Albany, Georgia-based Phoebe Physician Group, a subsidiary of Phoebe Putney Health System, operates primarily in a 41-county rural area where caregivers say missing a doctor’s appointment is increasingly acceptable to patients.
PROBLEM
That puts the overall no-show rate for physician groups at 12 percent—more than double the national average reported by the Medical Group Management Association of about 5 percent. In urban markets, for example, providers can pay for taxi vouchers; in South Georgia, that’s not an option. Automated reminder texts and phone calls don’t help.
Phoebe Physician’s size, large practice market, and difficulty recruiting staff in rural areas only exacerbated the problem. Frequent staff turnover and limited staff experience led to inconsistent scheduling, duplicate bookings, and varying appointment confirmation practices.
“You think your staff is sending these reminders, but they often don’t or they don’t do it effectively,” said Matthew Robertson, chief administrative officer at Phoebe Physician Group. “That’s why we decided to explore how AI technology could facilitate higher patient volumes while minimizing disruption for providers and improving the patient experience.”
PROPOSE
It all started with a conversation with Berkeley Research Group. Phoebe Physician told BRG about this issue and that reminder texts and calls were ineffective. And that the organization needed to eliminate the human element to free up staff time and ensure their work was actually getting done.
“BRG proposed an AI tool, Tool Development and Piloting MelodyMD, that they developed with Trajum ML,” Robertson explains. “The tool leverages machine learning to analyze years of patient visit data and predict the likelihood of a given patient not showing up for an appointment.
“When a new patient is scheduled, MelodyMD contacts Phoebe Physician’s scheduling system to analyze the patient’s no-show status and automatically creates an adjacent appointment slot if the no-show probability exceeds a set threshold,” he added.
Facing the Challenge
The tool’s developers examined data points to determine which correlated most strongly with the likelihood of a patient not showing up for an appointment. These included patient demographics, physician specialty, appointment wait times, appointment history, and insurance. As new patient visit data was added, the developers continued to refine their model.
“A key element that we’ve implemented over time involves making sure that we limit double bookings each day,” Robertson notes. “That means only patients who are highly likely to not show up are considered for double bookings. Then, exclusions are applied to specific clinics and appointment types.
“As the model was rolled out and tested, we adjusted the reminder process to improve communication with patients and ensure our team had enough time to fill new appointment slots,” he continued.
He added that AI tools also enable organizations to measure performance and make improvements at the following levels:
-
Patient access. Provide regular monitoring of utilization, no-show volume, completed visits, cancellations, cancellations within 24 hours, and rescheduled visits.
-
Referral management. Allows for regular monitoring of referral volumes, leakage and patient retention rates, as well as splitter and competitor volumes.
-
Supplier scorecard. Provide regular monitoring of relative value units of work, visit types, assessment and management coding, average number of visits per session, average days to schedule new and returning patients, no-show rates, and payer mix by provider.
-
Enhanced physician/healthcare provider productivity. Provide regular monitoring of relative unit values of work, visit types, assessment and management coding, and details of current Procedural Terminology by provider.
-
Human resources are not suppliers. Provide regular monitoring of full-time equivalent wages, productivity and overtime to ensure adequate staffing as needed.
RESULT
From January 2023 to February 2024, Phoebe Physician saw an average increase of 168 visits per week, which equates to approximately 7,800 additional visits and $1.4 million in net revenue from new patients.
Robertson said that while people are still not showing up and rural people’s perceptions of getting a doctor’s appointment remain a barrier, double booking has significantly reduced the impact of such events.
ADVICE FOR OTHERS
“It’s definitely important to talk to your providers early in the process and be transparent about what’s going on,” Robertson advises. “You want to be clear about the problem, the goals, and the potential impact you hope the AI technology will have. When we talked to 20 primary care providers and told them they had 84 no-shows a day—they were shocked. That got them on board to try a new solution.”
“You also want to make sure to consider health equity,” he continued. “For example, how do we make sure the AI tool doesn’t introduce implicit bias in scheduling? How can we fine-tune the algorithm to make sure underinsured and uninsured patients or those with certain types of insurance aren’t disproportionately impacted by having their appointments double-booked and potentially leading to longer wait times in the clinic?”
And of course, AI is only as good as the underlying data, he added.
“We worked with BRG to clean up and organize three years of patient data that we could then use to build an effective model,” he concluded.
Follow Bill’s HIT coverage on LinkedIn: Bill Siwicki
Email him: [email protected]
Healthcare IT News is a publication of HIMSS Media.
