Background: There is an increasing interest and potential for artificial intelligence (AI) use in neonatal intensive care. The core principle driving the design and conduct of this work has been to ensure that transformations enabled by AI are positive ones guided by the preferences and priorities of those who may benefit from innovative technology. Successful application of AI technologies could significantly improve the care of high-risk infants and their families.
A review of experience from other domains and needs expressed by family and clinician stakeholders at a previous PCORI-funded Engagement Conference held by the project team suggests that AI technologies could and should be employed to help improve consistency, continuity, and communication within the NICU.
Proposed Solution to the Problem: The project team will build upon its previous work and convene an ongoing multi-stakeholder community representing family, clinician, institutional, and technology communities. This group will work to build capacity for PCOR/CER into how AI technologies could be used to improve NICU care teams.
- Ensure that the transformations caused by AI use are positive, guided by the preferences and priorities of all those who could be affected by these innovative technologies
- Engage family and clinical stakeholders to build a joint understanding of issues involved in the formation of NICU nursing teams to ensure optimal care and collaboration with families through
- Identification and cataloguing of relevant information and outcomes that are important to families, bedside nurses, and institutional leadership
- Enumeration of potential barriers and facilitators to the use of advanced information technologies in addressing patient, family, and staff needs
- Utilize innovative methods to recruit new technology stakeholders to the community and apprise them of the issues and perspectives represented in the joint understanding above
- Raise awareness of the systematic biases that may result from AI use and develop approaches to identify and mitigate these effects. Special attention will be paid to the potential for these to exacerbate racial or social disparities.
- Consider and identify mechanisms to be used in future PCOR/CER to assess the impact of AI in modifying (in either positive or negative ways) the presence of social, racial, and other disparities in the provision of NICU care
- Define high-priority research agenda related to AI use in the NICU focused on improving care teams
- Build capacity among family, clinical, and technology stakeholders to perform high-quality PCOR/CER on the identified agenda
Activities: The team will conduct multi-stakeholder discussions regarding AI use in the NICU and hold a “hackathon” event in the second project year.
Projected Outcomes and Outputs: The work accomplished during this project will accelerate the use of patient-centered outcomes research principles to assess the adoption of artificial intelligence use in the NICU. It will build capacity for NICU family members to participate in PCOR/CER efforts and is likely to provide a valuable model for new technology integration across a variety of domains.
The project team, along with the stakeholder community, will be able to consider applications for further funding for development and evaluation of the ideas conceptualized during this project.
Patient and Stakeholder Engagement Plan: Engagement of the family stakeholders began even before creation of this proposal, along with other participants in the team’s previously funded PCORI Engagement Conference. They were members of the conference organizing group and have presented the results of that work at national conferences. The family stakeholders have participated in the design and creation of the present application.
Project Collaborators: Project collaborators include representatives of the University of Minnesota, the University of Vermont, the Black Women’s Health Imperative, as well as representatives of industry and other academic institutions.
James Gray, MD, MS
Mary Hitchcock Memorial Hospital
Last updated: February 2, 2023