Health systems spend years and hundreds of millions of dollars implementing EHRs, yet two healthcare CIOs say a significant share of that investment remains untapped.
More pressing, they argue, the gap between what EHRs can do and what organizations actually do with them may be the single most important variable in determining AI readiness.
Muhammad Siddiqui, Chief Digital and Information Officer of Richmond, Ind.-based Reid Health, estimates health systems use 60% to 70% of what their EHRs can actually do. He noted the figure is not drawn from a single metric.
“Most health systems are probably using somewhere around 60% to 70% of what their EHRs can actually do. I’d put us in that general range,” he told Becker’s. “That number is not based on a formal score alone. It reflects what we have turned on, what is truly adopted in day-to-day operations, how consistently workflows are being used as designed, and how much manual work still exists around the EHR. There is a real difference between owning functionality and realizing value from it.”
Michael Archuleta, CIO of Trinidad, Colo.-based Mt. San Rafael Hospital and Clinics, puts operational maturity at a similar level and points to what he sees as the central misunderstanding in healthcare IT.
“The biggest misconception in healthcare IT is that buying an EHR means you are capturing its value. You are not. If I had to estimate, most health systems are probably using 60% of their EHR functionality in a way that is operationally mature,” he told Becker’s. “The gap is not usually access to technology. The gap is optimization.”
For Mr. Archuleta, the distinction between a capability being technically active and genuinely embedded in clinical practice is where most organizations fall short.
“If a capability is live but clinicians are bypassing it, if workflows still feel fragmented, or if documentation burden remains unchanged, then from an operational standpoint that value has not been realized,” he said.
Both leaders pointed to similar categories of underutilized capability: advanced clinical decision support, workflow automation, scheduling and access tools, patient engagement and self-service, interoperability and analytics. Mr. Siddiqui noted the software is rarely the barrier.
“In most cases, the EHR is not the problem. The problem is that health systems have not had the time, operational discipline, governance or change capacity to fully standardize and get value from what is already there,” he said. “Local exceptions, legacy habits, limited build resources and competing priorities all slow that work down.”
Mr. Archuleta pointed to the institutional dynamics that allow the gap to persist. Health systems are simultaneously managing staffing shortages, cybersecurity threats, regulatory obligations, financial pressure, platform upgrades and day-to-day patient care. In that environment, optimization work rarely rises to the top of the priority list.
“Optimization work rarely feels urgent in the moment, which is exactly why it gets delayed. But that delay carries a cost,” he said. “Over time, organizations normalize inefficiency while sitting on platforms capable of much more.”
The concern for both leaders does not stop at operational inefficiency. Each said the EHR optimization gap poses a direct risk to how health systems are approaching artificial intelligence.
Mr. Siddiqui said AI deployed on top of fragmented workflows does not resolve the underlying problem.
“AI can be useful, but if the underlying workflow is fragmented, AI often just helps you move faster through the mess,” he said. “Health systems still need to do the basic work of workflow redesign, governance, template cleanup, decision support review and adoption management. If you do that work well, AI can add real value. If you do not, it becomes another layer on top of existing friction.”
Mr. Archuleta agreed, stating that AI will not fix broken workflows. Instead, it will “scale them.”
“If documentation practices are inconsistent, if governance is weak, if data lacks integrity, or if teams are underutilizing the core platform, adding AI on top of that environment simply accelerates the underlying problems,” he said.
He added that the organizations positioned to lead in AI are not necessarily the ones moving fastest. They are the ones optimizing workflows, strengthening governance, improving adoption and building trust in their data.
Mr. Archuleta framed EHR optimization not as a maintenance obligation but as a prerequisite.
“Maximizing the value of your EHR is not maintenance work. It is strategic infrastructure, and in many ways, it is the most important AI preparation work a health system can do,” he said.
At the Becker's 11th Annual IT + Revenue Cycle Conference: The Future of AI & Digital Health, taking place September 14–17 in Chicago, healthcare executives and digital leaders from across the country will come together to explore how AI, interoperability, cybersecurity, and revenue cycle innovation are transforming care delivery, strengthening financial performance, and driving the next era of digital health. Apply for complimentary registration now.
