OBJECTIVES: This systematic literature review (SLR) aimed to identify health state utility values among migraine patients. METHODS: A SLR was conducted in biomedical databases and Health Technology Assessment (HTA) websites from 2000-2014. Search terms included different combinations of migraine along with utility or quality-of-life measures. References of identified SLR/meta-analyses were investigated to identify additional studies. Each study was reviewed by two independent reviewers; disagreements were resolved by a third reviewer. RESULTS: The search yielded 29 publications (including a submission to the National Institute for Health and Care Excellence) from which 23 studies were extracted. The studies were mainly conducted in US (n=6), Canada (n=5), or UK (n=3). Sample sizes varied from 14 to >8000.  Age (mean: 21-47.5 years) was reported in 16 studies while disease duration (mean: 2.3-24 years) only in four. Proportion of females ranged between 70-80%. Pain and psychiatric disorders were the most common comorbidities. Migraine was primarily defined using the International Headache Society’s International Classification of Headache Disorders criteria (n=10). Most studies used validated utility instruments: EQ-5D (n=12), Health Utility Index-Mark 3 (n=7) or SF-36 (n=1) to assess utility values. Three studies used the Quality of Well Being scale, and two used custom questionnaires, using Time Trade Off or Standard Gamble techniques. Utilities values were reported using the following categories: by migraine pain severity (pain-free [0.87-0.96], mild [0.66-0.84], moderate [0.51-0.77] or severe [0.39-0.54]), by number of migraine days/month (or percentage reduction in migraine frequency: <50% to ≥75% associated with a utility increase: 0.001 to 0.010), or by migraine type (chronic [0.43-0.89]/episodic [0.62-0.91]). CONCLUSIONS: This SLR confirmed that increased migraine frequency and pain severity were associated with lower utility values among migraineures. A wide range of utility values were reported; potential differences across reported estimates are likely due to different populations or different methodologies implemented to derive estimates.