Background: Long-term opioid therapy (>12 months) is not effective for improving chronic non-cancer pain and function. Where patients are not experiencing pain relief with long-term opioids, the opioid should be tapered and discontinuation considered. Practitioners may find it challenging to tell patients experiencing pain that they are better off reducing or not taking medicines that do not help. This review aims to ascertain what is published about: (1) the interaction and (2) the nature of the relationship between practitioners and patients when prescribing opioids for chronic non-cancer pain in primary care. Method: A scoping review of English-language qualitative, quantitative or mixed-method studies in databases including: MEDLINE, Embase, PsycINFO, AMED, BNI, CINALH EMCARE and HMIC. The identified papers were reviewed to provide a descriptive summary of the literature. Results: The review identified 20 studies. The studies used a range of methods including interviews, focus groups, audio and video recordings of clinical consultations, telephone survey and data from patient records. One study reported that researchers had engaged with a patient advisory group to guide their research. Patients expressed the importance of being treated as individuals, not being judged and being involved in prescribing decisions. Practitioners expressed difficulty in managing patient expectations and establishing trust. Opioid risk and practitioner suspicion shape opioid prescribing decisions. There is a paucity of literature about how precisely practitioners overcome interactional challenges and implement personalised care in practice. Conclusion: The studies in this review ascertain that practitioners and patients often find it challenging to achieve shared decisions in opioid review consultations. Effective communication is essential to achieve good clinical practice. Collaborative research with PPI partners should be aimed at identifying communication practices that support practitioners to achieve shared decisions with patients when reviewing opioids for chronic non-cancer pain. [ABSTRACT FROM AUTHOR]