Journal of General Internal Medicine

, Volume 30, Issue 12, pp 1732–1732 | Cite as

Health Literacy During Aging

  • Lindsay C. KobayashiEmail author
  • Christian von Wagner
Letter to the Editor


Health Literacy Literacy Skill Visual Format Poor Vision Literacy Assessment 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

To the Editors

We appreciate Mattiesen et al.’s valuable letter regarding our paper. We agree that poor vision is an important risk factor for low health literacy when health information is provided in written or visual format. Hence, the effect of vision on health literacy is contextual. In our study, a medicine label printed at A4 size was used to assess health literacy; therefore, vision was important in this testing context (and in most others in health literacy research). Unlike in other instruments for assessing health literacy (e.g. the TOFHLA), vision was not formally tested as part of the English Longitudinal Study of Ageing (ELSA), and participants with poor vision were not a priori excluded from the study. However, participants could refuse the health literacy assessment if they either could not or chose not to complete it due to poor vision. At wave 2, 1.5 % (132/8780) and at wave 5, 1.6 % (96/5840) of respondents refused the health literacy test due to poor vision. Please note that the proportion of refusals due to poor vision is not the same as the proportion of participants in the study who had poor vision. Self-rated vision was assessed in the ELSA study interviews, and was distributed as follows: at wave 2, 15 % of our sample (803/5256) had ‘excellent’ vision, 36 % (1877/5256) had ‘very good’ vision, 39 % (2045/5256) had ‘good’ vision, 9 % (454/5256) had ‘fair vision’, and 1 % (76/5256) had ‘poor’ vision. The corresponding proportions for wave 5 were similar, at 14 % (727/5256), 35 % (1814/5256), 39 % (2050/5256), 10 % (538/5256), and 2 % (126/5256), respectively. When self-rated vision at wave 2 is added to our final multivariable model from the paper, having ‘fair’/‘poor’ vision is significantly associated with increased odds of health literacy decline over the follow-up (adjusted OR = 1.38; 95 % CI: 1.11-1.72; vs. ‘excellent’/’very good’/‘good’ vision). The inclusion of self-rated vision in the model does not change the observed ORs between age, any cognitive variables, and health literacy decline. Hence, self-rated vision does not appear to be a mediator or confounder of the relationships between age and health literacy decline, and between cognitive function/decline and health literacy decline. We agree that future work on health literacy should not exclude older adults with poor vision. Adults with poor vision are at risk of low health literacy when information is provided in visual formats and appear to be vulnerable to ageing-related decline in health literacy skills.

Copyright information

© Society of General Internal Medicine 2015

Authors and Affiliations

  1. 1.Health Behaviour Research Centre, Department of Epidemiology and Public HealthUniversity College LondonLondonUK

Personalised recommendations