How do we facilitate individuals with recurrent headache and migraine receiving evidence-based behavioral treatment?

Research on behavioral treatment for headache and migraine commenced in the 1970s driven by the emerging literature on biofeedback training. In 2000, the United States Headache Consortium, comprised of seven leading medical and headache associations, published evidence-based guidelines for the treatment of migraine. They reported that the evidence in support of the efficacy of behavioral treatment was ‘Grade A’ (“multiple well-designed randomized clinical trials, directly relevant to the recommendation, that yield a consistent pattern of findings”). ¹ In 2005, the American Headache Society brought together a group of leading psychologists on headache treatment to publish a review which summarized the results of meta-analytic reviews for behavioral treatment for migraine (thermal biofeedback, electromyographi biofeedback, cognitive-behavioral therapy [CBT] and relaxation training) and concluded that average improvement ranged from 33% to 55%, compared to 5% for no-treatment controls. ² In the 20 years since this review, new variations of behavioral treatment have been shown to be efficacious, and some studies have reported higher rates of improvement than the studies in the review. For example, one study that included CBT based on a functional model of headaches reported a reduction in headaches from pre- to post-treatment of 68% which increased to 77% at 12-month follow-up. ³ Behavioral treatment for headache and migraine is associated with a range of benefits including reduction in medication consumption and decreases in anxiety, depression, fatigue, irritability and anger. ⁴

Alternative delivery formats to therapist-delivered (clinic-based) have been investigated. One approach has been to use minimal-therapist-contact approaches (home-based) in which skills training is introduced in the clinic, but training primarily takes place in the home with the patient guided by printed materials and recordings. This approach reduces the number of clinic sessions to three or four. Most of the therapist-delivered programs have been converted into minimal-therapist-contact interventions including relaxation training, CBT, thermal biofeedback training, and combinations of these approaches. Meta-analytic reviews have reported that minimal-therapist-contact interventions for headaches can be as effective as therapist-delivered approaches. ⁵

Despite these positive results for behavioral treatment of headache and migraine, the limited evidence available suggests that referrals are very rarely made to psychologists for this form of intervention. For example, a large study in the UK reported that 98% of medical management of headaches takes place in general practice and that referrals to specialists were mainly limited to 2.1% to neurologists and 1.1% to general physicians, with the rate of referrals to psychologists too low to report. ⁶

This is ironic as the biopsychosocial model was published in 1977 as a replacement for the biomedical model and yet references to headache and migraine usually reflect a biomedical model. ⁷ Also, Lifestyle Medicine has been on the rise since the establishment of the American College of Lifestyle Medicine in 2004, which is based on recognition of the critical role of lifestyle in health and wellbeing. Reviews of lifestyle factors as they relate to migraine have been published. ⁸

There are many reasons that may underlie the low referral rate of patients with headache and migraine to psychologists for behavioral treatment, and these reasons vary from one setting/country to another. Reasons include: (i) headaches are labelled as ‘neurological disorders’ which ignores the psychosocial determinants of headaches; (ii) national clinical practice guidelines do not always include evidence-based behavioral interventions as a treatment option; (iii) ICHD-3 only includes diagnostic criteria rather than other factors important to headache management such as triggers; (iv) health professionals and members of the community tend to view psychologists as mental health practitioners rather than recognizing that this is only one of their roles; (v) not all clinical/health psychologists have had training specifically in headache assessment and treatment; and (vi) in some countries/settings, the cost of psychological services is high.

For the remainder of this Editorial, changes will be discussed that address the above six reasons suggested for the low referral rate. An obvious starting point for change is the international and national headache societies. Headache and migraine are labelled as ‘neurological disorders’ which deflects attention away from important questions such as why do migraine attacks occur when they do, why is the person experiencing migraine attacks at this time in their lives, why did the migraine problem begin when it did, and why is this person vulnerable to having migraine attacks? It ignores reactions to attacks despite the fact that reactions can have important implications as they can establish vicious cycles such as stress triggering headaches followed by reactions that increase stress levels. Behavior and lifestyle factors provide the psychosocial and developmental context for migraine and need to be taken into account in a comprehensive management plan.

A review of national clinical practice guidelines for headache and migraine would be appropriate as they are variable when it comes to advocating consideration of referral for evidence-based behavioral treatment. Some of the guidelines do recommend this option whilst others are limited to pharmacological interventions and others list behavioral strategies but under titles such as ‘home remedies’ therefore failing to recommend evidence-based approaches.

Work has been underway for some time developing the new version of the International Classification of Headache Disorders, ICHD-4, and consideration should be given to broadening the system to include factors other than diagnosis. ⁹ The Diagnostic and Statistical Manual of Mental Disorders (DSM-4) utilized a multiaxial system that included axes for psychosocial/environmental/contextual factors (Axis IV), and for severity/disability/functioning (Axis V). The International Classification of Diseases includes axes for disability (Axis II) and for environmental and personal lifestyle factors (Axis III). Broadening ICHD-4 might include adding axes for triggers and measures of disability/functioning. The Committee developing ICHD-4 has been very active in terms of consultation and periodically publishing articles about their progress, but the focus seems to be exclusively on diagnostic criteria rather than broadening the system. This is perhaps not surprising given that the Committee is comprised of 11 highly distinguished headache researchers, but they are all neurologists, consistent with a biomedical model of headaches rather than a biopsychosocial model. There is also the issue of funding the work of the Committee and the Committee Chair for the first three versions of the Classification noted that “the finances had to be found. The IHS had no money before the advent of triptans. The problem was managed with support from four major pharmaceutical companies”. ¹⁰ Parenthetically, it was mentioned in this paper that the Committee was aware of DSM-3 and “decided to use their methodology in a simplified version using only one axis”.

Another domain that merits consideration is education of health professionals and the community about the role of psychologists. Many view psychologists as focusing on mental health disorders, but this is only one application of psychological science. Psychology has been defined as ‘the study of human and animal behavior’ so that the psychology profession can potentially contribute to any disorder for which behavior and lifestyle factors are relevant. The belief that psychology is limited to mental health disorders deters medical practitioners referring headache patients to psychologists and discourages headache patients from wanting referrals to psychologists. It may lead to medical practitioners only considering referring headache patients to psychologists if there are significant mental health comorbidities but it should be emphasized that the studies demonstrating the efficacy of behavioral treatment for headaches have not set out to recruit subjects with such comorbidities and in fact some of the studies have used exclusion criteria pertaining to such comorbidities.

A further issue is that few psychologists have received training specifically in treatment of headache and migraine. Ways of changing this situation include advocating revision of course accreditation guidelines such that this training must be included in relevant courses. Another approach is increasing training opportunities via workshops at conferences and on-line courses. An alternative approach is to deal with some of the issues referred to here by delivering behavioral interventions via the internet and smartphone applications.

Finally, practical problems need to be addressed such as the affordability of behavioral treatment for headache and migraine, as in some countries, government and private health schemes do not provide adequate support. An economic analysis of how funding such treatment could lead to a net reduction in costs should help advocacy. International studies such as The Eurolight Project and Global Burden of Disease study have highlighted the enormous direct costs (healthcare costs) and indirect costs (lost productivity) associated with migraine. ⁴ Studies of behavioral treatment have consistently reported decreased use of medication (e.g., one study found a 70% reduction from pre- to post-treatment with CBT), ³ and the decreases in anxiety, depression and fatigue noted earlier, should assist increased productivity.

In conclusion, it is argued that there is a strong need for future research in two domains. First, more research is needed on how many patients diagnosed with headache and migraine are referred for evidence-based behavioral treatment, and the reasons for why more such referrals are not made. Second, there is a strong need for clinical trials that include both pharmacological and behavioral interventions (clinic-based and home-based), administered on their own and in combination, with the goal of determining relative efficacy but more importantly the ‘client-treatment matching’ goal of which approach works best for whom. One potentially interesting line of research would be to investigate stepped care models commencing with lower cost interventions and proceeding to higher cost interventions if the earlier approaches are not successful.