Changing physician knowledge, attitudes, and beliefs about migraine: evaluation of a new educational intervention

Use a presurvey of primary care providers (PCPs) enrolled in a continuing medical education (CME) program on headache management to ascertain their existing knowledge, attitudes, and beliefs regarding migraine and use a postsurvey to determine the extent to which the CME program has brought participant knowledge, attitudes, and skills closer to conformance with best evidence.

Migraine is a common and debilitating condition, which PCPs may not always manage satisfactorily. In an effort to improve management, the American Headache Society has developed a CME program called BRAINSTORM that encourages PCPs to adopt the US Headache Consortium Guidelines for headache care.

A 20-item questionnaire was developed that covered the essential elements of migraine care. The questionnaire was administered before and after a BRAINSTORM presentation to 254 consenting primary care clinicians attending a medical meeting at 1 of 6 sites. A control group of 112 comparable physicians who did not attend the presentation completed the same questionnaire. Prepresentation scores of attendees were compared to scores of nonattendees to assess the generalizability of results. Prepresentation scores on selected questions were used to assess participant baseline knowledge, attitudes, and beliefs. Pre- and postpresentation scores for attendees at all sites were compared using the Mantel-Haenszel statistic to assess the effectiveness of the BRAINSTORM CME. Pre- and postpresentation scores were compared by site using the Breslow-Day test to evaluate any differential impact based on CME location.

Prepresentation scores of attendees and nonattendees were found to be similar. No significant difference in performance was noted across sites. A chi-square analysis revealed a statistically significant difference between pre- and postpresentation scores for 16 of the test's 20 questions. In the pretest, all participants scored <66% on 2 questions related to prevalence, impact, and pathophysiology of migraine, 2 questions pertaining to history taking/physical examination, and 3 migraine management questions. Attendee scores improved to >66% posttest on all except 2 questions related to prevalence, impact, and pathophysiology of migraine.

Our results indicate that PCPs need to acquire greater understanding about the epidemiology and pathophysiology of migraine and may require guidance in history taking and physical examination of migraine patients. Improvement in scores posttest confirms that the BRAINSTORM program has a significant immediate impact on the knowledge, beliefs, and attitudes of participants. The program could be strengthened to improve emphasis in some areas where posttest scores showed no improvement.