Migraine Diagnosis and Treatment: Results from the American Migraine Study II

Abstracts & Commentary

Sources: Lipton RB, et al. Headache. 2001;41:638-645; Lipton RB, et al. Headache. 2001;41:646-657.

As a 10-year follow-up to their landmark American Migraine Study I completed in 1989 (Stewart WF, et al. JAMA. 1992;267:64-69), Lipton and colleagues now report on the results of another large population-based survey completed in 1999. The goal of the survey was again to identify patterns of migraine diagnosis, medication use, and migraine-associated disability. In the second report, Lipton et al report on the prevalence, sociodemographic profile, and overall "burden of migraine" in the United States.

A survey was mailed to 20,000 US households. The survey was designed to identify individuals with headaches, and determine headache severity, frequency, associated symptoms, and disability. Of the 43,527 people older than age 12, 29,727 responded giving a 68.3% response rate. A total of 3738 individuals could be diagnosed with migraine fulfilling International Headache Society (IHS) criteria. The total estimated prevalence in the United States was 27.9 million or 12.6% of the population. The prevalence of migraine by gender was 18.2% female, and 6.5% male. Migraine prevalence was 2-3% higher among whites than blacks and inversely related to household income. Prevalence increased from 12 years of age to 40 years of age. The most frequently reported migraine symptoms were "pulsatile" pain (85%), light sensitivity (80%), and nausea (73%). It was notable that only 59% complained of unilateral pain and 36% had migraine-associated aura.

Only 48% of individuals with survey-diagnosed migraine had ever been diagnosed by a physician. This compares to the 38% disparity described in the 1989 study. A total of 41% of migraineurs surveyed use prescription drugs for their migraine and this compares to 37% in 1989. There was roughly the same proportion of migraineurs using over-the-counter medications in 1999 (57%) and 1989 (59%). In terms of migraine-associated disability, the survey revealed that 57% of diagnosed and 45% of undiagnosed migraineurs experienced at least a 50% reduction in work/school productivity (P > .001).


The American Migraine Study II represents another pivotal neuro-epidemiological study. Lipton et al have shown that the prevalence of migraine has remained stable over the past decade. Gender differences, socioeconomic variables, and migraine characteristics have not changed as well. The increase in estimated total migraine from 23.6 million in 1989 to 27.9 million in 1999 is commensurate with the increase in total population and does not reflect some fundamental change in the disease state. Several methodological factors may have led to an underestimation or overestimation of migraine and are pointed out by Lipton et al. Since only "severe" headaches were counted as migraine, a large number of individuals with moderate headache could still fulfill IHS criteria but were excluded. Similarly, people on the other end of the spectrum with severe daily headaches were excluded by IHS definition of this being an episodic disorder. Nonetheless, meaningful conclusions can still be taken from this follow-up study. Despite the increase in awareness of migraine and the usefulness of the triptan class of drugs for acute migraine treatment, the burden of migraine remains significant. As it was 10 years ago, up to 50% of migraineurs are undiagnosed and even the ones who are diagnosed most are not given effective prescription medication. Lipton et al have demonstrated that the disability of migraine remains substantial. Primary care physicians are clearly not getting the message about the importance of migraine in their patients and this leaves the responsibility to neurologists to lead the way. —Jeffrey Reich

Dr. Reich, Assistant Professor, Neurology, New York Presbyterian Hospital-Cornell Campus, is Assistant Editor of Neurology Alert.