Designing Care Beyond the Visit: Key Takeaways from Becker’s Annual Meeting

At Becker’s Annual Meeting, Nabla hosted a panel on implementing AI tools to extend care beyond the visit, moderated by Dr. Matt Sakumoto (Chief Clinical Product Officer, Nabla), with panelists Dr. Tomi Kolade (Assistant Chief Medical Information Officer, UTHealth Houston) and Dr. Ryan Sadeghian (Chief Medical Information Officer and Medical Director of AI, University of Toledo).

The discussion surfaced a clear shift:

AI’s value won’t come from isolated tools—it will come from how well it coordinates care across the entire patient journey.

Here are five key themes that emerged.

1. AI must preserve clinical intent

As AI becomes more embedded in workflows, maintaining clinical accuracy and intent is critical. Tools that require clinicians to change how they practice risk adding cognitive burden and disrupting care rather than improving it.

Dr. Ryan Sadeghian emphasized the importance of context and validation:

“It all boils down to context and validation… you can’t just rely on the output.”

Even small misinterpretations can erode trust, especially in nuanced clinical domains.
The takeaway: AI should support clinicians, not shortcut or replace clinical reasoning.

2. Integration—not more fragmented tools—is what drives value

Across the panel, there was strong alignment on one point:
More tools ≠ better outcomes.

Dr. Matt Sakumoto underscored this reality by asking the audience how many organizations were using five or more AI vendors—more than half the room raised their hands, highlighting just how fragmented the current landscape has become.

Dr. Tomi Kolade described the ideal future state:

“It’s being able to do everything from one spot… not having to jump to different screens.”

Dr. Ryan Sadeghian reinforced the importance of embedding AI directly into the EHR:

“You want to feel like the technology is invisible… it’s in the background.”

As AI adoption accelerates, the challenge is no longer access to tools—it’s reducing fragmentation. Disconnected solutions introduce friction, risk, and cognitive load. True value comes from deep integration into existing workflows, where AI supports care without adding complexity.

3. Governance and partnership are becoming critical capabilities

As AI adoption accelerates, success increasingly depends on how organizations govern and scale it.

Dr. Kolade described the importance of structured oversight, including departmental representation and clear prioritization of initiatives.

Dr. Sadeghian shared a complementary model built around distributed ownership: empowering “trusted advisors” across the organization to drive projects and ensure accountability.

Both approaches reflect the same shift:
AI is no longer experimental—it requires intentional, system-wide governance.

That extends to vendors as well. As Dr. Sadeghian advised:

“Choose a partner, not just a product… don’t outsource your thinking.”

4. From automation to true care orchestration

Many health systems are already deploying AI—but often as point solutions.

Dr. Tomi Kolade drew a clear distinction:

“Automation is pretty much everywhere, but orchestration is longitudinal. It identifies who should work next on that task.”

The biggest breakdowns in care don’t happen during the visit—they happen before and after:

• Intake inefficiencies
• Delayed follow-ups
• Gaps in communication across the care team

Solving these challenges requires more than automation. It requires end-to-end coordination across the care continuum.

5. Early impact: efficiency, experience, and time reallocated

Even in early stages, AI is already improving the clinical experience.

Dr. Sadeghian shared: “I was having a conversation with some of our providers and one of them said to me ‘I started liking medicine once again.’”

Clinicians are saving time. More importantly, they’re spending that time differently:

• More present during patient visits
• More personalized communication
• Less time on administrative tasks

As Dr. Kolade noted, the goal isn’t just reducing total time, it’s improving the value of that time.

Looking ahead

The conversation made one thing clear:
AI in healthcare is moving beyond documentation and isolated use cases.

The next phase will be defined by:

• Systems that understand context
• Tools that support the entire care team
• Workflows that extend beyond the visit

For health systems, the challenge is no longer whether to adopt AI, but how to do so in a way that truly improves care delivery.

‍

‍