Development and validation of a functional health literacy test

Background: Three validated functional health literacy tests (FHLTs) have been developed for use in healthcare settings. However, none of these tests has been shown to be readily applicable for use in the general public. Objective: To develop and validate an FHLT for screening the functional health literacy level of the general public and patients in healthcare settings. Methods: Maze procedure, expert judgment, and interviews with 55 respondents were used to develop the 21-item FHLT, which was modified from public health education material entitled Guidelines on Taking Medicines published by the Singapore Health Promotion Board. The finalized FHLT was then distributed by convenience sampling among eligible English-speaking respondents from the general public and rheumatic patients in Singapore. Descriptive analysis was used to summarize the sociodemographic characteristics and health status of the respondents. Cronbach's α was used to assess the reliability of the test. Partial correlation coefficients were used to investigate the convergent validity between scores on the FHLT and the Rapid Estimate of Adult Literacy in Medicine (REALM), and the divergent validity between the FHLT score and education level. Test-retest reliability was assessed by intraclass correlation coefficient (ICC). Results: Data were analyzed from 223 respondents from the general public (mean [SD] age: 43.8 [14.2] years, 49.3% female, 45.7% with >12 years of education) and 200 rheumatic patients (mean [SD] age: 46.6 [14.7], 70.5% female, 45.5% with >12 years of education). Cronbach's α for the FHLT was 0.72 and 0.68 for the general public and rheumatic patients, respectively, suggesting adequate reliability. Convergent validity was demonstrated with a strong correlation between scores on the FHLT and REALM (0.65 [p < 0.01] for the general public, 0.68 [p < 0.01] for rheumatic patients). Divergent validity was shown by the weak correlation between the FHLT score and education level (0.33 [p < 0.01] for the general public, 0.28 [p < 0.01] for rheumatic patients). Test-retest reliability of the FHLT among rheumatic patients (n = 112, response rate = 56%) was shown to be high (ICC = 0.95). ICC was not calculated for the test-retest reliability among the general public because of a poor response rate of 9%. Conclusions: The 21-item FHLT has been shown to be a reliable and valid screening test for measuring functional health literacy levels of the general public as well as patients in healthcare settings. The availability of the FHLT would contribute to the effectiveness of health education programs and better health-related outcomes in the general population