Getty Images/iStockphoto

Are AI agents the key to optimizing online appointment scheduling?

Community Health Network deployed AI agents to help navigate consumers through the common logjams that keep them from using online appointment scheduling.

When online appointment scheduling first came onto the healthcare scene, it was touted as the key to opening the digital front door. But, according to Community Health Network's Maggie Gentry, the technology wasn't the cure-all the industry hoped it'd be.

Instead, online booking kept patients seeking appointments in a feedback loop, unable to bypass different registration barriers to ultimately book the slot they wanted, Gentry, the director of experience analytics at Community Health Network, explained.

"What's interesting about booking appointments, especially in healthcare, is that it's a gated process," she said. "It's very much controlled by the provider, very much controlled by the specialty, very much controlled by a referral process."

With appointment booking gated at so many junctures, it's easy for the online booking system to fail. For example, if an appointment requires a referral, the patient can't book the slot. If a certain specialty within a practice has gated its calendar, the system prompts patients to "start over" or "navigate again."

In many cases, patients will give up on the online booking system altogether. According to 2022 survey data, about 70% of patients said that even when they tried to book their appointments online, they still ended up having to call the doctor's office. Another 61% said they did not access healthcare in the past year because using an online self-scheduling system was too complicated.

This creates a bad situation on two fronts, according to Gentry. In the call center, staff are overworked and overwhelmed by an influx of patient calls that could've been handled by the online system. And for patients, this is a major source of frustration.

"If my intention is to sit down at a computer and schedule an appointment, the intentionality is there," Gentry stated. "I don't want to have to pick up the phone. If I wanted to pick up the phone, I would've started there."

That lapse in patient experience is big, as healthcare organizations face tighter margins and work to maintain market share. Healthcare consumers are accustomed to the experiences they have in other service sectors, like using Uber or making a dinner reservation on OpenTable. If consumers can easily make appointments on those platforms, why can't they in healthcare?

Community Health Network has turned the online appointment scheduling paradigm on its head by deploying AI agents to help patients bypass the logjams they encounter digitally. Using AI technology from Qualtrics, the organization has been able to streamline the online booking process for patients at its Walgreens-based urgent care centers.

AI agents proactively guide patient scheduling

The intercept agents can track when a consumer abandons the appointment-scheduling process. Just like a bot might flag an abandoned shopping cart on Nordstrom.com, these AI agents are able to determine when a user has given up on booking an appointment.

From there, the agent will ask how far the patient is willing to drive for care, or if they have a specific day or time they'd like the appointment scheduled. The agent has been designed to rank these choices, provide options and adjust those options if the location or time doesn't work.

Because Community Health Network deployed the agents in a small pilot, Gentry said her team was able to flag key friction points they hadn't initially considered.

"It wasn't just because they couldn't schedule the appointment -- there were other parts of the website that caused friction," Gentry explained.

For example, in the urgent care setting, patients must check a box confirming that their symptoms are non-emergent in order to obtain an appointment. If patients missed that box, they get stuck in that feedback loop, unaware of where they went wrong. The AI agents can identify the issue and show patients how to go back and check the non-emergent care box.

"It became a tool to assist and navigate instead of something that was reactionary," Gentry noted. "It was proactive and was able to handle those items in real time instead of having to redirect interactions to a different way of communication."

All said, Community Health Network's system receives around 4,300 views each month and has a 95% user satisfaction rate with the agent.

Designing AI implementation around a business problem

Still, implementing the system wasn't perfect.

According to Gentry, AI implementation -- especially in the call center -- required a lot of interpersonal work across different business leaders.

"We needed buy-in from operations, we needed executive sponsorship, and we really wanted folks to understand we're not here to replace their jobs," she recalled.

The team also needed to be laser-focused on the business problem at hand to ensure they got true return on investment.

"You've got to understand the business problem you're trying to solve. If you don't understand that, the tool you develop around it isn't going to be successful," Gentry advised.

"Our friction point was that constant looped experience for our patients and churn," she continued. "We wanted to keep our patients sticky, and we wanted to get them out of that looped experience. So how do we design for that? I think really putting the patient at the heart of the solution makes it pretty easy to design."

Gentry said her team looks forward to the more sophisticated "know me" work the AI will enable. For example, if a patient comes back and reviews their physician poorly, the AI can learn how to book the patient with a different clinician in the future.

Healthcare has been good at collecting this type of patient experience data in the past. But according to Gentry, AI capabilities are going to be the next frontier in personalizing healthcare and acting on patient experience data.

"It's really acting on feedback too, which has been monumental and is really going to change healthcare," she concluded. "We're great at collecting feedback, we're great at collecting data points, but being able to action that in real time is what really shifts it to an experience-first culture."

Sara Heath is an executive editor at Xtelligent Healthcare Media, where she covers patient engagement, healthcare policy and health IT.

Dig Deeper on Population health management