To our knowledge, this is the first publication addressing the relationship between literacy and multimorbidity. The results of the study suggest that low literacy is associated with the presence of multimorbidity in adults consulting in primary care in bivariate analysis, but this association is no longer present when controlling for age and family income.
Many previous studies have indicated that patients with low literacy were more likely to report poorer health than patients with adequate literacy [19
], while other studies found no relationship between literacy and health status when controlling for education [46
]. Many studies were conducted in older aged patients [21
] making generalization of results to other age groups difficult. One study did not adjust for confounding variables [21
]. In another study [20
], literacy was not evaluated with a validated tool but by the staff of the institution where the study was carried out. In all studies, health status was evaluated globally using self-rated health status categories. In this research, we used a more detailed instrument to measure patients’ disease burden than the studies supporting the existence of a relationship between health literacy and global health measures.
Other research has addressed the relationship between literacy and specific diseases with inconsistent results. One study reported that inadequate literacy (measured with the short form of the Test of Functional Health Literacy in Adults) was an independent predictor of diabetes mellitus and heart failure but not hypertension, arthritis or pulmonary disease, while adjusting for sociodemographic variables (age, sex, race/ethnicity, education and annual income), health risk behaviours (smoking habit and alcohol use), and body mass index [45
]. Another study using the same measure of literacy found that heart attack, stroke, hypertension, diabetes, arthritis and depression were all associated with the literacy level [48
]. A study accounting for hypertension, diabetes, obesity and depression concluded that only depression remained significantly associated with literacy after adjusting for confounders [46
]. Another study reported that individuals with low literacy had significantly higher rates of arthritis and hypertension, but no statistical differences were found in the prevalence of diabetes, pulmonary, or heart disease [25
A low level of literacy may be linked to certain health issues and not others. As our measure of multimorbidity evaluated chronic diseases as a whole as well as their severity, we may not have detected a link because of the presence of specific health issues not associated with literacy. An association between literacy and multimorbidity may exist when two or more specific diseases individually related to health literacy coexist in one person. The conceptualization and measure of multimorbidity could therefore have an impact on this association. That is why we conducted analysis using two distinctive conceptualizations and measures, in order to verify if results were different. It was not the case.
We found that multimorbidity was associated with age and family income in the multivariate models. The association of multimorbidity with age is well recognized [7
]. The relationship between socioeconomic status and multimorbidity has also been extensively documented [8
Our results do not allow us to rule on a potential association between health literacy, a more global concept than literacy [28
], and multimorbidity. A comprehensive measure of health literacy that considers other dimensions of the concept still needs to be developed. We could then verify if there is a link between health literacy and multimorbidity. Although we did not observe a direct association between literacy and multimorbidity, it is still important to continue taking this variable into account in patient care in order to tailor health information to patient needs and in a format they can understand [42
A limitation of this study is that participants were not randomly selected from the general population. We recruited patients from the waiting room of a single primary care setting. This method may over sample complex patients with several diseases or frequent attendees. However, we were able to recruit a group of patients with a good distribution of multimorbidity and literacy. Another limitation is the lack of statistical power to carry out multivariate analysis by individual disease. Although the study was conducted in one family practice, we expect the same results from similar primary care settings.
In conclusion, this study suggests that there is no relationship between literacy and multimorbidity when controlling for age and family income. Patients with multimorbidity may have specific diseases that are associated with low literacy. Further studies are needed to identify individual diseases and combinations of diseases linked to literacy while controlling for potential confounding variables. A possible association between health literacy and multimorbidity still needs to be explored when a comprehensive measure of health literacy is available.