Psychometric properties of the European health literacy survey questionnaire short-short form (HLS-EU-Q6) among Slovenian youth, adults, and older adults

Abstract Background Health literacy is an important determinant of health behaviours and self-rated health, making brief, valid, and reliable assessment tools essential. The six-item European Health Literacy Survey Questionnaire (HLS-EU-Q6) has rarely been validated across age groups and has not been validated in Slovenia. This study evaluated its validity and reliability among youth, adults, and older adults. Methods Data were drawn from the Health Literacy of Adults in Slovenia survey (HLS-SI19; N = 3,360), including youth (18–30 years), adults (31–64 years), and older adults (65 + years). Construct validity was assessed using confirmatory factor analysis and Item Response Theory (Samejima's Graded Response Model); convergent, discriminant, concurrent, concurrent-predictive, and predictive (criterion-related) validity, together with internal reliability, were assessed using correlations and hierarchical regression analyses. Results A one-factor structure demonstrated good fit among older adults but weaker fit among youth, adults, and the total sample. IRT analyses showed satisfactory item discrimination and measurement precision, although four items showed age-related differential item functioning, indicating partial measurement invariance across age groups. Internal consistency was satisfactory across all groups (Cronbach's α = 0.76–0.84). Discriminant validity relative to communicative and navigational health literacy, and concurrent validity with the HLS-EU-Q47 (r = 0.892), were supported in all age groups. Predictive validity appeared stronger among adults and older adults than among youth: among younger participants health literacy predicted only self-rated health, whereas among adults and older adults it also predicted more favourable attitudes toward vaccination. Conclusions The Slovenian HLS-EU-Q6 is a reliable and practical screening instrument, particularly for adults and older adults, although its construct and predictive validity among youth warrant further age-sensitive refinement.