Development and initial validation of the falls health literacy scale

Abstract

Objectives: (i) To develop the Falls Health Literacy Scale (FHLS), a health literacy tool specific to falls, (ii) to evaluate the FHLS's construct validity towards differentiating individuals with different fall-related health literacy, and (iii) to determine its reliability, construct validity and structure in an older population.

Methods: The initial FHLS, developed based on Sørensen et al.'s health literacy model, was first administered to 144 participants aged ≥18 years for feedback and scale improvement and preliminary analysis to determine the FHLS's construct validity in identifying individuals with different fall-related health literacy. After scale refinement, the FHLS was validated in 227 community-living people aged ≥65 years.

Results: Adult participants with more fall prevention knowledge scored higher on the initial FHLS than those with less fall prevention knowledge (p≤0.001). The final FHLS includes a 25-item subjective and a 14-item objective scale. Older people with ≥1 fall in the past year reported lower FHLS-subjective scores than those who had no falls (Cohen's [d]=0.29, confidence interval [CI]:0.03-0.56, p=0.03). Older people with lower levels of education had lower FHLS-objective scores than their more educated counterparts (d=0.51, CI:0.38-1.43, p≤0.001). Factor analysis of the FHLS-subjective generated six subscales, with CFA showing adequate model fit (RMSEA=0.077, CFI=0.883 and χ2/df =2.35). FHLS-subjective (25-item) showed good reliability, with Cronbach's alpha=0.93, mean inter-item correlation=0.34 (range -0.03-0.81) and intra-class coefficient =0.86 (95% CI:0.69-0.93).

Conclusion: The novel, context-specific FHLS displayed good construct validity and reliability. The FHLS holds promise as a screening tool to differentiate individuals with different degrees of fall-related health literacy, which may help guide fall prevention interventions.

Keywords: Accidental falls; Aged; Health behavior; Health promotion; Psychometrics.