Health literacy of patients on oral anticoagulation treatment- individual and social determinants and effect on health and treatment outcomes

Cardiovascular diseases are the leading cause of death worldwide. Atrial fibrillation (AF) is the most frequent arrhythmia, associated with high mortality and morbidity [1], and with an estimated prevalence in Spain of over 4.4%. Therapeutic management of AF requires modification of cardiovascular risk factors and use of drugs [1, 2] such as oral anticoagulants (OAC). Among these, both vitamin K antagonists (VKAs) and direct oral anticoagulants (DOACs) are effective in preventing embolic events [2]. Because the anticoagulant effect of DOACs is difficult to quantify, continuous monitoring and education are necessary to ensure therapeutic follow-up, reflecting the complexity of DOACs use [3, 4].

The recommendations for switching from VKAs to DOACs due to the suboptimal control of the international normalized ratio (INR) were eased in Spain in 2012. At present, however, the legal requirement for a visa and the high financial cost of DOACs prescribing make it difficult to obtain widespread access for patients [5]. For such reasons, VKAs continue to be the most frequently used OAC in Spain, despite the need for routine blood tests and an exquisite clinical control [6]. Further, it is vital that patients demonstrate adequate self-care, along with treatment concordance and sufficient knowledge to avoid or mitigate adverse effects, to maintain the effectiveness of this treatment.

With these requirements in place, health literacy (HL) could be an essential factor for the self-management of anticoagulation treatment, in line with other cardiovascular diseases [7]. Conceptually, HL encompasses different aspects such as empowerment, as well as competencies to understand, evaluate and use health information to make decisions, including the use of health services [8]. Some preliminary studies exploring the effect of HL on awareness about AF and medication concordance have demonstrated that people with inadequate or insufficient levels of HL have poorer health outcomes, have less knowledge about their health problems and diseases, use preventive services less, and suffer from higher mortality [9, 10].

So far, HL has been linked to the health conditions and outcomes of people with cardiovascular disease. Therefore, we believe that effective self-care, adherence to treatment with oral anticoagulants and prevention of complications could be directly related to the level of health literacy.

However, the existing literature has directly associated higher levels of HL with greater knowledge of the disease, but the association among HL, adherence to anticoagulant therapy and the occurrence of adverse events has presented highly variable results and studies have been scarce [11].

Because of this, we propose identifying the level of HL in the local population on anticoagulation treatment and exploring the relationship between HL and several health and clinical outcomes. In addition, our work will provide information about the relationship between the different dimensions that make up HL, including the social dimension as a novel aspect, and compliance with OAC treatment, INR control and the occurrence of complications.

In our study in patients with cardiovascular pathology, health literacy actively influenced the adequate self-management of anticoagulation treatment, the appearance of complications and the unscheduled use of health services. Patients with higher scores on HL dimensions had better control of OAC treatment and optimal frequency of visits. They also reported confidence about their information management skills, empowerment, and self-efficacy, presenting better self-care. These results coincide with those demonstrated by some authors [14‐17].

Therapeutic concordance with OACs was significantly related to the scores of all HL dimensions, with 54.8% of participants presenting an adequate control of OAC treatment. These results are similar to another study carried out in Valencia [18], where 53.9% presented good control, and slightly higher than the 47.5% of patients in Time in Therapeutic Range (TTR) in Granada [19]. Reading et al. [20] related inadequate HL to a lack of pharmacological adherence in AF, while another study associated a poor control of VKA (TTR < 50%) with limited HL, especially in patients over 65 [21]. However, other studies have reported equivocal relations between HL and treatment adherence; for example, HL as measured with the S-TOFHLA scale was associated with OAC treatment adherence but not with TTR [22, 23]. A 2017 study [24] using the SAHLPA scale linked inadequate level of HL to greater cognitive impairment and the need for help to take the treatment properly, but the multivariate analysis could not correlate an inadequate level of HL with having a TTR in the range. It is possible that these studies were unrelated because the scales were only used to evaluate recognition and ability to read certain words, not the different dimensions and skills that comprise HL. Because of this, patients may receive high scores in the scales yet lack more complex treatment-focused skills.

With respect to the occurrence of complications associated with OAC treatment, our study showed a significant association with scores in all HL dimensions, with higher-scoring patients experiencing fewer complications, lower emergency department attendance and hospitalization rates. Complications occurred in 49.2% of participants (228 patients), well above the 9% shown in a 2014 study [25]. This difference could be due to the origin of the information about the occurrence of complications, extracted from medical records in our case, while the 2014 study [25] used patient-reported complications which may have been affected by recall bias, particularly as the study noted that more than 50% of patients did not recognize emergency situations.

Overall, such results may have been expected as HL dimensions allow people to identify reliable and accurate health information, resolve doubts on their own or with peer support and learn about preventive health services, anticipating and mitigating the occurrence of complications. This perspective is also supported by other authors who associate lower HL with a reduced awareness of AF diagnosis [26], adherence to treatment [20], higher rates of re-hospitalization [27], risk of mortality [27‐29] and a decrease in physical and emotional health status [15]. This study [15] also uses the HLQ questionnaire but only evaluated two dimensions, “Understanding health information” and “Engaging with healthcare providers”, and suggests that HL is also associated with health behaviours in cardiovascular patients.

In addition to the above variables, a relationship between age, educational or academic level and social class was also observed with HL scores. Younger patients considered themselves capable of identifying accurate information from different sources, even on their own, remaining up-to-date and knowledgeable about health services. These results are consistent with other studies on patients with cardiovascular health problems, which negatively correlated the level of HL with the age of the participants [30‐32]. The powerful influence of ‘educational’ or academic factors on HL, explaining about 60% of the HL level [31], as well as ‘social class’ has been extensively described [14, 33, 34]. These findings coincide with our results where these variables presented significance in all dimensions of HL.

The health outcomes obtained in our study would align well with the WHO’s social determinants model, which proposes the ‘social origin’ of diseases, in turn causing health inequalities. Within this framework, HL has shown its central role as a determinant of health [35]. In fact, low levels of education limit access to better jobs, or nudging persons towards less secure and riskier jobs, with lower incomes and poorer levels of health throughout life [36, 37].

Our research has shown that different health determinants could contribute to inadequate control of anticoagulant therapy. Such determinants would be older age (over 75 years), basic or no education, limited HL (specifically the HL dimensions “Assessment of health information”, “Ability to find good health information” and “Understanding health information sufficiently to know what to do”), low social class, obesity, multi-morbidity and polypharmacy. This finding allows an early identification of a profile of patients, more vulnerable, who would be most at risk of developing complications, having worse health outcomes.

Further, our study is a pioneer in evaluating the nine dimensions that make up HL in patients on VKAs treatment (only one study has been found that evaluated the HL dimensions in this population and it only assesses two dimensions) and we noted the importance of the “social health support” dimension which showed a higher than expected average score. This dimension assesses a person’s social system and measures the following aspects: If I can get access to several people who understand and support me. If when I feel ill, the people around me really understand what I am going through. If I need help, I have plenty of people I can rely on. If I have at least one person who can come to medical appointments with me and if I I have strong support from family or friends.

In our study, this high score could be explained by the environment in which the study was carried out, a mainly rural área where inhabitants would benefit more from social support networks at neighbourhood and family level, less present or weaker in the urban environment, as shown in some studies carried out in Spain on patients with various pathologies [38, 39].

In addition to the novel aspects discussed above, our work establishes a strong relationship between the dimensions of HL and the social determinants that constitute the axes of inequality in health. It highlights the importance of considering HL as a determinant of health because it is a predictor of individual health status and enables people in vulnerable situations to be identified and health inequalities to be addressed more effectively.

For all these reasons, our study shows how HL intervenes in health behaviours and outcomes, modulating therapeutic concordance and the occurrence of complications in cardiovascular patients. Furthermore, taking into account the social determinants that influence HL, it would be appropriate to establish individual therapeutic measures and community interventions that act synergistically in HL to improve the prevention of cardiovascular disease.

The importance of the social environment in the health and safety outcomes of VKAs treatment is highlighted. People with higher scores on HL dimensions demonstrated better therapeutic control, lower emergency department attendance and fewer hospital admissions, highlighting the importance of intervening on this determinant because of its consequences on the control of patients with heart disease requiring VKAs treatment.