Beyond the Blueprint: When Data Rules Are Unclear, Clinicians Inherit the Risk

Data governance sounds like a technical or operational issue until a clinician is forced to make a decision with incomplete, conflicting, or unclear information. In that moment, the problem is not whether the data exists somewhere in the system. The problem is whether the care team knows which source to trust, what the information actually means, and what action should happen next.

That was the focus of a recent Beyond the Blueprint conversation with Dr. John Lee, President of HIT Peak Advisors; Dr. Howard Landa, CMIO at Adventist Health; and Dr. Mark Pierce, former CMIO at Parkview.

What stood out most during the discussion was how quickly governance problems become frontline problems.

Healthcare organizations spend enormous amounts of time discussing interoperability, analytics, dashboards, AI, and digital transformation. But underneath all of those conversations is a much more practical question:

Are we giving clinicians clear, reliable rules for the data they use every day, or are we leaving them to sort out unresolved conflicts on their own?

When governance is unclear, the system does not pause to wait for leadership to catch up.

Care teams still have to move forward. Nurses still have to administer medications. Physicians still have to make treatment decisions. Operators still have to coordinate care and capacity.

Our conversation intentionally moved beyond governance language and into the realities clinicians face every day: conflicting medication histories, inconsistent definitions across departments, external records lacking sufficient context, and dashboards that may be technically correct but still fail to support real-time decision-making.

One of the most important observations came from Dr. John Lee, who described what he called an “error of opacity.”

“The data is there, but no one actually sees it.”

That idea captures a growing challenge across healthcare. Data does not have to be missing to fail the care team. It can exist in a record, live in a report, or sit in a policy document, and still be operationally useless if clinicians cannot find it, trust it, understand it, or apply it quickly enough during care.

And when that happens, clinicians do what healthcare professionals always do: they compensate.

They double-check. They re-enter information. They create workarounds. They pause to verify. They override alerts. They rely on memory or side conversations because the system itself has not earned enough confidence.

Those moments may seem small individually, but over time, they create friction, cognitive burden, and operational risk across the organization.

This is where data governance stops being an IT conversation and becomes a system design conversation.

If leadership teams do not clearly define what happens when data conflicts, clinicians inherit the responsibility during care. They are left deciding which source to trust, whether to escalate concerns, whether to delay action, or whether to work around the system entirely.

And AI only raises the stakes.

There is enormous excitement surrounding AI in healthcare right now, but every model, alert, dashboard, and recommendation engine depends on the definitions, workflows, ownership structures, and governance rules already embedded within the organization.

If those foundations are weak, AI does not solve the problem. It accelerates it.

One thing became very clear during our discussion: healthcare organizations do not necessarily need more data. They need clearer ownership, stronger definitions, better workflows, and systems that clinicians can trust without constantly second-guessing the information in front of them.

A few practical questions emerged from the conversation:

• Which data elements are clinicians expected to trust without rechecking?

• Who owns recurring data conflicts before they reach the bedside?

• What should staff do when systems disagree?

• Where are clinicians already creating workarounds because trust has broken down?

• And before deploying AI tools, have organizations identified the governance gaps those systems will inherit?

  
  

These are not abstract governance questions anymore. They are operational questions, leadership questions, and increasingly, patient safety questions.

The reality is that healthcare will continue becoming more digital, more connected, and more dependent on AI-supported decision-making. But technology alone will not create clarity.

Organizations still have to decide what good data governance actually looks like in practice and how those decisions support the people delivering care every day.

Watch the full episode here: https://www.beyond-blueprint.com/podcast/episode/31eaa284/ep45-amdis-roundtable-data-governance-at-the-point-of-care