Changing Medication-Related Beliefs: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Objective: Medication-related beliefs, for example, beliefs that medicines are unnecessary or that side effects are likely, can influence medication behaviors and experiences, potentially impacting quality of life and mortality. At times, it may be useful to change medication-related beliefs, for example, to reduce patients' concerns about side effects when extensive evidence suggests side effects are rare. Currently we do not know the most effective methods to address medication beliefs. Method: Systematic review and meta-analysis of randomized controlled trials that measured medication-related beliefs in people prescribed medication for long-term condition(s). We extracted data on behavior change techniques (BCTs), belief measure, study and patient characteristics, risk of bias, and quality of description. Results: We identified 56 trials randomizing 8,714 participants. In meta-analysis, interventions led to small-to-medium effects (n = 36, Hedges' g = .362, 95% confidence interval [CI] [.20, .52], p < .001) in increasing beliefs about medication need/benefit and reducing concerns about medication (n = 21, Hedges' g = -.435, 95% CI [-0.72, -0.15], p < .01). Effect sizes were higher for interventions that reported a significant effect on adherence. Problem solving, information about health consequences, and social support (unspecified) were the most prevalent BCTs. Fourteen BCTs were associated with significant effects on need/benefit beliefs and four BCTs were associated with significant effects on concern beliefs. Conclusion: It is possible to modify medication-related beliefs using a range of interventions and techniques. Future research should explore the best ways to operationalize these BCTs for specific health conditions to support medication beliefs and improve adherence.