First‐order vs. second‐order structural validity of the Health Literacy Scale in patients with diabetes
Based on the World Health Organization definition, Nutbeam 4 proposed a model of health literacy that divided this into three types of literacy: functional health literacy (basic skills in reading and writing health information), communicative/interactive health literacy (cognitive and social skills in assessing and deriving health information), and critical health literacy (cognitive and social skills in analysing and using health information). Ishikawa et al. 5 used this model to develop the Health Literacy Scale (HLS) in Japanese patients with diabetes, which comprised functional, communicative, and critical subscales. The HLS is currently considered to be a comprehensive and useful instrument measuring health literacy for patients with diabetes 6.
However, a discrepancy in the structural validity of the HLS is emerging as a bothersome problem. Three subscales (factors) of the HLS were demonstrated in the original HLS study 6 and subsequent studies 7. Dwinger et al. 10 recently empirically demonstrated a two‐factor model, in which they combined the communication and critical subscales into one subscale due to the strong correlation between these subscales. This strong correlation between those two subscales is also a potential issue in the three‐factor model of the HLS.
A second‐order model is applicable when the first‐order factors are substantially correlated with each other 11. Based on this perspective, a second‐order model was proposed as the underlying structure of the HLS in the present study. To our knowledge, a second‐order model has never been investigated for the HLS. Therefore, the purpose of this study was to determine the structural validity of the HLS in Korean patients with diabetes based on the previously reported first‐order three‐factor and two‐factor models and a newly proposed second‐order model, to identify which model best represents the structure of the HLS.