Modified Rankin Scale (mRS)
- Introduced by
- Rankin scale original (Rankin, 1957); modified version by van Swieten et al.
- Year
- 1988
- Type
- Clinician-rated ordinal scale
- Family
- global disability rating
What does it do?
The Modified Rankin Scale is a simple 0-6 ordinal measure of global disability or dependency in patients with stroke and other neurological conditions.
Originally developed by Rankin in 1957 and modified by van Swieten and colleagues in 1988, it remains the most widely used global disability outcome in stroke clinical trials and clinical practice. Its simplicity, brevity, and strong prognostic association make it the gold standard for acute stroke outcome measurement and is mandated as a primary endpoint in virtually all stroke therapeutic trials.
The intuition behind the method
Stroke survivors experience variable degrees of disability ranging from complete independence to profound dependence requiring full care. A simple, universal disability scale enables rapid clinical assessment and prognostic stratification. The mRS ranks disability on a 0-6 scale from no symptoms to death, capturing global functional status in activities of daily living and instrumental activities. Unlike more detailed functional assessments, it distills complex disability into a single ordinal score that is intuitive and directly interpretable: 0-2 = good functional outcome (independent for daily living), 3-5 = dependent or dead. This simplicity enables very large sample sizes in trials and allows rapid triage and prognostic communication. The mRS has proven to be a robust predictor of long-term outcome, disability trajectory, and mortality.
History
The original Rankin scale was introduced by Rankin in 1957 as a simple global disability measure. It was refined by van Swieten and colleagues in 1988, who added the intermediate grades and clarified definitions, creating the Modified Rankin Scale. The mRS was adopted as the primary outcome measure in the NINDS rt-PA Stroke Study (1995) and has become the standard in all subsequent stroke trials. Its simplicity, interpretability, and strong predictive validity made it the near-universal primary endpoint in stroke research. The scale has been translated into 40+ languages. Standardized administration guides have been developed to improve consistency. In 2015, a detailed structured interview version was published to further enhance inter-rater reliability. The mRS remains fundamentally unchanged since 1988.
When to use it
mRS is the standard global disability outcome in acute stroke clinical trials—use for any randomized controlled trial testing thrombolytics, thrombectomy, neuroprotective agents, or other acute stroke therapies. It is essential in routine acute stroke care for prognostic communication and discharge planning. It is valuable in longitudinal stroke outcome studies examining long-term disability trajectory. It is used in health services research comparing stroke care pathways (comprehensive stroke centers vs. routine hospitals) and quality improvement initiatives. It provides rapid global assessment in resource-limited settings where detailed functional testing is not available. Less suitable when domain-specific disability information is needed (use stroke-specific QoL or Barthel Index for detailed functional assessment); mRS provides global rating only.
Limitations and considerations
- Ordinal rather than continuous scale; loss of granularity compared to detailed functional assessments (Barthel Index, modified Rankin detailed)
- Inter-rater reliability variable; depends heavily on clinician training and standardization. Untrained raters show moderate agreement; trained raters show good agreement
- Proxy and self-report mRS often disagree; self-report typically higher (better) than proxy rating by family. Source of rating (patient vs. proxy) must be documented
- Conceptual gaps: mRS does not distinguish patients within each grade (two mRS 3 patients may have vastly different functional capabilities); may miss clinically important transitions
Source conditions do not replace independent scientific review or validation by an analysis engine.
Strengths
- Simplicity and brevity (single 0-6 rating) enable extremely rapid assessment (2-5 minutes) suitable for busy acute settings
- Robust ordinal scale with clear categorical definitions enables consistent clinician rating with good inter-rater reliability when standardized administration is followed
- Strong prognostic power—mRS at baseline, discharge, and follow-up predicts long-term disability, quality of life, and mortality independent of other factors
- Universal applicability to all neurological conditions affecting disability (stroke, TBI, spinal cord injury, etc.); not disease-specific
- Primary endpoint in virtually all major stroke trials; enables direct comparison and meta-analysis across studies using standardized outcome definition
Applications
- Randomized controlled trials of acute and subacute stroke treatments (thrombolytics, thrombectomy, neuroprotective agents) as mandated primary efficacy outcome at 90 days post-stroke
- Routine acute stroke management for rapid prognostic stratification at admission (predicts discharge disposition, mortality), at discharge (guides rehabilitation level and discharge planning), and at follow-up
- Longitudinal natural history studies tracking functional disability trajectory months to years after stroke in relation to age, stroke severity, treatment, and rehabilitation
- Health services research comparing effectiveness of stroke care models, comprehensive stroke centers vs. standard hospitals, and health system-level outcomes
- Quality improvement and benchmarking programs tracking stroke center performance via proportion of patients achieving mRS 0-2 outcome
How it works
How it works
The mRS is assessed through brief clinician interview and observation of the patient's ability to perform activities of daily living. No formal testing is required. Administration takes 2-5 minutes. The clinician asks about the patient's ability to walk, perform self-care tasks (dressing, toileting, grooming), and maintain independence. Some structured interview guides exist to standardize assessment. For hospitalized patients, historical information from family or prior functional status is used to establish baseline. The mRS can be administered in person, by phone interview, or by structured proxy report from family if patient interview is not feasible.
The mRS is scored on a 0-6 ordinal scale: 0 = No symptoms at all; 1 = No clinically important symptoms (but aware of minor signs); 2 = Slight disability (unable to carry out all previous activities but able to look after own affairs without assistance); 3 = Moderate disability (requiring some help, but able to walk unassisted); 4 = Moderately severe disability (unable to walk unassisted and unable to attend to own bodily needs without assistance); 5 = Severe disability (bedridden, incontinent, requiring constant nursing care and attention); 6 = Dead. The final score is a single integer 0-6. No calculation is required; it is a categorical rating based on clinical judgment.
mRS score interpretation: 0-1 = no disability or minimal; 2 = slight disability but independent; 3 = moderate disability with some dependence; 4-5 = severe disability with major dependence; 6 = death. In stroke clinical trials, the primary endpoint often uses dichotomous outcomes: mRS 0-2 (good outcome/independent) vs. 3-6 (poor outcome/dependent) at 90 days post-stroke. This threshold defines functional independence. Proportions achieving mRS 0-2 are compared between treatment groups. Secondary analyses may examine mRS 0-1, shift in mRS distribution (proportion at each level), or time-to-event death or mRS 6. Prognostically, higher baseline mRS scores predict poor long-term outcome and higher mortality. At hospital discharge, mRS 4-6 predicts institutionalization or death; mRS 0-2 predicts community discharge.
Common pitfalls
- Do not rely on untrained personnel to rate mRS; standardized definitions and clinician training are essential for reliability. Ideally, central adjudication in trials by staff trained in mRS criteria
- Do not assume self-report and proxy ratings are equivalent; they systematically differ. In research, specify source (patient vs. proxy) and preferentially use patient self-report when possible
- Avoid using mRS as only outcome in detailed outcome studies; supplement with domain-specific measures (Barthel Index, stroke-specific QoL, NIHSS) to capture nuanced disability patterns
- Do not treat mRS as a continuous linear scale in statistical analysis; it is ordinal. Use ordinal or shift analyses rather than parametric ANOVA
Frequently asked questions
What is the difference between mRS 2 and mRS 3?
mRS 2 = slight disability, independent for daily living, but some activity limitation. Patient can perform self-care (dressing, grooming, toileting) independently; can ambulate, though gait may be abnormal. mRS 3 = moderate disability, some dependence in activities of daily living. Patient requires assistance for some self-care or mobility but is not bedridden. This is the clinical threshold for 'good outcome' (mRS 0-2) vs. 'poor outcome' (mRS 3-6) in trials.
Can mRS be assessed by phone or proxy interview?
Yes, phone and proxy assessment is feasible and widely used in research (especially for follow-up visits avoiding patient travel). However, agreement with in-person clinician rating is not perfect. Telephone and proxy mRS ratings tend to systematically differ from bedside assessment. In critical trials, in-person assessment is preferred; if phone/proxy is used, this should be documented and sensitivity analyses performed.
Is a low baseline mRS (e.g., mRS 0) in a stroke patient a problem?
Baseline mRS 0-1 (no or minimal pre-stroke disability) is actually prognostically favorable—indicates the patient had excellent pre-stroke function and likely has better stroke outcome potential. The concern is high baseline mRS (indicating pre-stroke disability), which predicts worse overall post-stroke outcome. mRS 0 baseline is not a problem; it's a favorable prognostic sign.
How is mRS used as a dichotomous outcome in trials?
mRS is converted to dichotomous outcome: mRS 0-2 (good outcome, independent) vs. mRS 3-6 (poor outcome, dependent). Proportions achieving mRS 0-2 are compared between treatment and control groups at 90 days (primary timepoint). Example: 40% of treatment group achieves mRS 0-2 vs. 26% of control group = statistically significant treatment benefit. This dichotomy is the standard trial primary outcome definition.
Related methods
Sources
van Swieten, J. C., Koudstaal, P. J., Visser, M. C., Schouten, H. J., & van Gijn, J. (1988). Interobserver agreement for the assessment of handicap in stroke patients. Stroke, 19(5), 604-607.
Open DOI referenceCopied.