Statistical considerations for design and analysis of a multiperiod crossover study to compare two treatments for migraine headache

The purpose of this paper is to discuss a long-term, multiperiod crossover study to compare two treatments for migraine headache. Principal attention is given to the analysis of an example in which patients randomly received a treatment sequence with test drug for three migraine headaches and placebo for one. An issue that requires attention for this example is the influence of a carryover effect for test drug that was greater when placebo was the subsequent treatment than when test drug was the subsequent treatment. A way to address this issue without excessive loss of power is to consider tests of "total treatment effects," which are weighted averages across headaches of differences between average response to test drug and average response to placebo. Since these weighted averages are linear combinations of treatment effects and carryover effects, their use requires an argument that carryover effects are at least partly a further form of treatment effects. For the example in this paper, this argument is realistic because the test drug could have provided much better relief to migraine headache than other treatment patients might have previously used. A second purpose of the paper is to present some alternative designs for situations like that represented by the example. The structure of variances for alternative specifications for treatment comparison is provided for models of interest for these designs.     

Genetics of migraine

Several historical reports focusing on the heredity of migraine, as well as recent studies on its epidemiology and molecular biology, have revealed evidence for a decisive role of genetic factors in the aetiopathogenesis of familial migraine. Indeed, family studies, segregation analyses and twin studies have shown that genetic factors play an important role in disposition towards migraine but could not explain the entire aetiopathogenesis. The influence of extragenetic factors, however, remains mostly unknown. Recent linkage analyses have provided evidence for genetic heterogeneity. A locus for Familial Hemiplegic Migraine (FHM), the only known type of migraine that follows autosomaldominant transmission, has been linked to chromosome 19p13 but genetic heterogeneity has also been shown, i.e., different types of migraine could be excluded from this locus. Further investigations should concentrate on identifying the FHM gene on chromosome 19p13, on linkage analyses with markers for different susceptibility genes, and on genomic analyses of highly informative pedigrees. This would lead to further clues to the pathogenesis underlying migraine and, thus, to therapeutic developments.     

Neurotology of migraine

Neurotologic symptoms are common with migraine, yet relatively little is known about the pathophysiology of such symptoms. Motion sensitivity with bouts of motion sickness occurs in about two thirds of patients with migraine. Episodes of vertigo occur in about one fourth of patients and, in some, vertigo is the only symptom (so-called "migraine equivalent"). Phonophobia is the most common auditory symptom, but fluctuating hearing loss and acute permanent hearing loss occur in a small percentage. Migraine can mimic Meniere's disease and so-called "vestibular Meniere's disease" is usually associated with migraine. The recent discovery of a mutation in a brain calcium-channel gene in families with hemiplegic migraine and in families with episodic vertigo and ataxia suggests a possible mechanism for neurotologic symptoms in patients with more common varieties of migraine. A defective calcium channel, primarily expressed in the brain and inner ear, could lead to reversible hair cell depolarization and auditory and vestibular symptoms. This hypothesis is currently being investigated in other families with migraine headaches and neurotologic symptoms. Hopefully, such studies will lead to improved diagnosis and better treatments in the future.     

Diagnosis of migraine

In this report we describe the circumstances surrounding an outbreak of mass hysteria among first year female university students. The outbreak was precipitated by a state of panic over the possibility of a fire which turned out to be harmless fumes from a locally used burning perfume. Twenty-three cases presented to the emergency room with symptoms of respiratory distress associated with marked emotional reactions. The outbreak was controlled by separating the patients, and the lack of media coverage facilitated this process. Cases with persistent and severe symptoms were those having physical and psychological problems prior to the occurrence of the outbreak. Individual vulnerability factors deserve special consideration in the management of mass hysteria since they are likely to influence the response to treatment.     

Familial occurrence of migraine without aura and migraine with aura

We report a study of 121 probands (patients) with migraine without aura (MO) and 72 probands with migraine with aura (MA), diagnosed according to the operational diagnostic criteria of the International Headache Society and selected from 35 general practices in Denmark. The probands were interviewed about the presence of MO and MA among their first-degree relatives. Compared with the general population, the first-degree relatives of probands with MO had a threefold increase of MO, and only one first-degree relative of one proband with MO had MA. First-degree relatives of probands with MA had a twofold increase of both MA and MO. Compared with the general population, few spouses had MO and MA. This threefold and twofold increase in family risk of MO and MA, combined with the lack of increased risk in spouses, strongly suggests that MO and MA are genetically determined.     

Behavioral management of migraine

Behavioral treatments for migraine have received increased attention over the past 25 years. In general, research has focussed on the six following areas: 1) comparative efficacy of treatments, 2) interplay of behavioral and pharmacological approaches, 3) development of delivery models that are cost-effective, 4) identification of characteristics associated with varied levels of response to treatment, 5) maintenance of effects and factors associated with long-term outcome, and 6) mechanisms of treatment. This paper briefly and selectively reviews the available literature in an attempt to point out the status of current research.     

Vertigo and migraine

Vertigo consists of a variety of syndromes and can be due to many etiologies. One of these causes is migraine, which in our experience is often overlooked, although migrainous vertigo is well known in the literature. Vertigo in migraine can occur as aura or during the headache phase, or independent of the attacks as aura without headache. The aim of this retrospective study was to analyze cases with vertigo and migraine: 23 (8%) of 298 patients with migraine examined in a neurological outpatient department also had rotational vertigo. 48% of these patients had vertigo independent from typical migraine headache. Two types of vertigo were found: permanent vertigo, and vertigo with the characteristics of paroxysmal positional vertigo. 57% of the vertiginous attacks lasted hours, 26% even days, and 17% minutes. Most of the patients had several attacks of vertigo, some involving up to 30 episodes. To recognize migraine as a cause of vertigo has therapeutic implications. Most of our patients with vertigo and migraine showed a good response to antimigraine therapy.     

Stabilization of asthma mortality

BACKGROUND: Rates of death from asthma in the United States have increased since 1978. OBJECTIVE: To identify and evaluate recent trends in asthma mortality. METHODS: Analysis of data from the National Center for Health Statistics identifying asthma (ICD 493) as the underlying cause of death in the 50 United States and the District of Columbia with rates of death from asthma by age, race, and sex and age-adjusted rates of death by race. The Bureau of the Census provided population data by age, race, and sex that permitted calculation of rates of death at 5 through 34 years of age. The Departments of Health of Australia, Canada, Great Britain, and New Zealand provided numbers of deaths from asthma and population data from which we have calculated rates of death. RESULTS: Rates of death from asthma in the United States increased from .8 per 100,000 general population in 1977 and 1978 to 2.0 in 1989 and have been 1.9 or 2.0 since then until an increase to 2.1 in 1994. A significant difference in regression over groups indicates a difference in average rates between 1979 through 1987 compared with 1988 through 1994. Rates of death from asthma have been much higher for white females than white males with an increasing disparity. Rates of death from asthma at 5 through 34 years of age have been much greater in blacks than whites with no significant change in rates across time from 1980 through 1994. Age-adjusted rates for blacks over all ages increased from 1.5 in 1977 and 1978 to 3.5 in 1988 with rates no higher than that until an increase to 3.7 in 1994. Age-adjusted rates for whites increased from .5 in 1977 to 1.2 by 1989 with none higher than that since then through 1994. Comparison of slopes indicates a significantly greater increase for blacks than whites (F = 68.296, P < .0001). Equality of slopes tests indicate significantly greater age-adjusted rates of increase for each race separately for 1979 through 1987 compared with 1988 through 1994. CONCLUSION: Since 1988 rates of death from asthma in the United States for most ages have stabilized at rates more than 50% higher than those of 1979, but there has been only a suggestion of stabilization of rates at 5 through 34 years of age, ages at which certification of death as due to asthma is most accurate. Rates of death have been much higher for blacks than whites, and among whites rates have increased more for females than males. These differences might be due to difference in prevalence or severity of differences in accuracy of diagnosis. Improvements in management would reduce asthma mortality.     

EEG study of cervical migraine

Trained musicians able to transcribe music make fewer erros in recognition of conventional melodies when the melodies are presented in the right ear, but show more right ear than left ear errors in the recognition of random note sequences. Other subjects, even though able to read music and play instruments, make more errors across all types of stimuli when stimuli are presented in the right ear. The ability to transcribe music, but not the ability to read music or to play instruments is associated with a sharp reduction in errors for stimuli presented in either ear.