Aim: To synthesise qualitative evidence on how patients, caregivers and healthcare professionals perceive and respond to cancer-related anorexia (CRA), and to develop a multi-level framework for improving CRA care.
Design: A qualitative meta-synthesis using the Joanna Briggs Institute (JBI) methodology, informed by the Social Ecological Model (SEM).
Methods: Seven databases were searched for qualitative studies from inception to April 2025. Studies were assessed using the JBI Critical Appraisal Checklist. Meta-aggregation was used to synthesise findings, and the ConQual method assessed confidence levels.
Data sources: PubMed, Embase, CINAHL, PsycINFO, Cochrane Library, CNKI and WanFang.
Results: Seventeen studies from 10 countries were included, reflecting the perspectives of patients, caregivers and healthcare professionals. Four synthesised findings were identified. At the individual level, CRA was linked to physical decline, emotional distress and changes in identity. The interpersonal level involved feeding-related tensions and caregiver burden. Organisational barriers included delayed care and poor cultural responsiveness. Policy-level factors such as limited insurance coverage and rural access further impeded care. Overall confidence in these synthesised findings was low to moderate.
Conclusion: CRA is not solely a biological condition but a multidimensional experience. Addressing CRA requires integrated and context-sensitive strategies across personal, relational, organisational and policy domains.
Implications: Nurses and clinicians should address not only physical symptoms but also the emotional and social dimensions of eating. Structured support for caregivers and improved service access, particularly in underserved settings, are needed.
Impact: This study provides a multi-level understanding of CRA. The findings support better patient care, caregiver support and more equitable healthcare policy design.
Reporting method: JBI methodology and ENTREQ guideline.
Patient or public contribution: No Patient or Public Contribution.
Trial registration: PROSPERO Database: CRD420251041265.

