Open AccessTranslation and validation of a multidimensional instrument to assess health literacy
Author(s) -
Dwinger Sarah,
Kriston Levente,
Härter Martin,
Dirmaier Jörg
Publication year - 2015
Publication title -
health expectations
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.314
H-Index - 74
eISSN - 1369-7625
pISSN - 1369-6513
DOI - 10.1111/hex.12252
Subject(s) - health literacy , confirmatory factor analysis , psychology , scale (ratio) , reliability (semiconductor) , construct validity , german , literacy , applied psychology , clinical psychology , psychometrics , structural equation modeling , health care , statistics , mathematics , geography , political science , pedagogy , power (physics) , physics , quantum mechanics , cartography , archaeology , law
Background The patients' task to find, evaluate and transfer health information to one's individual condition and life requires competences that are summarized by the term ‘health literacy’. Poor health literacy is associated with poorer health outcomes, like a higher rate of rehospitalization, lower receipt of screenings and a higher frequency of doctor visits. Objective Three levels of health literacy are distinguished: functional, communicative and critical health literacy. Aim of this study was to translate and adapt the ‘ F unctional C ommunicative C ritical H ealth L iteracy’ ( FCCHL ) questionnaire to G erman, and assess its psychometric properties. Methods/Design The FCCHL was sent to 9075 participants enrolled in a RCT on health coaching. 4040 participants responded. Besides descriptive and reliability analysis, confirmatory factor analysis was performed to test the questionnaire's postulated scale structure in a calibration ( N = 3000) and a validation sample ( N = 1040) for cross‐validation. Results The instrument was well accepted (missing values ≤2.1% per item) and showed acceptable or good internal consistency for the entire scale (α = 0.77) and the subscales (α = 0.75–0.80). The proposed three‐factor model did not fit the G erman data sufficiently. As the scales ‘communicative health literacy’ and ‘critical health literacy’ showed high intercorrelation (0.98), they were combined to a new scale called ‘processing health literacy’. The fit indices for the amended two‐factor model were satisfying in both subgroups. Conclusions Reliability and acceptance of the G erman FCCHL are satisfying. An amended two‐factor structure showed better validity than the original factor structure. Further research regarding the FCCHL and the underlying construct is needed.