The Reduced Multi-Dimensional Health Literacy Scale: A First Bilingual Validation Using RASCH Analysis.

This study aimed to evaluate the reliability and validity of the reduced health literacy scale (HLS) of Ishikawa, as used in translated German and Italian versions. METHODS: The reduced HL scale was applied to a convenience sample of 100 persons with type 2 diabetes (DM2) in a province of Northern Italy. Rasch analysis was used to evaluate the psychometric measures of the nine items, regarding communicative and critical HL. RESULTS: Person and item reliability was 0.87 for the German and 0.72 for the Italian version. Correlation values ranged from 0.55 to 0.76 (German version) and from 0.48 to 0.70 (Italian version). MNSQ values INFIT and OUTFIT expressed generally positive results. Possible redundancy was observed for one item in the German and Italian version, respectively, and in one for both versions. The values obtained in the analysis indicate a good fit, and no item had to be changed. In the German version, about 14% of persons demonstrate a high HL level (10% for the Italian version), while 39% expressed a low level of HL (56% for the Italian version). CONCLUSIONS: No particularly critical items of the reduced HL scale emerged in both versions applying Rasch analysis. The use of the reduced scale (only nine items) has the advantage of being integrated into longer questionnaires and into more structured assessment instruments. Furthermore, it could also be used as a rapid screening tool, allowing health professionals to address patients and their HL needs in a more targeted way. However, more psychometric testing with a larger sample is required.