Also known as modified Rankin Scale, mRS, Rankin scale
Clinician-assigned grade of global disability following stroke, 0 to 6.
Applies to / doesn’t apply to5
Applies to: Adults after stroke. The scale is assigned by a clinician from observation and interview. Structured versions have been validated in broader adult neurological and neurosurgical cohorts.
- Children and neonates — a separate paediatric stroke outcome measure exists.
- General disability outside neurological disease.
- As a measure of acute stroke severity. The mRS describes global disability and dependence; stroke severity is a different construct.
- As a prognosis, or to determine eligibility for any treatment.
- During the acute hospital phase, where its validity is questioned (ICC 0.675).
Choose the single grade whose full description best fits the person at the timepoint below. Read each description in full: a grade estimated without a formal assessment agrees poorly between assessors.
The time the grade describes. A pre-stroke grade and a post-stroke grade are different assessments.
Optional. Printed in the note. Grades based on a proxy’s account agree less well between assessors than grades from the patient’s own.
What this score estimates
Records which of seven described grades best fits a person’s global disability and dependence in everyday activities after a stroke, at a stated timepoint. There is no formula: the clinician selects one grade, 0 (no symptoms at all) to 6 (dead), and that grade is the result. Higher grades indicate greater dependence. The scale has no built-in timepoint, so the time the grade describes — currently, before the stroke, at 90 days, at 6 months or another time — is recorded with it. It is not a measure of acute stroke severity, not a prognosis, and not a determination of eligibility for any treatment.
Pearls & pitfalls
16
- Agreement between assessors on the exact grade — the figure this page shows — is moderate at best. Pooled across studies of traditional administration, unweighted kappa is 0.46 with exact agreement in 60% of assessments (Quinn 2009). In the largest multicentre study, with 15 raters across 3 centres, kappa was 0.25 and exact agreement 43% (Wilson 2005).
- Weighted kappa looks far better (about 0.90) because it forgives disagreement by one grade. One grade is the whole resolution of the instrument, so the unweighted figures are the relevant ones here.
- Estimating the grade without a formal assessment gave kappa 0.16 (Quinn 2009, randomised study). That is why every option on this page shows the full descriptor rather than a short label.
- Agreement varies by grade. Per-grade unweighted kappa: grade 0 0.70, grade 1 0.43 (the lowest), grade 2 0.51, grade 3 0.57, grade 4 0.66, grade 5 0.79. At 90 days the weakest boundaries are 0/1 (kappa 0.66) and 1/2 (0.70); the strongest is 2/3 (0.81).
- Structured interviews improve agreement in their developers’ hands, but improvements have not been reproduced consistently when tested by independent centres (Harrison 2013). This page shows the plain scale only.
- Agreement also varies with how the grade is obtained: proxy versus self-report (weighted kappa 0.63 against 0.81), telephone and postal administration, the individual rater and the rater’s specialty, and across countries independently of language.
- Two systematic reviews disagree about overall reliability (pooled kappa 0.46 rising to 0.62 with a structured interview in one; 0.56 rising to 0.78 in the other). The difference has not been reconciled; the figures above are Quinn 2009’s.
- There is no single canonical published wording. The descriptors here are the form printed by Wilson 2002 and Quinn 2010; other publications transpose clauses or shorten grade 4. Grade 4 deliberately keeps both of its “without assistance” clauses.
- Grade 6 (dead) was added for clinical trials and was not part of the original scale (Harrison 2013); van Swieten 1988 describes six grades, 0 to 5.
- Trial publications have applied words such as “good” and “favorable” to different, sometimes contradictory, ranges of the scale (STAIR XI). This page attaches no verbal label and no grouping to any grade.
Risk factors the score does not count:
- Acute neurological deficit or stroke severity
- Cognition, language or mood
- Participation and quality of life
- Comorbidity and socioeconomic circumstances, which also shape independence
- Change over time — each grade describes one timepoint
- Any laboratory value, vital sign, age or sex
What the guidelines say
5
Describes the modified Rankin scale as having six grades, 0 to 5.
View guideline source ↗States that grade 6 was added for clinical trial purposes and was not part of the original scale.
View guideline source ↗Specifies mRS assessment at 90 days after discharge, accepting assessments from 75 through 105 days.
Describes the modern instrument as a seven-level, clinician-reported measure of global disability, and documents that trial publications have applied terms such as “good” and “favorable” to different mRS ranges.
View guideline source ↗Reports pooled inter-observer agreement of kappa 0.46 for traditional administration and 0.62 with a structured interview.
View guideline source ↗Guideline positions are stated as published. Curie.MD Calc does not publish treatment direction.
Where this comes from
10
References
- van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19(5):604-607. PMID 3363593. First reliability study and first widely cited print reference for the modified scale; the modification is attributed to the UK-TIA group. View source ↗
- Rankin J. Cerebral vascular accidents in patients over the age of 60. II. Prognosis. Scott Med J. 1957;2(5):200-215. PMID 13432835. The original five-grade scale (I–V), with no grade 0 and no death grade. View source ↗
- Farrell B, Godwin J, Richards S, Warlow C. The United Kingdom transient ischaemic attack (UK-TIA) aspirin trial: final results. J Neurol Neurosurg Psychiatry. 1991;54(12):1044-1054. PMID 1783914. View source ↗
- Wilson JTL, Hareendran A, Grant M, et al. Improving the assessment of outcomes in stroke: use of a structured interview to assign grades on the modified Rankin Scale. Stroke. 2002;33(9):2243-2246. PMID 12215594. View source ↗
- Quinn TJ. Improving outcome assessment for clinical trials in stroke. MD thesis, University of Glasgow; 2010. Table 1 / Appendix A — the descriptor wording shipped here. University thesis; no stable PubMed or DOI record.
- Harrison JK, McArthur KS, Quinn TJ. Assessment scales in stroke: clinimetric and clinical considerations. Clin Interv Aging. 2013;8:201-211. PMID 23440256; PMC3578502. View source ↗
- Quinn TJ, Dawson J, Walters MR, Lees KR. Reliability of the modified Rankin Scale: a systematic review. Stroke. 2009;40(10):3393-3395. PMID 19679846. View source ↗
- Wilson JTL, Hareendran A, Hendry A, et al. Reliability of the modified Rankin Scale across multiple raters: benefits of a structured interview. Stroke. 2005;36(4):777-781. PMID 15718510. View source ↗
- Quinn TJ, Dawson J, Walters MR, Lees KR. Exploring the reliability of the modified Rankin Scale. Stroke. 2009;40(3):762-766. PMID 19131664. View source ↗
- Saver JL, Chaisinanunkul N, Campbell BCV, et al. Standardized nomenclature for modified Rankin Scale global disability outcomes: consensus recommendations from Stroke Therapy Academic Industry Roundtable XI. Stroke. 2021;52(9):3054-3062. PMID 34320814. View source ↗