CLINICAL QUESTION: Does an updated National Institutes of Health Stroke Scale (NIHSS) language assessment tool perform well in a sample of diverse, neurologically unimpaired English-speaking adults?
BACKGROUND: For more than 30 years, the NIHSS has been a vital tool for evaluating stroke patients worldwide. The original picture used in the NIHSS, cookie theft, was developed in 1972. The National Institute of Neurological Disorders and Stroke commissioned new visual stimuli for a contemporary global audience, including relevant and identifiable picture elements that depict easily recognizable and relatable scenes and objects. The new picture, the precarious painter, shows a scene with major and minor areas of interest for patients to describe. In addition, six new object illustrations were developed.
STUDY DESIGN: Prospective cohort validation study
SETTING: Convenience sample from the community at large and family members accompanying patients to Johns Hopkins Outpatient Center, the University of South Carolina, and Prisma Health
SYNOPSIS: The study enrolled 101 healthy, fluent English speakers, mostly from the U.S. In Phase I, a 50-participant subsample selected to reflect stroke age distribution and U.S. population demographics described the picture. Descriptions were used to generate content units (CUs), specifically nouns, verbs, modifiers, or prepositional phrases. There were 44 CUs identified, reported by at least 5% of participants. Descriptions were evaluated to examine four discourse variables: total CUs, left:right proportion of CUs, syllables, and syllables per CU. Performance was consistent across all discourse variables irrespective of demographic differences. In Phase II, performance with all participants showed no significant differences related to age, ethnicity, or education. A small but significant difference in left:right CUs by race was noted. Authors suggest this may be spurious, given the small sample size, or reflect an interaction between viewer culture and image content. The average performance and standard deviation were nearly identical between the two phases across all four discourse variables. Forty-one participants were asked to name object illustrations, resulting in high name agreement for six illustrations.
BOTTOM LINE: The new NIHSS visual stimuli, the precarious painter image, and six naming objects, have been validated to produce descriptive samples for language assessment supporting their use in a global population.
CITATION: Stockbridge MD, et al. New picture stimuli for the NIH stroke scale: a validation study. Stroke. 2024;55(2):443-451. doi: 10.1161/ STROKEAHA.123.044384.
Dr. Belcher
Dr. Belcher is a hospitalist at Duke Regional Hospital and an assistant professor of medicine at Duke University, both in Durham, N.C.