The use of triptans, opioids, and other rescue medications among patients with migraine prescribed calcitonin gene–related peptide inhibitors for migraine prophylaxis

BACKGROUND: Migraine attacks can be prevalent and debilitating, affecting 40 million adults in the United States. Patients can be prescribed prophylaxis (eg, β blockers, anticonvulsants, antidepressants) and rescue medications (eg, triptans, opioids [which are not recommended for use to treat migraine]) to treat and/or relieve migraine. Calcitonin gene–related peptide inhibitors (CGRPi) are the newest class of drugs approved for migraine prophylaxis. Patients taking CGRPi report reduced migraine frequency and severity, which should result in lower use of rescue migraine medications. OBJECTIVE: To compare the use of triptans, rescue medications, and opioids for patients with migraines prescribed preventive CGRPi (pCGRi) and patients who are not prescribed pCGRPi. METHODS: This retrospective cohort study used Komodo Health medical and pharmacy claims data from 2018 to 2023. Adults aged 18 to 60 years with a chronic migraine diagnosis and continuous enrollment for 12 months before and after index were included if they initiated a migraine medication (pCGRPi or non-CGRPi) during the index period. Outcomes included use of triptans, opioids, and other rescue medications in the post-index period compared with the pre-index period. Propensity score–matched cohorts were analyzed using adjusted logistic regression models with a treatment-by-time interaction term to estimate relative changes over time between cohorts. RESULTS: In adjusted analyses, compared with the non-CGRPi cohort, the pCGRPi cohort experienced lower odds in triptan use (interaction odds ratio [OR] = 0.57; 95% CI = 0.53-0.61; P < 0.001) and in opioid use (interaction OR = 0.92; 95% CI = 0.85-0.99; P = 0.033). Although rescue medication use increased over time, the relative pre-post change did not differ significantly between cohorts (interaction OR = 0.96; 95% CI = 0.89-1.04; P = 0.28). Sensitivity analyses restricted to neurologist-prescribed index medications showed consistent findings for triptans, but not for use of opioids or rescue medications. CONCLUSIONS: In the cohort prescribed pCGRPi, there was a statistically significant association with decreased of triptan use over time. Comparatively, the non-CGRPi cohort had statistically significant increases in the use of triptans, opioids, and other rescue medications over time. These findings support that prophylactic use of pCGRPi may be associated with a reduced reliance on migraine-specific rescue medications.