Predictive overbooking improves endoscopy scheduling
Click Here to Manage Email Alerts
A predictive overbooking system, which identifies patients likely to miss scheduled gastrointestinal endoscopy procedures and offers their appointments to others on short notice, effectively maximized scheduling at a Veterans Affairs outpatient endoscopy clinic.
These prospective study results have both clinical and operational significance, according to Mark W. Reid, PhD, a biostatistician in the department of psychiatry and behavioral neurosciences at the Cedars-Sinai Center for Outcomes Research and Education (CS-CORE).
“The main clinical significance of the study is the reduced waiting time for an appointment, which would allow for potentially life threatening polyps or lesions to be detected sooner,” he told Healio Gastroenterology. “By reducing wait times, we are also alleviating anxiety in veterans about having to wait a long time for their appointments, potentially improving their psychological quality of life. In terms of clinic operations, the predictive overbooking program allowed us to see more patients for endoscopy, with limited impact on staff overtime and overall care costs.”
Over 34 weeks, Reid and colleagues alternated between traditional “one patient, one slot” scheduling and predictive overbooking at an outpatient endoscopic clinic within the VA Greater Los Angeles Healthcare System.
For the 17 randomly selected intervention weeks, they used a previously designed and validated logistic regression model to assign a no-show risk score to scheduled patients using electronic health record-derived demographic, clinical and patient attendance data, and offered projected open appointments on a 2-week calendar to other patients using an active recruitment “Fast-Track” strategy. Then they compared scheduling strategies based on service use, staff workflow and clinic costs.
Overall, 2,446 patients completed appointments during the study period, 1,181 during control weeks, and 1,265 during the intervention weeks, 111 of whom accepted Fast Track appointments.
Almost all appointment slots were filled during weeks where predictive overbooking was used, with about one-third of an appointment left unused every day on average (0.35) compared with 2.5 appointments left unused during control weeks (P = .0001).
Predictive overbooking also significantly improved service use, which was defined as the ratio of completed appointments to capacity each day. On average, this increased from 86.4% during control weeks to 99.5% during intervention weeks, enabling the 111 additional patients to undergo procedures.
“Even though capacity was exceeded on several occasions during intervention weeks, the impact on clinic staff scheduled was minimal,” the researchers wrote. Intervention days were 34 minutes longer than control days on average (8.31 vs. 7.84 hours; P = .02). The intervention did cost an average of $26.13 more per day in overtime costs, but this was not significantly different than existing overtime costs. Each Fast Track patient cost the clinic an additional $19.52 to treat on average.
The key to applying the overbooking strategies of the airline industry to the endoscopy clinic is the accurate targeting of patients with a high likelihood of non-attendance, Gavin C. Harewood, MD, of the department of gastroenterology and hepatology at Beaumont Hospital in Dublin, Ireland, wrote in a related editorial.
“Although there is inherent uncertainty in patient behavior, when it comes to attendance for endoscopic procedures, past patient performance predicts future patient behavior with such accuracy that it allows precise overbooking to function effectively,” he wrote.
However, while airline passengers can take a later flight, “overbooked endoscopy patients, ... especially those who have taken a pre-colonoscopy preparation, must be facilitated on the same day ... Reassuringly, this study demonstrated that, with a precise overbooking approach, overburdening the endoscopy practice was minimal.”
Finally, Harewood emphasized that although overbooking appears promising, it is important “to ensure that serial non-attenders do not get left behind.” – by Adam Leitenberger
Disclosures: The researchers and Harewood report no relevant financial disclosures.