Beliefs about medications predict adherence to antidepressants in older adults

2012 
Background: Adherence to treatment is a complex and poorly understood phenomenon. This study investigates the relationship between older depressed patients’ adherence to antidepressants and their beliefs about and knowledge of the medication. Methods: Assessment was undertaken of 108 outpatients over the age of 55 years diagnosed with depressive disorder and treated for at least four weeks with antidepressants. Adherence was assessed using two self-report measures: the Medication Adherence Rating Scale (MARS) and a Global Adherence Measure (GAM). Potential predictors of adherence investigated included sociodemographic, medication and illness variables. In addition, 33 carers were interviewed regarding general medication beliefs. Results: 56% of patients reported 80% or higher adherence on the GAM. Sociodemographic variables were not associated with adherence on the MARS. Specific beliefs about medicines, such as “my health depends on antidepressants” (necessity) and being less worried about becoming dependant on antidepressants (concern) were highly correlated with adherence. General beliefs about medicines causing harm or being overprescribed, experiencing medication side-effects and severity of depression also correlated with poor adherence. Linear regression with the MARS as the dependent variable explained 44.3% of the variance and showed adherence to be higher in subjects with healthy specific beliefs who received more information about antidepressants and worse with depression severity and autonomic side-effects. Conclusions: Our findings strongly support a role for specific beliefs about medicines in adherence. Challenging patients’ beliefs, providing information about treatment and discussing side-effects could improve adherence. Poor response to treatment and medication side-effects can indicate poor adherence and should be considered before switching medications.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    31
    References
    35
    Citations
    NaN
    KQI
    []