outcome for patients who received CBT, pharmacotherapy, or combined treatment during acute treatment.79 The antidepressant was discontinued after 3 months, and at the end of the 1-year follow-up period, 66% of patients who received pharmacotherapy alone relapsed, compared to 28% of patients who received CBT alone or combined CBT and pharmacotherapy. Similar results have been reported in several other studies.73

CBT also appears to be efficacious in preventing relapse when implemented following the acute phase of medication treatment.73 For example, Fava and colleagues investigated the long-term effects of CBT added to antidepressant therapy of depressed patients who had already responded to the antidepressant.80,81 Following initial response to an antidepressant, patients were randomized to either CBT or a clinical management control condition, and the antidepressant was tapered. Over the next 4 years, 70% of patients in the clinical management condition relapsed, compared to 35% of those receiving CBT. By the 6-year follow-up, absolute rates of relapse no longer differed between the 2 conditions. However, CBT-treated patients tended to have a single relapse whereas the clinical management patients tended to have multiple relapses during the 6-year period. This protection from relapse provided by CBT appears to extend to patients with recurrent depression (history of 3 or more depressive episodes).82 A recent 2-year outcome study conducted by Fava and colleagues compared treatment of recurrent depression with either brief CBT or clinical management; antidepressant medication was tapered during the 2-year study.82 By 2 years, 80% of patients receiving clinical management had relapsed, compared to only 25% of those who were treated with short-term CBT.

Bipolar Depression.— The typical course of bipolar disorder is characterized by regular relapses to depression or mania/hypomania despite treatment with a wide array of mood-stabilizing medications, antipsychotics, and antidepressants.83 There is consistent evidence indicating that CBT is an efficacious adjunct to pharmacotherapy.73 Combined brief CBT and pharmacotherapy (mood stabilizers and concomitant pharmacologic agents), compared to pharmacotherapy alone, consistently and significantly reduces the number of manic, hypomanic, or depressed episodes and reduces the number of days hospitalized due to a mood episode.84,85 Thus, combined treatment for bipolar disorder has potential to not only enhance treatment outcome, but to reduce costs of care.

Anxiety Disorders.— CBT alone and pharmacotherapies alone are efficacious in the short-term treatment (12 to 15 sessions CBT, 3 months treatment) of anxiety disorders.73 However, CBT, but not pharmacotherapy, appears to offer long-lasting benefits long after the cessation of formal or acute treatment.62,63 Conversely, evidence for advantages of combined CBT and pharmacotherapy over monotherapy is mixed.73 Two large multicenter trials have reported deleterious effects of combined treatment of panic disorder, noting that some of the benefits of CBT provided during medication treatment are lost when medication is discontinued and thus concluded that the long-term effects of combined treatment appear to be inferior to CBT alone.86,87 Moreover, patients who are treated with CBT alone maintain their treatment gains.73,86,87 Otto and colleagues suggest that the deleterious effects of combined treatment are a function of the patient relying on medication as a “safety signal” (avoidance strategy), which competes with the focus in CBT on “relearning” that feared stimuli are safe and should not be avoided.73



Behavioral Activation for Depression.— Behavioral activation treatments for depression grew out of the notion that depression results from a lack of response-contingent reinforcement in the individual's environment.88 In general, these treatments focus on increasing the frequency of an individual's access to pleasant events (positive reinforcers) and decreasing the frequency of aversive events and avoidance behaviors.89 Typically, behavioral activation treatments focus on having the patient monitor his/her weekly activities, identify the quality and quantity of activities that may reduce or perpetuate depression, and generate ideas about pleasurable activities to employ during treatment. Subsequent treatment focuses on identifying concrete behavioral goals within major life areas such as relationships, employment, physical health, and recreational activities. Patient and therapist collaboratively devise a hierarchy of activities that the patient will find enjoyable, that promotes a sense of mastery and accomplishment, and that is congruent with the identified behavioral goals. The patient and therapist design weekly activity goals that the patient engages in independently, progressing from activities identified as “easy” to those that are more difficult. Activity schedules are used by the patient to schedule individual activities during the coming week and to monitor completion of designated activities, with space provided for ratings of pleasure and mastery derived from each activity.53 The ultimate goal is to have a majority of the patient's time allotted to pleasurable and productive activities that the patient is likely to complete, that facilitate the realization of identified life goals, and that enhance access to positive reinforcement in the patient's environment.

Cognitive Restructuring.— Depressive cognitions and beliefs are typically self-critical (“It's all my fault. I'm a failure.”), negative about the world or significant others (“Nobody cares about me.”), and hopeless about the future (“Things will never get any better. I will never be able to manage my problems.”) Individuals with depression see themselves as incapable, undesirable, and unworthy. They expect rejection, failure, and dissatisfaction and interpret their experiences in ways that confirm these negative expectations. Such cognitions and beliefs contribute to avoidance, social isolation, decreased activity, and decreased adherence to pharmacological or behavioral treatment for headache.

Cognitive restructuring is used to modify depressive cognitions that maintain symptoms of depression and contribute to behaviors that interfere with more adaptive functioning. The patient is taught to identify depressive thoughts, challenge the validity of underlying beliefs, and develop more rational and adaptive thinking patterns. For example, an individual prone to depression may interpret an essentially benign event (eg, not getting offered a position following a job interview) as confirmation of negative self-beliefs. The depressed individual may engage in self-statements such as, “They didn't select me for the job. They hated me. I'll never work again. I'm a failure. I am worthless. I can't do anything right. Why even bother to look for a job?” This pattern of thinking maintains feelings of hopelessness and despair. The patient is taught to identify the objective facts of the event (eg, the patient was not called back on one job interview), the automatic thought processes described above (eg, “They hated me.”), the underlying beliefs that the thoughts represent (eg, “I'm a failure. I'm worthless.”), and the cognitive errors inherent in these automatic thoughts (eg, overgeneralization of a single event, catastrophic thinking). Patients are then taught to adjust their thinking patterns in a more realistic manner by identifying disconfirming evidence for the identified thoughts, by questioning the validity of their underlying beliefs, and by generating more rational alternatives (“I did not get this job, but I have held jobs in the past. This one event does not mean that I am a failure.”) Behavioral “experiments” may be incorporated to help the patient objectively “test” the validity of their depressive thoughts.


Exposure-Based Treatments.— Exposure-based treatments for anxiety disorders promote prolonged exposure to feared stimuli, reducing the negative reinforcement obtained through avoidance behavior that ultimately serves to maintain anxiety. The many varieties of exposure-based treatments emphasize either imaginal or in vivo (“real life”) exposure to feared objects, places, bodily sensations, or situations. Imaginal exposure is used primarily with patients diagnosed with PTSD and generalized anxiety disorder, or in other cases where in vivo exposure is difficult or impossible to arrange. In vivo exposure is used primarily with panic disorder, agoraphobia, OCD, and specific phobias. Exposure-based exercises are tolerated best when the patient is educated about the procedure and potential reactions. Because exposure therapy can initially be perceived as invasive, it is necessary that the patient voluntary agrees to exposure. Reticence to complete exposure treatment may often be overcome by the therapist “cheerleading” and verbally reinforcing the patient's progress and helping the patient challenge assumptions that he/she cannot complete the procedure. Often, having the therapist model the designed exposure first increases the patient's willingness and self-efficacy in their own ability to complete the exposure. Exposure therapy is most effective when it is challenging yet graded in intensity (progressing to the most difficult items after less difficult items have been successfully completed), when practices are repeated frequently and spaced close together, when they last long enough to ensure a significant decrease in anxiety, and when the patient does not use subtle avoidance or distraction strategies during exposure.90 Specific adaptations of exposure-based treatments are discussed below.

For individuals with panic disorder, in vivo exposure targets feared bodily sensations (interoceptive exposure) and situations that are avoided out of fear of panic. The patient engages in activities designed to produce the feared sensations (eg, breathing rapidly through a straw to induce shortness of breath, spinning around in a swivel chair to produce dizziness) until they are tolerated with minimal distress. Subsequent exposure focuses on having the patient go to and remain in places that are typically avoided, such as grocery stores, malls, and other public places.

Imaginal exposure is used with generalized anxiety disorder patients to arrange prolonged exposure to worry themes. The patient identifies major themes of worry and is instructed to imagine the feared outcomes associated with these themes. The outcomes are described verbally in as much detail as possible (or written down in diary form) until anxiety decreases significantly. After worry exposure, the patient is taught to identify and discuss how he/she would cope with the feared outcomes.

Exposure treatment of OCD arranges for prolonged exposure to feared stimuli that are relevant to the patient's primary obsessions while prohibiting the patient from engaging in compulsive behaviors (exposure and response prevention). An OCD patient with contamination fears, for example, would be instructed to thoroughly handle “contaminated” objects (eg, trash, dirty clothes, belongings of other people) while refraining from hand washing and other compulsive behaviors. Obsessions that are not amenable to in vivo exposure (eg, religious obsessions, fears of harming others) are targeted using imaginal exposure.

Many PTSD patients attempt to avoid reminders and memories of their traumatic events, which ultimately perpetuates their anxiety and frequent reexperiencing of the trauma in the form of nightmares and flashbacks. Exposure with PTSD patients is often imaginal, aimed at having the patient repeatedly retell or write down the traumatic event in detailed form. In vivo exposure is sometimes used in patients who avoid specific situations associated with the traumatic event (eg, the road where they were in a traumatic car accident).

Social phobia and specific phobia are usually targeted using in vivo exposure. Patients with social phobia are gradually exposed to fearful social or performance situations. Group therapy is often used as a front-line treatment of social phobia, as it promotes prolonged exposure to other individuals and interpersonal interactions. Role-play exercises (eg, mock interviews, social interactions) are also used. Specific phobias typically are treated directly through in vivo exposure; exposure using pictures, videos, virtual reality equipment, or imagined stimuli is appropriate when in vivo exposure is not feasible for one reason or another.

Cognitive Restructuring.— Anxiety disorders are typically accompanied by fear-based thoughts of danger and vulnerability. Individuals with anxiety disorders typically engage in 2 types of cognitive errors: overestimating the likelihood of the occurrence of negative events and perceiving situations as much more catastrophic (intolerable, unmanageable, and dangerous) than they actually are. This is particularly true of the fear of panic attacks in panic disorder, of the feared outcomes of chronic worry characteristic of generalized anxiety disorder, of the obsessive fears in OCD, and of the fear of negative evaluation in social phobia. Individuals with anxiety are taught to more realistically appraise the likelihood that the feared events will actually occur and to identify and rely on adaptive coping skills. Generalized anxiety disorder patients, for example, are encouraged to recall times in which their feared outcomes did not occur and to monitor over time the outcomes of situations they have worried about chronically. Patients with PTSD often make negative assumptions about themselves and the dangerousness of the world as a result of their traumatic event. They are thus taught to develop objective beliefs about themselves independent of their traumatic event and to refrain from generalizing assumptions about the dangerousness of particular situations (and the world in general) from the outcome of the traumatic event they experienced.


Existing but limited empirical evidence highlights the importance of identifying and treating psychiatric disorders in headache patients. More research is needed on the effects of treating comorbid psychiatric disorders on headache outcomes, and vice versa. Because both psychiatric disorders and headache disorders are amenable to CBT, it is likely that treatments for these 2 disorders can be combined and/or modified relatively easily. The major components of CBT include educating patients about the roles of behavior and cognition in their disorders and working collaboratively with them to develop alternative means of behaving and thinking that reduce associated symptoms and functional impairment. In depression, the main foci are on identifying and implementing behaviors that are positively reinforcing and on modifying cognitions that are self-critical, negative, and hopeless. In anxiety, patients work to cease avoiding feared stimuli (exposure therapy) and to modify thoughts regarding danger and vulnerability. We hope that practicing clinicians and clinical researchers will use and evaluate the strategies and techniques outlined here to improve the functioning of headache patients who suffer from comorbid psychiatric disorders.

Conflict of Interest:  None