Validation of international stroke scales for use by nurses in Greek settings

Background: Improving stroke outcomes by educating nurses in state-of-the-art stroke nursing skills is essential, but unfortunately, to date, there are limited validated stroke assessment scales for routine clinical and research use in Greece. Objective: The aim of this paper is to validate and culturally adapt three internationally recognised stroke scales for use in Greece. Methods: A critical appraisal of the international literature was undertaken to identify suitable scales to assess stroke impact: neurological, functional status and level of dependence. We identified: Scandinavian Stroke Scale (SSS), Barthel Index (BI) and modified Rankin Scale (mRS). They were formally translated and culturally adapted from English to Greek. Their validity was tested using Cronbach's alpha and Median Discrimination Index, while construct validity was checked by Principal Component Analysis (PCA). These were used on 57 consecutively selected patients with stroke from a Greek hospital, mean age 67.7 (+/- 6.7 SD) years, range 54-85 years, length of stay, 8.5 (+/- 2.7 SD) days. Results: All three scales show high internal consistency. The Cronbach's a on admission/discharge for the SSS ranged from 0.86 to 0.88. The BI's reliability ranged from 0.95 to 0.93. The Median Discrimination Index was 0.70 (SSS) and 0.83 (BI). PCA showed that although a significant general factor (F1) explains most of the variance (57.0% on admission and 56.4% on discharge) a second factor (F2) of less significance was also highlighted. The convergent validity of the three scales was confirmed. Discussion: The stroke tools selected showed high reliability and validity, thus making these suitable for use in Greek clinical/academic environments. All three scales used are almost routinely undertaken in stroke studies internationally and form a backdrop for bio-statistical, functional and social outcome post-stroke. Conclusions: The Greek version of the stroke tools show that both SSS and BI have high internal consistency and reliability and together with the mRS could be used in any Greek stroke care delivery setting.