The founders of macro-eyes, a machine learning company that simplifies personalized patient care, announced the introduction of Sibyl, a predictive scheduling solution that cuts the financial and operational damage from patient No-Shows without relying on patient behavior change.
We’ve all called to book a medical appointment to be told that the first available slot is in 5 to 6 weeks. That day, 10 appointments may go empty, even 20; often more. No one shows up to ~15% of scheduled appointments. At many sites, No-Shows can constitute nearly 40% of appointments.
No-Shows and lack of optimization in scheduling costs healthcare providers billions, hits morale, strains operations and has implications on care that can cost lives. We developed Sibyl to solve the problem with cutting-edge machine learning and deliver long-needed, massive improvement in cutting the damage from No-Shows. Sibyl is AI that learns when to schedule individual patients to increase overall utilization , ” said Benjamin Fels, CEO of macro-eyes. Healthcare is increasingly data-driven, scheduling is not. It’s mission-critical infrastructure, yet the decision-making that determines scheduling doesn’t benefit from data-driven insight or predictive analytics.
Sibyl is a predictive scheduling solution that machine learns the appointment times that are best-fit for both the patient and provider, increasing utilization overall. The software functions as an add-on to existing scheduling systems, showing schedulers appointment recommendations for each patient.
It’s extremely difficult to change patient behavior,” explains Fels. “Likely the reason No-Shows continue to cost providers $150B each year. Sibyl offers a proven approach based on solid science. We use patterns in behavior to learn when patients are most likely to show and the mathematics of optimization to build schedules that enable the greatest access to care.”
Sibyl uses macro-eyes core AI, refined over years at leading academic medical centers in NY and California, to analyze appointment histories and thousands of data points across provider, location and time of care. “The schedule is like a puzzle, and Sibyl is an expert at fitting together the schedule to minimize gaps,” Fels explains.
Sibyl works like x-ray glasses for the schedule, seeing through the chaotic calendar to understand where there are gaps that would otherwise be impossible to see. By integrating predictive analytics with schedule optimization, Sybil provides a peerless tool for healthcare organizations, improving the bottom line as well as the patient experience.
Sibyl at Work
During the software’s late-stage testing, macro-eyes worked with 20 clinics across the United States to analyze 2 million appointment records. The anonymized records contained reams of information, including scheduled appointment times, but the test temporarily eliminated whether or not those appointments were kept. Sibyl churned through the records and generated its own recommended schedule. With that done, the real-life show/no-show results were compared side-by-side with Sibyl’s results.
The outcome? Sibyl predicted actual patient outcomes with 76% accuracy. Sibyl incorporated more than 60 factors to build each prediction. Several of the no-show factors turned out to be patient ZIP code, appointment start time, age, and appointment history. Sibyl is prediction + optimization . Sibyl demonstrated schedule optimization that would increase utilization by 23% without increasing investment to add hours or providers. For one group of clinics, that would translate to nearly $10 million in revenue.
At least one other scheduling platform exists for clinics and hospitals, but its core algorithms are rules-based. It ascertains an “average patient” profile and then makes recommendations based on this profile rather than learning, adapting, and making ranked predictive recommendations, as Sibyl does. Sibyl delivers the most accurate, effective results of any healthcare scheduling platform on the market.
Sign up for the free insideBIGDATA newsletter.
Speak Your Mind