Canonical text (the exact bytes fingerprinted)

TRUVACE RECORD VERSION
record: TRV-2026-0508
version: 1
kind: certified
reason: Certified into the record
timestamp: 2026-07-22T06:09:14.307224Z
status: published
lens: p_space
sector: health
headline: Multiturn Large Language Model-Based Conversational Agents for Patients With Cancer and Caregivers: Scoping Review
dek: 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 a…
gain_title: (none)
problem_title: LLM-based multiturn chatbots for cancer patients and caregivers lack consistent reporting on safety risks and mitigation, and lack evaluation of conversational continuity and memory.
trace_subject: (none)
gain_reading: (none)
gain_evidence: (none)
problem_reading: LLM-based multiturn chatbots for cancer patients and caregivers lack consistent reporting on safety risks and mitigation, and lack evaluation of conversational continuity and memory.
problem_evidence: Reporting on safety risks, mitigation strategies, prompt design, model parameters, and adherence to LLM reporting guidelines was often limited or absent.
quick_read: As of July 2026, a scoping review of literature from January 2022 to January 2026 identified only eight studies describing LLM-based chatbots that support multiturn dialogue for patients with cancer and informal caregivers. Most were prototype systems using ChatGPT-based models, some with retrieval-augmented generation, designed for information provision or emotional support.

The findings matter because oncology involves complex informational and emotional needs across the disease trajectory where continuity and safety are critical, yet the review found no evaluation of conversational continuity or context retention and inconsistent safety reporting. It remains uncertain whether these agents can maintain context, improve clinical outcomes, or be deployed safely without stronger memory mechanisms and transparent reporting.
limitation: Evidence base is very small and early-stage, with only eight included studies focused on prototype development and heterogeneous evaluation, limiting generalizability about effectiveness or safety.
tag: Evidence-backed problem
key_points: Review searched six databases for January 2022 to January 2026 plus IEEE Xplore and ACM Digital Library in May 2026 for multiturn LLM chatbots for cancer patients and informal caregivers. | Eight studies met inclusion criteria, with ChatGPT-based models most common and retrieval-augmented generation used in several studies. | Chatbots were primarily designed for emotional support or information provision, but most were prototypes with limited clinical outcome evaluation. | No studies evaluated conversational continuity or context retention, and only 2 studies reported any conversational memory mechanism.
rundown: The review followed Joanna Briggs Institute methodology and PRISMA-ScR guidelines, with two reviewers independently conducting selection and extraction across PubMed, Embase, Scopus, Web of Science, CINAHL, PsycINFO, IEEE Xplore and ACM Digital Library.

Evaluation approaches varied widely across the eight studies, including response quality, psychological outcomes, and user experience, while reporting on prompt design, model parameters, and LLM reporting guideline adherence was often limited or absent.
sources:
- peer_reviewed | JMIR Cancer | https://doi.org/10.2196/96241 | 2026-07-21
prev: 0000000000000000000000000000000000000000000000000000000000000000

sha256

35ed37609b751152f7c9c44bf5cee95f35300b5acd8351bb2cd740b67c96cbe9

previous

0000000000000000000000000000000000000000000000000000000000000000