A retrospective real-world evidence assessment of association of the Migraine Disability Assessment Score (MIDAS) with healthcare utilization among patients with episodic migraines in the U.S.




LaVallee, John Christopher

Journal Title

Journal ISSN

Volume Title



The impact of migraine-related disability on healthcare resources utilization (HRU) is unclear. The objective was to determine the relationship between Migraine Disability Assessment Scores (MIDAS) and HRU among people with episodic migraines (EM) treated in a real-world setting.

Data were obtained from the DRG RWD Repository. Adults with EM and ≥ 1 completed MIDAS survey, 01/2016 to 12/2018, were included. Descriptive statistics and inferential tests assessed the relationship between MIDAS scores and HRU. Results were then stratified by physician type, patients treated only in primary care (PC) and patients with at least one specialist (SC) visit.

In total, 756 patients were included; mean age was 47.8 (14.9); 87.7% were female; 49.6% resided in the South. MIDAS distribution was, in order of lowest disability to highest disability: 31.7% Grade I; 13.9% Grade II; 19.4% Grade III; 14.9% Grade IV-A; and 20.0% Grade IV-B. Grade I patients were the oldest (51-years) and included a higher percentage of men (38%-vs-31% women). A quasi-linear trend was seen in the proportion of patients with a headache diagnosis and MIDAS severity (47% to 51%). The subset of patients with at least one SC visit was 533 (70.5%), leaving 223 (29.5%) PC patients, with no differences in MIDAS categories by physician type.

Utilization for 6-months post-index was assessed. No major trends were detected comparing MIDAS scores and HRU. The most common preventative medications were antidepressants (23.4%); lower for SC patients (18.8%) than PC patients (34.5%). The most common relief medications included triptans (28.7%), again lower for SC patients (24.6%) than PC (38.6%). Opioids were used by about 18% of patients for both practitioner types. A higher proportion of patients with SC received care in an office setting (81.8%-vs-73.5%) and/or the ER (10.1-vs-6.3%). Secondary analyses assessing Lost Productivity Time (LPT) found similar trends.

For the 492 patients with at least two MIDAS scores, 49% (N=243) showed a reduction in their MIDAS category, indicating improvement.

No major trends were detected comparing MIDAS scores and HRU. Medication utilization was lower for the SC group, while the percent of patients with an office visit or ER visit was higher.


LCSH Subject Headings