3 Proven Strategies for Getting AI Tools Into Clinical Practice Without Resistance

Health systems are navigating an explosion of AI tools — from ambient clinical scribes to AI-powered decision support. But the technology itself is only part of the challenge. Two clinical leaders from the University of Iowa Health Care and Fisher-Titus Medical Center recently shared what actually drives successful adoption in HealthLeaders' The Winning Edge webinar series.

Their advice maps to three core disciplines every healthcare leader should build before deploying AI.

1. Define ROI Before You Deploy — For Both Finance and Clinical Outcomes

Assessing value means engaging clinical stakeholders early and anchoring the evaluation around a specific problem the tool is meant to solve. Some AI tools have hard ROI: coding AI improves revenue capture directly. Others, like AI scribes, have softer ROI — but James Blum, Chief Health Information Officer at University of Iowa Health Care, notes they've become a workforce recruitment and retention tool in competitive markets.

Fisher-Titus, which serves rural communities with persistent open positions, has found AI tools that automate clinical functions actually address staffing shortfalls directly.

2. Earn Clinician Buy-In — Don't Force It

Unlike the EMR rollout era, where technology was largely mandated from above, AI adoption works best when clinicians are invited to identify their own pain points first. Executives — CMOs, CNOs, CMIOs — should champion the "why" behind each tool, not issue mandates. Addressing the fear that AI will replace staff is an essential part of that conversation.

3. Streamline Governance — Treat AI Like Any Other IT Tool

Both organizations have found that establishing separate AI governance structures creates unnecessary complexity. Instead, they apply the same acquisition, cybersecurity, and performance evaluation frameworks they already use for IT tools. The exception: an AI oversight group with training in AI-specific performance characteristics to ensure tools are appropriate for the patient population.

For leaders in any industry, the governance takeaway is instructive: building parallel bureaucracies for AI slows adoption without improving safety. Integration into existing review processes is faster and more sustainable.

🔗 Read the full article on HealthLeaders Media