UNDERSTANDING AND SUPPORTING ORAL MEDICATION ADHERENCE IN BREAST CANCER

Oral therapies are widely prescribed for the treatment of different types of cancer, including breast cancer, but, despite their convenience and apparent simplicity, relevant rates of non-adherence to oral medications have been reported. Nonadherence can lead to suboptimal clinical outcomes and increased economic burden on healthcare system. The aim of this PhD thesis is to comprehensively address non-adherence in breast cancer, enhancing understanding of this phenomenon and supporting patients, particularly those with metastatic disease, in maintaining adherence to their oral therapies. In light of healthcare constraints and limited resources, special emphasis is placed on identifying individuals most likely to benefit from targeted interventions or additional support. These objectives were addressed through a series of studies employing diverse design and methodologies. Three literature reviews were conducted: a systematic review of interventions aimed at improving adherence, a scoping review on predictive models of non-adherence, and a systematic review on the contribution of artificial intelligence to adherence. These reviews synthesised and organised existing knowledge, providing a foundation for the subsequent phases of this PhD project. In addition, two original studies focused specifically on metastatic breast cancer patients. The first was a qualitative study comprising focus groups that explored patients’ perceived barriers and facilitators to adherence, as well as their perspectives on digital tools for adherence support. The second was a pilot optimisation study assessing the feasibility and perceived usefulness of three intervention components (i.e., educational material, personalised reminders, and behavioural feedback) delivered according to a full factorial design, while also collecting data on adherence and relevant psychological variables. Across all studies, adherence was confirmed as a multifactorial phenomenon, shaped by clinical, psychological, and relational factors. Interventions including prompts and behavioural feedback appeared most promising, while predictive models, even those using machine learning, showed limited accuracy and faced validation and implementation challenges. For metastatic patients, treatment side effects were a key barrier, whereas supportive patient-clinician relationships facilitated adherence. Digital tools were cautiously welcomed when enabling timely communication or peer support. Unexpectedly, metastatic breast cancer patients showed high adherence, partly explained by strong beliefs in the necessity of their treatment. Finally, a pilot multicomponent intervention combining educational material, reminders, and behavioural feedback proved feasible, low-cost, and valued by patients. In summary, this thesis advances the understanding of medication non-adherence in breast cancer, with a particular focus on patients with metastatic disease. The findings underscore the importance of tailored, patient-centred strategies that consider clinical, psychological, and relational factors, providing a foundation for future research and interventions aimed at supporting adherence and enhancing patient well-being.