Diagnostic criteria for migraine

A validity study

Merikangas, Kathleen Ries PhD
Dartigues, J. F. MD, PhD
Whitaker, Amy BA
Angst, Jules MD

Genetic Epidemiology Research Unit, Yale University School of Medicine, New Haven, Connecticut (Dr. Merikangas and Ms. Whitaker); The Psychiatrische Universitatsklinik, Zurich. Switzerland (Prof. Angst); and the Unite INSERM, Universite de Bordeaux, Bordeaux, France (Dr. Dartigues).

Address correspondence and reprint requests to Dr. Kathleen R. Merikangas, Genetic Epidemiology Research Unit, Yale University School of Medicine, 40 Temple Street—Lower Level, New Haven, CT 06510–3223.

Supported in part by Research Scientist Development Award MH00499 from the National Institute of Mental Health; the John D. and Catherine T. MacArthur Foundation, Mental Health Network I on Psychobiology of Depression and Other Affective Disorders; Grant #3.873.0.88 from the Swiss National Science Foundation; and a grant from Glaxo Laboratories, France.

Neurology 44(6):p S11-S16, June 1994.

Article abstract

To identify the optimum combination of symptoms for the International Headache Society (IHS) diagnostic criteria for migraine, the criteria were systematically assessed for validity using an epidemiologic sample from Zurich, Switzerland. The indicators of validity used included subjective distress, occupational impairment, family history of migraine, and treatment. The symptoms that provided the best discrimination between migraine and other headache subtypes were photophobia, phonophobia, and osmophobia, in combination with gastrointestinal symptoms. The evaluation of the validity of the IHS classification of migraine is impeded by several factors, including: the presence of multiple headache syndromes within an individual, the tendency for headache characteristics to change over a lifetime, the effects of headache treatments in obscuring syndromes, and the lack of generalizability of findings based on clinical samples. The methods used in this study serve as a model for applying statistical techniques for evaluating the validity of diagnostic criteria. The findings, however, should be replicated in additional studies to determine their generalizability to specific demographic and clinical subgroups.