Skip to main content
v2026.11,986 entries · CC-BY 4.0

Modified Rankin Scale (mRS 0–6): Stroke Disability Grades, Structured Interview (mRS-SI), Utility-Weighted mRS (UW-mRS), Ordinal Shift Analysis, and LOINC Coding

Technical guide to the Modified Rankin Scale (mRS 0–6): stroke grades, structured interview (mRS-SI), utility-weighted UW-mRS, and LOINC 75859-9.

Written and maintained by CASRAI Editorial Board

Published

On this page

Short answer: The Modified Rankin Scale (mRS) is a 7-point ordinal global disability rating scale—ranging from 0 (no symptoms at all) to 6 (dead)—used worldwide as the primary 90-day functional outcome endpoint in acute stroke clinical trials and stroke center quality registries. First described by Dr. John Rankin in 1957 and modified to its modern 0–6 structure by van Swieten et al. (1988), the modified rankin scale in clinical research is analyzed via binary dichotomization, proportional odds ordinal shift analysis, or patient-centered Utility-Weighted mRS (UW-mRS) weights.

What Is the Modified Rankin Scale? The 0–6 Ordinal Scoring Definitions

While instruments such as the National Institutes of Health Stroke Scale (NIHSS) quantify specific focal neurological deficits on physical examination and Timed Up and Go (TUG) tests measure physical gait speed, the Modified Rankin Scale evaluates global functional independence and participation in daily activities. Every score from 0 to 6 is defined by strict functional boundary rules:

mRS Score Official Grade Description (van Swieten et al., 1988) Operational Boundary Criterion in Structured Interviews UW-mRS Utility Weight (Chaisinanunkul et al., 2015)
0 No symptoms at all Complete absence of any new post-stroke neurological symptom, subjective complaint, or physical sign. 1.00
1 No significant disability despite symptoms; able to carry out all usual duties and activities Patient notices mild residual symptoms (e.g., mild numbness or slight dysarthria) but performs 100% of pre-stroke work, leisure, and household activities. 0.91
2 Slight disability; unable to carry out all previous activities, but able to look after own affairs without assistance Cannot return to at least one pre-stroke work, social, or leisure role, but can live alone independently for ≥1 week (handling meals, finances, and basic household needs) without another person’s help. 0.76
3 Moderate disability; requiring some help, but able to walk without assistance Requires another person’s assistance for instrumental activities of daily living (shopping, cooking, finances, housekeeping), but walks without physical help from another person (using a cane, quad cane, ankle-foot orthosis, or walker counts as walking unassisted). 0.65
4 Moderately severe disability; unable to walk without assistance and unable to attend to own bodily needs without assistance Cannot walk without physical assistance from another person or requires assistance with basic bodily needs (dressing, bathing, toileting), but does not require continuous 24-hour nursing supervision and can be left alone safely for several hours. 0.33
5 Severe disability; bedridden, incontinent, and requiring constant nursing care and attention Requires constant presence or near-continuous attention of a caregiver day and night; cannot be left unattended safely. 0.00
6 Dead All-cause mortality occurring on or before the specified trial follow-up timepoint (typically Day 90). 0.00

Structured Interviews (mRS-SI, RFA, smRSq) and Inter-Rater Reliability

In unstructured clinical interviews, clinicians frequently disagree when distinguishing between mRS 1 vs. 2 (whether a patient has truly resumed all previous activities) and mRS 2 vs. 3 (whether a patient requires help with instrumental activities of daily living). Systematic reviews by Quinn et al. and Wilson et al. (2002) demonstrated that unstructured mRS grading yields only moderate inter-rater reliability (unweighted κ ≈ 0.56), which reduces statistical power in randomized controlled trials.

To prevent endpoint misclassification, stroke clinical trial protocols mandate two methodological controls:

  1. Validated Structured Interview Algorithms: Investigators use the Modified Rankin Scale Structured Interview (mRS-SI) (Wilson et al., 2002), the Rankin Focused Assessment (RFA) (Saver et al., 2010), or the 5-item yes/no Simplified mRS Questionnaire (smRSq) (Bruno et al., 2010). Structured interviewing raises inter-rater reliability to unweighted κ ≈ 0.74–0.81 and weighted κ > 0.90, and validated telephone administration shows high concordance with in-person assessment for decentralized 90-day trial follow-ups.
  2. Centralized Video Certification and Blinded Adjudication: Phase III neurovascular drug and mechanical thrombectomy trials require all site raters to complete standardized training and pass video-case certification before scoring trial participants, with 90-day assessments performed by raters blinded to treatment assignment.

Modified Rankin Scale in Clinical Research: Dichotomization, Ordinal Shift, and UW-mRS

When designing primary efficacy endpoints and evaluating statistical significance vs. clinical significance in acute stroke trials, biostatisticians choose among three analytical paradigms:

  • Fixed Binary Dichotomization: Collapses the 7-point scale into a binary responder outcome—most commonly mRS 0–1 (“excellent outcome” in intravenous thrombolysis trials such as NINDS tPA and ECASS III) or mRS 0–2 (“functional independence” in early endovascular thrombectomy trials). While simple to interpret, binary cutoffs discard clinically meaningful transitions across the rest of the scale (such as shifting a severe stroke patient from mRS 5 to mRS 3).
  • Ordinal Shift Analysis (Proportional Odds / Cochran–Mantel–Haenszel): Recommended by the Stroke Therapy Academic Industry Roundtable (STAIR), ordinal shift analysis evaluates improvement across all seven levels of the mRS simultaneously using proportional odds logistic regression (reporting a common odds ratio) or the assumption-free Mann–Whitney / van Elteren test.
  • Utility-Weighted Modified Rankin Scale (UW-mRS): Because equal 1-point steps on the ordinal mRS do not represent equal intervals in health-related quality of life (the transition from mRS 3 to 4 represents a much larger loss of utility than 1 to 2), Chaisinanunkul et al. (2015) derived patient-centered utility weights from EQ-5D and Patient-Reported Outcome Measures (PROMs) (`1.00, 0.91, 0.76, 0.65, 0.33, 0.00, 0.00`). Pivotal late-window thrombectomy trials (such as the DAWN trial) adopted 90-day mean UW-mRS as their primary endpoint.

EHR, Registry, and HL7 FHIR Interoperability (LOINC 75859-9)

For Comprehensive Stroke Centers participating in the American Heart Association Get With The Guidelines–Stroke registry and Joint Commission quality reporting, both pre-stroke baseline mRS and 90-day post-stroke mRS must be captured in structured EHR flowsheets. In HL7 FHIR R4 clinical data exchange and NINDS Stroke Common Data Elements (CDEs), the total score is mapped to Regenstrief LOINC code 75859-9 (Modified rankin scale [Score]), allowing automated extraction from EHR flowsheets into stroke registries and clinical trial Electronic Data Capture (EDC) systems.

Common Misunderstandings About the Modified Rankin Scale

  • “Using a cane or walker places a patient at mRS 4”: False. Under standard mRS scoring rules, a patient who walks using a cane, crutch, ankle-foot orthosis, or rolling walker without physical support or hands-on assistance from another person is classified as able to walk unassisted (eligible for mRS 3 or lower). Score 4 applies only when the patient cannot walk without another person’s physical help or cannot attend to basic bodily needs.
  • “Only stroke-related disability counts toward the mRS score”: In standard clinical trial conventions (including the Rankin Focused Assessment), post-stroke mRS rates the patient’s overall functional disability from all causes (which is why trials record pre-stroke baseline mRS at enrollment and often require baseline mRS 0–1 or 0–2 as an eligibility criterion).

Frequently Asked Questions

What is the Modified Rankin Scale and how is it used in clinical research and healthcare?

The Modified Rankin Scale (mRS) is a 7-level ordinal clinician-reported outcome scale (scored from 0 for no symptoms to 6 for death) originally developed by Dr. John Rankin in 1957 and modified by van Swieten et al. in 1988. It serves as the primary 90-day functional disability endpoint in acute ischemic and hemorrhagic stroke clinical trials and Comprehensive Stroke Center quality registries.

What standards or regulatory requirements govern the Modified Rankin Scale?

In randomized stroke trials and regulatory submissions, the Modified Rankin Scale is governed by NINDS Stroke Common Data Elements (CDEs), mandatory assessor video certification, structured interview protocols (such as the mRS-SI or Rankin Focused Assessment) to minimize inter-rater variance, and CMS / Joint Commission Comprehensive Stroke Center quality reporting (STK-OP-1 / 90-day mRS follow-up).

How does the Modified Rankin Scale integrate with EHR and health informatics workflows?

In Electronic Health Record (EHR) and clinical trial data exchange workflows, the total Modified Rankin Scale score is standardized under Regenstrief LOINC code 75859-9 (Modified rankin scale [Score]) and transmitted as an HL7 FHIR Observation resource alongside baseline pre-stroke mRS and 90-day post-discharge follow-up assessments.

Sources

Follow CASRAI

Research-administration guidance, standards updates and independent tool reviews.

Ask CASRAI · free to try

Ask about Modified Rankin Scale (mRS 0–6): Stroke Disability Grades, Structured Interview (mRS-SI), Utility-Weighted mRS (UW-mRS), Ordinal Shift Analysis, and LOINC Coding

2 free questions a day, no account.

An AI assistant for research: funder rules, compliance, methods and publishing. It links the sources behind its answers and labels answers that come from a web search.

Answers draw on CASRAI's guides and dictionary plus the federal and funder documents we index: Federal Register, Grants.gov, Regulations.gov and UKRI.

Works on this site and inside Claude, Cursor and the AI tools you already use.

Everything CASRAI publishes — this page, the dictionary, the guides and the news — stays open to read, with no account and no card.

Referenced across the research world

University of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logoUniversity of Cambridge logoColumbia University logoCrossref logoUniversity of Edinburgh logoHarvard University logoUniversity of Oxford logoPrinceton University logoStanford School of Medicine logoUniversity College London logoORCID logo
  • University of Cambridge logo
  • Columbia University logo
  • Crossref logo
  • University of Edinburgh logo
  • Harvard University logo
  • University of Oxford logo
  • Princeton University logo
  • Stanford School of Medicine logo
  • University College London logo
  • ORCID logo

View CASRAI adoption →