Abstract
BACKGROUND: Suboptimal medication adherence remains a major cause of allograft failure after kidney transplantation. Previous studies have focused on isolated factors rather than integrated mechanisms. Based on the COM-B model, this study investigates the mediating roles of medication beliefs and regulatory emotional self-efficacy (RESE) between medication literacy, social support, and medication adherence. METHODS: A cross-sectional survey included 351 kidney transplant recipients (KTRs) from a tertiary hospital in Changsha (April-July 2025). Participants completed a general information questionnaire, the Basel Assessment of Adherence to Immunosuppressive Medications Scale, the Chinese version of the RESE Scale, the Social Support Rating Scale, the Chinese Medication Literacy Scale, and the Beliefs about Medicines Questionnaire-Specific. Data were analyzed using SPSS and AMOS for descriptive, correlational, hierarchical regression, and mediation analyses (bootstrapping with 5000 samples). RESULTS: The medication non-adherence rate in KTRs was 37.6%, primarily due to missed doses (33.3%). Medication literacy, social support, medication beliefs, and RESE were significantly correlated with adherence (p < 0.01). After controlling for demographic variables, these factors explained 47.2% of the variance in adherence. Path analysis showed that medication literacy (β= -0.219) and social support (β= -0.180) directly reduced non-adherence and also indirectly improved adherence through medication beliefs and RESE. Specifically, medication literacy had indirect effects via medication beliefs (β= -0.034, 11.6%) and RESE (β= -0.039, 13.4%); social support exerted indirect effects through medication beliefs (β= -0.113, 35.0%) and RESE (β= -0.030, 9.3%). All bootstrap 95% CIs excluded zero. CONCLUSION: Medication adherence among KTRs remains suboptimal. Within the COM-B framework, this study confirms that medication literacy and social support not only directly affect adherence but also exert indirect effects through the dual mediating pathways of medication beliefs and RESE. These findings suggest that clinical interventions should adopt a multidimensional approach, focusing not only on enhancing medication knowledge and support systems but also specifically addressing patients' medication beliefs and emotional self-efficacy. A multi-path synergistic strategy is recommended to optimize intervention effectiveness.