Sponsored

Administrative strain to strategic gain: 4 ways AI is reshaping hospital operations

Advertisement

Administrative costs now account for a quarter of all healthcare spending, a fact that has moved hospital operations into the spotlight as executives look for new ways to protect margins.

During a recent discussion at Becker’s Healthcare 13th Annual CEO + CFO Roundtable, leaders from Franciscan Health, Access Healthcare, now part of Smarter Technologies, and VillageMD shared how their organizations are pursuing transformation through AI, human-centered design and enterprise-level change management.

Here are four key takeaways from the session:

1. Not a back-office issue

The panel emphasized that administrative inefficiency is a major driver of financial distress for hospitals, not a secondary concern.

Frank McHugh, CFO and interim president and CEO of Franciscan Health Olympia Fields, highlighted that many healthcare organizations are on an average into one or 2% operating margin, but about 50% still do not have a positive operating margin.

He added that traditional cost control tactics are outdated and that executives must be deeply involved in patient throughput, care coordination and chronic disease management to protect revenue.

“The modern financial executive today has to be involved in clinical operations in terms of statusing patients, getting key performance metrics in place and really treating patients with the right care, in the right time at the right location,” said Mr. McHugh.

2. Fueling friction

When asked about sources of cost and delay, Mr. McHugh pointed to denials and manual processes as the biggest culprits. He estimates that 20% of claims are generally denied, adding that fragmented systems and inconsistent payer rules create bottlenecks across eligibility, prior authorization and coding workflows.

Mike Roberts, former CIO and CTO of VillageMD, echoed that sentiment and emphasized the need to break down silos internally and externally.

“There is information asymmetry,” said Mr. Roberts. “It’s about knowledge bases existing in individual departments with individual people that take people to do things because only they know how to do it.”

3. Human in the loop

All three panelists agreed that AI has transformative potential but said fear and poor communication often stall progress. Mr. Roberts urged organizations to frame AI as a tool for enabling staff, not replacing them.

“We should be thinking about how we make people two x or three x more productive so that we don’t need to add more staff,” Mr. Roberts said. “We can absorb that attrition and redeploy that capital elsewhere.”

Sid Mehta, Chief Growth Officer at Access Healthcare, now part of Smarter Technologies, added that the nature of the task can and should dictate the deployment of technology.

“The goal is to figure out where you can impact and how you create a career path for people. So you’re on the journey together and the clear call for action is combined communication and human in the loop,” said Mr. Mehta.

4. Results and measurable wins

Hospitals using AI-powered scribes, for example, have seen physicians increase daily visits by several patients, which in turn boosts access and billing speed.

“Not only does it speed up the efficiency of the bill, it also enables that physician to see anywhere from three to five more patients a day,” Mr. McHugh said.

The panel also emphasized metrics like cost to collect, cash acceleration and denial rates as core to tracking success. To achieve results, Sid Mehta emphasized that while vendors are often pressed for a strong return on investment (ROI), typically 5-to-1, the true measure of success lies in scalability. He said organizations should focus on enterprise-level outcomes.

As AI evolves rapidly, he cautioned against heavy upfront capital investments that may become obsolete. Instead, he encouraged health systems to partner with vendors that offer flexible, scalable solutions and support long-term transformation.

Looking ahead, panelists forecasted that administrative operations will shift from fragmented tasks to episode-based models, with AI agents managing end-to-end workflows. Mr. Roberts predicted new roles like “agent orchestrator” and “AI curator” will emerge.

Mr. Mehta summarized the moment as a tipping point with organizations inserting AI into a process that’s already been built.

“There is going to be contextualizing around technology,” Mr. Mehta said. “I really feel like this is the tip of the iceberg. It’s an exciting time and the momentum is on our side.”

Advertisement

Next Up in Strategy

Advertisement