INTRODUCTION

Patients with cancer are at increased risk of thrombotic complications from both the disease and its treatments, with antithrombotic therapy (ATT) usually continued in the last phase of life where the benefit is less clear and there is high risk of harms. Physiological changes toward the end of life increase the risk that ATT will cause serious bleeding events, but discussion between clinicians and patients of ATT risks and benefits is sub-optimal. This realist synthesis explores shared decision-making (SDM) to: (a) provide insights into why prescribing continues in end-of-life care; (b) build a conceptual platform for optimizing ATT prescribing for persons living with cancer towards end-of-life.

METHODS AND FINDINGS

We conducted a realist synthesis using context-mechanism-outcome configurations and 'if…then' statements. A total of 17,036 citations identified across 10 databases (Medline, EMBASE, APA PsycInfo, CINAHL Complete, CDSR, CENTRAL, EPISTEMONIKOS, Web of Science Core Collection, Assia, Google Scholar). Ninety-one papers included following reverse chronology quota record screening (from: database searches (n = 56), consortium experts (n = 35)). Included papers: quantitative (n = 40), qualitative (n = 17), mixed-methods (n = 2), evidence syntheses (n = 16), commentaries (n = 9), case reports (n = 7). Exclusion criteria: persons <18 years, non-English language, not focused on SDM or deprescribing in palliative care. An analytic appraisal journal was used with realist logic to synthesize insights from included papers (contents from 43/91 included in this paper). The concept of 'prescribing inertia' was used to formulate explanatory theories about clinician reluctance to deprescribe and the mechanisms underpinning SDM, including (a) the meaning medications have to end-of-life patients (e.g., 'life preserving' or 'symptom management') and public awareness of medications (e.g., high-profile chemotherapy versus low-profile ATT) are determinants of: (i) clinician motivation to engage patients around deprescribing, (ii) patient understanding, volition and participation in SDM; (b) SDM for ATT deprescribing requires sensitive engagement with patients and families without removing positivity around survival and continuing clinician interest in their welfare; (c) multi-disciplinary clinical decision-making about timing and suitability of deprescribing in end-of-life care requires specialized consensus-driven processes and evidence-based decision support tools; (d) if patients are healthy enough, empowerment interventions outside clinical encounters (e.g., health literacy apps) may increase patient and family readiness to engage in deprescribing conversations; (e) organizational investments can facilitate discussion of deprescribing (e.g., improved electronic medical record prompts, clinician communication skills training, data presentation to clinicians of actual ATT risks). Limitations: despite robust screening and selection, the sample of included papers does not reflect the entirety of eligible source material and did not include a systematic search for papers focusing on low- and middle-income country countries.

CONCLUSION

Implementation of ATT deprescribing is enabled or constrained by (a) the meaning of medications to patients; (b) clinician engagement and understanding; (c) multi-disciplinary clinical decision-making processes (including support tools); (d) patient empowerment; (e) organizational investment. Addressing these multi-level factors, including the development of SDM tools, can address the prescribing inertia that may cause devastating impacts on patients and their families as well as moral distress amongst healthcare staff. This study was performed as part of the Horizon-Europe funded SERENITY project.