In-office discussions of migraine: results from the American Migraine Communication Study.

BACKGROUND: Research indicates that successful migraine assessment and treatment depends on information obtained during patient and healthcare professional (HCP) discussions. However, no studies outline how migraine is actually discussed during clinical encounters. OBJECTIVE: Record naturally occurring HCP-migraineur interactions, analyzing frequency and impairment assessment, and preventive treatment discussions. DESIGN: HCPs seeing high volumes of migraineurs were recruited for a communication study. Patients likely to discuss migraine were recruited immediately before their normally scheduled appointment and, once consented, were audio- and video-recorded without a researcher present. Separate post-visit interviews were conducted with patients and HCPs. All interactions were transcribed. PARTICIPANTS: Sixty patients (83% female; mean age 41.7) were analyzed. Patients were diagnosed with migraine 14 years and experienced 5 per month, on average. APPROACH: Transcripts were analyzed using sociolinguistic techniques such as number and type of questions asked and post-visit alignment on migraine frequency and impairment. American Migraine Prevalence and Prevention Study guidelines were utilized. RESULTS: Ninety-one percent of HCP-initiated, migraine-specific questions were closed-ended/short answer; assessments focused on frequency and did not focus on attention on impairment. Open-ended questions in patient post-visit interviews yielded robust impairment-related information. Post-visit, 55% of HCP-patient pairs were misaligned regarding frequency; 51% on impairment. Of the 20 (33%) patients who were preventive medication candidates, 80% did not receive it and 50% of their visits lacked discussion of prevention. CONCLUSIONS: Sociolinguistic analysis revealed that HCPs often used narrowly focused, closed-ended questions and were often unaware of how migraine affected patients' lives as a result. It is recommended that HCPs assess impairment using open-ended questions in combination with the ask-tell-ask technique.

Healthcare provider-patient communication and migraine assessment: results of the American Migraine Communication Study, phase II.

OBJECTIVE: The American Migraine Communication Study I (AMCS I) revealed communication deficits illustrated by differing healthcare professional (HCP) and patient reports about issues such as impairment and frequency. AMCS II was designed to assess an intervention using open-ended questions about impairment and 'ask-tell-ask' sequences to confirm headache frequency in days versus attacks. RESEARCH DESIGN AND METHODS: HCPs who participated in AMCS I completed an internet-based intervention. Researchers were sent to HCPs' offices, and patients likely to discuss migraine were recruited immediately prior to normally-scheduled appointments. Post-consent, visits were recorded without a researcher present. Separate post-visit interviews were conducted with all parties. All interactions were transcribed. MAIN OUTCOME MEASURES: Transcripts were analyzed using validated sociolinguistic techniques, and study results were compared to AMCS I. RESULTS: HCPs assessed impairment in 90% of interactions compared to 10% in AMCS I (p<0.0001) and used open-ended questions to assess impairment in 55% of visits (95% CI: 0.4261-0.6598). Impairment between attacks was discussed in 37% of visits vs. 0% in AMCS I (p<0.0001). HCPs completed full ask-tell-ask sequences in 29% of visits (95% CI: 0.1921-0.4070). AMCS II contained more discussions of migraine preventive therapy with appropriate candidates compared to AMCS I (74 vs. 50%; p=0.069) without statistically increasing median visit length (9:36 vs. 11:00; p=0.668). Post-visit, HCPs and patients were often aligned about impairment and frequency and reported high levels of satisfaction. CONCLUSIONS: Although further research with a larger sample is needed, a brief, internet-based intervention appears to promote positive communication changes not associated with increased visit length.