Gen AI and ROI: Key takeaways from HIMSS24

  • Posted on March 21, 2024
  • Estimated reading time 3 minutes
Key takeaways from HIMSS24

The inspiring theme of last week’s HIMSS24 healthcare conference was ‘Creating Tomorrow’s Health’ and the predominant topic of discussion among the healthcare organizations (HCOs) in attendance was ‘Generative AI in Healthcare.’ It’s been about a year since generative AI emerged and HCOs came to HIMSS looking to share experiences and learn more about gen AI in healthcare. The question I heard and was asked most often at HIMSS was, “how can ROI be measured from gen AI in healthcare?” My answer to that question is in the following paragraphs.

HIMSS24 started with a bang on Day One, when 700+ attendees packed a panel discussion called AI’s Role in Modern Healthcare: The Wins, The Woes and What’s Next. I was joined onstage by four other panelists from providers and academia, and we emphasized the importance of putting leadership and a clear vision for AI before technology.

I shared a personal checklist I use when helping clients plot an initial AI journey, which consists of thinking through the Six “S” capabilities of AI. AI is very good at Summarizing, Synthesizing and Standardizing data and outputs. What should you summarize, synthesize and standardize with gen AI? Then think about how you use those capabilities to Support, Supplement and Simplify the job of a human in your organization, be it a clinician, a patient services agent or a back-office worker. If HCOs do this well, they will see ROI in lower attrition, more efficient billing and claims, and lower administrative expenses and operating costs. Benchmark these costs at the outset and revisit them as you deploy gen AI to different job personas and departments.

On Day Two, it was my honor to join Jean Francois Saint-Pierre from the World Health Organization (WHO) to present The World Health Data Hub: Breaking Down Barriers to Health Information. Avanade and Microsoft partnered with the WHO to help the organization stand up a data model and evolve it in an agile fashion. The goal: to create the world’s first comprehensive, end-to-end data solution for global health to reduce fragmentation, streamline processes, identify and resolve gaps and inequalities, and ensure data is accessible, findable and usable for all stakeholders.

Jean François’ takeaways after such a transformational project were that there’s never a ‘perfect’ time to start a large data & AI project, set the right expectations, and build acceptance and adoption from the bottom up and not the other way around. Jean Francois gave a compelling step-by-step guide to how the project progressed and where it stands today, leading attendees to suggest this could be an effective model for other large data and gen AI projects.

In the evening of Day Two, I hosted a private dinner discussion between Dr. David Rhew, Chief Medical Information Officer for Microsoft, and Kaveh Safavi, a Senior Managing Director in Accenture’s healthcare practice, during which we discussed responsible deployment of AI and how to leverage gen AI for maximum benefits. We emphasized that proper skilling of GenAI users is needed.

One core skill required by gen AI users is the ability to create effective prompts. Research shows that a well-crafted series of prompts leads to significantly better gen AI results and ROI. With this in mind, Microsoft has created prompt libraries, which help users learn how to best prompt gen AI models, and Microsoft is releasing guidance on how to create the most effective prompts as part of its Copilot ecosystem. Avanade and Microsoft are also planning to educate and upskill clients’ prompting abilities with what we are calling “prompt-a-thons.”

I come away from HIMSS24 inspired by the innovation and dedication to deploy AI in healthcare responsibly and, in turn, improve health outcomes across the globe.

Learn more about Avanade’s digital health solutions and AI.

Avanade Insights Newsletter

Stay up to date with our latest news.

Share this page
Modal window