Aim: Within a treated migraine population, to evaluate if the sub-group meeting criteria for high disease-specific total costs is significantly different to the sub-group with medium and/or low-costs, and to identify the associated risk factors. Methods: Data from the Household Component of Medical Expenditure Panel Survey (MEPS-HC, 2008-2012), a nationally representative survey of non-institutionalized civilians in the US, were analyzed. Key inclusion criteria were migraine diagnosis (ICD-9 code: 346.XX) and prescribed treatment for migraine. Patients were categorized into high (>top 10th percentile), low (<bottom 10th percentile), and medium (between high and low) cost sub-groups per migraine-specific total costs. Logistic regression models were applied to identify predictors of high vs medium and medium vs low-costs. Preventive eligibility, defined as (i) past/current use of migraine preventives or (ii) overuse of acute medications, was compared to non-preventive eligibility. Results: Within the treated migraine cohort (n = 1,735), the mean age was 39 years, 80% were female, and the majority were in the medium-cost sub-group (n = 1,360) (low-cost n = 190, high-cost n = 195). Significant predictors of high vs medium-costs were low SF-12 Physical Composite Scores (OR = 0.95; 95% CI = 0.92-0.97), low SF-6D health utility index scores (OR = 0.019; 95% CI = 0.002-0.193), preventive eligibility-i (OR = 0.019; 95% CI = 0.002-0.193), and preventive-eligibility-ii (OR = 3.10; 95% CI = 1.62-5.91). Statistically significant (p < 0.05) predictors of medium vs low-costs included anxiety, Fleishman score, preventive-eligible-i, and preventive-eligible-ii. Conclusions: Among patients treated for migraine, distinct characteristics, including patient-functioning measures and comorbidities, are predictive of high vs medium-costs, and medium vs low-costs. Preventive eligibility is a predictor of being in the higher cost sub-groups; however, preventive treatments that improve functioning and reduce acute medication use have the potential to reduce migraine-specific costs. Limitations: The results are limited to a population that is diagnosed and treated for migraine. Over-the-counter medication use, and migraine headache frequency and severity were not captured.
Aim: Within a treated migraine population, to evaluate if the sub-group meeting criteria for high disease-specific total costs is significantly different to the sub-group with medium and/or low-costs, and to identify the associated risk factors. Methods: Data from the Household Component of Medical Expenditure Panel Survey (MEPS-HC, 2008-2012), a nationally representative survey of non-institutionalized civilians in the US, were analyzed. Key inclusion criteria were migraine diagnosis (ICD-9 code: 346.XX) and prescribed treatment for migraine. Patients were categorized into high (>top 10th percentile), low (<bottom 10th percentile), and medium (between high and low) cost sub-groups per migraine-specific total costs. Logistic regression models were applied to identify predictors of high vs medium and medium vs low-costs. Preventive eligibility, defined as (i) past/current use of migraine preventives or (ii) overuse of acute medications, was compared to non-preventive eligibility. Results: Within the treated migraine cohort (n = 1,735), the mean age was 39 years, 80% were female, and the majority were in the medium-cost sub-group (n = 1,360) (low-cost n = 190, high-cost n = 195). Significant predictors of high vs medium-costs were low SF-12 Physical Composite Scores (OR = 0.95; 95% CI = 0.92-0.97), low SF-6D health utility index scores (OR = 0.019; 95% CI = 0.002-0.193), preventive eligibility-i (OR = 0.019; 95% CI = 0.002-0.193), and preventive-eligibility-ii (OR = 3.10; 95% CI = 1.62-5.91). Statistically significant (p < 0.05) predictors of medium vs low-costs included anxiety, Fleishman score, preventive-eligible-i, and preventive-eligible-ii. Conclusions: Among patients treated for migraine, distinct characteristics, including patient-functioning measures and comorbidities, are predictive of high vs medium-costs, and medium vs low-costs. Preventive eligibility is a predictor of being in the higher cost sub-groups; however, preventive treatments that improve functioning and reduce acute medication use have the potential to reduce migraine-specific costs. Limitations: The results are limited to a population that is diagnosed and treated for migraine. Over-the-counter medication use, and migraine headache frequency and severity were not captured.
Authors: Ariane K Kawata; Neel Shah; Jiat-Ling Poon; Shannon Shaffer; Sandhya Sapra; Teresa K Wilcox; Shweta Shah; Stewart J Tepper; David W Dodick; Richard B Lipton Journal: Headache Date: 2021-02-16 Impact factor: 5.887