Clinical Report: A User-Friendly Interface for the Preclinical Assessment of AI

Background

The rapid expansion of patient portal messaging has raised concerns regarding clinician workload and burnout. As healthcare systems explore AI solutions to manage message flow, rigorous testing is essential to ensure that these tools enhance care quality without compromising safety or equity.

Data Highlights

No numerical or trial data provided in the source material.

Key Findings

• A dedicated testing platform is critical for evaluating AI tools in patient communications.
• AI-generated messages must maintain clarity and accuracy to prevent misunderstandings.
• Integration of AI in live EHR systems poses risks to data integrity and patient safety.
• A sandbox environment allows for safe experimentation with AI outputs without impacting live systems.
• Feedback and iteration are essential for developing effective AI messaging solutions.

Clinical Implications

Healthcare providers should be aware of the potential benefits and risks associated with AI in patient messaging. Ensuring that AI tools are rigorously tested before implementation is crucial for maintaining patient safety and care quality.

Conclusion

The development of AI tools for patient portal management requires careful consideration and testing to ensure they meet clinical standards and enhance communication without compromising safety.

Related Resources & Content

1. Steitz BD, Unertl KM, Levy MA, Appl Clin Inform, 2021 -- An analysis of electronic health record work to manage asynchronous clinical messages among breast cancer care teams
2. Garcia P, Ma SP, Shah S, et al., JAMA Netw Open, 2024 -- Artificial intelligence-generated draft replies to patient inbox messages
3. Hristidis V, Ruggiano N, Brown EL, et al., J Med Internet Res, 2023 -- ChatGPT vs Google for queries related to dementia and other cognitive decline: comparison of results
4. Sendak MP, Ratliff W, Sarro D, et al., JMIR Med Inform, 2020 -- Real-world integration of a sepsis deep learning technology into routine clinical care: implementation study
5. Shortliffe EH, Sepúlveda MJ, JAMA, 2018 -- Clinical decision support in the era of artificial intelligence
6. JAMA Network Open — Patient Perspectives on AI-Drafted Electronic Portal Messages
7. JAMA Network Open — What Patients Want From AI-Drafted Portal Messages—Empathy in the In-Basket
8. npj Digital Medicine — Accessing AI mammography reports impacts patient follow-up behaviors: the unintended consequences of including AI in patient portals
9. DIGITAL HEALTH — Implementing artificial intelligence (AI)-supported communication tools in healthcare: System-level perspectives
10. HTI Rules - ONC - Office of the National Coordinator for Health Information Technology
11. Patient Perspectives on AI-Drafted Electronic Portal Messages
12. What Patients Want From AI-Drafted Portal Messages—Empathy in the In-Basket
13. Accessing AI mammography reports impacts patient follow-up behaviors: the unintended consequences of including AI in patient portals
14. Utilization of Generative AI-drafted Responses for Managing Patient-Provider Communication | npj Digital Medicine
15. A scoping review of studies on secure messaging through patient portals: persistent challenges and potential solutions | npj Health Systems