A predictive model for medication adherence in older adults with heart failure

Although many studies have examined the predictors of medication adherence (MA), further empirical research is required to clarify the best model for predicting MA for older adults with heart failure (HF). Thus, we hypothesized a model in which information (knowledge), motivation (social support and depressive symptoms), and behavioural skills (barriers to self-efficacy) would be associated with MA in patients with HF.

Using a cross-sectional survey, 153 adults aged ≥ 65 years taking medication for HF were recruited from a university hospital in Korea. Data were collected based on the information-motivation-behavioural skills (IMB) model constructs and MA. In the hypothesized path model, self-efficacy was directly related to MA (β = -0.335, P = 0.006), whereas social support was indirectly related to MA through self-efficacy (β = -0.078, P = 0.027). Depressive symptoms were directly related to MA (β = 0.359, P = 0.004) and indirectly related to MA through self-efficacy (β = 0.141, P = 0.004). The hypothesized MA model showed a good fit for the data. Knowledge, social support, and depressive symptoms accounted for 44.3% of the variance in self-efficacy (P = 0.004). Left ventricular ejection fraction, knowledge, social support, depressive symptoms, and self-efficacy explained 64.4% of the variance in MA (P = 0.004).

These results confirmed the IMB model's suitability for predicting MA in older adults with HF. These findings may guide and inform intervention programmes designed to alleviate depressive symptoms in older adults with HF and enhance their HF knowledge, social support, and self-efficacy, with the ultimate goal of improving their MA.