A randomized controlled trial in Japan tested home-visit nurses using a guideline-informed AI chatbot during simulated heart failure cases. Nurses reported lower mental effort and task difficulty, with higher confidence compared with controls. While the trial used simulation rather than live clinical endpoints, it points to a near-term operational use case: AI as cognitive support to standardize reasoning and reduce fatigue in decentralized settings. For digital health teams, the result strengthens the case for evaluating AI tools on human factors and workflow burden, not only diagnostic accuracy.