Multiturn Large Language Model-Based Conversational Agents for Patients With Cancer and Caregivers: Scoping Review
Clinical Snapshot
PICO Framework
| P — Population | Adult patients with cancer and informal caregivers |
| I — Intervention | Large language model (LLM)-based multiturn conversational chatbots designed for oncology support (information provision, symptom consultation, emotional assistance) |
| C — Comparator | No formal comparator — scoping review mapping the research landscape; studies may include comparisons to standard care or other chatbot types |
| O — Outcomes | System design characteristics, intervention purposes, evaluation approaches, safety considerations, transparency of LLM-related components, and conversational capability (e.g., context retention, conversational memory) |
Bottom Line
This scoping review maps the nascent landscape of large language model-based multiturn chatbots designed for patients with cancer and their caregivers. Despite comprehensive searching across eight databases, only eight studies met inclusion criteria — a finding that itself underscores how early-stage this field remains. Most identified studies describe prototype systems rather than clinically evaluated interventions. ChatGPT-based architectures predominate, with retrieval-augmented generation emerging as a common design feature. Chatbots are primarily oriented toward emotional support and information provision, yet no study evaluated the conversational capabilities — context retention, memory, continuity — that are fundamental to genuine multiturn interaction. Safety reporting was inconsistent and often absent. For senior clinicians, the message is clear: LLM-based conversational chatbots in oncology are not yet ready for clinical deployment. The evidence base lacks clinical outcome data, validated safety frameworks, and regulatory-grade transparency. Institutions considering piloting such tools should do so only within formal research governance structures, with prospective ethics approval, robust safety monitoring, and explicit informed consent. Future research must prioritise clinical outcome evaluation, standardised safety reporting, and transparent LLM documentation before these tools can be responsibly integrated into cancer care pathways.
Key Findings
Effect Size: Not applicable — scoping review; descriptive findings only. Eight studies met inclusion criteria from a search spanning January 2022 to May 2026 across eight databases.
Primary Outcome: Mapping of the research landscape of LLM-based multiturn conversational chatbots for patients with cancer and informal caregivers across five domains: system design, intervention purpose, evaluation approaches, safety considerations, and LLM transparency
Nnt Or Sensitivity: Not applicable — no clinical outcome data synthesised. Key descriptive findings: ChatGPT-based models were the most commonly used LLMs; retrieval-augmented generation (RAG) was applied in several studies; chatbots were primarily designed for emotional support or information provision; no studies evaluated conversational continuity or context retention; only 2 of 8 studies reported any conversational memory mechanism; safety risk reporting, mitigation strategies, prompt design, and model parameter transparency were frequently absent.
Confidence Interval: Not applicable — no quantitative synthesis performed
Clinical Application
Clinical implementation of LLM-based multiturn chatbots in oncology is not yet feasible based on current evidence. The field is characterised by prototype-stage development, absence of clinical outcome data, inconsistent safety frameworks, and limited evaluation of core conversational capabilities (context retention, memory). Significant development, validation, and regulatory work is required before deployment in clinical settings. In Australia, the Therapeutic Goods Administration (TGA) has issued guidance on software as a medical device (SaMD), and LLM-based chatbots intended for clinical decision support or symptom management may require regulatory classification under the TGA's Digital Health framework. The Australian Digital Health Agency's National Digital Health Strategy 2023–2028 identifies AI-enabled tools as a priority area, but no specific regulatory pathway for oncology chatbots has been established. Cancer Australia and the RACGP have not yet issued guidelines on LLM-based conversational agents in cancer care. PBS implications are not directly relevant at this stage. Australian oncology nurses and general practitioners should be aware that no evidence-based chatbot system currently meets the threshold for clinical recommendation. Institutional governance frameworks for AI tools in healthcare, such as those being developed by the Australian Commission on Safety and Quality in Health Care, will be essential before any deployment. Adult patients with cancer across disease trajectories and their informal caregivers, particularly those with complex informational, symptom management, and psychosocial support needs. The findings are most relevant to oncology nurses, digital health implementers, and health informaticians considering chatbot deployment in cancer care settings.
Abstract
BACKGROUND: Large language model (LLM)-based conversational agents are increasingly used in health care, yet their capacity to support genuine multiturn dialogue remains underexplored. In oncology, where patients and caregivers experience complex informational and emotional needs throughout the disease trajectory, conversational agents may support information provision, symptom consultation, and emotional assistance. However, research specifically examining multiturn conversational agents designed for patients with cancer and informal caregivers remains limited. OBJECTIVE: This scoping review aimed to map the research landscape of LLM-based multiturn conversational chatbots developed for patients with cancer and informal caregivers, focusing on system design, intervention purposes, evaluation approaches, safety considerations, and transparency of LLM-related components. METHODS: This scoping review followed the Joanna Briggs Institute methodology and PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses Extension for Scoping Reviews) guidelines. Six databases-PubMed, Embase, Scopus, Web of Science, CINAHL, and PsycINFO-were searched for studies published between January 2022 and January 2026, with supplementary searches conducted in IEEE Xplore Digital Library and ACM Digital Library in May 2026. Studies were included if they described LLM-based chatbots designed for patients with cancer or informal caregivers that supported multiturn conversational interaction. Two reviewers independently conducted the study selection and data extraction. RESULTS: Eight studies met the inclusion criteria. Most studies focused on prototype development, with limited research evaluating clinical outcomes. ChatGPT-based models were the most commonly used LLMs, and retrieval-augmented generation techniques were applied in several studies. Chatbots were primarily designed for emotional support or information provision. Evaluation approaches varied widely, including response quality, psychological outcomes, and user experience. However, no studies evaluated interaction-level characteristics such as conversational continuity or context retention, and only 2 studies reported any conversational memory mechanism. Reporting on safety risks, mitigation strategies, prompt design, model parameters, and adherence to LLM reporting guidelines was often limited or absent. CONCLUSIONS: This scoping review identified only 8 studies on LLM-based multiturn conversational chatbots for patients with cancer and informal caregivers. The field remains at an early stage, characterized by prototype-oriented development, heterogeneous design and evaluation approaches, and inconsistent safety and transparency reporting. Future development should prioritize genuine conversational capability, safety management, and transparent reporting.
References
- 1.Jeong, Y., Cha, H., & Suh, E. E. (2026). Multiturn large language model-based conversational agents for patients with cancer and caregivers: Scoping review. JMIR Cancer. https://doi.org/10.2196/96241
Related Research
The American journal of pathology
An Integrated Machine Learning and Genomic Framework for Precise Detection of Gastric Cancer
3 Aug 2026
Current opinion in chemical biology
Prediction of protein-protein interactions and co-complex models with deep learning
3 Aug 2026
European journal of radiology
A CT-based deep learning model to differentiate between benign and malignant adrenal lesions
3 Aug 2026
This content is for educational purposes for healthcare professionals only and does not constitute clinical advice. Clinical decisions should be based on individual patient assessment, current guidelines, and appropriate specialist consultation. Editorial Standards · Privacy Policy · Terms of Service