Change in Health Literacy over a Decade in a Prospective Cohort of Community-Dwelling Older Adults

• Published in: Journal of general internal medicine : JGIM 2021-04, Vol.36 (4), p.916-922
• Main Authors: Curtis, Laura M., Kwasny, Mary J., Opsasnick, Lauren, O’Conor, Rachel M., Yoshino-Benavente, Julia, Eifler, Morgan, Federman, Alex D., Altschul, Drew, Wolf, Michael S.
• Format: Article
• Language: English
• Subjects: Adults; Aged; Chicago - epidemiology; Cohort Studies; Correlation analysis; Cross-Sectional Studies; Health education; Health Literacy; Humans; Independent Living; Internal Medicine; Longitudinal studies; Medicine; Medicine & Public Health; Middle Aged; Multivariable control; Older people; Original Research; Prospective Studies; Regression analysis; Regression models; Surveys and Questionnaires; Time measurement
• ISSN: 0884-8734; 1525-1497
• DOI: 10.1007/s11606-020-06423-8

Background Health literacy is often viewed as a static trait in longitudinal studies, which may over or underestimate an individual’s ability to manage one’s health. Objectives We sought to examine health literacy over time among older adults using three widely used measures. Design A prospective cohort study. Participants Community-dwelling adults ages 55 to 74 at baseline with at least one follow-up visit ( N = 656) recruited from one academic internal medicine clinic and six community health centers in Chicago, IL. Measures Health literacy was measured using the Test of Functional Health Literacy in Adults (TOFHLA), Newest Vital Sign (NVS), and Rapid Estimate of Adult Literacy in Medicine (REALM) at baseline and up to three follow-up time points. Results In unadjusted analyses, significant changes since baseline were found beginning at the second follow-up (mean ( M ) = 6.0 years, SD = 0.6) for the TOFHLA ( M = − 0.9, SD = 0.95, p = 0.049) and the REALM ( M = 0.3, SD = 2.5, p = 0.004) and at the last follow-up ( M = 8.6 years, SD = 0.5) for the NVS ( M = − 0.2, SD = 1.4, p = 0.02). There were non-linear effects of baseline age on TOFHLA and NVS scores over time (piecewise cubic spline p = 0.01 and p < 0.001, respectively) and no effect on REALM scores ( B = 0.02, 95% CI − 0.01 to 0.04, p = 0.17) using multivariable mixed-effects linear regression models, controlling for race, education, income, and comorbidity. Conclusion We found a negative relationship between age and health literacy over time as measured by the TOFHLA and NVS. Health literacy barriers appear to be more prevalent among individuals in later life, when self-care demands are similarly increasing. Clinicians might consider strategies to assess and respond to limited health literacy, particularly among patients 70 and older. REALM performance remained stable over 10 years of follow-up. This questions whether health literacy tools measure the same attribute. Prospective health literacy studies should carefully consider what measures to use, depending on their objective.