Acute treatment optimization in episodic and chronic migraine: results of the American Migraine Prevalence and Prevention (AMPP) Study.

BACKGROUND: The migraine Treatment Optimization Questionnaire (mTOQ) was developed to assess response to acute treatment in persons with migraine. The original validated form used yes or no response options.
OBJECTIVES: This study aims to (1) assess the psychometric properties of a 6-item version of the mTOQ (mTOQ-6) using ordinal response options; (2) compare treatment optimization using the revised mTOQ-6 for both episodic and chronic migraine (EM and CM, respectively); (3) identify demographic, headache, and treatment features associated with treatment optimization.
METHODS: The American Migraine Prevalence and Prevention (AMPP) Study is a longitudinal, US population-based study. Annual questionnaires were mailed to a sample of 24,000 severe headache sufferers identified by screening a panel constructed to be representative of the US population. The current study included respondents to the 2006 AMPP Study survey who met modified International Classification of Headache Disorders-3 beta criteria for migraine; persons with CM (≥15 HA days/month) or EM (<15 HA days/month) were included. Acute treatment optimization was measured with the mTOQ-6. A single factor latent variable model was used to assess item characteristics. This model was expanded through structural equation models (SEM) to incorporate a contrast between persons with CM and EM on the scaled treatment optimization scores. We estimated both an unadjusted SEM and a SEM adjusted for demographic features, headache characteristics, and acute treatment.
RESULTS: Migraine criteria were met by 8612 persons (539 for CM and 8073 for EM) who completed the mTOQ-6 as part of the 2006 AMPP Study survey. When compared, those with CM exhibited worse treatment optimization across all domains of the mTOQ-6. For example, 35.1% of CM and 44.6% of EM respondents reported being pain free at 2 hours "half the time or more" with their usual migraine medication. Latent variable model parameters indicated excellent psychometric properties of the mTOQ-6. Scaled treatment optimization scores obtained from the unadjusted SEM were significantly lower (indicating worse treatment optimization) for persons with CM (3.25) compared to persons with EM (4.01), b = -0.76, P < .0001; scores remained significantly lower for CM after adjustment with a wide array of demographic and disease severity covariates. Poor treatment optimization was associated with cutaneous allodynia, major depression, and the use of nonsteroidal anti-inflammatory drugs. Better treatment optimization was associated with the use of triptans and preventive medications.
CONCLUSION: Estimates of the latent variable scores for the mTOQ-6 revealed persistent low levels of treatment optimization for both EM and CM, though treatment optimization is worse for CM.