Arabic cross cultural adaptation and validation of the National Institutes of Health Stroke Scale

Abstract

IntroductionThe National Institutes of Health Stroke Scale (NIHSS), the most commonly used tool to quantify neurological deficit in acute stroke, was initially developed in English. We present our experience in developing and validating an Arabic version of the NIHSS (arNIHSS). MethodsA)Scale development phase: 6 bilingual neurologists translated the scale to Arabic. Items 9 and 10 were modified to suit the Arabic language and culture. A panel of 11 Arab neurologists reviewed the final product and an Arabic language expert did final editing.B)Scale validation phase: 10 examiners (four neurology residents and six nurses), who had no experience with the NIHSS, were trained to use the arNIHSS. Patients with acute stroke were recruited at two academic institutions in Egypt. Each patient was examined on admission by 3 examiners using the arNIHSS and at 24hours by one of the three examiners. The agreement between the first three examinations was used to calculate the interrater agreement. The agreement between the admission and the 24-hour arNIHSS performed by the same examiner was used to calculate the intrarater agreement. Construct validity was evaluated by correlating the arNIHSS on admission with the infarct volume on initial the diffusion weighted imaging (DWI) using the Alberta Stroke Program Early CT score (DWI-ASPECTS) and the functional outcome at 3months assessed by the modified Rankin Scale (mRS). ResultsIn 6months, 137 patients were recruited (mean age±standard deviation 62±12years; 48 women). For interrater agreement, weighted kappa value ranged from 0.36 to 0.66 and intraclass correlation coefficient (ICC) for the whole scale was excellent at 0.95 (95% confidence interval [CI] 0.94–0.97). For intrarater agreement, weighted kappa ranged from 0.52 to 1.0 and the ICC was 0.94 (95% CI 0.87–0.98). The construct validity of the arNIHSS is demonstrated by its correlation with the DWI-ASPECT and the 3months mRS score (Spearman correlation −0.46 and 0.58 respectively; P<0.001 for both). ConclusionWe developed and validated a culturally adapted Arabic version of the NIHSS. Further validation in other Arab countries is recommended.