Cohort Definition Name : Earliest event of Migraine, including history of migraine
Contributor name : Azza Shoaibi’, ‘Jill Hardin’
Contributor OrcId : 0000-0002-6976-2594’
Logic Description : Earliest occurrence of Migraine indexed on occurrence of Migraine condition (or observation of history of migraine-in the one of the cohorts versions) or symptoms (Headache , Aura) or a use of antimigraine drug (or the 1st time). Patients entering the cohort with an antimigraine drugs or migraine symptoms are required to have a diagnosis or observation of Migraine or history of migraine within 60 days. Cohort exit is the end of continuous observation.
Recommended study application : target or outcome
Assertion statement : This cohort definition was executed on at least one real person-level observational health data source and resulted in a cohort with at least 1 person.
Target Clinical Description : Migraine is a recurring headache syndrome typically accompanied by a mixture of other neurologic dysfunction symptoms. It is usually an episodic headache associated with sensitivity to light, sound, or movement; however, it may vary considerably among patients. The understanding of pathogenesis has evolved from an intracranial vasoconstriction theory to a complex neuronal dysfunction leading to a sequence of changes intra- and extra-cranially. Several neuropeptides, including CGRP, are involved in the vascular response and neuroinflammation. Migraine triggers may include stress, bright lights, hunger, physical exertion, hormonal changes during a menstrual period, sleep changes, and diet. The disease burden is relatively high, with an approximate prevalence of 15% among women and 6% among men. The burden appears to be even higher in the US, with the prevalence reported at 20% and 10%, respectively. The highest burden is in the age subgroup of 18-44 years old and is inversely associated with income and educational attainment. Migraine may result in work-related disability and productivity loss.
Evaluation conclusion : We created phenotype algorithms to identify migraine phenotype in observational data. We used Cohort diagnostics and phevalutor to help gain insights and evaluate the performance of the cohort definitions. We performed the evaluation across a network of claim data sources and 1 EHR US data source. The data sources are:
IBM® MarketScan® Commercial Database (CCAE), Optum’s longitudinal EHR repository (Optum EHR), Optum’s Clinformatics® Data Mart (DOD), IBM® MarketScan® Multi-State Medicaid Database (MDCD), IBM® MarketScan® Medicare Supplemental Database (MDCR), Japan Claims Database (JMDC), Clinical Practice Research Datalink (CPRD) , IQVIA® Australia Longitudinal Patient Data (LPD) database (Australia), IQVIA® Disease Analyzer (DA) France database (France), QVIA® Disease Analyzer (DA) Germany database (Germany), IQVIA® Adjudicated Health Plan Claims Data (formerly PharMetrics Plus) - US database (PharMetrics), IQVIA® Ambulatory EMR (EMR).
We recommend one version of the cohort for incidence migraine and we recommed the other to reprsenet patients with migraine (prevalent cohort) . Both cohorts have an entry event criteria of 1. migraine diagnosis (including the history of migraine for the prevalent cohort but not the incidence) or 2. use of antimigraine drugs or 3. symptoms of headache or aura but requiring a diagnosis of migraine within 60 days. The index date correction was incorporated by allowing an entry criterion based on symptoms or drugs after observing a significant (more than 10%) of patients with symptoms and/or drug proceeding the migraine diagnosis.
The incidence rate of migraine using the recommended cohort was in line with expectations, with higher rates among females and higher rates for ages 39-45. The observed difference in rates between females and males ranged between 2 to 6 times while prior studies reported that migraines occur in up to 17 percent of females and 6 percent of males each year (UP to Date).We examined the sensitivity of the recommended cohort by assessing the overlap of the two cohorts with a cohort of antimigraine users. As illustrated in cohort diagnostics, the overlap ranged between 11 to 43% across data sources and averaged around 26%, with the poorest overlap in MDCR. The poor overlap may be explained by the fact that antimigraine drugs have other indications, especially in databases like MDCR with an older population. However, it may indicate poor sensitivity, at least in some data sources. The index event breakdown showed that the majority of patients enter the cohort due to the code “Migraine, unspecified” which does not allow us to assess the distribution of diverse subtypes (ie Migraine with or without aura). 13-42% of patients started an antimigraine drug around index (with the lowest rate in MDCR) and 44% to 63.1% had a neurological examination or a neurology visit on the day of index in US data sources. In addition, the most common condition around index is in line with expectations (pain, Anxiety disorder, Dizziness and giddiness, vomiting, nausea, visual and sleep disturbances) specially in JMDC . These characteristics suggest that patients captured by the two definitions truly have migraine (acceptable PPV). Phevaluator results indicate that the PPV ranged from 0.94 (in CCAE) to 0.36 (incident 11703 - EHR) and sensitivity ranged from 0.94 (prevalent – CCAE) to 0.68 (incident with index date misclassification 11703 – JMDC).