The NIH Stroke Scale is unreliable in untrained hands

Abstract

Background and Purpose: The NIH (National Institutes of Health) Stroke Scale helps to define candidates for thrombolytic treatment and other important clinical decisions. We evaluated its use by last-year medical students who had no specific training in use of this scale. Methods: The scale (principles, utilization, methods for scoring, main pitfalls) was presented between 2 stroke lectures. Following a 30-minute study period, a case from the NIH training videotape (case 1, tape 2, total score 4) was presented, and the 13 items were scored by 42 voluntary students. The analysis focused on possible errors of potential clinical relevance. Results: Only 15 students (36%) scored all items correctly. Twenty (48%) gave results at least 2 points apart from the correct total score, and 14 students had scores ≥6, which could lead to the inappropriate use of recombinant tissue-type plasminogen activator (rt-PA). At least 10% of the students made mistakes in most (8 of 13) items of the scale. Conclusions: Graduating medical students are not ready to use the NIH Stroke Scale without specific training. The magnitude of the mistakes found in this study is enough to inappropriately include or exclude patients from rt-PA treatment. Training programs should be considered a sine qua non for qualification for rt-PA use in daily clinical practice. Copyright © 2002 by National Stroke Association