An epidemiological case-control study of migraine and its associated comorbid conditions

BACKGROUND: The economic impact of migraine to society has been well documented. However, few studies have identified the risk of migraine-specific comorbid conditions to explain the higher healthcare resource utilization by migraineurs. Because comorbidity has the potential to increase treatment costs, estimating its risks and its economc impact becomes imperative. OBJECTIVES: To estimate the risk for migraine by presence of a defined set of comorbidities, and to quantify the association of estimated healthcare costs between the case-control groups with and without comorbidity. METHODS: A matched case-control was designed using data from 1998 Idaho Medicaid claims. Cases were identified by physician-diagnoses and migraine-specific medication use. A 1:1 matching was performed on age, race, gender, and residence. Presence of comorbidity was determined by ICD-9 diagnostic categories. Total healthcare resource utilization was estimaed for all physician, hospital (inpatient and outpatient), and emergency services, including prescription medicines paid by Idaho Medicaid. Multivariate logistic regression was used to compute the odds ratio controlling for the matching variables. Multiple-regression models were used to quantify the association of total healthcare costs. RESULTS: Migraine patients were approximately five times more likely (Odds Ratio = 5.34; Confidence Interval95%–4.38 to 6.52) than their controls to be diagnosed with comorbid condition(s). A model with variables of age, gender, presence of comorbidity and migraine, explained approximately 21% of the variance in healthcare utilization (p < 0.05). CONCLUSIONS: Migraine patients are at a higher risk for defined comorbid conditions resulting in significantly higher healthcare costs as compared to non-migraine patients.