Skip main navigation

Use of the Barthel Index and Modified Rankin Scale in Acute Stroke Trials

Originally publishedhttps://doi.org/10.1161/01.STR.30.8.1538Stroke. 1999;30:1538–1541

    Abstract

    Background and Purpose—The Barthel Index (BI) and the Modified Rankin Scale (MRS) are commonly used scales that measure disability or dependence in activities of daily living in stroke victims. The objective of this study was to investigate how these scales were used and interpreted in acute stroke trials.

    Methods—We identified from MEDLINE the major efficacy trials with neuroprotective drugs, thrombolytic drugs, and anticoagulants in acute ischemic stroke published between January 1995 and December 1998. We selected those trials that used the BI and/or MRS as outcome parameters.

    Results—Fifteen trials fulfilling the inclusion criteria were identified. The BI was used in 13 and the MRS in 8. In 4 trials mean and median scores of the BI were used, and in 1 trial median scores of the MRS were compared. Primary end points included the BI in 7, the MRS in 6, and both the BI and MRS in 3. With regard to the BI, a variety of sum scores between 50 and 95 were used as cutoff scores to define favorable outcome. Favorable outcome on the MRS was defined as either ≤1 or ≤2.

    Conclusions—Among the efficacy trials in acute stroke, we found remarkable differences in the choice of primary end points and in the definition of favorable outcome on both the BI and MRS. This lack of consensus strongly hinders the design, interpretation, and comparison of acute stroke trials. In general, it may be easier to define poor outcome instead of favorable outcome. Poor outcome could be defined if any of the following end points are reached: death, institutionalization due to stroke, MRS >3, or BI <60.

    Substantial efforts are being made to develop drug treatments that may minimize brain damage and improve outcome in patients with ischemic stroke. In clinical trials, the Barthel Index (BI) and the Modified Rankin Scale (MRS) are commonly used scales to assess outcome.

    The BI was developed in 19651 and later modified by Granger and coworkers2 as a scoring technique that measures the patient’s performance in 10 activities of daily life. The BI is considered a reliable disability scale for stroke patients.3 The items can be divided into a group that is related to self-care (feeding, grooming, bathing, dressing, bowel and bladder care, and toilet use) and a group related to mobility (ambulation, transfers, and stair climbing). The maximal score is 100 if 5-point increments are used, indicating that the patient is fully independent in physical functioning. The lowest score is 0, representing a totally dependent bedridden state. The MRS measures independence rather than performance of specific tasks.4 In this way, mental as well as physical adaptations to the neurological deficits are incorporated. The scale consists of 6 grades, from 0 to 5, with 0 corresponding to no symptoms and 5 corresponding to severe disability (Table 1).

    Although both scales are easy to use and have an acceptable degree of reliability,34 there is apparently no consensus on how these scales should be used to determine outcome in clinical trials. The aim of our study was to investigate how both scales were used and interpreted in recently published clinical drug trials in acute ischemic stroke.

    Methods

    We searched through MEDLINE for the major pivotal randomized trials (phase III trials) in acute stroke in which the efficacy of either thrombolytics, neuroprotective drugs, or antithrombotic compounds was investigated and the final results were published between January 1995 and December 1998. We selected those trials that used the BI and/or MRS either as primary or secondary end point. We assessed how the scores obtained from these scales were presented and interpreted.

    Results

    We identified 15 trials that together enrolled 9082 patients (Table 2). There were a variety of primary end points, including stroke scales, mortality, BI, MRS, and the Glasgow Outcome Scale (GOS). Four trials defined >1 primary end point. Assessment of the primary end points occurred at 3 months in 10 studies. In 1 study patients were scored at 1 month; in 4 other trials they were scored at 6 months.

    The BI was used in 13 trials and served as primary outcome measure in 7 trials. Two trials compared mean scores, and 2 others used median scores. In regard to definition of a favorable outcome, sum scores of ≥50, ≥60, ≥75, ≥85, and ≥95 were used. Three trials used a so-called combined BI/Rankin Scale, giving a sum score of 110 for patients showing complete recovery. In some trials, the BI was subdivided into different categories, of which the number varied from 3 to 6 categories (Table 3).

    The MRS was used in 8 trials and was a primary outcome parameter in 6. In 1 trial median scores of the MRS were compared between the 2 treatment groups. Favorable outcome on the MRS was defined in 3 trials as a score ≤1 and in 5 trials as a score ≤2. In the second European Cooperative Acute Stroke Study (ECASS II) trial, the proportion of patients with a 3-months MRS of ≤1 was used to assess efficacy of recombinant tissue plasminogen activator (r-tPA).15 There was no statistically significant difference with placebo. However, by shifting MRS grades from ≤1 to ≤2 (not a predefined end point) to indicate favorable outcome, a significant result in favor of r-tPA was obtained.

    Discussion

    Over the last 4 years, a number of pivotal acute stroke trials with thrombolytics, anticoagulants, and neuroprotective drugs were published. All failed to demonstrate a significant clinical beneficial effect on predefined primary end points, except for the National Institute of Neurological Disorders and Stroke (NINDS) and Fraxiparine in Ischemic Stroke Study (FISS) trials.1216 Remarkably, the criteria by which outcome was measured and the measuring instruments that were used differed considerably between the trials. The lack of consensus regarding what should be considered a clinically meaningful effect on outcome in stroke patients strongly hinders the design, interpretation, and comparison of acute stroke trials.

    The BI was the most commonly used scale for assessing activities of daily living. However, the criteria for classifying patients with a favorable outcome varied substantially from trial to trial; a variety of sum scores from ≥50 to ≥95 were used. In fact, many of the cutoff scores were arbitrarily chosen and have never been validated. Curiously, although the BI is widely used, few studies have been conducted on the clinical relevance of the sum scores. Granger and coworkers2 found that a score of 60 was a pivotal score at which patients move from assisted independence to dependence. In practical terms, with a score of ≥60, most patients were independent for essential personal care, such as moving around unassisted, sphincter control, eating, and personal toilet. A score of 85 usually corresponded to independence with minimal assistance.2 This means that the majority of patients were able to get dressed and to move from armchair to bed unassisted. Kay and coworkers20 and Dennis and colleagues21 also found that a score of <85 corresponded to a state in which patients reported needing help in performing activities of daily living, with a sensitivity of 94% to 95% and a specificity of 80% to 86%. Therefore, it is difficult to defend why in the NINDS and ECASS II trials a BI of ≥95 (defined as minimal or no disability) was chosen to define favorable outcome.1516 Why not use a score of ≥85, which has been shown to correspond to an acceptable level of autonomy in the majority of patients, or a score of 100 if one predicts that the therapy should result in complete recovery? Some investigators have even invented a combined BI/Rankin Scale with a sum score of 110, although such a scoring system has also not been functionally justified.61415 The BI is an ordinal (noncontinuous) scale. Therefore, parametric statistical methods cannot be used. We found it cumbersome that prestigious journals accepted the presentation of results in mean or median BI values,6141517 which are completely inappropriate statistical end points. A number of studies distinguished between different categories of disability on the BI.6891116 This may allow assessment of a global shift toward independence in subsequent categories of scores rather than reliance on a single score that dichotomizes outcome into favorable or poor. However, the cutoff scores used to define these different categories should also be validated and not arbitrarily chosen.

    In contrast to the BI, we found fewer variations between the trials in use of the MRS. Only 2 cutoff scores were used: ≤1 in 3 trials investigating r-tPA and ≤2 in 5 other trials. A striking example of how data can be differently interpreted by post hoc analyses was recently demonstrated in the ECASS II trial.15 It was possible, simply by shifting the cutoff score from the predefined ≤1 grades to ≤2 grades on the MRS, to obtain a statistically significant effect in favor of r-tPA. At first glance, a cutoff score of ≤2 (able to look after own affairs without assistance) appears to be more meaningful than a score of ≤1 in terms of independence (Table 1). Kay et al20 found that self-reported dependence had a sensitivity of 85% and specificity of 87% against the MRS dichotomized at ≥2 and a sensitivity of 94% and specificity of 70% against the MRS dichotomized at ≥3.

    Concordance between outcome parameters is also important in the evaluation of trial results. Although cutoff scores were chosen on a pragmatic basis, this was one of the strengths of the NINDS trial, in which each of the 4 outcome parameters (BI, MRS, National Institutes of Health Stroke Scale, GOS) showed consistency in the degree of benefit with r-tPA.16 Some investigators advocate the use of self-reported dependence as a valid means of dichotomizing stroke patients for the purpose of clinical trials.2021 We should be careful with this definition of dependence. Many factors unrelated to the intervention, such as psychological and socioeconomic factors, may influence this subjective statement. For example, a wheelchair-bound patient may have a level of income or a health insurance system that allows the acquisition of expensive technical aids to compensate for restrictions to perform desired activities, thereby making the patient “independent.” The BI and MRS are reliable measures that provide a more objective assessment of functional recovery after stroke.

    Another important issue is that in the analysis of the trials, different types of stroke were mixed together. Bamford and coworkers22 showed that the 6-month combined outcome of being dead or dependent (defined as MRS score >2) differed greatly between patients with total anterior circulation infarcts and those with lacunar or partial anterior circulation infarcts. Efficacy in a controlled trial should cover the entire study population. However, although effective randomization should ensure a good balance of stroke subtypes in the treated and placebo groups, it would also be appropriate to analyze outcome separately in function of stroke subtypes. If patients with anterior circulation lesions are selected, one should distinguish between total anterior circulation infarction, partial anterior circulation infarction, and lacunar infarction.22

    Instead of trying to define favorable outcome, for which there will be no consensus, we believe that it may be easier to define poor outcome. Rather than using a single scale, we suggest a definition of poor outcome if any of the following occurs: death, institutionalization due to stroke, MRS >3, or BI <60. Whereas there is no consensus regarding the definition of favorable outcome, there will be less disagreement that each of these 4 conditions corresponds to an unfavorable outcome, and it is a more objective instrument than self-reported dependence.

    Table 1. Modified Rankin Scale

    Level Description
    0 No symptoms
    1 No significant disability, despite symptoms; able to perform all usual duties and activities
    2 Slight disability; unable to perform all previous activities but able to look after own affairs without assistance
    3 Moderate disability; requires some help, but able to walk without assistance
    4 Moderately severe disability; unable to walk without assistance and unable to attend to own bodily needs without assistance
    5 Severe disability; bedridden, incontinent, and requires constant nursing care and attention

    Table 2. Pivotal Trials in Acute Ischemic Stroke Published in Final Form Between January 1995 and December 1998

    Clinical Trial (Year of Publication) Compound Tested No. of Randomized Patients Primary End Point(s) Time Interval Between Stroke Onset and Assessments Favorable Outcome on BI Favorable Outcome on MRS Results of Trial
    FIST5 (1996) Flunarizine 331 Neurological status, BI, MRS 6 mo ≥50 (independence) ≤2 No improvement of neurological and functional outcome
    PASS6 (1997) Piracetam 927 Neurological status 1 mo Comparison of mean scores+110 (complete recovery) No effect on neurological outcome
    RANTTAS7 (1996) Tirilazad 658 BI, GOS 3 mo ≥60 No improvement of functional outcome
    LUB-INT-98 (1997) Lubeluzole 721 Mortality 3 mo ≥75 ≤2 No significant reduction in mortality but improvement of functional outcome
    LUB-INT-59 (1998) Lubeluzole 725 Mortality 3 mo ≥75 ≤2 No effect on mortality or functional outcome
    ASSIST10 (1997) Selfotel (CGS 19755) 509 BI 90 d ≥60 Trials stopped prematurely because of unfavorable risk-benefit ratio
    Ebselen in Acute Stroke11 (1998) Ebselen 302 GOS 1 and 3 mo ≥75 (mild disability) No effect at 3 mo, but significantly more good recovery or moderate disability at 1 mo
    FISS12 (1995) Low-molecular-weight heparin 312 Self-reported dependence 6 mo ≥85 (independence) Significantly less death+dependence at 6 mo; no effect at 3 mo
    TOAST13 (1998) Low-molucular-weight heparinoid 1281 GOS, BI 3 mo ≥60 No improvement of functional outcome
    ECASS14 (1995) r-tPA 620 BI, MRS 3 mo Comparison of median scores; use of a combined BI/Rankin Scale (110 points) Comparison of median scores+≤1 No improvement of functional outcome; may be beneficial in a well-defined subgroup
    ECASS II15 (1998) r-tPA 800 MRS 3 mo Comparison of median scores including use of a combined BI/Rankin Scale (110 points)+≥95 ≤1 (post hoc also ≤2) No statistical benefit on predefined outcome
    NINDS16 (1995) r-tPA 291 (part 1) +333 (part 2) BI, MRS, GOS, NIHSS 3 mo ≥95 ≤1 r-tPA improves clinical outcome at 3 mo
    MAST-E17 (1996) Streptokinase 310 Death+severe disability on MRS 6 mo Comparison of mean scores ≤2 Increase in mortality
    MAST-I18 (1995) Streptokinase 622 MRS 6 mo ≤2 Increase in 10-d mortality; no significant beneficial effect at 6 mo
    ASK19 (1996) Streptokinase 340 Mortality+BI 3 mo ≥60 Increased mortality and morbidity

    FIST indicates Flunarizine in Stroke Treatment; PASS, Piracetam in Acute Stroke Study; RANTTAS, Randomized Trial of Tirilazad Mesylate in Patients with Acute Stroke; LUB-INT, Lubelazole International; ASSIST, Acute Stroke Studies Involving Selfotel Treatment; FISS, Fraxiparine in Ischemic Stroke Study; TOAST, Trial of Org 10172 in Acute Stroke Treatment; MAST-E, Multicenter Acute Stroke Trial–Europe; MAST-E, Multicenter Acute Stroke Trial–Italy; and ASK, Australian Streptokinase Trial.

    Table 3. Subcategories of the BI used in Pivotal Acute Stroke Trials

    PASS6 NINDS16 LUB-INT 5 and LUB-INT 98,9 Ebselen in Acute Ischemic Stroke11
    6 categories: 4 categories: 3 categories: 3 categories:
    110 (complete recovery) 95–100 (minimal or no disability) 100–75 (independent or mildly dependent) 110–75 (mild disability)
    100–85 (nearly complete independence) 90–55 <75 (moderately or severely dependent) 70–50 (moderate disability)
    80–60 (assisted dependence) 0–50 0 (dead) 45–0 (severe disability)
    55–40 (institutional care) Dead
    <40 (constant care)
    Dead

    Abbreviations are as defined in Table 2.

    This study was supported by a grant from the Academic Hospital Groningen (to Dr Sulter).

    Footnotes

    Correspondence to Professor J. De Keyser, Department of Neurology, Academisch Ziekenhuis Groningen, PO Box 30.001, 9700 RB Groningen, Netherlands. E-mail

    References

    • 1 Mahoney F. Barthel D. Functional evaluation: the Barthel Index. Md Med J.1965; 14:61–65.Google Scholar
    • 2 Granger CV, Devis LS, Peters MC, Sherwood CC, Barrett JE. Stroke rehabilitation: analysis of repeated Barthel Index measures. Arch Phys Med Rehabil.1979; 60:14–17.MedlineGoogle Scholar
    • 3 D’Olhaberriague L, Litvan I, Mitsias P, Mansbach HH. A reappraisal of reliability and validity studies in stroke. Stroke.1996; 27:2331–2336.CrossrefMedlineGoogle Scholar
    • 4 van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJA, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke.1988; 19:604–607.CrossrefMedlineGoogle Scholar
    • 5 Franke CL, Palm R, Dalby M, Schoonderwaldt HC, Hantson L, Eriksson B, Lang-Jenssen L, Smakman J. Flunarizine in Stroke Treatment (FIST): a double-blind, placebo-controlled trial in Scandinavia and the Netherlands. Acta Neurol Scand.1996; 93:56–60.CrossrefMedlineGoogle Scholar
    • 6 De Deyn P, De Reuck J Deberdt W, Vlietinck R, Orgogozo JM, for the Members of the Piracetam in Acute Stroke Study (PASS) Group. Treatment of acute ischemic stroke with piracetam. Stroke.1997; 28:2347–2352.CrossrefMedlineGoogle Scholar
    • 7 The RANTTAS Investigators. A randomized trial of tirilazad mesylate in patients with acute stroke (RANTTAS). Stroke.1996; 27:1453–1458.CrossrefMedlineGoogle Scholar
    • 8 Grotta J, for the US and Canadian Lubeluzole Ischemic Stroke Study Group. Lubeluzole treatment of acute ischemic stroke. Stroke.1997; 28:2338–2346.CrossrefMedlineGoogle Scholar
    • 9 Diener HC, for the European and Australian Lubeluzole Ischemic Stroke Study Group. Multinational randomised controlled trial of lubeluzole in acute ischemic stroke. Cerebrovasc Dis.1998; 8:172–181.CrossrefMedlineGoogle Scholar
    • 10 Davis SM, Albers GW, Diener HC, Lees KR, Norris J. Termination of acute stroke studies involving selfotel treatment. Lancet.1997; 349:32. Letter.CrossrefMedlineGoogle Scholar
    • 11 Yamaguchi T, Sano K, Takakura K, Saito I, Shinohara Y, Asano T, Yasuhara H, for the Ebselen Study Group. Ebselen in Acute Ischemic Stroke: a placebo-controlled, double-blind clinical trial. Stroke.1998; 29:12–17.CrossrefMedlineGoogle Scholar
    • 12 Kay R, Wong KS, Yu YL, Chan YW, Tsoi TH, Ahuja AT, Chan FL, Fong KY, Law CB, Wong A, Woo J. Low-molecular-weight heparin for the treatment of acute ischemic stroke. N Engl J Med.1995; 333:1588–1593.CrossrefMedlineGoogle Scholar
    • 13 The Publications Committee for the Trial of Org 10172 in Acute Stroke Treatment (TOAST) Investigators. Low molecular weight heparinoid, Org 10172 (Danaparoid), and outcome after acute ischemic stroke: a randomized controlled trial. JAMA.1998; 279:1265–1272.CrossrefMedlineGoogle Scholar
    • 14 Hacke W, Kaste M, Fieschi C, Toni D, Lesaffre E, von Kummer R, Boysen G, Bluhmki E, Höxter G, Mahagne M-H, Hennerici M, for the ECASS Study Group. Intravenous thrombolysis with recombinant tissue plasminogen activator for acute hemispheric stroke: the European Cooperative Acute Stroke Study (ECASS). JAMA.1995; 274:1017–1025.CrossrefMedlineGoogle Scholar
    • 15 Hacke W, Kaste M, Fiechi C, von Kummer R, Davalos A, Meier D, Larrue V, Bluhmki E, Davis S, Donnan G, Schneider D, Diez-Tejedor E, Trouillas P, for the Second European-Australasian Acute Stroke Study Investigators. Randomised double-blind placebo-controlled trial of thrombolytic therapy with intravenous alteplase in acute ischaemic stroke (ECASS II). Lancet.1998; 352:1245–1251.CrossrefMedlineGoogle Scholar
    • 16 The National Institute of Neurological Disorders and Stroke rt-PA Stroke Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med.1995; 333:1581–1587.CrossrefMedlineGoogle Scholar
    • 17 The Multicenter Acute Stroke Trial–Europe Study Group. Thrombolytic therapy with streptokinase in acute ischemic stroke. N Engl J Med.1996; 335:145–150.CrossrefMedlineGoogle Scholar
    • 18 Multicentre Acute Stroke Trial–Italy (MAST-I) Group. Randomised controlled trial of streptokinase, aspirin, and combination of both in treatment of acute ischemic stroke. Lancet.1995; 346:1509–1514.CrossrefMedlineGoogle Scholar
    • 19 Donnan GA, Davis SM, Chambers BR, for the Australian Streptokinase (ASK) Trial Study Group. Streptokinase for acute ischemic stroke with relationship to time of administration. JAMA.1996; 276:961–966.CrossrefMedlineGoogle Scholar
    • 20 Kay R, Wong KS, Perez G, Woo J. Dichotomizing stroke outcomes based on self- reported dependency. Neurology.1997; 49:1694–1696.CrossrefMedlineGoogle Scholar
    • 21 Dennis M, Wellwood I, Warlow C. Are simple questions a valid measure of outcome after stroke? Cerebrovasc Dis.1997; 7:22–27.CrossrefGoogle Scholar
    • 22 Bamford J, Sandercock P, Dennis M, Burn J, Warlow C. Classification and natural history of clinically identifiable subtypes of cerebral infarction. Lancet.1991; 337:1521–1526.CrossrefMedlineGoogle Scholar

    eLetters(0)

    eLetters should relate to an article recently published in the journal and are not a forum for providing unpublished data. Comments are reviewed for appropriate use of tone and language. Comments are not peer-reviewed. Acceptable comments are posted to the journal website only. Comments are not published in an issue and are not indexed in PubMed. Comments should be no longer than 500 words and will only be posted online. References are limited to 10. Authors of the article cited in the comment will be invited to reply, as appropriate.

    Comments and feedback on AHA/ASA Scientific Statements and Guidelines should be directed to the AHA/ASA Manuscript Oversight Committee via its Correspondence page.