Calc Function

    • Calcs that help predict probability of a diseaseDiagnosis
    • Subcategory of 'Diagnosis' designed to be very sensitiveRule Out
    • Disease is diagnosed: prognosticate to guide treatmentPrognosis
    • Numerical inputs and outputsFormula
    • Med treatment and moreTreatment
    • Suggested protocolsAlgorithm





    Chief Complaint


    Organ System


    Patent Pending

    Modified NIH Stroke Scale/Score (mNIHSS)

    Shortened, validated version of the NIHSS.


    Note: numbering scheme is intentional and reflects differences from the original NIHSS (i.e., 1A, 7 and 10 are eliminated, and 3 and 4 are combined).

    When to Use
    Why Use

    The mNIHSS can help physicians quantify the severity of a stroke in the acute setting.

    The Modified National Institutes of Health Stroke Scale (mNIHSS) was designed to eliminate the parts of the NIHSS that had poor interrater reliability while maintaining the original score’s utility in assessing stroke severity.

    • Lower is better; increasing mNIHSS scores are correlated with more severe strokes and worsened clinical outcomes.
    • The mNIHSS performs as well as the original score in predicting patients at high risk of hemorrhage if given tPA and which patients are likely to have good clinical outcomes.
    • The mNIHSS has superior interrater reliability (<90%) compared to the original NIHSS (~66%).
    • The mNIHSS is more reliable in multiple settings, including calculating scores from medical records, when used via telemedicine, and when used in clinical trials.

    Points to keep in mind:

    • Currently, the mNIHSS is used much less frequently than the NIHSS in both the clinical setting and in trials.
    • Many guidelines make reference to the NIHSS rather than the mNIHSS, including those making recommendations concerning tPA administration.
    • The NIHSS only takes an average of 6 minutes to complete, so some who question the clinical utility of altering a well-validated and widely used scale.

    There are nearly 800,000 cases of acute stroke in the United States every year, with 130,000 associated deaths (4th leading cause of death in Americans).

    The mNIHSS can help physicians determine the severity of a stroke, predict clinical outcomes and can help guide management.

    The mNIHSS has the same correlation with clinical outcomes as the NIHSS but with better interrater reliability.


    Please fill out required fields.

    Next Steps
    Creator Insights


    In patients who present with symptoms concerning for ischemic stroke:

    • Consult Neurology immediately (if available) for all patients presenting with ischemic stroke.
    • Evaluate whether the patient is a potential candidate to receive intravenous thrombolysis (tPA).
    • Determine the onset of stroke symptoms (or time patient last felt or was observed normal).
    • Consider further imaging including CT, CT angiography and MRI/MRA to evaluate for hemorrhagic stroke.
    • In appropriate circumstances and in consultation with both neurology and the patient, consider IV thrombolysis for ischemic strokes in patients with no contraindications.
    • Always consider stroke mimics in the differential diagnosis, especially in cases with atypical features (age, risk factors, history, physical exam), including:
      • Recrudescence of old stroke from metabolic or infectious stress;
      • Todd’s paralysis after seizure;
      • Complex migraine;
      • Pseudoseizure, conversion disorder.

    Critical Actions

    • The mNIHSS performs as well as the NIHSS in predicting clinical outcomes, but it is important to recognize that individual cases will vary and that management decisions must be made in consultation with the patient whenever possible.
    • Whenever possible, patients with acute stroke should be transferred to a stroke center for their initial evaluation and treatment, as the holistic care (medical optimization, early initiation of PT and OT, patient and family education and discharge planning) is associated with improved clinical outcomes.


    Addition of the selected points.

    Facts & Figures

    In 2001, Lyden et. al proposed to modify the NIH Stroke Scale, 'deleting poorly reproducible or redundant items (level of consciousness, face weakness, ataxia, dysarthria) and collapsing the sensory item into 2 responses,'' with the goal of both simplifying the scale, improving its reproducibility and providing more relevance to each assigned point.

    This modified NIH Stroke Scale (mNIHSS) has now been evaluated in several trials (see below) and has begun to achieve acceptance in the stroke community as a reliable tool — even more reliable and valid, in some studies, than the NIHSS itself (better Kappa scores, suggesting higher levels of agreement in point assignments between raters).

    Language/Aphasia Test

    Evidence Appraisal

    • Several trials have prospectively validated the mNIHSS and found it to perform as well as the original score in predicting which patients are at high risk of hemorrhage if given tPA and which patients are likely to have good clinical outcomes.
    • The mNIHSS has superior interrater reliability, with studies showing that >90% of its items have excellent agreement between raters, as opposed to ~66% of the items in the NIHSS.
    • The mNIHSS appears to be more reliable in multiple settings, including calculating scores from medical records, when used via telemedicine, and when used in clinical trials.
    • The mNIHSS was developed using individual item analysis of the NIHSS to identify items that had poor reliability or were redundant. The mNIHSS was then evaluated by scoring it against video recordings originally made for NIHSS scoring certification for participants in the NINDS trials. These tapes consisted of of 11 stroke patients being evaluated by 4 different examiners. (Lyden 2001)
      • The number of items with poor interrater reliability decreased 6% when the mNIHSS was applied (from 20% with the NIHSS to 14%).
      • The ORs for predicting clinical outcomes at three months were essentially identical for the mNIHSS compared to the original score.
    • A review article examining several studies comparing the mNIHSS to the NIHSS found that mNIHSS has consistently superior interrater reliability in a variety of settings.
      • 71% vs. 54% of items had excellent agreement.
      • Only 5% vs. 12% of items had poor agreement.
    • A study found that when using the medical record to estimate a patient’s stroke score that the mNIHSS gave an estimate that was closer to the patient’s actual score than the the mNIHSS (70% vs. 62% for being within 1 point of the actual score). (Kasner 2003)


    Dr. Patrick D. Lyden

    About the Creator

    Patrick D. Lyden, MD is chair of the Department of Neurology, Director of the Stroke Program and holds the Carmen and Louis Warschaw Chair in Neurology at Cedars-Sinai. Previously, he was a professor and vice chairman for clinical neurology at UCSD and served as the Clinical Chief of Neurology and Director of the Stroke Center at UCSD Medical Center. Dr. Lyden has published more than 200 journal articles and abstracts and edited a textbook on stroke intervention.

    To view Dr. Patrick D. Lyden's publications, visit PubMed

    Content Contributors