Portuguese adaptation and psychometric validation of the Osteoporosis Knowledge Assessment Tool in community-dwelling...

Given the critical role of knowledge in prevention and management of osteoporosis and osteoporotic fractures, a culturally adapted instrument to assess osteoporosis-related knowledge among citizens is essential for patient education.

This study aimed to translate, culturally adapt, and psychometrically validate the Osteoporosis Knowledge Assessment Tool (OKAT) for Portuguese adults aged 50 years and older.

The study involved two phases: translation and cross-cultural adaptation, and evaluation of measurement properties. An expert committee developed a pre-final Portuguese OKAT, which was tested in adults aged 50-74 years, recruited from primary care and from a blood donation centre. Reliability was assessed via internal consistency (α, ω, KR-20), test-retest reproducibility (r, ICC), and item-level agreement (Cohen's Kappa). Item difficulty, discrimination, and readability were also evaluated. Known-groups validity was examined by comparing scores across sex, age, education, and prior osteoporosis diagnosis.

Phase A identified five items with semantic or conceptual challenges, resolved through expert consensus. In phase B, 153 participants (mean age 62 ± 7 years; 54% female) completed the Portuguese OKAT, showing borderline acceptable to acceptable internal consistency (α = 0.698; ω = 0.700; KR-20 = 0.698) and good test-retest (n = 20) reliability (ICC = 0.84; r = 0.84; Cohen's Kappa = 0.375-0.89, with four items yielding non-computable or negative values). Item difficulty and discrimination were appropriate, and readability was suitable (Flesch Reading Ease = 68.35). Women had statistically higher knowledge scores, but no significant differences were found for age, education, or prior diagnosis groups.

The Portuguese version of OKAT demonstrated generally acceptable psychometric performance, supporting its use in adults aged ≥50y. Future research should confirm these findings in more diverse and representative populations.