Methodologies and Data Used in Migraine Cost‐of‐Illness Studies: A Scoping Review for the New Zealand Context

Migraine causes significant costs to health systems and society, which have not been quantified in New Zealand (NZ), although Māori and Pacific peoples have higher rates of undiagnosed migraine. This scoping review evaluated cost‐of‐illness (COI) studies measuring direct and/or indirect costs of migraine from countries with comparable health systems to NZ. A systematic search including Medline, EMBASE, Scopus and EconLit until 28 May 2025 identified 6266 records. After screening and eligibility assessment, 134 studies were included, of which 105 assessed direct and 67 assessed indirect costs. Data were extracted on study characteristics, cost analysis, methodology and reporting. Narrative synthesis involved a critique of study approaches based on potential for selection bias, measurement bias and confounding. Studies were published from 1992 to 2025, predominantly from the United States (45%) and funded by industry (64%). Data sources included administrative health and insurance claims datasets (for a top‐down costing approach), national or panel, patient group and workplace surveys (for a bottom‐up costing approach) and a mix of sources. Top‐down approaches were subject to selection bias and often lacked key cost inputs. Bottom‐up approaches could reduce selection bias and collected more extensive data. Measurement of lost productivity lacked standardisation. Only 20% of studies performed sensitivity analyses. Comprehensive data on migraine prevalence, direct and indirect costs are needed for a future COI study in NZ and these data must be representative of Māori and Pacific Peoples or will underestimate the burden in these populations. Bias and confounding can be reduced by using an accurate case definition of migraine and controlling for comorbidities.