Current issues of ACP Journal Club are published in Annals of Internal Medicine


Review: Several nonpharmacologic treatments are effective for migraine, especially when used with other agents or for prophylaxis

ACP J Club. 1999 Jan-Feb;130:11. doi:10.7326/ACPJC-1999-130-1-011

Source Citation

Pryse-Phillips WE, Dodick DW, Edmeads JG, et al. Guidelines for the nonpharmacologic management of migraine in clinical practice. CMAJ. 1998 Jul 14;159:47-54.



How effective are nonpharmacologic therapies (biofeedback; relaxation; cognitive-behavioral therapy; psychotherapy; hypnosis; physical measures; physiotherapy, osteopathy, or chiropractic; transcutaneous electrical stimulation or acupuncture; occipital or supraorbital nerve block; riboflavin; or naturopathy or homeopathy) for migraine?

Data sources

Studies were identified from MEDLINE (1975 to 1996) by using the terms migraine and alternative medicine, acupuncture, biofeedback, chiropractic, hypnosis, and herbal medicine. Bibliographies were checked, and experts were consulted.

Study selection

English-language, randomized, controlled trials were selected if they studied adults with migraine and nonpharmacologic interventions.

Data extraction

Data were extracted on study design and quality, interventions, and outcomes. Data were not combined. Clinical practice guidelines were developed.

Main results

Effective diagnosis with appropriate history and physical examination and patient education and support are important aspects of migraine management. Trigger factors for migraine must be identified and addressed.

Findings from randomized controlled trials support the use of several non-pharmacologic treatments: prophylactic biofeedback (comparable in effectiveness to prophylactic pharmacotherapy); relaxation, including progressive muscle relaxation, breathing exercises, or directed imagery, especially when used for long-term prophylaxis (mixed but generally favorable results); cognitive-behavioral therapy alone or in conjunction with other therapies (mixed results); psychotherapy with biofeedback (1 trial); and chiropractic manipulation (reduced frequency and severity of migraine). For other interventions, effectiveness has not been verified in controlled trials or trial data have not been conclusive.


Many nonpharmacologic treatments are effective, or somewhat effective, for migraine headache. Most are more effective when used in conjunction with other treatments or for prophylaxis.

Source of funding: Glaxo Wellcome, Inc.

For correspondence: Dr. W.E. Pryse-Phillips, Division of Neurology, Health Sciences Centre, St. John's, Newfoundland A1B 3V6, Canada. FAX 709-737-6656.


Headache is one of the most common problems encountered in general practice and the practice of neurology. Still, self-care is the most prevalent form of treatment of headache, and only a few persons with headache consult a physician. I am somewhat less optimistic than the authors of this review that most practitioners can reliably distinguish tension-type headache from migraine, if only because most of my patients have a "mixed bag" of headache patterns that render neat classification schemes somewhat difficult to use in practice (1).

Most would agree that nonpharmacologic treatments could complement, or be substituted for, pharmacologic treatments for a disorder as complex as migraine. The problem is that nearly every form of behavioral modification therapy works for somebody with migraine, yet no approach works for everybody. The reader may believe that the addition of meta-analysis to randomized controlled trial evidence consistently provides a framework for rational therapy, but an intuitive, case-by-case, patient care protocol may be the most effective pathway in the delivery of medical care for patients with headache.

Should changes in daily caffeine intake be viewed as a pharmacologic therapy? Caffeine is highly effective for migraine headache in many patients, provided that it is omitted from the diet between headaches. However, an additional analgesic is often necessary. Perhaps it is hyperbole to speak of caffeine habituation, but some persons with migraine have been using caffeine as a "medicine" for prolonged periods without realizing the consequences of regular dosing and the benefits that might accrue from abstinence between headaches. Giving up one's morning cup of coffee is an insuperable challenge for some persons.

It is now appropriate to identify commercial sponsors in peer-reviewed publications, and we should congratulate Glaxo Wellcome, Inc., for providing an unrestricted educational grant for important research that challenges the prevalent view that pharmacologic therapy is the best and only way to treat a disorder that is quintessentially human (2).

Matthew Menken, MD
Robert Wood Johnson Medical SchoolSomerset, New Jersey, USA


1. Hopkins A. Clinical Neurology: A Modern Approach. Oxford: Oxford Univ Pr; 1993: 113-28.

2. Sacks O. Migraine. Berkeley: Univ of California Pr; 1992:140-63.