Grant: CIHR Fellowship

Background: Artificial intelligence (AI) can quickly process a lot of information, which is why it is receiving significant attention in healthcare. It can help make administrative tasks and care processes more efficient and can also help clinicians make decisions. What is Missing: Despite the projected benefits of AI, its use has been limited. Most research focuses on how AI is designed and how users interact with it, but not enough on how to implement AI and make it work well in real-world healthcare settings. Additionally, not much research includes perspectives of diverse people who will be impacted by and/or use AI in healthcare settings. There are also worries about equity since not everyone may benefit equally from AI. I plan to use several well-known implementation science and digital health equity frameworks to conduct a three-phased research project to develop a responsible guide for adopting and evaluating AI in healthcare in Canada and the United States. A Canada-US advisory committee made up of patient partners, clinicians, decision-makers and AI experts will oversee the following plans. Specifically, I plan to: (Phase 1) Review academic studies and healthcare reports about how AI is used, focusing on adoption strategies, factors that influence adoption, and evaluations; (Phase 2) Look at public opinions and gather diverse perspectives on AI in healthcare posted on popular social media platforms (YouTube, X, and Reddit); (Phase 3) Conduct interviews to investigate experiences of patients, caregivers, clinicians, decision-makers, and AI experts on using AI in real-life healthcare settings in Canada and the US. Impact: Findings from the above three phases will be combined into a complete guide for how to implement and make AI work well in healthcare settings. This is expected to address a critical gap, ensuring the responsible adoption and evaluation of AI in real-world North American healthcare settings, accounting for equity considerations.