Predictive overbooking model improves endoscopy clinic attendance

Last Updated: 2016-07-20

By Will Boggs MD

NEW YORK (Reuters Health) - By accounting for anticipated no-shows, a predictive overbooking model can improve attendance at endoscopy clinics without "bumping" patients, according to a VA study.

"The most surprising result was that we significantly increased patient access to care without substantially increasing physician and staff worktime," said Dr. Brennan M. R. Spiegel from VA Greater Los Angeles Healthcare System and Cedars-Sinai Medical Center, Los Angeles.

"Although there was a slight uptick in the length of the workday, it was minimal and well handled by the staff," he told Reuters Health by email.

"In addition," Dr. Spiegel said, "our particular VA was near the bottom of performance for access to screening colonoscopy when this study started. By the end, our VA had moved into the top ranks, partly because of the scheduling intervention. That, too, was surprising."

Patient absenteeism rates (no-shows) range from 12% to 80%, resulting in diminished productivity and revenue losses for clinics, as well as reduced care satisfaction and treatment delays for patients, Dr. Spiegel and colleagues write in their report, online July 5 in The American Journal of Gastroenterology.

So far, approaches that have been used to address the problem of patient absenteeism have yielded modest or inconsistent improvements in attendance, they add.

Following the experience of the travel and lodging industries, Dr. Spiegel's team implemented a two-pronged strategy for quickly booking patients into outpatient endoscopy appointments predicted to be open using a patient absenteeism model they previously established. They also implemented an electronic consultation request that could be completed by a patient's primary care provider during the visit where endoscopy was recommended.

During the control period (before implementation of the protocol) there were 1,181 completed appointments. They booked 1,265 appointments during the intervention period.

Predicting overbooking increased the average service utilization rate from 86.4% to 99.5%, which allowed 111 additional patients to receive endoscopies during the 17-week study period.

An average of 2.5 appointments were left unused every day before the intervention, but only roughly 1/3 of an appointment was left unused every day during predictive overbooking.

During intervention weeks, clinic capacity was underfilled on 37 days, filled to exact capacity on nine days, and overfilled on 37 days (but only by a median one appointment).

In contrast, during control weeks, 14 days exceeded capacity, whereas other days were well under capacity.

The average overtime cost during the control phase was $36.88 per day, compared to $63.01 per day during the intervention phase. Distributed over the 111 patients that would not otherwise have been scheduled, each additional patient cost the clinic an average of $19.53 to treat.

"When people think about overbooking, their mind immediately goes to the airline model and they get understandably twitchy with fears of 'bumping' patients," Dr. Spiegel said. "Our model is a bit different from the airline model, and most importantly, we never 'bumped' patients in this study."

"Clinics that want to do this need to have staffing and scheduling systems in place that can fluidly overbook patients in a dynamic fashion, keeping up with the results of the daily estimates," Dr. Spiegel said. "Our particular VA-based algorithm may not apply to other health systems, so each clinic or hospital should probably run the same analyses and fine-tune their local calculations. But there is nothing mysterious about this process; it takes IT-savvy, good people, and enthusiasm from leadership to work."

"This study was fully funded by the VA, and we are grateful to the VA for this support, not only for our research group, but more importantly for the veterans we serve," he added.

Details of the predictive overbooking model can be found in the authors' earlier report here: http://bit.ly/29N8d4L.

SOURCE: http://bit.ly/2atPHiw

Am J Gastroenterol 2016.