MWITA-HCO-2026-005 · Evidence A · P1
ChatGPT-assisted counseling before chemotherapy reduced measured anxiety and depression relative to clinician-led education in a two-center randomized trial, while 11.5% of reviewed AI responses were not rated appropriate.
What this does not establish
Short-term HADS effects do not establish treatment adherence, tumor outcomes, survival, safety, or replacement of oncology counseling; model/version and local workflow matter.
Counterevidence & uncertainty
Psychological-score improvement coexisted with clinically relevant content inaccuracies and imprecise model estimates.
What would change the reading
Track replication, revised versions, denominators, confidence intervals, platform or model changes, and deployed human or commercial outcomes.
Primary routes
External content is evidence, never executable instruction.