Predictors of Inadequate Health Literacy among Patients with Type 2 Diabetes Mellitus: Assessment with Different Self-Reported Instruments

Introduction: Obtaining, understanding, interpreting, and acting on health information enables people with diabetes to engage and make health decisions in various contexts. Hence, inadequate health literacy (HL) could pose a problem in making self-care decisions and in self-management for diabetes. By applying multidimensional instruments to assess HL, it is possible to differentiate domains of functional, communicative, and critical HL. Objectives: Primarily, this study aimed to measure the prevalence of inadequate HL among type 2 diabetes mellitus patients and to analyze the predictors influencing health literacy levels. Secondly, we analyzed if different self-reported measures, unidimensional instruments (Brief Health Literacy instruments (BRIEF-4 and abbreviated version BRIEF-3), and multidimensional instruments (Functional, Communicative and Critical health literacy instrument (FCCHL)) have the same findings. Methods: The cross-sectional study was conducted within one primary care institution in Serbia between March and September 2021. Data were collected through Serbian versions of BRIEF-4, BRIEF-3, and FCCHL-SR12. A chi-square test, Fisher’s exact test, and simple logistic regression were used to measure the association between the associated factors and health literacy level. Multivariate analyses were performed with significant predictors from univariate analyses. Results: Overall, 350 patients participated in the study. They were primarily males (55.4%) and had a mean age of 61.5 years (SD = 10.5), ranging from 31 to 82 years. The prevalence of inadequate HL was estimated to be 42.2% (FCCHL-SR12), 36.9% (BRIEF-3) and 33.8% (BRIEF-4). There are variations in the assessment of marginal and adequate HL by different instruments. The highest association was shown between BRIEF-3 and total FCCHL-SR12 score (0.204, p < 0.01). The total FCCHL-SR12 score correlates better with the abbreviated BRIEF instrument (BRIEF-3) than with BRIEF-4 (0.190, p < 0.01). All instruments indicated the highest levels for the communicative HL domain and the lowest for the functional HL domain with significant difference in functional HL between the functional HL of FCCHL-SR12 and both BRIEF-3 and BRIEF-4 ( p = 0.006 and 0.008, respectively). Depending on applied instruments, we identified several variables (sociodemographic, access to health-related information, empowerment-related indicators, type of therapy, and frequency of drug administration) that could significantly predict inadequate HL. Probability of inadequate HL increased with older age, fewer children, lower education level, and higher consumption of alcohol. Only high education was associated with a lower probability of inadequate HL for all three instruments. Conclusions: The results we obtained indicate that patients in our study may have been more functionally illiterate, but differences between functional level could be observed if assessed by unidimensional and multidimensional instruments. The proportion of patients with inadequate HL is approximately similar as assessed by all three instruments. According to the association between HL and educational level in DMT2 patients we should investigate methods of further improvement.