Migraine Disability Assessment Questionnaire

1. Abstract

The Migraine Disability Assessment Questionnaire (commonly known by its acronym, MIDAS) is a brief, psychometrically validated, self-administered questionnaire designed to quantify headache-related functional disability across personal, occupational, and social domains over a retrospective three-month (90-day) recall window. Developed by Richard B. Lipton, Walter F. Stewart, and colleagues in the late 1990s, the instrument was conceived to overcome the limitations of measuring headache burden purely through physiological indices such as attack frequency or episodic pain intensity. MIDAS comprises five core evaluative items measuring lost time due to complete absence (absenteeism) and marked reduction in efficiency (presenteeism, defined as productivity decreased by 50% or more) across three critical behavioral domains: employment or educational attendance (Items 1 and 2), household chores and domestic management (Items 3 and 4), and family, social, or leisure engagements (Item 5). Two additional supplementary diagnostic items (Questions A and B) assess headache frequency in days and average pain intensity on an 11-point numeric rating scale (0–10), providing treating clinicians with essential context without contributing to the overall disability score.

The total MIDAS score is derived from the direct arithmetic summation of days reported across the five core items (theoretical range: 0–90 lost or significantly compromised days per item, capped conceptually across domains, yielding total aggregate scores typically categorized into four empirically validated functional disability grades: Grade I [0–5 days, little to no disability], Grade II [6–10 days, mild disability], Grade III [11–20 days, moderate disability], and Grade IV [21+ days, severe disability]). Extensive international psychometric evaluations have demonstrated robust properties, including high internal consistency (Cronbach’s alpha coefficients consistently ranging from 0.76 to 0.86), exceptional test-retest reliability across both clinic-based and population-based cohorts (Spearman rank correlation coefficients and Pearson r typically between 0.75 and 0.84), and substantial convergent validity when correlated against prospectively maintained daily headache diaries and the Headache Impact Test (HIT-6). The instrument serves as a cornerstone in stratified headache care, guiding therapeutic escalations from simple analgesics to triptans, calcitonin gene-related peptide (CGRP) antagonists, or preventative pharmacological regimens.

2. Keywords

Migraine Disability Assessment, MIDAS, headache-related disability, functional impairment, absenteeism, presenteeism, psychometrics, headache diary, cephalalgia, migraine grading

3. Authors

The Migraine Disability Assessment Questionnaire was conceptualized and developed by an international team of neuro-epidemiologists and clinical neurologists:

• Richard B. Lipton, M.D. — Department of Neurology, Epidemiology, and Population Health, Albert Einstein College of Medicine and Montefiore Medical Center, Bronx, New York, United States.
• Walter F. Stewart, Ph.D., M.P.H. — Department of Epidemiology, Johns Hopkins Bloomberg School of Public Health, Baltimore, Maryland, and the Center for Health Research, Geisinger Health System, Danville, Pennsylvania, United States.
• K.M.A. Welch, M.B., Ch.B., F.R.C.P. — Department of Neurology, University of Kansas School of Medicine, Kansas City, Kansas, and Rosalind Franklin University of Medicine and Science, North Chicago, Illinois, United States.
• Dutch Adaptation & Validation: Pijn Kennis Centrum Maastricht (Pain Knowledge Center Maastricht), Maastricht University Medical Centre, the Netherlands (1998).

Correspondence regarding the original epidemiological development of the instrument was historically directed through the Department of Neurology at the Albert Einstein College of Medicine.

4. Purpose

The primary clinical and research objective of the Migraine Disability Assessment Questionnaire is to provide an objective, standardized, and interpretable metric of the burden of illness imposed by migraine and other primary headache disorders over an extended, ecologically valid interval. Historically, clinical assessments of migraine severity prioritized biological parameters, such as the number of headache attacks per month, the peak intensity of pain measured on visual analog scales (VAS), and associated autonomic symptoms including nausea, photophobia, and phonophobia. However, these physiological metrics routinely fail to capture the profound multidimensional impact of recurrent headache episodes on an individual’s actual functional capacity, role performance, and socioeconomic integration.

MIDAS was engineered specifically to bridge this communicative and clinical gap by translating subjective pain experiences into a concrete, tangible behavioral metric: lost productive time, expressed simply as days of lost or substantially reduced activity. By evaluating both complete inactivity (such as work absence or complete abandonment of family duties) and partial decrements in function (productivity reduced by 50% or more), MIDAS encapsulates both the visible and invisible costs of headache disorders.

In routine clinical practice, MIDAS functions as a catalyst for physician-patient communication. Patients often minimize or inadequately communicate their functional losses during brief outpatient consultations, while clinicians may under-prescribe disease-specific acute treatments (such as triptans) or prophylactic medications in the absence of explicit disability documentation. When integrated into the initial assessment protocol, MIDAS facilitates stratified care—a treatment paradigm demonstrated in randomized clinical trials to produce faster headache relief, lower rates of adverse events, and greater economic efficiency compared to traditional stepped-care approaches. Under a stratified care model, individuals scoring in Grade III or Grade IV disability are immediately considered candidates for aggressive acute intervention and preventative migraine therapies.

In public health, occupational health, and clinical trial research, MIDAS serves as a standardized outcome measure. It enables health economists to compute direct and indirect socioeconomic costs—particularly the economic burden of presenteeism, which often exceeds the economic toll of overt absenteeism in migraine cohorts. Furthermore, longitudinal administration of MIDAS permits researchers to evaluate whether novel pharmacological compounds, behavioral interventions, or lifestyle modifications succeed in restoring real-world functioning over three-month intervals.

5. Psychological Construct

The psychological and behavioral construct operationalized by MIDAS is headache-attributed functional disability, defined within the conceptual parameters of the World Health Organization’s International Classification of Functioning, Disability and Health (ICF) as activity limitations and participation restrictions resulting directly from a health condition. Rather than assessing affective distress, psychological coping styles, or generalized health-related quality of life (HRQoL), MIDAS isolates behavioral output limitations across three specific life domains:

1. Occupational and Academic Functioning (Items 1 and 2)

This subscale captures the economic and occupational toll of primary headache. It bifurcates lost workplace utility into two discrete constructs:

• Absenteeism (Item 1): Total physical absence from paid employment or educational instruction due to headache symptoms. This constitutes absolute role cessation.
• Presenteeism / Underproductivity (Item 2): Days in which the patient reported to the workplace or classroom, yet experienced a functional capacity reduction of at least 50% attributable to cephalalgia and associated neurological manifestations (e.g., cognitive slowing, sensory hypersensitivity). By explicitly instructing respondents to avoid double-counting days enumerated in Item 1, this item ensures discriminant validity between attendance failure and on-the-job performance decay.

2. Domestic Role Performance / Household Work (Items 3 and 4)

A persistent limitation of historical socioeconomic burden instruments was the systematic neglect of unpaid domestic work, disproportionately underestimating the burden of migraine in homemakers, part-time workers, and individuals who manage household infrastructures. This dimension assesses:

• Domestic Role Inactivity (Item 3): Days on which domestic responsibilities (e.g., cooking, cleaning, child rearing, grocery shopping, home maintenance) were completely omitted.
• Domestic Role Inefficiency (Item 4): Days on which domestic labor output was degraded by half or more, isolating reduced performance in private, non-monetized settings.

3. Interpersonal, Familial, and Leisure Participation (Item 5)

The fifth item isolates the psychosocial, interpersonal, and recreational impact of migraine. Migraine attacks are intermittent and unpredictable, frequently forcing individuals to cancel social gatherings, family milestones, sporting events, or leisure pursuits. Unlike work settings, where contractual obligations may enforce physical presence despite severe pain, social engagements are often the first activities sacrificed by migraine sufferers attempting to conserve energy for vocational demands. Item 5 captures this discrete facet of psychological isolation and leisure disruption.

Supplementary Clinical Indices (Questions A and B)

Questions A and B quantify headache frequency (total headache days during the recall period) and mean episodic intensity (0–10 numeric rating scale). Psychometrically, these supplementary items do not measure disability per se; instead, they function as clinical modifiers that contextualize the severity of the primary five-item construct. For instance, an individual with 30 headache days but only 3 disability days demonstrates distinct clinical coping dynamics compared to an individual with 4 headache days that yielded 4 days of complete incapacity.

6. Theoretical Framework

The conceptual framework underpinning MIDAS draws from behavioral medicine, health economics, and the biopsychosocial model of chronic pain. Historically, headache medicine was governed by biomedical reductionism, which presupposed that subjective pain intensity correlates linearly with life disruption. Clinical observations repeatedly contradicted this assumption: two individuals experiencing identical neurovascular activation and visual analog scale pain ratings can exhibit vastly divergent degrees of vocational absenteeism and psychosocial dysfunction.

To reconcile these divergences, Lipton, Stewart, and Welch operationalized functional disability within a behavioral framework grounded in the quantification of temporal disruption. Drawing from the work of Nagi (1965) and the evolving disablement paradigms that culminated in the WHO ICF framework, the authors posited that pathology (e.g., neurogenic inflammation, trigeminovascular sensitization) produces an impairment (cephalalgic pain, sensory hypersensitivity), which subsequently leads to functional limitations (reduced reading speed, inability to tolerate computer screens or physical exertion) and ultimately manifests as disability or social role limitation (inability to attend work, neglect of familial care).

A critical psychometric insight embedded within the MIDAS framework is the 50% threshold heuristic for presenteeism. Psychometric investigations during instrument construction indicated that asking patients to rate continuous, fractional degrees of work efficiency loss (e.g., 10%, 25%, 65%) introduced profound cognitive burden, subjective bias, and recall noise. By establishing a discrete, operationalized behavioral anchor—”productivity reduced by half or more”—the scale transformed a nebulous perceptual estimate into a binary cognitive decision rule. If a patient accomplished less than half of their expected tasks, the day is tallied; if they managed more than half, the day is excluded.

Furthermore, the selection of a three-month (90-day) retrospective recall window represents a carefully calculated psychometric compromise. Migraine is fundamentally an episodic disorder with substantial intra-individual temporal variability; evaluating a 7-day or 14-day interval risks capturing an atypical period of temporary remission or, conversely, an acute cluster of status migrainosus. Conversely, a 12-month recall period is profoundly degraded by human cognitive decay and recency effects. Methodological investigations confirmed that a 90-day recall period accurately reflects long-term migraine patterns while maintaining acceptable concordance with daily prospective diary recordings.

7. Validity

The psychometric validity of the MIDAS questionnaire has been evaluated across diverse international clinical trials, headache referral centers, and large-scale general population samples, supporting its construct, convergent, and predictive utility.

1. Construct and Factorial Validity

Construct validity was initially established by demonstrating that MIDAS total scores correlate robustly with physician global assessments of disease severity and independently discriminate between clinical diagnostic categories established by the International Headache Society (IHS). Patients diagnosed with chronic migraine or high-frequency episodic migraine consistently exhibit statistically significant, higher MIDAS scores compared to those with infrequent episodic tension-type headache.

2. Convergent Validity (Comparison to Daily Headache Diaries)

The gold standard for validating retrospective recall instruments in pain medicine is the prospective daily diary. In their seminal validation studies, Stewart, Lipton, and colleagues (2000) compared retrospective MIDAS responses against 90-day prospective daily diary recordings in both clinic-based and population-based cohorts. The correlation between the MIDAS score and diary-derived measures of disability was exceptionally strong:

• Work/school days missed (MIDAS Item 1 vs. Diary): Spearman’s correlation coefficient $rho = 0.74$ to $0.79$.
• Household work missed (MIDAS Item 3 vs. Diary): Spearman’s $rho = 0.65$ to $0.72$.
• Overall disability days composite: Pearson correlation coefficients typically exceeded $r = 0.63$ ($p < 0.001$), demonstrating that respondents do not substantially over- or underestimate their aggregate lost time when utilizing the MIDAS format.

Additionally, convergent validity with alternative health-related quality of life measures, such as the SF-36 Health Survey, revealed significant inverse correlations between the MIDAS total score and the SF-36 Physical Functioning ($r = -0.42$), Role Physical ($r = -0.56$), and Social Functioning ($r = -0.48$) subscales.

3. Predictive and Clinical Validity

The predictive validity of MIDAS has been established through clinical trials evaluating stratified care management strategies (such as the Disability in Strategies of Care [DISC] study). Stratification of treatment regimens based on MIDAS baseline grades (Grades I/II assigned to non-steroidal anti-inflammatory agents or simple analgesics; Grades III/IV assigned to triptans) produced significantly superior clinical outcomes—including 2-hour headache response rates and reduction in disability hours—compared to stepped-care paradigms. This demonstrates the capacity of the instrument to predict therapeutic response and inform clinical decisions.

8. Reliability

The reliability of MIDAS has been extensively demonstrated through evaluations of internal consistency and test-retest reproducibility across multiple languages, clinical environments, and cultural contexts.

1. Internal Consistency

The internal consistency of the five core items has consistently met the accepted standards for psychometric instruments utilized in individual clinical decision-making. In the primary development cohorts, Cronbach’s alpha ranged from $0.76$ to $0.86$. Independent linguistic validations—including the French, German, Italian, Spanish, Turkish, and Dutch versions—have replicated these findings, with alpha coefficients consistently falling between $0.73$ and $0.85$. These values confirm that while the items evaluate distinct functional spheres (occupational, domestic, social), they collectively tap into a coherent unidimensional latent construct of headache-induced behavioral disruption.

2. Test-Retest Reliability

Given that MIDAS is designed to evaluate stability and therapeutic change over time, test-retest reliability is a critical psychometric metric. In test-retest investigations with re-administration intervals ranging from 5 to 21 days (a period sufficiently short to prevent authentic clinical shifts yet long enough to minimize direct memory effects), the questionnaire demonstrated exceptional stability:

• Spearman rank correlation coefficients for individual items ranged from $0.67$ to $0.84$.
• The total MIDAS score yielded test-retest correlation coefficients between $r = 0.80$ and $r = 0.84$ in population-based studies, and up to $r = 0.88$ in specialty neurology clinics.
• Categorical stability across the four disability grades (Grade I through Grade IV) revealed weighted kappa ($kappa$) coefficients ranging from $0.70$ to $0.79$, reflecting high agreement in clinical grade assignment across repeated administrations.

9. Factor Analysis

Structural evaluations of the five core MIDAS items through both Exploratory Factor Analysis (EFA) and Confirmatory Factor Analysis (CFA) have established the dimensional architecture of the instrument.

1. Exploratory Factor Analysis (EFA)

Early principal component analyses and exploratory factor extractions consistently yielded a robust one-factor or two-factor solution, depending on the extraction criteria utilized. When eigenvalues greater than 1.0 were applied, a single dominant general factor emerged, accounting for approximately $58%$ to $68%$ of the total variance across items. In this unidimensional configuration, all five items exhibited high factor loadings:

• Item 1 (Work missed): Factor loadings ranging between $0.70$ and $0.81$
• Item 2 (Work reduced by half): Factor loadings ranging between $0.74$ and $0.83$
• Item 3 (Household work missed): Factor loadings ranging between $0.78$ and $0.86$
• Item 4 (Household work reduced by half): Factor loadings ranging between $0.79$ and $0.88$
• Item 5 (Social/leisure missed): Factor loadings ranging between $0.71$ and $0.79$

2. Confirmatory Factor Analysis (CFA) and Structural Fit

Subsequent psychometric studies utilizing Confirmatory Factor Analysis evaluated whether a unidimensional model (single “Disability” latent construct) or a correlated two-factor model (Factor 1: Work/School Productivity; Factor 2: Domestic & Social Life) best represents the observed covariance matrix. Although a two-factor model frequently yields slightly higher statistical fit indices due to localized residual correlations between identical behavioral metrics (i.e., absolute absence vs. reduced efficiency within the same environmental domain), the unidimensional model demonstrates acceptable-to-excellent goodness-of-fit indices:

• Comparative Fit Index (CFI): $0.96$ to $0.98$
• Tucker-Lewis Index (TLI): $0.94$ to $0.97$
• Root Mean Square Error of Approximation (RMSEA): $0.045$ to $0.068$
• Standardized Root Mean Square Residual (SRMR): $0.032$ to $0.045$

These findings substantiate the practice of aggregating the five items into a single composite score representing overall headache-related disability.

10. Instrument / Measurement Tool

The Migraine Disability Assessment Questionnaire is structured as follows:

• Test Type: Patient-Reported Outcome Measure (PROM), self-administered paper-and-pencil or digital questionnaire.
• Administration Format: Self-report questionnaire completed by the patient or administered via clinical interview.
• Target Population: Adults (18+ years) and older adults presenting with primary headache disorders (e.g., episodic migraine, chronic migraine). A pediatric variant (PedMIDAS) exists separately for children and adolescents.
• Recall Period: Past 3 months (90 days).
• Item Count: 5 primary scored items, plus 2 unscored supplementary clinical items (total 7 items).
• Response Scale: Number of days (0-90 days for Questions 1-5).
• Supplementary Response Scale: Question A: Number of days (0-90 days); Question B: 0-10 numeric rating scale (0 = no pain at all, 10 = pain as bad as it can be).
• Scoring Algorithm: Arithmetic sum of the numerical values recorded in Questions 1, 2, 3, 4, and 5:
  $$\text{MIDAS Total Score} = \sum_{i=1}^{5} Q_i$$
• Disability Grading System:
  • Grade I (Little or No Disability): 0 to 5 days
  • Grade II (Mild Disability): 6 to 10 days
  • Grade III (Moderate Disability): 11 to 20 days
  • Grade IV (Severe Disability): 21 or more days (subdivided in specialized research into Grade IV-A [21–40 days] and Grade IV-B [41+ days])
• Scoring Notes: Questions A and B are recorded purely for clinical diagnostic evaluation and are excluded from the MIDAS disability score calculation. Items 2 and 4 explicitly require respondents to exclude days already counted in Items 1 and 3, respectively, preventing statistical inflation of total lost days.

11. Permissions & Fee and Test Year

The original English version of the Migraine Disability Assessment Questionnaire was developed and published in 1998 by Richard B. Lipton and Walter F. Stewart, with definitive population-based psychometric validation published in 2000 and 2001. The Dutch version was adapted in 1998 by the Pijn Kennis Centrum Maastricht (Pain Knowledge Center Maastricht) in collaboration with Maastricht University Medical Centre.

The MIDAS questionnaire was developed with partial educational grant support and was historically copyrighted by Astra-Zeneca LP. However, MIDAS is widely distributed and made freely available for non-commercial clinical use, individual patient care, and academic research without fee. Academic researchers and clinicians typically do not require formal licensing permissions for clinical applications, provided the instrument is cited accurately and used without item alterations. For commercial use, sponsored pharmaceutical clinical trials, or electronic integration into proprietary software systems, researchers should verify licensing guidelines through the appropriate copyright holders or original academic authors.

12. References

• Lipton, R. B., Stewart, W. F., Sawyer, J., & Edmeads, J. G. (2001). Clinical utility of an instrument assessing migraine disability: The International Burden of Migraine Study (IBMS). Cephalalgia, 21(5), 564–569. https://doi.org/10.1046/j.1468-2982.2001.00216.x
• Lipton, R. B., Stewart, W. F., Stone, A. M., Láinez, F. T., & Sawyer, J. P. (2000). Stratified care vs step care strategies for migraine: The Disability in Strategies of Care (DISC) study—a randomized trial. JAMA, 284(20), 2599–2605. https://doi.org/10.1001/jama.284.20.2599
• Stewart, W. F., Lipton, R. B., Kolodner, K., Sawyer, J., Lee, C., & Liberman, J. N. (2000). Validity of the Migraine Disability Assessment (MIDAS) score in relation to a diary-based measure in a population with a wide range of headache frequency. Cephalalgia, 20(10), 860–870. https://doi.org/10.1046/j.1468-2982.2000.00139.x
• Stewart, W. F., Lipton, R. B., Whyte, J., Dowson, A., Batenhorst, A., Giles, F. R., & Sawyer, J. (1999). An international study to assess reliability of the Migraine Disability Assessment (MIDAS) score. Neurology, 53(5), 988–994. https://doi.org/10.1212/wnl.53.5.988
• Stewart, W. F., Lipton, R. B., Dowson, A. J., & Sawyer, J. (2001). Development and testing of the Migraine Disability Assessment (MIDAS) Questionnaire to assess headache-related disability. Neurology, 56(6 Suppl 1), S20–S28. https://doi.org/10.1212/wnl.56.suppl_1.s20

13. Items of the Scale (Questionnaire)