The Course of Frequent Episodic Migraine in a Large Headache Clinic Population: A 12-year Retrospective Follow-up Study

Carl G. H. Dahlöf, MD, PhD; Maria Johansson, PhD; Susanne Casserstedt, PhD; Tina Motallebzadeh, PhD


Headache. 2009;49(8):1144-1152. 

In This Article


Study Population

The Gothenburg Migraine Clinic (GMC) was founded in 1991 as Sweden's first tertiary headache clinic. Patients who consult at GMC are usually referred from primary care and regional hospitals, but a minority are self-referred. All patients have been seen by the first author (C.D.) or his colleagues (4 in total over 15 years, all trained and supervised by C.D.).

At the GMC, a standardized questionnaire based partly on the ICHD diagnostic criteria has been developed. This questionnaire is called the Migraine File, and it is utilized for all patients at their first visit. The information obtained is entered into a database (Microsoft Office Access 2003) to facilitate statistical analysis. As of September 2005, 5456 patients (4159 women and 1297 men) with migraine (with or without aura) had been evaluated at the GMC, and data from these patients have been collected and stored in the Migraine File database.

Eligible to participate in the study were men and women in the Migraine File database diagnosed before December 31, 1996, with migraine without aura (MwoA 1.1) or migraine with aura (MwA 1.2) according to ICHD criteria.[26] From a total population of 2812 eligible patients so diagnosed and with a baseline attack frequency of 1 to 6 attacks per month, 374 subjects (200 women, 174 men) were randomly selected by means of a list of random sampling numbers and divided into 5 groups depending on age of migraine onset (0–9, 10–19, 20–29, 30–39, 40–49). The upper limit for age of onset was based on the fact that migraine prevalence peaks in middle-age and decreases thereafter.[2,31,32] Our goal was to enroll 40 women and 40 men in each of the 5 age groups. Although the male : female migraine prevalence ratio in the general population is 1 : 2–3, we used a ratio of 1 : 1 in this study because we planned to analyze the male and female groups independent of one another.

The randomly selected participants were first sent a letter to inform them about the purpose of the study and their right to decline participation without this affecting their future relationship with the GMC.

Study Questionnaire and Interview

The 374 randomly selected subjects were contacted by telephone and interviewed with the aid of a newly constructed questionnaire partly based on the Migraine File. The interviews were carried out between September 2005 and December 2006. The information obtained in 2005/2006 was used to identify 2 subgroups: persistent migraineurs and past (resolved) migraineurs. We then compared baseline information in the Migraine File from the subjects' initial visits to determine if there were any differences between these 2 subgroups.

In the interview, all participants were asked about right or left handedness, the year of migraine onset, height, and weight. This was followed by the question "Do you still suffer from migraine?" If the answer was "No," the subjects were asked when their attacks ceased and their opinion as to why; then the interview was closed. If the answer was "Yes," the migraineurs were asked about their medical consultations, any family history of migraine, premonitory symptoms, aura, headache lateralization, phonophobia, photophobia, nausea, vomiting, triggering factors, character (pulsatile vs non-pulsatile) and intensity of pain, and frequency and duration of attacks. Further, there were questions about past and present treatment, allergies, smoking and drinking habits, and occurrence of head trauma. Women were asked about menstruation, pregnancy, and the use of contraceptives.

Each interview took approximately 30 minutes to complete. After 5 unsuccessful attempts at contact, a subject was regarded as lost to follow up. Subjects who were deceased, no longer registered as a Swedish citizen, or had moved abroad were also regarded as lost to follow up.

Statistical Analysis

Differences considered to have potential clinical relevance arbitrarily were set at 10%. Where this was exceeded, a statistical analysis was performed and P values calculated using chi-squared tests. The hypotheses that the statistical tests were based upon were defined as H0 = remission of migraine is independent of the variable and H1 = there is an association between remission of migraine and the variable. Statistical significance was set to ≤.05. As this was intended to be primarily a descriptive study and not designed to measure differences in detail, no adjustments were made for multiple analyses. Data were processed with Microsoft Access and Microsoft Excel.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.