Adherence with migraine prophylaxis in clinical practice.

OBJECTIVE: To characterize adherence with antidepressants, antiepileptic drugs, and beta blockers as prophylaxis against migraine in typical clinical practice.
METHODS: Using a large US health insurance claims database (calendar years 2003 to 2005), we identified all patients with migraine who began prophylaxis with selected antidepressants, antiepileptic drugs, or beta blockers ("study agents"). Patients not continuously enrolled for 6 months prior to start of prophylaxis ("pretreatment") and for 6 months subsequently ("follow-up") were excluded. Treatment cohorts were constituted based on the type of prophylaxis received. Adherence with migraine prophylaxis was examined by type of agent received using medication possession ratios (MPRs), defined as total days with medication divided by total follow-up days. MPR < 0.80 was considered indicative of nonadherence.
RESULTS: A total of 4,634 patients met all entry criteria and received antidepressants (n = 1,803), antiepileptics (n = 1,896), or beta blockers (n = 935) on their index date. Over the next 6 months, the mean (SD) number of prescriptions for study agents was 2.7 (1.9) for antidepressants, 2.9 (2.0) for antiepileptics, and 2.8 (2.0) for beta blockers, totaling 91.0 (71.4), 98.7 (75.6), and 96.7 (73.0) therapy-days, respectively. Mean MPR at 6 months was 0.48 for antidepressants, 0.51 for antiepileptics, and 0.51 for beta blockers. By the end of the follow-up, 73.4%, 70.2%, and 67.6% of patients who initiated migraine prophylaxis with antidepressants, antiepileptics, and beta blockers, respectively, were designated nonadherent (ie, MPR < 0.80).
CONCLUSION: Our findings suggest that many patients who begin migraine prophylaxis with antidepressants, antiepileptics, or beta blockers are no longer taking these medications at 6 months.